Free Banking Volume II History
Edited by
Lawrence H. White Associate Professor Department of Economics University of Georgia, USA
An Elgar Reference Collection
The International Library of Macroeconomic and Financial History Series Editor: Forrest H. Capie Professor of Economic History and Head of the Department of Banking and Finance The City University Business School, London
1.
Major Inflations in History Forrest H. Capie
2.
Multinational and International Banking Geoffrey Jones
3.
Monetary Regime Transformations Barry Eichengreen
4.
Financing Industrialization (Volumes I and II) Rondo Cameron
5.
Financial Crises (Volumes I and II) Michael Bordo
6.
Price Controls Hugh Rockoff
7.
Protectionism in the World Economy Forrest H. Capie
8.
Commodity Monies (Volumes I and II) Anna J. Schwartz
9.
Debt and Deficits (Volumes I, IT and III) Lakis C. Kaounides and Geoffrey E. Wood
10. Central Banking in History (Volumes I, II and III) Michael Collins 11. Free Banking (Volumes I, II and III) Lawrence H. White Future titles will include: War Finance Larry Neal Stock Market Crashes and Speculative Manias
Free Banking Volume II
© Lawrence H. White 1993. For copyright of individual articles please refer to the Acknowledgements. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without the prior permission of the publisher.
Published by Edward Elgar Publishing Limited Gower House Croft Road Aldershot Hants GUll 3HR England Edward Elgar Publishing Company Old Post Road Brookfield Vermont 05036 USA
British Library Cataloguing in Publication Data Free Banking. - (International Library of Macroeconomic & Financial History;No.ll) 1. White, Lawrence H. II. Series 332.1
ISBN 1 85278 597 7 (3 volume set)
Printed in Great Britain at the University Press, Cambridge
Contents Acknowledgements
vii
An introduction to all volumes appears in Volume I.
PART I
PARTU
SECONDARY ACCOUNTS OF FREE BANKING THOUGHT 1. S.G. Checkland (1975), 'Adam Smith and the Bankers', in Andrew S. Skinner and Thomas Wilson (eds), Essays on Adam Smith, Oxford: Clarendon Press, 504-23 2. Tyler Cowen and Randall Kroszner (1987), 'The Development of the New Monetary Economics', Journal of Political Economy, 95 (3), June, 567-90 3. Scott Sumner (1990), 'The Forerunners of "New Monetary Economics" Proposals to Stabilize the Unit of Account', Journal of Money, Credit and Banking, 22 (1), February, 109-18 4. Lawrence H. White and George A. Selgin (1990), 'LaissezFaire Monetary Thought in Jacksonian America', in Donald E. Moggridge (ed.), Perspectives on the History of Economic Thought, 4, Aldershot, UK: Edward Elgar, 20-39 5. George A. Selgin and Lawrence H. White (1990), 'LaissezFaire Monetary Theorists in Late Nineteenth Century America', Southern Economic Journal, 56 (3), January, 774-87 THE 'FREE BANKING' ERA IN THE UNITED STATES 6. Fritz Redlich (1947), 'Free Banking', The Molding of American Banking: Men and Ideas, Part I, 1781-1840, New York: Hafner, 187-204, 292-7 7. Hugh Rockoff (1974), 'The Free Banking Era: A Reexamination', Journal of Money, Credit and Banking, 6 (2), May, 141-67 8. Robert G. King (1983), 'On the Economics of Private Money', Journal of Monetary Economics, 12 (1), July, 127-58 9. Arthur J. Rolnick and Warren E. Weber (1983), 'New Evidence on the Free Banking Era', American Economic Review, 73 (5), December, 1080-91 10. James A. Kahn (1985), 'Another Look at Free Banking in the United States', American Economic Review, 75 (4), September, 881-5 11. Hugh Rockoff (1985), 'New Evidence on Free Banking in the United States', American Economic Review, 75 (4), September, 886-9
3
23
47
57
77
93 119 146
178
190
195
vi
Free Banking II
PART III OTHER EXPERIENCES 12. Richard H. Timberlake (1987), 'Private Production of ScripMoney in the Isolated Community', Journal of Money, Credit, and Banking, 19 (4), November, 437-47 13. Tyler Cowen and Randall Kroszner (1989), 'Scottish Banking Before 1845: A Model for Laissez-Faire?', Journal of Money, Credit, and Banking, 21 (2), May; 221-31 14. Lawrence H. White (1990), 'Scottish Banking and the Legal Restrictions Theory: A Closer Look', Journal of Money, Credit, and Banking, 22 (4), November, 526-36 15. Eugene N. White (1990), 'Free Banking During the French Revolution', Explorations in Economic History, 27 (3), July, 251-76
201
212
223
234
PART IV PRIVATE CLEARINGHOUSES 16. Charles W. Muon (1975), 'The Origins of the Scottish Note Exchange', Three Banks Review, 107, 45-60 17. Richard H. Timberlake, Jr. (1984), 'The Central Banking Role of Clearinghouse Associations', Journal of Money, Credit, and Banking, 16 (1), February, 1-15 18. Donald J. Mullineaux (1987), 'Competitive Monies and the Suffolk Bank System: A Contractual Perspective', Southern Economic Journal, 53 (4), April, 884-98 19. George A. Selgin and Lawrence H. White (1988), 'Competitive Monies and the Suffolk Bank System: Comment', Southern Economic Journal, 55 (1), July, 215-19 20. Donald J. Mullineaux (1988), 'Competitive Monies and the Suffolk Bank System: Reply', Southern Economic Journal, 55 (1), July, 220-23 21. Gary Gorton and Donald J. Mullineaux (1987), 'The Joint Production of Confidence: Endogenous Regulation and Nineteenth Century Commercial-Bank Clearinghouses', Journal of Money, Credit, and Banking, 19 (4), November, 457-68 22. Steven Horwitz (1990), 'Competitive Currencies, Legal Restrictions, and the Origins of the Fed: Some Evidence from the Panic of 1907', Southern Economic Journal, 56 (3), January, 639-49
330
Name Index
341
263
279
294
309
314
318
Acknowledgements The editor and publishers wish to thank the following who have kindly given permission for the use of copyright material. Academic Press, Inc. for article: Eugene N. White (1990), 'Free Banking During the French Revolution', Explorations in Economic History, 27 (3), July, 251-76. American Economic Association for articles: Arthur J. Rolnick and Warren E. Weber (1983), 'New Evidence on the Free Banking Era', American Economic Review, LXXIll (5), December, 1080-91; James A. Kahn (1985), 'Another Look at Free Banking in the United States', American Economic Review, LXXV (4), September, 881-5; Hugh Rockoff (1985), 'New Evidence on Free Banking in the United States', American Economic Review, LXXV (4), September, 886-9. Elsevier Science Publishers for article: Robert G. King (1983), 'On the Economics of Private Money', Journal of Monetary Economics, 12 (1), July, 127-58. Macmillan Publishing Company for excerpt: Fritz Redlich (1947), 'Free Banking', The Molding of American Banking, Part I, 1781-1840, 187-204. Ohio State University Press for articles: Hugh Rockoff (1974), 'The Free Banking Era: A Reexamination', Journal of Money, Credit, and Banking, 6 (2), May, 141-67; Richard H. Timberlake, Jr. (1984), 'The Central Banking Role of Clearinghouse Associations', Journal of Money, Credit, and Banking, XVI (1), February, 1-15; Richard H. Timberlake (1987), 'Private Production of Scrip-Money in the Isolated Community' , Journal of Money, Credit, and Banking, 19 (4), November, 437-47; Gary Gorton and Donald J. Mullineaux (1987), 'The Joint Production of Confidence: Endogenous Regulation and Nineteenth Century Commercial-Bank Clearinghouses', Journal of Money, Credit, and Banking, 19 (4), November, 457-68; Tyler Cowen and Randall Kroszner (1989), 'Scottish Banking Before 1845: A Model for Laissez-Faire?', Journal of Money, Credit, and Banking, 21 (2), May, 221-31; Scott Sumner (1990), 'The Forerunners of "New Monetary Economics" Proposals to Stabilize the Unit of Account', Journal of Money, Credit, and Banking, 22 (1), February, 109-18; Lawrence H. White (1990), 'Scottish Banking and the Legal Restrictions Theory: A Closer Look', Journal of Money, Credit, and Banking, 22 (4), November, 526-36. Oxford University Press for excerpt: S.G. Checkland (1975), 'Adam Smith and the Bankers', Andrew S. Skinner and Thomas Wilson (eds), Essays on Adam Smith, 504-23. Royal Bank of Scotland pIc. for article: Charles W. Munn (1975), 'The Origins of the Scottish Note Exchange', Three Banks Review, 107, 45-60.
viii
Free Banking II
Southern Economic Journal for articles: Donald J. Mullineaux (1987), 'Competitive Monies and the Suffolk Bank System: A Contractual Perspective', Southern Economic Journal, 53 (4), April, 884-98; George A. Selgin and Lawrence H. White (1988), 'Competitive Monies and the Suffolk Bank System: Comment', Southern Economic Journal, 55 (1), July 215-19; Donald J. Mullineaux (1988), 'Competitive Monies and the Suffolk Bank System: Reply', Southern Economic Journal, 55 (1), July, 220-23; George A. Selgin and Lawrence H. White (1990), 'Laissez-Faire Monetary Theorists in Late Nineteenth Century America', 56 (3), January, 774-87.
University of Chicago Press for article: Tyler Cowen and Randall Kroszner (1987), 'The Development of New Monetary Economics', Journal of Political Economy, 95 (3), June, 567-90.
Every effort has been made to trace all the copyright holders but if any have been inadvertently overlooked the publishers will be pleased to make the necessary arrangement at the first opportunity . In addition the publishers wish to thank the library of the London School of Economics and Political Science and the Marshall Library, Cambridge University for their assistance in obtaining these articles.
Part I Secondary Accounts of Free Banking Thought
[1] Excerpt from Andrew S. Skinner and Thomas Wilson (eds), Essays on Adam Smith, 504-23
XII
Adam Smith and the Bankers S. G.
CHECKLAND""
S
attention has been given to Adam Smith's views on monetary questions at the level of high theory. But Smith also commented at length on the operation of banking systems both actual and ideal. It jg necessary, of course, to relate these two levels of discussion to one another. In so doing there appears the classic dichotomy of outlook between the intellectual and the practising banker. This has been present at least since William Paterson and John Law and is today as lively as ever. Smith's treatment is very largely based upon the Scottish banks,t as had been that of Sir James Steuart. 2 This, of course, is not surprising, for it was Smith's native system and had been largely formed in his lifetime. Moreover, it could fairly be claimed to be the most advanced pattern of banking in Europe, and presumably the world. OME
II
To establish a perspective on Smith's ideas about banking it is necessary to take account of the institutional evolution of the Scottish system. Great changes took place in it during his creative lifetime, beginning shortly "fter the 1745 Rebellion, when Smith was in the last two years of his tenure of the Snell Exhibition that had carried him from Glasgow College to Balliol, and continuing through his period in the Glasgow chair (1751-63), the famous continental tour (1764-6), and the years in which he wrote The Wealth of Nations, first in Kirkcaldy (1767-73) and then London (1773-6). Those parts of the book which dealt with banking underwent only the minor revisions after the first edition; nor did Smith take up the subject again in any other form. At the time of the 'Forty-five' there were two 'public' banks in Scotland: the Bank of Scotland (founded in 1695) and the Royal Bank of Scotland (founded in 1727)' These were the only institutions strictly entitled to be called banks. As public banks they enjoyed the status of limited liability, by • Professor of Economic History in the University of Glasgow. I The principal treatment of the Scottish banks is in WN II.ii.41-78. 2 Sir James Steuart, An Inquiry into the Principles of Political Oecol/omy (1767; cd. Andrew S. Skinner, 1966), Book IV, Ch. iii to xxii.
Free Banking 1J
4
Adam Smith and the Bankers
505 virtue of being state creations, the one by Act of the Scottish Parliament and the other by Royal Charter. Scotland had thus two such public banks while England had only one. This was to mean that neither of the Scottish banks could claim the central position enjoyed by the Bank of England to the south. After a period of open warfare the two banks had established a kind of truce. Each was wary of the other, keeping substantial holdings of its rival's notes in case it might be necessary to take either aggressive or defensive action. The Bank of Scotland had considerable Jacobite leanings; the Royal Bank had the opposing sympathy, having been the creation of a Whig clique led by Campbell, Duke of Argyll, part of the programme of Walpole's government for the pacification of Scotland. There were also the private bankers, mostly based in Edinburgh. They inevitably became powerful in the parlours of the public banks. using the purchase of shares as a means of entering the bank directorates, and in varying degrees dominating them. The financing of the tobacco trade was not without its significance in this connection. 3 The bills of exchange drawn by the French Farmers-General provided a very important element of liquidity in Scottish banking, passing through the hands of the private bankers. Perhaps the greatest of these merchant bankers was William Alexander: he was the first merchant member both of the directorate of the Royal Bank and of the Edinburgh Town Council (Price. i.608). The private bankers did not issue notes. but used those of the public banks, made available to them through credits. A third order of banking institution had appeared in 1747-50 when in Aberdeen and Glasgow local groups of merchants set up provincial banking companies. The two public banks composed their differences. entering into a pact of mutual support in 1752; they then set about the attempt to destroy the new provincial banking companies. The immediate incentive to do this was the issuing of notes by such companies; the public banks believed this to be lawfully their monopoly. The note issue was the basis of the business of the public banks, the private bankers having accepted the role of the public banks in this regard. The public banks succeeded in their destructive tactics in the case of Aberdeen, but the two Glasgow companies. the Ship Bank a nd the Arms Bank, fought off their attackers. Meanwhile the private bankers had further flourished. Thus, when Sir James Steuart was setting out his views on banking in his I1/quiry into the Principles of Political Oeconomy. between 1763 and 1767. the Scottish system had a kind of classic unity, with two tiers of institutions. The public banks were at its centre. They issued notes to borrowers on the basis of heritable (real) property. and on personal property. They were not 3
Jacob Price, Fra1/ce a1/d the Chesapeake, a History of the French Tobacco Monopoly, and of Its Relationship to the British alld Americall Tobacco Trades (1973), i.
I6]4-I79I
539·
Free Banking II
Essays on Adam Smith 506 discounters of inland trade bills, and so had no direct connection with commerce; to these banks alone Steuart argued should be entrusted 'the great national circulation' (Principles, 485). They provided the credits and the bank notes on which the 'exchangers' (Steuart's name for the private bankers) could operate. The exchangers in turn performed the function of bill discounters, of course charging interest on their discounts, plus a commission of per cent or more. They were the risk-bearers of the system, the link between the publicly created banks and the world of commerce. 'This set of men', said Steuart, 'are exposed to risks and losses, which they bear without complaint because of their great profits' (Principles, 484). Steuart envisaged that the private bankers could be controlled and when necessary sustained by the public banks. In any case casualties among the exchangers could never be serious for the system as a whole: 'These exchangers break from time to time; and no essential hmt is thereby occasioned to national credits.' Steuart, while delineating the resp('ctiv~ functions of public banks and private bankers, took no account of the fact that the latter were more than bankers, being typically merchants and commission agents as well, using public bank resources in their own transacti,ons, .often speculatively. Steuart does not discuss the provincial banking compantes. But though Steuart, in one sense, conceived of a kind of ideal system based upon division of function, he could criticize the way these functions were performed, cspecially by the public banks. In his discussion of thc long and difficult exchange crisis in the 1760s, when Scotland was drained of the precious metals, and bills on London rose to a high premium, Steuart condcmned the public banks for their attempt to right the situation by reducing thcir credits. Their correct course, Steuart believed, would have been to maintain credit and incomcs, by means of vigorous borrowing abmad (Principles, 507). Steuart presumably did not know that the public banks had indced sought foreign credits in Holland, but had found thc cost prohibitive. 4 Steuart's discussion of banking went beyond the structure of Scottish institutions and the way they had behaved in the recent past. He envisaged a general system of public responsibility for the money supply, discharged through a public office, in the light of criteria derived from the economy as a whole: the tax systcm should be uscd to reinforce the policics that lay bchind monctary action. 'Thc statesman', wrote Steuart, 'ought at all timcs to maintain a just proportion between the produce of industry and thc quantity of circulating (.quivalcnt in the hands of his subjects for the purchase of it' (Principles, 323). Should the circulating medium bc inadequatc
t
4 Dank of Scotland (hereinafter DS) Minlltes, 20 March 1764; DS Ollt Letter l1ooh, 6 July 1764. Dank of Scotland, Edinburgh, National Register of Archives (Scotland), Survey 945, 1/1/5.
5
6
Free Banking II
Adam Smith and the Bankers
the public office should 'supply the deficiency of the metals by such a proportion of paper credit as may abundantly supply the deficiency'. Presumably in Steuart's eyes the Scottish public banks had some such responsibility, though they were owned by shareholders and enjoyed no special position created by the state other than the limitation of their liability. Indeed he urged that the two banks should merge 'in order to form a really national bank' (Principles, 513). Scarcely had Steuart committed his two-part scheme of Scottish banking to paper, a reasonable enough description of the roles of public banks and private bankers, when it began to lose its coherence. The public banks, having long desisted from discounting inland (i.e. Scottish) bills, reentered upon that practice from 1761.5 Already certain of the private bankers had pulled ahead of the rest, creating dominating positions both in the Scottish money market and within the public banks. The Glasgow banking companies, and those formed in Dundee and Ayr in 1763, and in Perth, Aberdeen, and Dumfries (all in 1766), did not trade on the basis of Steuart's division of function, but sought to operate as autonomously as possible, offering a full range of banking services, and relying heavily on their own note issues. Finally the situation was further complicated by the taking in of deposits by the public banks on a new scale after 1766, when the promissory note replaced the old Treasurer's bond; all Scottish banking institutions then became rivals for deposits. 6 The tidy rationale of Sir James Steuart of the mid-1760s was destroyed by the growth of the economy and the banking responses these provoked, especially in the provincial towns. There was now a three-part distinction of scale and function between public banks, private bankers, and provincial banking companies. Steuart had had too little time on his return to Scotland to acquaint himself directly with the on-going developments in Scottish banking; by contrast his writings of 1777 on the working of the corn laws show that he had been able to inform himself in depth in practical terms of the real anatomy of the situation. 7 It is also probable that because the view Steuart took of banking fitted with his general economic philosophy, it cut him off from complex and changing realities. In the 1760s there had been a good deal of public debate in Scotland about banking: it culminated in the Act of 1765. This banking statute and its preliminaries were of fundamental importance to Scottish practice. There were three principal questions. Should entry to banking be free? Should the use of paper substitutes for specie continue to be permitted without regard to the smallness of the units in which it was issued? Was 5 DS Mil/utes, 27 Mar. 1761; Royal Bank of Scotland (hereinafter RBS) Mil/utes, 25 June 1761 . 6 BS, Balances taken from ledgers. 7 'Memorial on the Corn Laws', 14 Oct. 1777, in Skinner, ed., 737-8.
Free Banking II
508
Essays
011
Adam Smith
the w,e of the optional clause a legitimate banking practice? (The public banks and the Glasgow banking companies had taken to inserting in their notes a clause which, if activated by endorsement to the note, allowed the is!'ming bank to postpone redemption in specie for a period of six. month!';, interest being payable at 5 per cent in the interval; WN Il.ii.98.) All three questions centred upon the note issue. One of the responses to the attempts by the public banks to curtail credit during the long exchange crisis was the issuing of petty notes by a great many parties, who th\ls entered upon quasi-banking activities. The small-notes mania of the 1750S and 1760s produced promises to pay for 10 shillings,s shillings, and eyen I shilling. It was easy to put such notes into circulation because of the chronic shortage of silver coin, making people willing to receive payment, especially of wages, in whatever medium was available. The two public banks had made no real effort to supply this need, though the Bank of Scotland issued notes for 10 shillings from q60 R and the British Linen Company had paid its cottage spinners and weavers this way. How far the private hankers joined in this kind of issue is not clear, though Mansfield & Co., one of the largest, appears to have issued 10 shilling notes. There was, of course, no legal restriction whatever on the issue or the size of notes in Scotland. It was this great facility in issuing notes that made Scotland the most striking example of free entry into banking. Nothing whatever was required by way of capital provision or any other safeguard. To this the public banks took great exception, for they claimed with some justice that it was impossible to conduct any sort of reasonable monetary policy under such conditions. They desired, if banking companies could not be put down entirely, that they should be made to desh,t from issuing notes, acting as Sir James Steuart had envisaged his second tier of banking institutions, accepting the 'leadership' of the public banks and using their notes only. The size of notes that should be permissible related to two questions: how was a convenient hand-to-hand currency to be provided in an economy in which the precious metals had already been replaced by paper money to a unique degree, and how far should the loss of specie be allowed to go as paper money displaced it ever lower down the scale of transactions? The optional clause of course represented the fundamental question, how far was it permissible for the state to allow banking concerns, by 'agreement' with their note-holders, to depart from the principle of convertibility of notes on demand into specie? The clause had first come into being as a device necessary when there was bank war: a concern threatened by a rival with a run had only to decline, under the clause, to pay specie. But the optional clause could have a further, far-reaching effect: it could permit the bankers greatly to economize on their specie holdings, at the same time 8
BS Jl1illlltes, 25 June 1760.
7
8
Free Banking II
Adam Smith and the Bankers yet further reducing the precious metal element in the economy. Moreover, it was soon also discovered that the clause could be used by bankers in times of monetary stringency temporarily to exempt themselves from pressure and so to continue fairly generous credits. In this way short-term influences on the monetary situation could be cushioned, and in larger crises the bankers could liquidate their position in a more orderly manner. Conversely, the freedom to do such things freed the bankers from the strict discipline of the gold standard. In 1764 the government in London became aware of the need to intervene in the banking situation in Scotland. It did so through the Privy Council. The Lord Privy Seal was a Scot. He had received correspondence from Provost Ingram of Glasgow on behalf of the local banking companies there, the Ship, the Arms and the Thistle Banks. The Provost enclosed a memorial, together with 'thoughts', to both of which documents Sir James Steuart had appended lengthy and penetrating notes. 9 James Oswald, Adam Smith's boyhood friend, was a member of the Council, together with another prominent Scot and friend of Smith, Gilbert Elliott.lO The knowledge that official action was pending caused the two public banks to make a joint approach to the government. They proposed that they should have 'an exclusive privilege of banking and of issuing printed notes'.l1 If pressed, their delegates were to offer to pay an agreed annual sum to be added to the developmental funds available to the Trustees for Improving Fisheries and Manufactures in Scotland. In so doing the Scottish public banks were asking for a position stronger even that that of the Bank of England, for though the Bank had a monopoly in the sense of a prohibition of other banking companies with more than six partners, at no time had the Bank had a sole right of note issue of the kind the Scottish banks were now seeking. If awarded such a monopoly the public banks were prepared to abandon the optional clause. Alternatively, the public banks proposed that a man or partnership might not issue' notes without having ready to produce £10,000 in land or in the funds. The committee of the Privy Council that dealt with the delegates of the Scottish public banks was led by Oswald and Elliott. Adam Smith may well have been in London at this time (January 1764) for it was at the end of January that he and Henry, Duke of Buccleuch, met there preparatory to their departure for France. It is an intriguing conjecture whether Smith was consulted. In any case the answer given to the delegates was thoroughly Smithian. 'The right of banking', pronounced the Committee in the grand Enlightenment manner, P Partners in the Glasgow banking companies, Memorial, with tlotes by Sir Jallles Steuart of Coltness, 4 Feb. 1763; Thoughts concerning banks atld the paper currency of Scotla1ld, with Notes by Sir James Steuart, 1764. Both in Mitre of Caldwell. Selectiolls of the family papers preserved at Caldwell, Maitland Club part 2, vol. i (1883). 10 BS Milllltes, 1 Feb. 1764. 11 Ibid., 6 Jan. 1764.
Free Banking II
510
Essays on Adam Smith
'is a matter not of Publick favour but of Right to every Subject in Common.'12 The delegates were further informed that should the two public banks proceed at law in an attempt to put down the banking companies, the government itself would introduce new legislation to make such compani('s explicitly legal. The delegates were also told that the optional clause Illust go. Finally, it was proposed that not only should the 'small notes' cease, hut that Scotland should have no notes under £5. This of course was the view held by Adam Smith. The public banks were thus rebuffed. The only thing saved from the wreckage was the £1 and the guinea notes, for the government did not persist with the idea of a £5 minimum for Scotland. In England notes under £t were made illegal in 1775 and those under £5 were banned ill 1777. But the ending of the optional clause did not mean that the Scottish banking system was one that rested upon the instant convertibility, in full, of notes into specie on demand. The Scottish bankers had developed other devices for minimizing such a requirement, plain though it was in law. They used the award of credits against clients to oblige them to be very moderate in their specie demands; a banker typically required borrowers to draw only his notes, and, if possible, to pay in only the notes of others. In this way the Scottish system was one of more or less continuous partial suspension of cash payments. In any case businessmen operating 011 any scale knew that in times of crisis the outcome depended not on the meagre specie supply in Scotland, but the availability of London credits. Meanwhile the exchange crisis continued. In 1764 the two public banks had sought to sustain the Edinburgh-London exchange by loans from the Bank of England and from Amsterdam. Neither effort succeeded. The Bank of Scotland favoured an all-out war by the two public banks against other issuers, but the Royal Bank would not agree, partly because its protege, the British Linen Company, was a vigorous issuer of notes. IJ As the Scottish specie supply reached dangerously low levels both banks bought silver in England and carted it by wagon to Edinburgh. This was very costly. In the course of 1762 the Bank of Scotland had brought from London upwards of £100,000 at a cost of about £4·IOS. per cent, £3.4s. per cent being paid for exchange and £1.6s. per cent for carriage. 14 Bctwecn 1764 and 1769 the Royal Bank spent over £20,000 in this way, equal to the profits of a note circulation of £80,000. 15 The publie banks also moved against 'English riders', that is men who collected specie for export; moreover they adopted strong sanctions against anyone who aided such activities. 16 By mid-summer 1769 the public banks had brought about 12 IJ 14
15 16
Ibid., I Feb. 1764. Ibid., 5 Dec. 1764,2 Jan., 5, 16 June 1766; RBS Minutes, 22 May 1765. Select Committee 0/1 Banks of IsSlie (1841), 303. HOS lH.inlltcs, 4 Dec. 1771. Also WN 1.286. BS fl1il/llles, 13 Mar. 1764; HBS filiI/lites, 18 December 1767.
9
Free Banking 11
10
Adam Smith and the Bankers
5I 1
a considerable credit reduction; the exchange crisis was a good deal relieved. In this situation and in this year the firm of Douglas, Heron and Co., popularly known as the Ayr Bank, was formed, with powerful backing from the landed nobility of the south-west of Scotland, including Adam Smith's protege, the Duke of Buccleuch. It at once began to isssue notes on a large scale by making available easy credits. The two public banks knew that they could not put down the Ayr Bank by their own efforts. Nor could they expect any help from the government in disciplining the system. They were also aware by this time that they themselves were under-capitalized. Retreat was the only tactic: they greatly reduced their lending and note issue, leaving the field very largely to the Ayr Hank. By December 1771 the issue of the Royal Bank had been brought down to £22,753. 17 Between 1769 and 1772 the Scottish banking system consisted of the two public banks, the British Linen Company, the Ayr Bank, about twenty private bankers (all general merchant houses, among which two, Sir William Forbes and Co. and Mansfield and Co., greatly outdid the rest), and four banking companies operating as partnerships in Glasgow together with five others in different towns. There were initiatives afoot for further provincial banking companies. Moreover, at all levels banks and bankers had or sought London connections, either with the Bank of England as in the case of the public banks or with London bankers. All had learned the importance of London-held assets or London credits. The situation was thus a fairly complex one. The Ayr Bank ran its hectic course, collapsing in June 1772. All but four private banking houses went with it. The shareholders of the Ayr Bank were confronted with an appalling deficit. The only way to prevent legal proceedings against them, and the sale of large parts of their landed estates, was to find a means of borrowing at long term in London. This was done through the sale of annuities, authorized by Act of Parliament. In this painful matter Adam Smith was of service to the Duke of Buccleuch and others. 'My attention', he wrote, 'has been a good deal occupied about the most proper method of extricating them.'18 The annuities were a very expensive way of raising the necessary funds, but there appeared to be no alternative. By the end of 1772 Scottish banking had been through ten extraordinary years, involving a prolonged exchange crisis, a major banking statute, a proliferation of new provincial banking companies, a purging of the private bankers, and the brief life cycle of a venture of extraordinary size, daring, and ineptitude. There followed important new developments in Scottish banking, which 11 18
RBS 1I1imlfes, 4 Dec. 1771. Letter of 5 Sept. 1772. in John Rae. Life of Adam Smith (1895), 253.
Free Banking II 512
Essays on Adam Smith
bring the story down to The Wealth of Nations in 1776. Under the influence of the Bank of Scotland the note exchange, briefly begun not long before the Ayr Bank failure, was revived and extended so that by 1774 the Scottish note exchange in Edinburgh unified the system: by weekly and latcr twiceweekly exchanges the argument, so often used in theoretical dcbate, that no bank could force into 'the Circle' more notes than the public would willingly hold, was given greater reality.19 The bank of Scotland now embarked upon a branch system: by 1780 it had seven branches and by 1790 eighteen. The Royal Bank, though it did not open branches except in Glasgow in 1783, became Edinburgh agent for a number of provincial banking companies, providing them with credits. Both public banks set ahout strengthening their capital position. These developments, taken together, signified a new and more constructive approach on the part of the two puhlic hanks to their role within the system. But Scottish banking at the level of the public banks continued to be highly political. There were rumours in 1774 that the Bank of Scotland, through one of the private banks (probably Mansfield and Co.) was trying to secure influence on the Court of the Royal Bank. The private bankers, especially the important ones, had large holdings of the stock of hoth public banks: this was both a good investment and a means of excrcising control through the election of directors. A proposal for the union of thc two public banks, as Steuart had suggested, came to nothing. 20 But one intrigue did succeed. Henry Dundas, in 1776, was in the early stages of building up his political control of Scotland. One path to doing so lay through the Royal Bank. Dundas persuaded William Ramsay and Patrick Miller, senior partners in Mansfield and Co., to sell a considerable part of their holding of Royal Bank stock in order to bring in new shareholders, and through them new directors. As a result of this tactic Sir Lawrence Dundas was ousted as Governor, and the Duke of Buccleuch, so lately involved in the Ayr Bank debacle, put in his place. 21 It is against this background, extending over Adam Smith's years from the age of twenty-two to the age of fifty-three, that his treatment of banking must be placed. III
The crcation of a successful banking sector requires that there be at least some degree of comprehension of the system by legislators together with reasonably adequate rules of conduct for its participants. The statesman I. BS Alil/fltes, 8 July, 9 Dec. 1771, DS Oflt Leiter Boo!I, 19, 2S Junc, 8, 10,17 July 1771; 29 J unc 1774· 20 Anon., A Letter to the Proprietors of the Balik of Scotland (1777). 21 \Villinm Ramsay to the Duke of Duccleuch, 22 July, Apr. 1790. Scottish Record Office, GD I 13/283/8.
11
12
Free Banking II
Adam Smith and the Bankers
51 3
and the theorist start their reasoning from the needs of the economy; the banker typically starts from the conditions of profitability and safety of his own business, given its demand liabilities and the soundness of the local and regional trading with which his principal assets in the form of advances lie. At the theoretical level it is necessary to identify the elements of which the system consists and the relationship of those elements to one another and, in interaction, their links with the economy as a whole. In the years in which The Wealth of Nations was forming, the fundamental theoretical problem could be expressed in the terms: how were the proper limits to the creation of bank notes, to replace or supplement the coin provided by the state, to be determined? The other elements of monetary debate, including the terms of entry into banking, related directly to this. In seeking an answer theorists had to define the role of money in the economy. Was its supply governed by automatic principles so that the issue of paper money had ouly a displacement effect? Did paper simply push sl'xie out of circulation, without altering the size of the total money supply from what it would 'naturally' have been, namely such as was required by the level of trade and incomes, and without inducing further effects? Or was the money supply capable of being positively changed upward and downward by the action of the bankers or a public authority, through expansion or contraction of lending, with far-reaching repercussions upon employment, incomes and stability? Scotland had already been through this argument, some seventy years before The Wealth of Nations, when the protagonists had been William Paterson and John Law,22 the former favouring a system of convertibility of notes into specie on demand, the latter advocating a policy of st:tte manipulation of the money supply. 'l'he banker seldom tried to answer such questions. I lis mind was rooted in his business. This was true over the miscellaneous range of public banker, private banker, and provincial banker. The basic question for all, in the state of banking in Smith's time, was: how much of one's own note issue could be got into the circle and kept there? This gave the banker his interest-free loan from the public, typically some multiple of the capital he had provided. But his note issue depended to a significant degree upon the terms upon which he was prepared to make advances in discounts or any other form. A weak banking concern or one trying to make a beginning would be tempted to accept the riskier kind of borrower, or to extend further in the upswing, thus exposing itself to losses. Under the simplified system envisaged by Sir James Steuart the public banks would act in the public interest, and would have some degree of control over other banking houses. But with the refusal by government to limit free banking in any 22 John Law, MOlley ami Trade Considered: with a Proposal for Supplying the Natioll zuith fdolley (1705.1750); William Paterson, The Occasion of Scotlalld's Decay ill Trade (1705).
Free Banking [J
Essays on Adam Smith way, at a time when a number of new banking enterprises were coming into being, this ability was much weakened, if not destroyed. IV
At the general theorelical level Adam Smith took the natural or equilibristic view of the money supply.23 It could not properly be varied by bank or state action, except perhaps within very narrow limits. Certainly it should not be used as a tool of public policy. Smith was, of course, very much aware of the temptation to use the state in monetary matters. The situation in various of the British colonies had attracted his attention, especially in America where certain colonial governments had not only issued their own notes, but had made them payable not on demand but several years after issue; to get such instruments into circulation, and keep them there, they were made legal tender and receivable in payment of taxes. Such action could only add to inAationary pressure; Smith cordially approved of the Act of the British Parliament in 1764 forcing colonial legislatures to stop such actions (WN Il.ii.IOO-103). But simple abdication by the state in monetary matters was not enough. Though the state was not itself to be a direct participant, it had a responsibility to provide an appropriate legislative environment for banking, as indeed for every economic activity. In Smith's thinking two sets of rules were thus required: one for the state, for the furtherance of an automatic monetary system, and one for the bankers, so that they might behave in a way appropriate to their own interests and yet promote an optimal banking situation. The principles the state must maintain were, in Adam Smith's view, of two kinds. One asserted the rightness of free banking, but the other imposed certain constraints on banking action. The state should assume no supervision over entry into the banking business. But it should encourage the erection of as many banking enterprises as possible, and it should give monopolies to none (WN Il.ii.I06). Thus no banking concern should be in a position dominant over others, much less empowered to control their conduct. For with a Plultiplicity of modest banking concerns the consequences of the error; or speculations of a few, resulting in their ruin, would be dispersed throughout the system so that no serious damage could result. Free entry and vigorous competition were thus the formulae. This view of Smith's is difficult to reconcile with the real situation in Scotland, with the two public banks very much larger and more powerful than any other part of the system, and bound by a pact of mutual assistance. It is perhaps surprising that the extraordinary tale of the Ayr Bank, which lJ See Hobert V. EagJy, 'Adam Smith and the Specie Flow Doctrine', Scottish JOllmal of Political Ecollomy, vol. xvii. 1970.
13
14
Free Banking II
Adam Smith and the Bankers conducted operations on a major scale with a naivete passing into knavery, involving Smith's own patron and others of his friends in heavy losses, did not cause him to amend his simplistic atomistic view of the Scottish banking structure. The competitive system, as envisaged by Smith, could have the further beneficent effect that the banks, in mutal competition, would press their note issues as far as they would go in terms of acceptability to the community, thus economizing on sterile specie holdings. Paper money could thus be an acceptable substitute for metallic currency because it replaced the more expensive instrument of commerce (gold and silver) with the less expensive (paper) (WN II.ii.26). Smith therefore approved of Scotland's abandonment of a high proportion of its stock of coin, fur when necessary, 'goods will always bring in money'. Yet certain constraints imposed by the state were necessary. There were three of these. The first had to do with the size of notes. Smith was a strong opponent of the Scottish small notes, and would, indeed, have restricted the size of notes to a minimum of £5. This he justified on the ground that in this matter of petty notes the natural liberty of a few 'beggarly bankers', which might endanger the security of the whole society, 'ought to be restrained by all governments' (WN II.ii.92-4). Such notes, by penetrating the pockets of 'many poor people' were a threat to such humble folk through the frequent bankruptcy of their issuers. Smith does not discuss the problem of shortage of the means of payment, in small denominations, which gave rise to such note issues. But there was another reason for condemning notes under £5. The general principle was held by Smith that the smaller the size of the notes in circulation the further went the displacement of the precious metals by paper. Such petty notes might cause an almost total loss of specie, forcing it out of its final repositories, tills, and pockets. At this point Smith echoed the fears of Bume for a country that had almost none of the metals, its commerce and industry 'suspended upon the Daedalian wings of paper money'.24 Indeed it is hard to see how a system of convertibility could be operated at all under such conditions. Smith believed that the Act of 1765 had had a good effect, in relieving the dangerous scarcity of gold and silver in Scotland (WN II.ii.92). Smith, indeed, would have stopped the displacement of specie by notes at the level of 'dealers' (i.e. merchants and wholesalers), and not let it reach the level of consumers (WN II.ii.89-90). Smith saw the money supply as being different in kind at the levels of wholesale dealing versus retail. But to have implemented such a view in Scotland after 1765, confining notes to a minimum size of £5, would also have acted as a severe restraint on entry 24 (WN II.ii.86); David Hume, Writillgs RBS Minlltes, 12 May 1773.
011
Economics, ed. Eugene Rotwcin (1955), 69.
Free Banking II
Essays on Adam Smith
to the banking business, for the £1 notes were an essential element in the' profitability of the proliferating provincial han king companies. The second infraction of natural liberty advocated hy Smith in connection with hanking had to do with the optional clause. For Smith, the Act of 1765 was right in denying the liberty to offer or accept a bank note, the convertibility of which was optional on the issuer. This remained true even though the acceptor was willing. For to some degree the optional clause had become yet another means of minimizing gold and silver holdings, contributing to the specie denudation of Scotland heyond the point of legitimacy. Moreover, it had also heen used as a means of disciplining borrowers: those who insisted on gold instead of notes would havc it hinted to them that if they did not abate their demands the clause would be operated against them. 2s Nor should any banking concern havc the 'liberty' of such self-defence against factitious attack. Smith, indeed, h:1<1 nothing to say about the legitimacy or effects of bank war through the' collection of a rival's notes followed by a sudden demand for payment, a not infrequent Scottish occurence. There were two arguments that might have been expected to make Smith sympathetic to the optional clause. The banks, by having available such a means of gaining time against sudden demands, would thereby be able to reduce the amount of sterile coin held in the system. But of course Smith thought in terms of an only partial displacement: the optional clause, like the small notes, might well cause it to be carried too far. Secondly, by using the clause the banks might be able to bring about a smoother contraction in times of difficulty. But this argument had no real weight against the view held by Smith that full and instant convertibility was necessary as an essential and continuous control on over-issue. Smith's policy views on the money supply, then, rested on the belief that it was good to replace much of the sterile gold and silver in the system, exporting it and receiving real goods in exchange. But though the total money consisting of notes and specie together was to be left to natural forces, the encroachment of paper upon specie was to be controlled. This was to be done by the state prescribing a minimal size of note and forbidding the optional clause. By such means the optimal (or at least a better) relation between paper and specie would be achieved, with the former used for large transactions and the latter for small ones. Specie would therefore be available in the system at two principallevcls: as reserves in the banks, uneconomized by the optional clause, and among retailers and wage payers, undisplaced by notes under £5. There was enough of the mercantilist in Smith to make him uneasy about an economy which did not have at least this much specie about. The third monetary constraint approved by Smith was that of a state 2S HBS Millutes,
IZ
May 1773.
15
16
Free Banking 11
Adam Smith and the Bankers imposed maximum on interest rates (WN II.iv.14). The case for a ceiling rate had two elements. It would prevent extortion through usury. But of much more general interest was Smith's belief that it was right to fix a maximum rate by law so that 'prodigals and projectors' would not be able to bid rates to high levels, being those 'who alone would be willing to give this high interest', so that 'sober people', who would bid for the use of money in terms of its probable returns in productive use, would be deprived. This would result in a great part of a country's capital passing out of productive into wasteful hands. But this ceiling rate should be somewhat above the price which is commonly paid for the use of money by sound borrowers; no law could hold the rate below this level. Smith thought the existing legal limit to be 'perhaps, as proper as any'. He thus approved a system in which the rationing of money by price was subject to a cut-off point of 5 per cent. Robert Scott-Moncrieff, agent of the Royal Bank in its Glasgow office from 1783, probably the biggest discounting concern in Britain outside London, described the effect of an interest ceiling on the banker: 'Every other trader', he wrote to head office, 'when his article is on demand can raise the price, but a poor banker has no such means of lessening the demand or profiting by it-every customer thinks he has a right to badger him and tear it from him.'26 Steuart, in a perhaps paradoxical reversal of roles relative to Smith, condemned 'the absurdity of fixing the rate of interest by law among trading men in a trading nation'.27 In spite of the self-equilibrating character of the circulating medium as assumed by Smith, the total money supply had somehow got too big in the 1760s, giving rise to a sustained exchange crisis. How was the 'automatic' thesis to be reconciled with this? Smith's account of the behaviour of the banks has to be pieced together. The story begins, presumably about 1762, with the banks having 'over-traded a little' (WN II.ii.6S). Businessmen had then demanded more. The banks and bankers declined to accommodate them. The projectors then had recourse to accommodation bills, drawing and redrawing upon one another and upon London, a practice called 'raising money by circulation' (WN Il.ii.66-7). At this point, in Smith's account, the banks re-enter the picture. 'The bills which A in Edinburgh drew upon B in London, he regularly discounted two months before they were due with some bank or banker in Edinburgh; and the bills which n in London redrew upon A in Edinburgh, he as regularly discounted either with the bank of England, or with some other bankers in London' (WN ILii.69). Smith in using the phrase 'some bank or banker in Edinburgh' makes no distinction between the two public banks and the private bankers. At this time, in the early 1760s, it would probably be the private bankers in Edinburgh who were making the advances in the form of 2. Scott-Moncrieff Letters, RBS. 9 Oct. I80!. 27 Note to Thol/ghts Concemillg Banking, Caldwell Papers (Maitland Club) i.
212.
Free Banking II
5 IB
Essays on Adam Smith
discounts, uut going to the public banks in their turn for credits, in the manner envisaged by Steuart. Presumably it was the public banks which made the principal effort to limit credit in the first phase descrihed hy Smith, hut then found themselves in the second phase unable to resist the pressure of the private bankers who were themselves in trouhle. The Scottish banks and hankers are thus presented as having tried to restrain credit, and then acting as discounters of a flood of accommodation paper. In the end, instead of standing firm, the public banks had yielded to the importunities of the business world, when they ought to have brollght the whole expansion up short. Though Smith does not say so, the Ayr Bank was one of the worst offenders because its borrowers used its notes to speculate in the London exchanges, buying London bills with them and thus causing them to be returned quickly on the Bank. The banks and bankers had between them augmented the money supply above what was 'natural': "rhe greater part of this paper', wrote Smith, 'was, consequently, over and above the value of the gold and silver that would have circulated in the country, had there been no paper money.' The result had been adverse exchanges and a drain of specie. Smith does not avail himself of the view that the exchange crises from 1762 were due to other, non-banking factors, as Steuart had argued (Principles, 506), including poor harvest and capital movements in the form of the withdrawal of English funds, but insists upon the responsibility of the projectors backed by the bankers. Smith does not take proper account of the efforts by the public banks to control the situation, which in turn created the conditions out of which further new banking companies, enjoying free entry, came. Instead, in describing the behaviour of the public banks when confronted in 1769 with the Ayr Bank, he implies that they had been following a highly inflationary policy, and welcomed the willingness of the Ayr Bank to take on bad borrowers as providing them an escape from an impossibly over-extended situation. In this way 'they were enabled to get very easily out of that fatal circle, from which they could not otherwise have disengaged themselves without incurring a considerable degree of loss, and perhaps some degree of discredit' (WN lI.ii.74). This indictment of the banks and bankers as capable of creating inflationary conditions on such a scale is never explicitly assimilated to Smith's gcneral monctary theory. It may be that he regarded such inflation as of short term and subject to self-correction through specie loss abroad. But Scotland lived under such conditions more or less continuously for some ten years from 1762. Nor does Smith consider whether in that period inflation contributed to growth of output. The fact was established, however that not only had 'the circulation ... been overstocked with paper money', but it had been increased beyond its natural level as defined by the specie that would have been present on a
17
18
Free Banking 11 Adam Smith and the Bankers
specie system only. The ability of the bankers to create such a situation, and their actual behaviour in doing so arose because the banks as entities had lacked both an understanding of their situation and a code of conduct to guide them. Trouble had arisen because 'every particular banking company has not always understood or attended to its own particular interest' (WN Il.ii.53). Note that at this point Smith seems to place the blame on 'banking companies'. It followed from this that a code of sound practice derived not from generalized monetary theory, but from the immediate and multiple transactions of the banks, would ensure the correctness of the money supply. Smith believed that the practice 'of the banking trade ... is capable of being reduced to strict rules' (WN V.i.e.32). Indeed the availability of such rules made banking one of the few activities capable of being treated in terms of standard practice to such a degree that it could safely be entrusted to joint-stock companies. Smith's code contained four elements. If these were continuously adhered to two consequences would follow. Every banking enterprise would be sound, and the total money supply would vary strictly with the needs of the economy-it would be continuously correct. First, there was the real bills doctrine: bankers should discount only those bills that represented the financing of the movement of real goods at short term (WN ll.ii.59). Each bill would thus be self-liquidating, for the delivery of the corresponding goods would provide for its discharge and for the reflux of the bank notes that had been advanced. Secondly, bankers in their cash credit advances (a form of overdraft) should insist that the account be frequently 'turned over', with repayments and new borrowing (WN II.ii.6I). The banker would then be able to control his total advances at any given time by curtailment of renewals. Thirdly, the general principal of short-term lending only, embodied in the first two rules, was reasserted: only where repayment was certain within three or at most eight months should a loan he made (WN II.ii.6o). Finally, a banker should relate a loan not to the total resources of a borrower, but only to that part of his capital which he would otherwise be 'obliged to keep by him unemployed and in ready money for answering occasional demands' (WN ll.ii.59-63,86,93). That is to say, the banker should go no further than relieving a borrower of the need to keep a cash float. These rules were not of course original with Smith: all but the fourth were the standard canons of Scottish banking, though not of course always or necessarily the practice. Was this code capable, as Smith suggested. of promoting an optimal money supply over time? Was it true that if each bank followed 'correct' practice the money supply would look after itself? Only two of the rules, the first and the last, had anything to say about how much should be loaned. The first was the more realistic, for it could be related to bills of lading and
Free Banking II
520
Essays on Adam Smith
other documentR. But it did not take account of changeR in the priceR of commoditieR during the lifetime of a bill; a fall would mean that the Rl1ITlR entered in the bill of exchange would be lesR than the value of the gnodR when Rold. The fourth rule, related to the cash float, would he diflicult to apply, for Ruch a concept is hard to define and meaRure. There waR, of course, one element of clarity in it: that the banker Rhould certain Iv not finance fixed capital, and only a Rmall proportion of variahle ca'pital, certainly not, for example, the wages bill. Rules two and three had to do with the length of life of loans: they merely asserted that they should he short in order that the banker could continuously review his Ritllation. tn general, no attempt was made to build a bridge between the money supply created under these rules and that which would, in aggregate, yi('ld fll11 employment. It was assumed that the 'natural' money supply that would he called for on a pure specie basis (e.g. that appropriate to the nccd~ of the economy) would be forthcoming under the operation of these rules. It would Reem highly likely however, that Rhould the bankers have confin('d their lending to Smith's austere programme the economy might \\'dl not have had the money supply neceRsary to promote full employment. How would the matter look from the banker's point of view? For them the critical questions always were: when should credit contraction be brought about, what means should be used (e.g. refusal of renewal of discounts versus reduction of advances on cash accounts or on other baseR), who should be subjected to pressure (e.g. which borrowerR should be made to repay), and how much, in aggregate, was it necessary to draw back? Of course it could be argued that if Smith's four canons of bank lending were followed there would be no upswing, but merely a steady progression. Unfortunately Smith docs not discuss the causes of the trade cycle, presumably finding them to lie in departures from his rules. The implication in Smith is that if every lending transaction were in the terms prescribed the system would be stable at full employment. But from the point of view of the public banks, with their experience of crises, brought about or at least aggravated by other bankers who were competing for note issue via loans, it was necessary to bring in an element of generalized control. It was no good elaborating a system of ideal internal rules; loans did get made on other terms than those set out by Smith and it was necessary to be able to cope with the resulting situation when it arose. Hence the public banks' constant recurrence to the note issue as the most important condition of entry to banking. V
There were a number of further loose ends m Smith's thinking about banking. One had to do with the collective, imitative, and mutually exciting
19
20
Free Banking II
Adam Smith and the Bankers
521
activities of bankers. Smith lays down the general proposition that 'though the principles of common prudence do not always govern the conduct of every individual, they always influence that of the majority of every class or order' (WN II.ii.36). This statement is of course a fundamental element of the liberal creed: the assertion of the free-standing individual, capable of rational choice within the parameters in which he finds himself. Yet Smith in his treatment of the inflation of the 1760s and early 1770S describes the bankers as collectively supporting inflation by discounting a flood of accommodation paper. 'The Scotch banks, no doubt', wrote Smith, 'paid all of them very dearly for their own imprudence and inattention' (WN II.ii.S6). Adam Smith does not directly discuss the question of the right of the public to insist upon disclosure by note issuers of their situation. His four canons for bankers were all domestic and interior: the bankers should be left to apply the rules to themselves, without state or public scrutiny. Steuart believed otherwise, at least so far as non-public banks, with no legal identity and no permanently committed capital, were concerned. All note issuers not trading with public authority, said Steuart, 'ought to be obliged to keep books open to inspection by the public'.28 Such an insistence upon democratic scrutiny might be thought a fitting part of the system of natural liberty, improving the data upon which an individual could base his transactions with such bankers. An individual who did not like the behaviour of a banker as revealed in his books could then decline to accept his notes or otherwise trade with him, turning to another whose procedures were preferred. But at this point a difficulty arises. Would not such scrutiny by the public lead directly to state intervention? For a person aggrieved by the behaviour of a banker, judging him to be behaving dangerously, or of having so behaved, might well invoke the law. It would then be necessary for the judges to be provided with a set of principles. This would involve the state in laying down canons of sound banking. The outcome, therefore, of disclosure, could well be a state-enforced code of banking, a thing that would on general grounds be abhorrent to Smith. So stood the dilemma of disclosure. Steuart, for his part, does not enter into the difficulties of enforcing the opening of the books of bankers. Especially in a note-issuing situation such a requirement could make every banker vulnerable to attack by his rivals or his enemies, and might well unsteady public confidence in times of monetary disturbance. The view implied by Smith, of course, prevailed: Scottish and English banking companies proceeded on the basis of no required public disclosure down to the Companies Acts of the midnineteenth century, and with the legal minimum thereafter. With the affairs of the note-issuing bankers inviolate from public scrutiny there could be no question of requiring other business concerns not providing money 28
Ibid.,
226.
Free Banking II
522
Essays on Adam Smith
substitutes to disclose their position: the veil of secrecy was C::lst over busincss in gener::ll in the interest of the system of n::ltural liherty. The bankers were related to the state in yet another way. Thi!! too Smith passes over. It had to do with the remitting from Edinburgh to London of the public revenues. The right to hold public balances and tr::lnsmit them was highly valuable, for it increased the disposable resources of b::lnkers t.1l the extent of the average holding of such monies. It was also important from the point of view of note issue, for taxes were paid in bank notes, so that the remitting bank received large sums of the notes of other banks and W::IS accordingly in a position to substitute its own notes in the Circle ::Ind indeed, if so inclined, to harass its rivals. The Bank of Scotl:md, lacking political favour, had no standing with regard to the revenlle: holding and remittance of the public funds was shared between the Royal Dank and the private bankers Sir William Forbes and Co., both highly acceptahle to the ruling party. The Royal Bank had the account of the Receiver General of the Customs, while Forbes and Co. enjoyed the Excise Account. Forbes and Co. had inherited their preferred position from the firm out of which they had come, namely Coutts and Co. of Edinburgh, who had remitted the excise since 1742. Though the state had no business to dispense in Scotland in terms of issuing and managing the public debt, so profitable to the Dank of England, it did have the revenue to dispose of, and did so in terms of the patronage system. In this important sense the state was involved in Scottish banking, operating not by principle but by favour. 29 Finally, there is the question of Smith's estimate of the Scottish money supply in the I 770S. His view was that 'in the present times the whole circulation of Scotland cannot be estimated at less than two millions, of which that part which consists in gold and silver, most probably, docs not amount to half a million' (WN II.ii.43). This means that at least £1,500,000 must be accounted for in terms of bank notes. But calculations would put Scottish notes in the hands of the public at some £900,000. Any possible margin of error or variation could hardly bring this above a million. It seems probable that Smith, if he inquired about the size of note issue of the various banks, was given a figure representing those notes signed by directors and paid to the cashier. But the Scottish banks always held very large sums in their tills. It is this till money that would seem to have entered Smith's figure of issue to the public. VI
Smith did not confront Steuart's views on banking; in this as in all othcr aspccts of thc economy hc studiously ignored Steuart. Had hc done so, 29
WilJiat11 Ramsay to Duke of Buccleuch, 26 May 1790, Scottish Hecord Office, GD
113/283/7.
21
22
Free Banking II
Adam Smith and the Ballkers
52 3
taking account of the non-homogeneity of Scottish banking institutions, his treatment might have gained in realism. It has frequently been said that a good many aspects of Smith's thought were not rigorous in the integrative sense; that his mixture of theory and institutional treatment yield a picture that is highly elliptical. This is certainly true of banking. It was necessary, of course, that his views on the subject be consistent with his general approach to the economy, resting as it did on the market mechanism operated by men who enjoyed free entry and minimal supervision. The banking expression of Smith's system of natural liberty was a set of institutions composed of many enterprises, none capable of monopolistic power or even leadership, each guided by prudential rules, and all trading within an environment of law provided by the state. Acting in aggregate they would, Smith implied, provide an optimal money supply or an effective approximation to it. But in the light of banking conditions in the Scotland of Adam Smith's time, not to speak of the years to follow when matters approximated even less to his assumption, his view of banking omitted important aspects of reality which, if properly attended to, might have damaged his view of economic processes. Both Steuart and Smith can be accused of allowing the demands of their general systems of thought to affect their presentation of banking. Both left some important theoretical issues unbroached and both left some unresolved. But taken together they provide a most illuminating dichotomy, perhaps the best theoretical starting-point for a consideration of pre-industrial banking, and not without relevance for later times.
[2] The Development of the New Monetary Economics
Tyler Cowen and Randall Kroszner
This paper looks into the history of economic thought to examine the forerunners of the "new monetary economics." This approach elllphasi:tes the role of regulations on private financial intermediation in determining the particular institutional arrangements that contemporary monetary theory treats as data. The "new view" investig'lles the possibility that under laissez-faire the unit of account and means of payment, traditionally bundled together in the item called ··nlOney," may become separated. The earlier writers who share this perspective have been overlooked by historians of economic thought as well as by recellt contributors to the new monetary economics. An examination of these theorists is of more than historical interest, for man)' of their insights and analyses are relevant to modern monetary theory.
Close on the heels of the "new classical" revolution in monetary theory is another intellectual trend rapidly gaining momentum. This new approach has been dubbed "the new monetary economics" (Hall 1982) and "the BFH system" (Greenfield and Yeager 1983), taken from the initials of three important innovators: Fischer Black (1970), Eugene Fama (1980), and Robert Hall (1982, 1983). Other recent contributors to this approach include Neil Wallace, Thomas Sargent, Robert Greenfield, and Leland Yeager. The new monetary economThis is a revised version of a paper originally emitled "The Developmem of the Legal Restriuifllls Theory of Money." In addition to the participants in seminars at George MaSflll lJlliversity ami New York University, we would like to thank Richard Cothren, George Stigler, Bill Woolsey, Leland Yeager, and an anonymous referee for valuable COllllllelllS and/or discussions. Kroszner acknowledges support from the National Science Foulldation. The usual disclaimer applies. [Juunlul C, l~g7
uf Pulil,(ul £'0',0"". IY87. vol. 9~. no. 3) loy The Univer>ity of Chicago. All right. r.",,,,.d. 0022-380818719503-0003$01.50
24
Free Banking II JOURNAL OF POLITICAL ECONOMY 568 ics-henceforth NME-combines features of Wallace's (1983b) "legal restrictions theory of money" with the possibility of separating the unit of account and medium of exchange functions of money. as will be outlined below. Notwithstanding the recent interest in the NME, there has been very little effort to discover its roots. Sargent and Wallace (1982) have noted certain similarities between the legal restrictions theory and the real bills doctrine of the British "Banking School." They have, however. overlooked a number of important writers who are in many respects much closer to the legal restrictions and BFH points of view. Since the NME has a long tradition starting in the eighteenth century and stretching into the early part of the twentieth century. our paper shall examine its development through the history of economic thought. The relevance of our present endeavor is twofold. First. many of the ideas discussed by the early theorists of the NME mold are relevant to modern economic theory and are just starting to enter the current debate (as discussed below). Second, nearly all the figures in the history of the NME have been unjustly overlooked or completely ignored and deserve a more prominent place in the history of monetary theory.· Our paper opens with a skeletal description of the NME. which is intended as a synthesis rather than a representation of anyone author's views. As a general introduction to the NME forerunners' framework, some comparisons and contrasts with other approaches to monetary theory, including the quantity theory and the Austrian evolutionary approach, are offered. The following section briefly considers some of the earlier and less systematic discussions of NME themes such as monetary separation and the abstract unit of account. Next, we examine the first writings concerning the NME. spanning approximately the two decades prior to the First World War, when this set of ideas was debated extensively among such writers as William Whittick, Alfred Westrup, BerYamin Tucker, Hugo Bilgram, Arthur Kitson, J. Greevz Fisher, and Henry Meulen. These discussions concern the results of free banking as well as both the feasibility and desirability of the separation of monetary functions. We then I None of the major histories of monetary or economic theory, such as Viner (1937), Mints (1945), Schumpeter (1954), or Blaug (1978), discusses this active group of Amer· ican and British writers. Three surveys that focus specifically on free banking thought- Vera Smith (1936), Brown (1982), and White (1984b)-and Hayek's competitive currencies proposal (1978) only cite Henry Meulen in passing. Finlay (1972) contains a brief discussion of Arthur Kitson and Meulen but misrepresents their monetary theory. In his encyclopedic compilation of all economic thinkers who advocated some form of currency stabilization, Fisher (1934) devoted two sentences to William Whittick (p. 54) and provided a short description of Kitson's contributions (pp. 51, 122-24).
Free Banking II NEW MONETARY ECONOMICS
foclis on Meulen's thorough analysis of these ideas. The concluding section summarizes the contributions of this neglected group of theorists and raises some of the issues that remain unresolved for the NME. Characterization of the New Monetary Econom·ics
We shall characterize the NME by the following seven propositions. 1. "Legal restrictions on private intermediation" in the financial seClOr are crucial to the analysis of monetary phenomena (Wallace 1983b, p. I). Highly critical of the assumptions of monetary theory that render it narrowly institution-bound, the new school asserts that alternative institUlional frameworks "would make current monetary theory almost completely invalid" (Black 1970, p. 9). 2. As a direct consequence of the emphasis on legal restrictions, the "new view" finds that "money is exactly a creation of regulation" (Hall 1982, p. 1554).2 In a laissez-faire economy, completely devoid of governmental interference in the financial sector, "money in the usual sense would not exist" (Black 1970, p. 9). 3. Thus, in such a world, the quantity theory of money (as well as the liquidity preference theory [Black 1970, p. 20]) loses meaning. The quantity theory is valid only under special institutional arrangements that arise because of legal restrictions on intermediation; the quantity theory relations are an "artifact" of regulation (Hall 1982). 4. The provision of monetary services is conceptually distinct from the provision of what we typically think of today as "money" (Fama 1980, p. 49), and the laissez-faire case in which the private sector provides only the former is a useful analytical benchmark for monetary theory. In addition, proponents of the NME find the welfare implications of laissez-faire in financial intermediation to be favorable. 5. The traditional functions of money (unit of account and means of payment) need not be intertwined in a single entity called "money." The unit of account might be purely abstract or take a real commodities form. 3 In either case, it could exist separately from the media of payment. The latter might take the form of real asset claims (Fama 1980, p. 43). 2 This Slatt:Illt:1II rdi:rs to the continued existence of money in a developed economy, Ihe cillcrgcncc of lIIoncy ilSdf, and so Ihis view should nOI be confused wilh Knapp', (I U~4) "SlalC theol"Y of money." He al"gued that money comes illlo being and gains i1s "validity" cxdusivdy thwugh the decrees of the state. Ellis (1934, pp. 13-41) contains a concisc sllllllnary and sympathetic criticism of Knapp's "chanalislIl." MOl"e on this i,sllt: laler. ~ Tht: uOlion of an austract accounting unit will be developed in the next section.
1101
25
Free Banking II
26
57 0
JOURNAL OF POLITICAL ECONOMY
6. In the absence of legal restrictions in the banking sector of a developed economy, the disentangling of the monetary functions would eradicate many of the problems attending changes in the value of the unit of account. The banking and financial sector would thus have "no special control over prices or real activity" (Farna 19RO, p.47). 7. This independence of the financial an9 real sectors would eliminate those sources of macroeconomic instability arising from the "money side." In this way, the macroeconomic properties of a laissezfaire world would bear a striking likeness to those of a harter ecollomy. Fama (1980) has even called it a system of "sophisticated barter." Analytic Overview: Some Comparisons and Contrasts
Before permitting the forerunners to speak for themselves, we shall explore some comparisons and contrasts of their ideas with, first, the doctrines of the quantity theorists of their time and, second, the modern NME theorists. 4 Both the turn-of-the-century quantity theorists and the NME forerunners were strongly influenced by the bimetallist controversies, which focused on the ability of monetary disturhances to generate deleterious macroeconomic effects as well as disruptive changes in an economy's "standard of value" or unit of account. In this way, both approaches are acutely concerned with developing a financial system that would minimize disturbances arising from shifts in the supply of and demand for money. The two schools, however, part company in their respective analyses of the institutions that would achieve such a goal. First, for the NME forerunners, "money as we know it," while performing important functions in the economy, is a fundamentally disequilibrating force. Tying the money supply to gold or some other commodity base through convertibility would subject the entire economy to the vagaries of supply and demand conditions of that commodity. Breaking the tight link to gold through central bank management of a gold exchange or fiat standard could not achieve stability either (see additional "political" reasons below). The forerunners argue that monetary problems arise precisely because a special commodity or governmentally issued paper money has such a central position in the economy. The pivotal role of the circulating asset or asset claim is what gives "money" its powerful ability to wreak havoc in the economy. In this way, the forerunners found the quantity theory • Once again, we are providing simply a thumbnail synthesis here. not exegeses of individuals· works.
Free Banking II NEW MONJ::TARY ECONOMICS
57 1 to be highly misleading for the analysis of monetary phenomena because t he theory does not take account of this underlying source of disclJ uilibrat ion. Second, in order to achieve a set of institutions that would permit "money" lO be dislodged from its disruptive position in the economy, the forerunners believed that total deregulation of the monetary systelll would be required. In contrast, the quantity theorists considered sm:h a situation infeasible and typically believed that some form of celllral bank control is necessary to eliminate monetary disturbances. For the forerunners, legal restrictions are an important obstacle to stabilization, for they help to maintain the focal position of "lIloney." State COllt rol and regulation were identified by the forerunners with Illollopoly, privilege, and manipulation of economic varhlbles contral")' to the public interest. In addition to the criticism in point 1 above, this perspective led the forerunners to have little faith in proposals for government management of the system. They disagreed with those also wary of state intervention who advocated the gold stalldard becallse sllch a cOlllmodity hase could not be insulated from state interlerence (as Clark [1984] has recently documented f()t· the period in which the f()rerunners wrote). This second "political" level of criticism ari5es from the forerullners' libertarian and "public choice" analysis of government involvement in financial regulation (see ~IcElroy 1981, 1982). While many of the quantity theorists also harbored suspicions of government power, they viewed the possible abuses of such control over monetary institutions as far less troublesome than the effects of complete del·egulation. Third, unlike the quantity theorists, the forerunners wished to sepanlte the means of exchange from the unit of account in all transactions, Notions of separation of monetary functions, however, are not unique to the Nl\IE forerunners. Variations in the value of the unit of account motivated Jevons (1876) to devise his partial indexation scheme involving "multiple legal tenders."s Similar considerations also led Irving Fisher (1912, 1920) to develop his "compensated dollar" program. Such proposals do not address the concerns of the forerunners enumerated above. A "money" would still have a precariOilS place in the economy. and such indexation schemes would still be vulnerable to political manipulation. In addition. such plans en-
" Thc onlill,lry Kold stalld,lrd was to be used 1(lr everyday Ir,msacliuns, whcrc,ls ,Ill illllcXl'd "",hular slam!;lrd" w()uld be Ihe slillulard f()l' ,III illlenemp()fil: C()nll'ill:lS ()"cr ::Imolllhs ill duralion . .Ic"()IlS linmd Ihal "we cUllIe 10 rCKard as alm()SI Ilc('cssary Ih,lI union ,,( fUllnillllS whidl is. ill lhe mOSI, a m,lIler ()f l'Illlvcniencc. 'liltl lllily llOl '1Iw'Iys be dcsirahlt,' (I M71i, p. 16).
27
28
Free Banking II JOURNAL OF POLITICAL ECONOMY
counter traditional index number problems, for example, type of index, choice of weights, and revisions over time. fl In order to resolve the problems they posed, the forerunners dcvdopcd thc concept of the ahstract unit of account. Under the abstract IInit scenario, both market goods and instruments employed as media of exchange have their own price in terms of the unit of cKCOllnt. The unit of account is "abstract" in the sense that it does not correspond to any existing commodity or exchange medium. The quantity theory is not relevant to such a framework, for, first, the notion of "the" price level is no longer particularly meaningful and, second, every good could be said trivially to follow a quantity theory relation with the appropriate choice of a price leveL' Since exchange media have their own price in terms of this unit, changes in their supply and demand would exercise no special influence over economic activity. The transactional inconveniences resulting from distinct prices for medi~ of exchange are clearly a major disadvantage of the abstract unit. Although the issue was briefly noted during the debates among the forerunners and their critics, it did not receive systematic attention until White (1984a, 1985), McCallum (1985), and O'Driscoll (1985) made the point central to their criticisms of the modern NME theorists. Despite the many similarities outlined in our paper, the forerunners deviated from the modern NME theorists in one important rcspect that makes the forerunners more akin to certain branches of the older quantity theory: They typically adopted an evolutionary approach to the development of monetary institutions. This approach was pioneered by Menger (1871/ 1981, pp. 257-85; 1892) and von Mises (1912/1971) of the Austrian school. H According to the Austrian evolutionary theory, money emerges as the outcome of a decentralized sequential decision-making process. The purchasing power "(;rcl"nlil"lri ami Ye
Free Banking II NEW MONETARY ECONOMICS
573 of IllOlle)' arises out of barter as the undesigned consequence of individuals' illleractions in the marketplace.!! The pivotal role of confidCllcc ill tlae llIolletary institutions, which thereby generate the expectations that determine the media's values, is an important element of Austrian monetary theory, which the forerunners also adopted (see Ellis 1934; Frankel 1977, pp. 36-41; Simmell978, esp. pp. 190-203; Laidler and Rowe 1980). We shall describe below the attempts by Kitson and Meulen to extend this evolutionary approach to explain Ihe elllergence of Ihe abstract unit and the separation of monetary functions as a further stage in the development of financial institutions. 1\·luch of the controversy among the forerunners and their criti(s surrouuds the explanation of a pmcess by which an abstract unit could arise and be sustained in the economy, a challenge White (lYIHa) has put to the modern NME theorists. It is the adherence to the evolutionary approach that led the forerunners to views of a few specific aspects of the NME world differelll from those of the BFH thinkers and legal restrictions theoriSIS. Advocates of the BFH system typically envisaged some form of positive state intel"vention in order to institute the monetary reform: redefinilion of Ihe unit of account. In contrasl, the forerunne.-s and Ihe Minnesota legal restrictions theorists wished to focus on the consequences of relaxing governmental constraints in order to examine what will arise. III The formal models that the modern legal restrictions Iheorisls have employed (e.g., Kareken and Wallace 1980; Sargent aJl(1 Wallace 198~) start de novo and Ihen trace the path that gives rise to lIIonetary Of
29
Free Banking II
30
.JOURNAL OF 1'0l.lTICAI. ECONOMY
!l74
reature of the overlapping generations models and Wallace 1198:\al for a reply.) The equalization of the rates of return on different stores of value is a key to the welfare implications of these models. The forerunners. more akin to the BFH theorists. stressed the gains to he secured by providing a stable unit of account and the diminished macroeconomic influence of changes in the supply of and demand for exchange media. It is these features. the forerunners argued. that result from laissez-faire. The equalization of rates of return on various stores of value plays no role in their model. Instead. they perceived the welfare gains of laissez-faire as accruing from the t\\·o other functions of money. the unit of accollnt and the medilllll of exchange.
Early Experiences and Discussions of Monetary,Separability Some degree of separation of the means of payment from the accounting unit is a typical feature of monetary systems from primitive times to the French Revolution (see Einzig J9fifi). Einaudi (195~, p, 257) describes these systems consisting of an "imaginary" ullit of account separat.ed from the media of exchange: In former times. because of the existence of money of account, men every day set a price on the florins. scudi. ems. doblons, sequins. and testoons which they recei\'ed and paid out. Every day, in every single transaction. it was lIIade dear to their minds that the money with which they paid. e\'en bank money or paper money. was a colllmodity like any other. that its price was governed by the market and. like any other price. was the result of an infinite numher of economic and noneconomic forces which determine t.he general equilihriulII of all prices. II This curious characteristic did not go unnoticed by early monetary theorists in both Europe and the United States. ~Ionlesquieu (I i48! 1899. PI'. 378-80) was one of the first to consider a system with complete separation. He described the "macute" system reportedly employed by certain African lribes in which the macUle was adopted as a purely abstract unit of account of valuation in exchange. 12 II Einillldi (1953. p. 233) goes on to explain. "Money of account was not created by decrce but grcw almost spontaneously out of men's habit of kceping accounts inlllone· tary units." I\fore recent medicval and renaissance scholarship (L'mc and Mucllcr 1985) supports t.his view that abstract accounting units evolved Ollt of a "long process of historical change" (Einaudi 1953. p. 246). 12 MOl1lcs<)uicu explains: "Thc negroes on the coast of Africa havc a sign of \'alue without moncy. It is a sign merely ideal. foundcd on the degrcc of cstcem which thev
31
Free Banking II NEW MONETARY ECONOMICS
575
As a method for better understanding the consequences of monetary disturbances, Scottish mercantilist Sir James Steuart (l810, pp. 1-21) suggested analyzing a system based on an abstract accounting unit separated from the exchange media. The unit, labeled "money of account" (p. 2), was compared to both the African macute (pp. 6-7) and a yardstick of unvarying length (p. 13), a metaphor that is frequently found in the work of the writers we will be discussing and has made its way into the modern literature (e.g., Greenfield and Yeager 1~IH3, p. 3(1». John Stuart Mill also took note of the macule system in his Principles (1 ~7l/ 1929, pp. 483-84) but did not elaborate on it. r-,'Iill's t:olllemporary John Gray (1848, esp. chap. 4) offered a reform proposal based on monetary separation. This separation, he argued, would ellsure macroeconomic stability because Say's law would be in for<:e, as in a bane.· ecollomy.l;! In the lJ Ilited States, Stephen Colwell H presented an institutional approadl to monetary theory hinging on the distinction between "l1loney of accoulll" and the "means of payment" (1859, pp. 2-3, 101, 459-tiO). He reviewed a variety of proposals for monetary reform that addressed separation issues (including the debate between Rohert Morris and Alexander Hamilton Oil the appropriate coinage systcl1l for the )'oullg United States [pp. 88-89]).15 In addition, Colwell desuibed the American and Canadian colonial experiences in whit:h the "monies of account" of France and England persisted long aftcr lhe corresponding means of payment were no longer available. A so-call cd hookkeeping barter system, which lasted into the 1800s, was dcveloped ill the American colonies and provides an example of how accoullling operated under monetary separation (see Baxter l!lflti).lti
fix ill lheir millds for all merchandise, in proportion 10 the necd they have of it. A ccnai .. clJmmodity or mcrchandise is worth three macoutes; another, six maCOllles; allolhcr, tell macolltes; that is, as if they said simply three, six, alld ten. The price is formcd by a comparison of all merchandise with each other. They have therefore no panintlar mOlley; but eilch kind of merchandise is money 10 the other" (1IlUU, p. 37H). Thc cightcclIlh-cclIlury Italian economist Ferdinando Galiani (1750/1977, p. 7!!) alsu 1I0ted tIlt: allstrilct 'Iualities of the macute. , I~ Somewhat Iouel·, ill the German literature Simmd (IU07/1978, pp. IUII-U3) was ililercslcd ill "cilses ill which a certain kind of money is used as a standard of valuc but 11(1( for ,,("Iual paymclIl" alld examined examples ill Europe, Africa, alld Illdia. H ()II (:ohvell's lifc, works, ;lIul rel;lIiollship wilh IlcllI·Y C...'ey, sce C;ucy's "MclllIIiI' of SlcphclI Colwcll" (I H7l!), rcad hdin'c Ihc Amcl·icillI l'hilosol,hit:Oll Assodillillll II1cclillKs. Sce also Kuhey (I U31l) for a summary of Colwell's eculIOIllic cUlltriblitillllS. ,r. Colwell illso provi,lcd rdercnces to thc dcvcluplllcl\l of the idea of IllUllclary sepilrabilily ill Ihe French liter.llure, including the Marquis Garnier and Charlcs Cuquelill (pp. 75-77, 330). III Although these early precursors of the NME frequently provided ferlile illsiKhlS, lhey did not develop a systematic approach to monetary theory nor form ;1 cuherelll school of thuuKht. The loose citation practices that characterize much lale lIillCICClllh-
Free Banking II
32
JOURNAL OF POLITICAL ECONOMV
Forerunners of the New Monetary Economics A.
Late Nineleeuth-Century ReJ01'1nerS
William Whiuick and Alfred Westrup were the fIrst Americans to attempt to integrate the ideas of monetary separation with a Iheory of free banking.17 Whittick's views are most fully expounded in his Valtle and an Invariable Unit oJ Value (1896). While this pamphlet is long on polemics and short on content, Whittick had considerable affinities with the NME perspective. His preferred solution was a system offree bankingl!! in which "(I) money should be simply a scale of value, (2) its unit entirely arbitrary, (3) free from any commodity incumbrance, and (4) it must be poised 'in a just equilibrium between f1uctualing values'" (1896, pp. 84-85). The media of exchange will arise from the monetization of all exchangeable wealth (p. 91). Like most olher writers in the tradition of the NME, Whiuick saw the continued existence of the gold standard as a result of monopolistic intervention (p. 92). He claimed that the precious metal cannot be a standard of value because the value of gold itself is continually varying, but rather is merely a standard of quantity and fineness because the dollar was equal to 25.8 grains of gold (p. 41). Similarly, Westrup wrote that "Ihe misapprehension and confusion that befogs the minds of those who insist on the 'standard of value' idea, is the result of not viewing money and the denominator, 'dollar,' separate and distinct from each other. [Here Westrup footnotes Mill.] Each has its independent function. The one is to aid and facilitate the distribution of wealth; the other is a means by which we can express more or less value" (1895, p. 100). Westrupl9 supported both free banking and a mutual banking scheme taken from P.-]. Proudhon and William B. Greene (see Proudhon 1927). His plan involved the exchange of claims to real assets (1895, p. 112), while economic calculation would occur in terms of abstract units rather than commodity units. Both Whittick and Westrup claimed that their alternative would increase debtor-creditor
and early twentieth-century economic writing prevent us from detcrmining the extent of the influence of these earlier authors on the Bilgram-Kiuon-Mculen circle discusscd below. Nonetheless, these early writers do deserve recognition for their priority in constructing important parts of the foundation of the NME approach. 17 Claims to the priority of Whiuick and Westrup are documented in the following issues of Liberty; June 27, 1891, p. 2; June 29, 1895, p. 2; August 20, 1895, p. 6; and August 24, 1895, p. 7. I" Whittick argued that it is best to minimize government interference into monetary affairs because "legislation is essentially stationary, while society is naturally evolutiona ry" (1896, p. 98). 19 Many of his views had been developed earlier (see Westrup 1886, 1891).
Free
Banking
II
577 equilY, ease economic calculation, destroy the "banking monopoly," and eliminate business depressions.:!0 A series of debates over variants of the NME appeared in a periodical named Liberty.21 The NME approach to monetary questions first appeared in Liberty on April 4, 1891 (p. 3) in a reprint of an article entitled "Evolution in Finance," clipped from the Galveston News.:!:! The piece inquired: "Why is it that the admirers of evolution as regards gold as the standard are not likewise avowed admirers of evolmion as to the means of payment? They must know that when parties are willing to refer to the standard, but when law compels the aClllal presence of the standard, law prevents the action of evolution by free contract." Further on, the article referred to "the tendency of evolution to separate the two functions [of money]. Banking science can and does separate them where the evolution is permitted to occur in the form which it will assume in freedom." Unfortunately, the rest of lhis reprint presents no explanation or justification for this claim, which is of central importance for the NME. 23 The next major development of the legal restrictions theory came with the publication of A Study of the Money Questi071 (1894) by Hugo Hilgram. 2.' He lamented the confusion in monetary theory that had arisen from ignoring the simple proposition: "The unit of value is not necessarily money, nor is money necessarily a denominator of value" (1894, p. 17). While the medium of exchange and the unit of account had evolved together through social processes "initiated in the early NEW MONETARY
~:C:ONOMICS
'It A gcm:ral weakness shared by Whiuick and Westrup is, however, IIml lhey idclllilicd frec banking Wilh lhe aClual separalion of monelary funClions WilhoUl I)ro\·iclillg an ade'lllillc explanalion of how (or whelher) one mighl lead 10 lhe olhcr. This aS~Crli .. n rc.·civcs IIIl1rc allcnlilln laler. "' I.ibclly, puhlishcd from H!tll to HIOH, was edilcd by Uel~jamin Tuckcr. It was dCl'lllcd III lhc discussion of lhe polilical, social, ami economic issues of the day, primaril)' from an "individualisl" point of view (see McElroy 19HI, 19H2). .•• Weslrup (tH9!>, p. 74) asserled Ihal Ihe Gulvestotl News was Ihe mosl pupular and wi.ltly rt'acl daily Ilaper in Texas and had advocaled free banking fllr a num1Jcr of years . • :1 This daim is impurlilnl nol only for ils possible pulicy implicalions bUl also because il pnullises a slliluioll III Ihe nagging quesliun why individuals huld money. If IIIlIney W(I"t III disappear in lhe absence of government intervention, lhen we need nOl auempl III derive ilS continued exislence from individual oplimizing behavior in a pure markel sClling. (The origin of lIloney, however, can slill indeed be explained by an invisible hand IIIcl:hanism; see Menger (1892).) Thus Ihe absence of money from Ihe basic mmlcl of ~eneral e(luilibl'ium may be a virtue of Ihe Iheory ralher lhan a failure. This a"s~lIcc woul
33
34
Free Banking II JOURNAL OF POLITICAL t:CONOMY
stages of human history," Bilgram argued, "this evolution of mOlley has incidenlally led to the adoption of the commodity of which money was lI1ade ,IS a universal value denominator .... This is, however, no excuse for so completely confollnding these two concepts as to lise the sallie wOl"{llo denolllinate both" (p. (9). llilgram identified "money" as simply any asset that th rough tacit or explicit agreement happens to be rendered acceptable "for the purpose of mediating exchanges" (p. 23); he argued that we do not need any quantity restrictions on the supply of money: "If a bill were introduced in congress for regulating the number of locomotives to be made and used in this country, lest an excessive number might be made, tending to increase the traffic beyond the needs of the people, the sanity of the author of this bill would justly be questioned. Yet, we accept without question the decree of congress which limits the amount of money, although money is by far the most important of all tools" (p. 49). In a close parallel, Wallace, after making a similar point with more contemporary examples, poses the questions: If it makes sense to control their [private bank notes and deposits] quantities, why not those of other credit instruments? For example, most economists would not favor a proposal to constrain the dollar volume of mortgages on single family residences to grow at a prescribed rate. Almost certainly, most would say that it is a necessary feature of a wellfunctioning credit system that the number of mortgages be determined in the market and not be set administratively. But if this is right for one set of private credit instruments, why is it not right for all? [1983b, p. 6] llilgram proposed a modified form of freedom in banking, in which the government's role is to mandate that all banks are insured ami to monitor the solvency of each hank (i.e., ensure that the forms of weah h hacking the moncy isslIe arc sOllnd) in order 10 prevent fraud (18!H, pp. 19-(2). Beyond this. money "shollid he regulated excillsively by the law of sllpply and demand" (p. (3).2!". No legal tender laws would be required since all notes could be accepted or refused as the transactor sees fit (p. 31). It is important to note that for llilgram the unit of account was not abstract but rather lOok a real commodity form (pp. 40-41). As will be discussed below. this point led llilgram (along with Tucker) to oppose the abstract unit of account framework. 25 Bilgram eqllivocated on hi~ defen~e of freedom in banking (e.g., 1H!l4. p. ,ljOj, a matter that led to a fierce debate in Lihrrty (.July 28. 1894. p, 5; August 25. 11194. p. 5; Decemher 2!I, 1894, p, 2; April 6, I H95. p. 7),
35
Free Banking II NI::W l\IONI::TAKY ECONOMICS
ll.
579
,hi/ill)' Kitson and flis Critics
The puhlication of Kitson's A Scientific Solution of the Money Question (I HH5) was an important step in the formulation of the "abstract unit or au:oum" branch of early NME thought. Kitson 26 called for the issuance uf media of exchange backed by all forms of wealth (p. 297), whilt: an abstract unit of account (p. 178) would be used for economic Gilculatiun and pricing. 27 Thus the Whittick-Westrup-Kitson approach extended work in the NME tradition by making the unit of acculIlIl purely abstract. Short I)' after the publication of Kitson's book, Bilgram wrote a lcllgthy review of it for Liberty.21i The heart of Bilgram's critique lay in his claim that when "comparing values, or economic (lualllities, the mediating unit must also have value" (Liberty, April 20, 1895, p. 3). Bilgram found Kitson's abstract unit inadequate for use in economic calculation and reaffirmed his advocacy of some form of real comlIIodily unit of
lid",,,.
36
Free Banking II
580
JOURNAL OF POLITICAL ECONOMY
period. In this way, the abstract unit can be derived (rom a scquential process, which ultimately refers back to an original commodity value, a process analogous to von Mises's description of the origin of money (191211 971, esp. PI'. 108-23). It was not lIntil.J. Greevz Fisher that the Westrllp-Whiuick-Kitsoll . system encountered its most penetrating critic. Fisltcr:1O was a frequent correspondent with LibpTty, and Tucker printed many of his lengthy letters on h;ll1king al\(I monetary theory. Althollgh Fisher favored free banking (LibeTty, June 27, 1891, p. 3), he also thought that many other advocates of free banking seriously exaggerated the significance of legal restrictions on financial institutions.:H Fisher was also highly critical of the view that free banking would substantially lower or eliminate the rate of interest (as was claimed by Tucker, among others), for he defended a time preference theory of interest (Liberty, July 28, 1894, p. 8; April 1897, p. 7). In addition, he rejected the Kitsonian schema for the separability of monetary functions as well as Kitson's heavy emphasis on the monetary system as the source of most economic problems. Fisher charged that "separability" proposals were impractkal and would not evolve from a state of laissez-faire. He argued that, through an invisible hand process, one commodity (in this case, gold) emerges as the most salable of all commodities and, hence, becomes money (Liberty, AEril 1897, p. 7). Salability is thus the preeminent feature of money.' 2 Although all commodities are media of exchange in some sense (though their relative transactions velocities may differ), only one commodity (the most salable one) will hecome the denominator of values. Fisher closed the main statement of his argument by asserting that "the standard of value is simply one member of the larger class of exchangeables. All commodities are media of exchange; one commodity is a standard of value" (Libl'rty, August 1897, p. 7). The weak link in Fisher's argument is his claim that the most salable of all commodities must become the denominator of values. Fisher simply asserted that this happens "inevitably.,,:-I:" He providecf no specific arguments for this c1aim. 31 ~o Fi~her was a prominent figllre in the British free-thollght movemcnt (sce Royle !!Ilm, PI" li:i, 22[,). ~I Fisher slilfed that "there is nl) legal ohstade, nothing. in I;",t. whatever ('X('''pt the Inconveniences of bulk, fluctuation of vallie, and other inherent defects. to p.-('vent the introduction and circulation of promises of wheat, cotton, oil. iron. or other COIllmodities" (Lib"ty, June 27, 1891, p. 3). ~2 The evolutionary explanation expounded by Fisher is virtually identical to the Austrian theory of money. See the discussion of the relation of the Austrians and the forerunners above. 3~ Unfortunately, Tucker did not permit Kitson to make any further replies to the criticisms printed in Liberty. See Open Ileview (May 1910, p. 7) for Kitson's explanation of Tucker's hostility. S·' White (1984a, 1985) endeavored to lill this gap by emphasizing the trallSactional
Free Banking II NEW MONETARY ECONOMICS
I n May 1909, Kitson initiated a periodical called the Open Ret/iew Ihat colltinued publication until the end of 1910. 35 In this journal, Kitson and Meulen argued for free banking and the separation of mOIlt:tary functions. (Since most of Meulen's thoughts were later incorporaled into his book, we shall wait to examine them until the next seclion.) Kitson's articles were also quite similar to his other published works. Milch of his work in the Open Review was devoted to his critiqut: of Ihe gold standard. 36 Kitson also penned two brief manifestos, ''The Nt:w School of Finance" (OPen Review, 1909-10, pp. 327-35) alld "Tht: Siandard of Value Fraud" (Open Review, May 1910, pp. 6~), which outline the Kitsonian formula of free banking and the separahility of monetary functions. Whilt: lllauy critics of Kitson wrote letters to the Open Review, the quality of Ihese critiques is almost uniformly poor. Only Henry Seymour (OPen Review, May 1910, pp. 14-15) succeeded in raising cogent poinls against Kitson. Seymour argued that the gold standard, while imperfect, should not be eliminated. Much of the erratic nature of the valut: of gold is caused by the policies of the Bank of England, not the gold standard itself. Seymour claimed that laissez-faire in banking undergirded by the gold standard is the most stable system possible. Furtht:rmore, if people are seriously concerned about inHation or ddlalion, contract indexation is always possible. Meulen wrote a brief reply to Seymour's letter. While conceding some of Seymour's points, Meulen claimed that an abstra~t unit of account is nonetheless
illnJlI\'Cllielll:es uSSIKialed wilh a distinct price for cuch of thc mcdia of CXdlilll!;C. Scc al", III, C,,1I11111 (I !III5) uilli O'Uriscull (I !lH5). :I~. The 01"111 Utview was a forum for the discussion of colllemporary political and economic issues: land reform, foreign affairs, and, certainly, bunking reform. At first, the Opm Review was almost completely the work of Kitson. As publication cominued, articles and lellers from other writers were printed with increasing frequency. Among the contrihutors were Meulen (who lOok over as acting editor when Kitson lOured India [Olml/levifw, I!lO!l-I(), p. H!ll), Ilenry Seymour, Egmont Hake (two other ligures active in lJanking reiilrm), and the literary critic G. K. Chesterton. (The first two volumes of the Open Review have been collected in a book [Open Review, J909-J9JUl, which does not dislinguish between different issues and offers inconsistent pagination.) :11; Kitsoll's critique of the gold standard is well summarized by the following excerpt from Ihe Banking and Currency Reform League statement of principles (1910, p. 4): "We have lied down our entire exchange system to a ridiculously inadequate gold busis. Hence, whenever the proportion of gold in the Bank of England reserves becomes slllall in relation to the superstructure of credit (which may occur simply because gold has been exported in the form of foreign investments, or because production and the volume oJ' credit have increased at home), we raise the Bank Rate in order to cause the gold 10 relurn, that is, we make credit and exchange medium dearer, thereby penalisin!; our producing classes, bankrupting great numbers of our manufacturers and causilll{ Iluaillilies of g(HHls 1Il be thrown upon Ihe market al ruinous prit"es in thc CIIdeavour 10 oblain exchange medium."
37
Free Banking II
38
JOURNAL OF POLITICAL ECONOMY
superior to a laissez-faire gold standard because only theforlller can perfectly register changes in the value of gold. 1II addition to editing the 0t)("1 Review, Kitson founded the Bankillg and Currency Reform League in December 1909 with Meulen as league secretary (Open Review, May 1910, p. 5). As the cover of the league's statement of principles (1910) announced, the league devoted itself to "the repeal of the Bank Charter Act and such laws as now interfere with freedom of banking." This unsigned statement is an excellent summary of the Kitson-Meulen position. The phrase "legal restrictions" is used as a means of describing the forces that have led Great Britain to its present monetary institutions and away from the free banking/separation of monetary functions alternative. For example, the statement argued that "we might long since have outgrown the necessity for the use of gold itself as an exchange medium, had not our legal tender and banking laws prevented its substitutjon in the channels of exchange" (p. 4). The central concerns of the Banking and Currency Reform League were the evolution of money and the consequences of legal restrictions for this evolutionary process. Despite the brief length of the statement of principles, it is rich with historical examples and evidence since the statement provi(les a brief sketch of British banking history as interpreted within an NME framework. Like the Ot)en Review, the Banking and Currency Reform League never quite flourished. The work of Kitson and Meulen was probably far too abstruse to appeal to a popular audience. Nonetheless. it was activities of this nature that led Frederick Soddy (1931) to duh Kitson "the doyen of British banking reform.',:n C.
lIenry Meulen: The Last of a Tradition
Henry Meulen is the most sophisticated and most interesting early contributor to the theory of legal restrictions. H is views on money and banking Gill be found in his study Free lJanliing (1934), which remains the only comprehensive book-length statement of NME.:~R Free lJallk~7 Kitson continued to argue for banking reform until his death in 1937, although he deVeloped his original set of ideas only up until the First World War. His position then moved much closer to that of Major Douglas's "social credit" scheme (see Fisher 1920, pp. 123-25). After the outbreak of the war, Kitson no longer felt that free banking was a viable solution to financial problems and then advocated complete nationalization of the banking system. (See the preface of Kitson [1917J for his reasons for altering his views.) We are examining here only Kitson's prewar thought. 3R The 1934 book is a revised and expanded version of his shorter 1917 monograph /ndrLftriai JrLfliu through Banking R,fonn. Of all the writers we have examined, Meulen placed the greatest emphasis on the idea of "legal restrictions" being the central focus of his monetary theory: he even lIsed the phrase repeatedly.
Free Banking II NEW MONETARY ECONOMICS
illg contains a lengthy exposition of Meulen's theory of the evolution
of financial institutions under laissez-faire as well as several chapters on the history of money and banking. The cornerstone of Meulen's thought is that the separation of monetary functions and the disappearance of outside money are the natural outcome of laissez-faire: "The credit system described in the previous chapter was an a priori one, deduced from a knowledge of the needs of commerce. I will now endeavour to demonstrate inductively that the ideal system described represents the fulfillment of the historical strivings of men towards improvement of the exchange system, and that the inequity of our present social relationships is for the most part due to the deHection of the previous course of progress by unwarralltable State interference" (Meulen 1934, p. 66). Before examining Meulen's explanation of the evolution of such a system, we shall flrst look at his conception of the ideal end state. Chapter 12, "An Invariable Unit of Value," contains Meulen's description of how a system characterized by the separation of monetary functions might operate. Meulen begins by noting a problem with the gold standard: "The obvious defect of the use of a valuable metal to lIIeasure values is the very fact that the metal is itself the ol~ject of hUlllan desire-a notoriously fluctuating force-and that it may, lIIoreover, be artificially monopolized. It is precisely as though our yard-stick were made of gutta-percha and could sometimes be stretched to 40 inches in length, and at others reduced to 30 inches. No llIeasure should vary in that quality which it is designed to measure" (pp. 237 -3H). Most of the previolls plans designed to deal with this prohlelll involved some form of a tabular standard (p. 23H), but like (;reenheld and Yeager (1983, p. 3(9), Meulen explicitly distinguished his proposal from any form of indexation. He claimed that his systelll, unlike an index number scheme, "establish[es] a simple unit of value ill which all commodities may be priced, and which shall relllain unaffected by the price fluctuations of anyone commodity whatsoever" (1934, p. 239). Instead of being a claim to a certain weight of gold, ballk notes would be a claim for their "face worth" of gold according to the market conditions of the metal. Under the gold standard, notes are claims to a varying worth of gold and, hence, a varying worth of commodities. I\Ieulen developed an argument similar to Kitson's that the "bank note pound" can become an invariable unit of account simply by defining its price as equal to one (p. 240). The price of gold in terms of hank note pounds would be permitted to fluctuate freely ill accord with supply and demand for the precious metal. All relative prices would he expressed in terms of the standard unit of account (pp. 2'10--11). Thus the unit of account hears a relatiollship to gold no
39
40
Free Banking II
584
JOURNAL OF POLITICAL ECONOMY
different from its relationship to other commodities. Changes in the supply of and demand for gold do not affect the general price level; they only change the price of gold expressed in terms of rhe IIllil of account (p. 240). Media of exchange would consist of paper creclit claims issued "elastically" by the private sector (see chap. <1 011 media of exchange). Meulen claimed several advantages for his proposed reform. First. it would stabilize the value of contracts and other intertemporal transactions (p. 238). Second, banks would be more secure since they co\lld "issue credit on a much wider range of security than is the case today" (p. 277). Third, increases in the supply of and demand for gold would not inAuence real economic activity (p. 242), and internaliOlwl gold drains would be no problem for either bank security or general economic activity (p. 277). As we have already indicated, whether the separability of monetary functions would ever actually evolve under laissez-faire is a critical <Juestion for the NME. Meulen's work is particularly interesting hecause he also presented an account of how such a system might evolve through market processes. Chapters entitled "The History of Exchange" and "A Review of Scottish Banking" contain Meulen's theory of the evolution of financial institutions. In these chapters, Mellien interwove a general account of the evolution of money with examples from-monetary history. He carefully examined Scottish banking history to 1845, a period marked by a considerable degree of laissez-faire with regard to financial institutions (see White 1984b). Crucial to Meulen's position is the proposition that a system of free banking provides a continual incentive to economize on the use of gold. 39 This can be achieved through the enhanced reputation of the bank for security,10 the use of the option c1ause,'11 and improvements in monitoring the financial soundness of banks. (See Meulen [1934, p. 81] for 39 It was well known in the "Banking School" literature that this was an important feature of the Scottish free banking system (see Meulen 1934. c1mp. 8). See also White (1984b, PI'. 43-44). who noted that Scottish reserve ratios dropped from 10-20 pcrcent in the second half of the eighteenth century to between 0.5 percent and 3.2 percent in the first half of the nineteenth century. ·10 Klein (1!174) and lIayek (1978) have underscored the importmlCc of reputation and quality differentiation in private monetary competition. See Brown (1982) for a summary of these arguments and further references. 11 The "option clause" refers to a contractual agreement betwecn b'lJlk alld depositor that notes may not be redeemable on demand in the case of a declared "emergency"; however, the note would hear interest for the dural.ion of the suspension. See Cameron (1967. p. 68 ff.) for a more detailed exposition. Meulen (1934. p. 12~1) claimed that Scottish banks were so secure that option clause notes circulated at par before the option clause was outlawed in 1765. We will discuss later the importance of the option clause to the evolution of monetary institutions. See Cowen and Kroszner (1987) for further analysis of the significance of the option clause prohibition.
Free Banking II NEW MONETARY ECONOMICS
further discussion.) Meulen also noted the imernational ramifications of such a system (p. 324): Those countries with free banking will tend to lose their gold to countries without free banking because gold is granled a "monopolistic" status in the latter coumries. Since Iht! privale sector demonstrates a continued tendency to economize on the use of gold, the result of free evolution would be Ihat media of exchange would be backed only by the reputation of the issuer: 1:.1 "In an advanced state of society, increasing use is made of contract and promise in all human relationships .... The process is one of gradual specialization and division of labor. Accordingly, with increased civilization, the transfer of intrinsically precious money is replaced by the transfer of documentary promises .... The fulfilllIlem of these conditions necessitates the use of a money which is imrinsically worthless" (p. 63). Thus confidence in the system will grow to the point at which bank notes would not be backed by any gold at all (p. 309). Competition and the desire to attract new customers and keep old ones provide strong incentives for banks to enhance their reputations. As a bank's reputation improves, it will rely less and less on gold reserves. Another important factor is the real resource savings resulting from the elimination of gold backing for currency (p. 293).'13 Meulen asserted that this trend was present in eighteenth-century Scotland: "The early Scottish bankers introduced the note with the option clause .... In accepting these notes at par, producers showed that they did not require gold in order to effect exchanges .... The bank then ceased to be regarded as a storehouse whence gold could be withdrawn for export abroad, and the banker increasingly assllllled the function of integrity valuer, pure and simply, the demand for gold being directed to the professional goldsmith" (pp. 82-83). Thus Scotland was "following the path of progress to the theoretical ideal which I have previously outlined" (p. 82). This trend was arrested, however, when the option clause was outlawed in 171i!1 (p. 88). He also cited the influence of English banking regulations on the evolillion of Ihe Scottish system since an important part of Scottish finan'cial interlllediation involved de.ding with the English system,
I:! T"ou~h small (h.lllge mighl slill be reprt:senled by cheap metal tokens (Meulen I\J:H, 1" Ii~l). 1:1 Similar argulllents on how free banking leads to the economilatiun uf specic can be IOllml ill Ad.ulI Smith (1776/HI37. pp. 276-!H). which alsu Icd Smith to praise Scotbud's ballkillg system, although he (pp. 30Y-!O) was much less sanguine abollt the dlclts of thc option dause lhan I\leulen. On "r.·ec" banking in Scotland. see Whitc ( 1!lIHb) ;111<1 COWCII alld Kroszner (1987). IlicreasiliK lruSl all" cOlilidcnce ill mOlletary nl hall~c is a thclile also clllph'lsiLcd by Sillllllci (HI7H) alld thc Austrialls (Ellis I!I:H; Fa alll.d 1~177; Llidler .lIId J{owc I!JHU).
41
Free Banking II
42
586
JOURNAL OF POI.ITICAL ECONOMY
thereby presumably inhibiting the Scottish system frolll evolving along strict laissez-faire lines (p. 324). According 1.0 Meulen, the emergence of a distinct and separate lInit of accollnt also follows from the econolllization of !-{old (p. R~). As Kold is used less for monetary purposes, its nonmonetary uses have a progressively greater influence over its value. As the valuc of gold becomcs increasingly determined by its nonmonetary uscs, it tends to hecome no more efficient than any other commodity as a unit of account or standard of value.'H "Gold thus tend[s] to fluctuate in price, and consequently to become unfit to serve as a standard of value. This very fact, however, as will be shown later, stimulated human ingenuity to the discovery of a better [standard] of value" (p. 83). Meulen's vision of this process paralleled Kitson's (cited above), in which the value of gold on the previous market day becomes the abstract unit of account and the price of gold, as well as the price of all other commodities, floats in terms of this abstract unit on subsequent days. Meulen went beyond Kitson, however, by attempting to explain how this separation could occur through market evolution. Further along in the exposition (pp. 291-93), Meulen stated that an abstract unit of account will be used under laissez-faire because it is more efficient to guarantee holders of notes access to a certain worth of gold rather than a certain weight of gold. White (1984a) has argued that the separation of monetary functions would involve considerable inconveniences to consumers because the value (in terms of the numeraire) of exchange media would no longer be constant. But Meulen believed that banks would be willing to bear the risk of fluctuations in such values in order to reap the gains from dispensing with their gold reserves (1934, p. 293).'15 Concluding Comments Early NM1~ theorists (particularly Meulen) were well on their way to deveiopinK a coherent alternate framework fi»' analyzing many of the central isslles of monetary theory. In addition, these early thinkers offered interesting original contrihutions in at least four areas: (I) the Ina
Free Banking II NUV MONt:TAKY t:CONOMICS
of IlHHl~lary functions, (3) the likelihood of such a separation evolving IInder laiss~z-faire, and (4) the nature of the unit of account (abstract, as opposed to embodiment in a commodity or bundle of commodities). The inability to provide a convincing explanation of the transition to a syst~m wilh a unit of account separate and distinct from the m~ans of payment is a major weakness of the theoretical apparatus de\'dop~d by the early new monetary economists. Meulen had argued thal the progressive demonetization of gold would lead to its abandonment as an accounting unit, but this claim is of questionable validity. Only if the nonmonetary demand for gold were highly unstable need this r~sull f()llow. Ther~ are several possible paths that the economy could follow in the evolution toward a cashless payments system, some of which are discussed in more detail in Cowen and Kroszner (1986). For example, if eqllity shares port folio-ciominate dollar-denominated <:hecking ami savillgs deposits, it is possihle fiu' the dollar to remain as all ahstract IInit of account while disappearing completely as a means of payment. In this case, unlike the usual BFH approach, the abstract unit scenario would require 110 discrete change in the economy's unit of accoulll. An alternative possibility involves the government or the private bank clearinghouse system engineering a deliberate switch of the accounting IInit. In this scenario, the new unit of account could be abstract or a "real" commodity or commodity basket. Investigations into such issu~s as the NME raises. however. lay virtually dormant until Fischer Hlack's (I Y70) article eventually revived interest in this line of inquiry. Hi' References Anderson, UenjHmin M.• Jr. Tile Valll.e of Money. New York: Macmillan. 1917. Banking and Currency Reform League. A Stalemenl of Principles. London: Lelchwirth. 1910. Baxter. W. T. "Accounting in Colonial America." In Sludies in lhe Hislory of r\cc01l11ling. edited by A. C. Littleton and Basil S. Yamey. London: Sweet & Maxwell, 1956. l\ilgrmll. Hugo. A Study of the Money Queslion. New York: Humboldt. 1894. Bilgr;lIll, Hugo. and Levy. Louis E. Tile Causes of Business Depressions as Disdu.I'j,e/ b.'Illll AII(lIJ~is of lile Basic Principles of Econ01l1ics. Philadelphia: LippinCOil. H1l4. ,,, Wc h'I\'c found no systematic allempts to develop a legal restrictions approach hctwccil l\Iculen (HI34) and Black (1970). Benjamin M. Anderson (1917), however, ducs havc •• Hinitics to the NME (see Salerno 1985). In the interim period, one can find urcasional investigations that touch on NME themes. See the references provided ill Hla('(, (1\170, pp. 9-11) for a list of relevant, ailleit fragmentary, commenls by Tobin, Virlrc)', 011111 olhers.
43
Free Banking II
44
588
JOURNAL OF POLITICAL ECONOMY
Black, Fischer. "Banking and Interest Rates in a World without Money: The Effects of Uncontrolled Banking." J. Bank Res. I (Autumn 1970): 9-20. Blaug, Mark. Economic Theory jn Retrospect. 3d ed. Camhridge: Cambridge Univ. Press, 1978. Brown, Pamela. "Constitution or Competition? Alternative Views on Monetary Reforms." Literature of Liberty 5 (Autumn 1982): 7-52. Brunner, Karl, and Meltzer, Allan H. "The Uses of Money: Money in the Theory of an Exchange Economy." A.E.R. 61 (December 1!J71): 7H4-805. Calomiris, Charles, and Cone, Kenneth. "A Note on Competitive Payment Systems." Manuscript. Evanston, III.: Northwestern Univ., 1984. Cameron, Rondo E. Banking in lhe Early Slages of InduslrialiUllion: A Silldy ill Comparative Economic History. New York: Oxford Univ. Press, 1967. Carey, Henry C. "A Memoir of Stephen Colwell." In Miscellaneous Work.!. Philadelphia: Baird, 1872. Clark, Truman A. "Violation of the Gold Points, 1890-1 !J08." ].P.E. 92 (Oc· tober 1984): 791-823. Colwell, Stephen A. The Ways and Means of Paymtml. Philadelphia: Lippincott. 1859. Cowen, Tyler, and Kroszner, Randall. "The Evolution of a Cashless Payments System." Manuscript. Cambridge, Mass.: Harvard Univ., 1986. - - - . "Legal Restrictions on Financial Intermediation during Scotland's 'Free' Banking Era." Manuscript. Cambridge, Mass.: Harvard Univ., 1!l87. Dorfman, Joseph. The Economic Mind i71 American CiviliUllion. Vol. 3. 18651918. New York: Viking, 1949. Einaudi, Luigi. "The Theory of Imaginary Money from Charlemagne to the French Revolution."·)n Enterprife and .'inltiar C/UI1lI{t': Rrr/fliIlK-f ill Economic History, edited hy Frederic C. Lane. Homewood, III.: Irwill (for American fcon. Assoc.), 1953. Einzig, Paul. Primitive Money in lis Elhnological. Hiflorical awl Economic Aspects. 2d ed. New York: Pergamon, 1966. Ellis, Howard S. German Monetary Theory, 1905-1933. Cambridge, Mass.: Harvard Univ. Press, 1934. Fama, Eugene F. "Banking in the Theory of Finance."]' Monelary EWII. Ij (January 1980): 39-57. Finlay, John L. Social Credit: The English Origins. Montreal: McGiII-Queen's Univ. Press, 1972. Fisher, Irving. The Purchasing Power of Money: lis Vt't""lIIillatioll alld Relalioll III Credit, Intnest and Crises. New York: Macmillan, 1911. - - - . Stabilizing the Vollar: A Plan 10 Stabilize lhe Gm,.,.al Pricr l.rTlelwilhollt Fixing Individual Prices. New York: Macmillan, 1920. - - - . Stabl, Money: A History of th, Movt'ment. New York: Adelphi. 1934. Frankel, S. Herbert. Money: Two Pllilosophies. Oxford: Hlackwell, 1977.
Friedman, Milton. "Financial Futures Markets and Tahular Stamlanls." ].P.E. 92 (February 1984): 165-67. Galiani, Ferdinando. Vella Moneta. Naples: Raimondi. 17!i0. English eel.: On Money. Translated by Peter R. Toscano. Ann Arbor. Mich.: Univ. ~ficro films, 1977. Gray, John. Lectures on th, Nature and Use of Money. Edinburgh: Hlack, 184R. Greenfield, Robert L., and Yeager, Leland B. "A Laissez-Faire Approach to Monetary Stability."]' Manry, Credit and Banking 15 (August 1983): 302-15. Hall, Robert E. "Monetary Trrnds in the Uniltd Stales a"d Ih" U"i(rd Ki/lgdom: A Review from the Perspective of New Developments in Monetary Economics."]' Econ. Lil,rature 20 (December 1982): I 552-5£j.
45
Free Banking II NEW MONETARY ECONOMICS
589
- - . "Optimal Fiduciary Monelary Syslems." J. Mlmetary Econ. 12 (july IUH3): 33-50. Hayek, Friedrich A. von. Denatio7Ullisati071 of Money: A71 Analysis of the Theory Illld l'mL"tice uf Cuncurrent Currencies. 2d ed. London: Insl. Econ. Affairs, HI7H.
Jevuns, William Stanley. Money aud the Mechanism of Exchange. New York: ApplelOn, 1876 . .Jl)lIe~, Ruhen A. "The Origin and Development of Media of Exchange." .J.1'.t:. IH, no. -t, pt. 1 (Augusl 1976): 757-75. Karcken, Juhn H., and Wallace, Neil, eds. Models of Monetary Economics. MinlIeilpolis: Fed. Reserve Bank Minneapolis, 1980. KilSOll, Anhur .•i Scientific Solutio" of the MOlley Question. BoslOn: Arena, Hi~15.
- -...I Fl'Cllldlllelit Stcmdard: An Exposure of the Fraudult71t Chq.racter of Our
;\lullelul'Y S/cllldurd. London: King, 1917.
llel~jalllin. '"The Competitive Supply of Money." J. Money, Credit alld Bllllkillg 6 (Novemher 1974): 423-53. Klmpp, Georg F. The State Theol, of Money. London: Macmillan (for Royal
Klein,
hon. SOc.), I U24. Laidler, D'lVid, and Rowe, Nicholas. "Georg Simmel's Philosophy of Money: A Review Anide for Economists." J. Econ. Literature 18 (March 1980): 971Il5.
Lalle, Frederic C., and Mueller, Reinhold C. Mouey and Banking iii Medieval /llid Rellai.HallCe Venice. Vol. 1. Coins ""d Moneys uf Account. B"ltimore: Johns Hopkins Univ. Press, 1985. J.ibel'ty. Edited by llel~jamin Tucker. Weslport, Conn.: Greenwood Reprint, IU70.
McCallum, Bennen T. "The Role of Overlapping-Generations Models in Monetary Economics." Carnegie-Rochester COIif. Ser. Public Policy 18 (Spring 1983): 9-44. - - . "Bank Deregulalion, Accounting Syslems of Exchange, and the Unil of Accoulll: A Critical Review." Working Paper no. 1572. Cambridge, Mass.: N.ll.E.R., March 1985. McElroy, Wendy. "Be~iamin Tucker, Individualism and Liberty." Literature of I.iba/y" (Aulllllln I!IMI): 7-39. - - . I.ibnty, IH81-19(JlJ: A Comprehellsil/e Jlldex. St. l'aul, Minll.: Coughlin, 19M2. Menger, Karl. Gl'lmtisiitze der Volkwirtschaftlehre. Vienna: Braum(iller, 1871. English ed.: Pri1ICipies ofEconornics. New York: New York Univ.l'ress, 1981. - - . "On the Origin of Money." £Con. J. 2 (june 1892): 239-55. Meulen, Henry. Free Banking: An Outline of a Policy of Individualism. London: Macmillan, 1934. Mill, John Stuarl. Principles of Political Ecotlomy. with Some of Their Applications to SocialPllilosophy. 7th ed. London: Longmans, Green, Reader, and Dyer. 1871. Reprint. London: Longmans. Green, 1929. Mints, Lloyd W. A History of Banking Theory in Great Britain and the United States. Chicago: Univ. Chicago Press, 1945. Momesquieu, Baron de [Charles de Secondal]. Tire Spirit of Laws. 2 vols. Rev. ed. New York: Colonial. 1899. Niehans, Jurg. Tire 7'lteury of Money. Baltimore: Johns Hopkins Univ. Press, 197K. O'Driscoll, Gerald ('., Jr. "Money in a Deregulaled Financial System." Fed. lleJI!T1Je Ilunk /Jallus Econ. Rev. (May 1985), pp. 1-12.
46
Free Banking II
590
JOURNAL OF POLITICAL ECONOMY
Open Review, edited by Arthur Kitson. London: Palmer, Red Lion Court, 1909-10. Patinkin, Don. Money, Interest, and Prices: An Integration of Mlmelmy and Value Theory. 2d ed. New York: Harper and Row, 1965. Proudhon, Pierre-Joseph. ProudllOn's Solution of tile Social Problem. New York: Vanguard, Hl27. Rohey, Ralph W. pflrclwsinl{ Pllwp-r: An Intrlltiltction to Qlwli/ll/ivI' Cr,.t!il Glmlrol IlfUI't/ /III tI,I' "I1'I''''·;I'.f IIf SII'/IIII'" A. Golwell. New York: rrcllIice-Jl;aIl. 1938. Roylc, I·:dwm·d. Ilflt/;wlf, SI'f:lllflriJ/.f, flllIllll'/lIIbiirmu: /'''I"tf",· ,.,.,.1' 'J'h,,,,/{Irt ;/1 LJritain. 1866-1915. M;anchcster: M:lllchellter Univ. Press. I!IMII. Salerno. Joseph. "The Monetary Economics or Beryalllin M. Anderson." Manuscript. Newark. N.J.: Rutgers Univ .• 1985. Sargent. Thomas J.. and Wallace, Neil. "The Real-Bills Doctrine versus Iht' Quantity Theory: A Reconsideration."J.P.E. 90 (December 1982): 1212-3(j. Schumpeter. Joseph A. History of Economic Analysis. New York: Oxrord Univ. Press, 1954. Simmel, Georg. Philosophie des Geldes. 2d ed. Leipzig: Duncker & HUll1hlot, 1907. English ed.: The Philosophy of Money. Translated by Tom Bouoll1ore and David Frisby. London: Routledge & Kegan Paul. 1978. Smith. Adam. An Inquiry into the Nature and Causes of the Wenlth of Nntiolls. 1776. Reprint. Edited by Edwin Cannan. New York: Modern Library. 1937. Smith, Vera C. The Rationale of Central Banking. London: King. 1936. Soddy, Frederick. The Role of Money: What It Should BI', Cllllirastl'd fIIillr Wlrnlll Has Become. l.ondon: Routledge, 1934. Steuart-Denham, Sir James. Principles of Banks and Bmlkill/{ of MOII,.y. London: Davis, 1810. Viner, Jacob. Studies in the Theory of Internalional Trade. New York: Jlarper, 1937. von Mises, Ludwig. Theorie des Geldes und der Unlnrtfslllittlr.. Munich: Ihll1cker & Humblot, 1912. English ed.: The Theory of Mo"I'.~ fIIld Grl'dit. Irving,oll' Hudson, N.Y.: Found. Econ. Education, 1971. Wallace, Neil. "A Comment on McCallum." Carn(!gi(!-Ror:llt~sl"r Cflllj: Sf'r. Public Policy 18 (Spring 1983): 51-56. (a) - - - . "A Legal Restrictions Theory of the Demand ror 'Mone)" and Ihe Role or Monetary Policy." Fed. Reserve Bank Miml/!OIJOli,f Q. Rm 7 (Winler 1983): 1-7. (b) Westrup, Alrred B. The Financial Problem: Its Relation 10 Labor Rrfimn mId Prosperity. Dallas: privately published, 1886. - - - . Citizens' Money: A Critical Analysis in Ihe Light IIf Ft'tt! Tmdl' ill Ilnllkirrg. Chicago: Mutual Bank Propaganda, 1891. - - - . The New Philosophy of Money. Minneapolis: Leonard, 1895. While, Lawrence H. "Competilive Payments Systems and the lJnil of Ac· count." A.KIl. 74 (Seplcmher 19144): 699-712. (n) - - - . Free Bn.tlkitlg in LJrilrtin: Themy, EX/Jerimce, {/ml /)I'bfllr. 1800-1845. Cambridge: Cambridge Univ. Press, 1984. (b) - - - . "On Transactions Costs in Cashless Payments Systems." Manuscript. New York: New York Univ., 1985. . Whittick, William A. Value and an Invariable Unit of Value: An Imporlallt Discovery in Economics. Philadelphia: Lippincott, 1896. Yeager, Leland B. "Stable Money and Free-Market Currencies." Cata J. 3 (Spring 1983): 305-26. - - - . "Comment on Cowen and Kroszner." Manuscript. Auburn, Ala.: Auburn Univ., 1985.
[3] NOTES, COMMENTS, REPLIES
The Forerunners of "New Monetary Economics" Proposals to Stabilize the Unit of Account A Note by Scott Sumner A recent paper by Cowen and Kroszner (1987) traced the historical roots of the "new monetary economics." This approach has two basic tenets. First, our current monetary system is a creation of government regulation. Second, an ideal monetary system would separate the function of unit of account from the medium of exchange. Greenfield and Yeager (1983) propose(j a monetary system in which the government's only role would be to define the unit of account in terms of a comprehensive basket of commodities. They named this regime a "BFH system" after three important theorists: Black (1970), Fama (1980), and Hall (19M3). A related approach, described by Wallace (1910) as the "legal restrictions theory of money," emphasizes the importance of legal barriers against the private issue of interest-bearing currency in creating the demand for non-interestbearing fiat money. Because the term "new monetary economics" covers several distinct schools of thought, this paper will focus solely on the forerunners of BFH system. In the early works of Arthur Kitson and Henry Meulen, Cowen and Kroszner (henceforth C-K) claim to have found antecedents of the modern BFII proposals for aeating an abstract unit of account separated from the media of exchange. This paper will demonstrate that although Kitson and Meulen did show some affinities to the new monetary economics approach (particularly the legal restrictions theory), the macroeconomic implications of their policy proposals fell far ~hort of the modern BFH system. In fact, it is not clear they had any coherent policy at all. Contrary to the assertions of C- K, there is no convincing evidence that Kitson and Meulen proposed a system where the unit of account would be separate from the media of exchange. C-K also noted that "the forerunners, more akin to the BFH theorists, stressed the gains to be secured by providing a stable unit of account and the diminished macroeconomic influence of changes in the supply and demand for exchange media" (p. 574). It does appear that the avoidance of deflation was the SCOTT SUMN~({
i,,' associale proJessor oj economics. Bendey College.
journal oj Alolley. Credil. and Banking, Vol. 22, No, 1 (February 1990) Copyright © 1990 by the Ohio State University Press
48
Free Banking II 110
:
MONEY. CREDIT. AND BANK ING
primary motivation (lfthe RFH forerunners. It is not at all clear. however. that t he system proposed by Kitson and Meulen would have provided a sta hie ulli t oi" account. This paper will also demonstrate that the "compensated dollar" plans oi" \VilIiams (IK92) and Fisher (1920) i"eaturcd a unit oi" account which corresp(lnds much more closely to the ideas of modern BFH theorists than docs the KitsonMeulen system. In fact, proposals by Black (1981), Hall (1982). and Greenfield and Yeager (1983) share many of the strengths and weaknesscs of thc WilliamsFisher approach.
J. Henry Meulell C- K call Henry Meulen the "the most interesting and most sophist icatcd contributor to the theory of legal restrictions," and term his treatise Free Ranking (1934) "the only comprehensive book-length statement of N ME." ( 19K7. p. 5X2). Meulen's work does contain a proposal for the denationalization of money. But the unit of account discussed in Meulen's book is fundamentally difi"erent frol11 the sort of unit envisioned by modern RFH theorists. And it is not clear that Meulen's proposal would actua By separate the unit of account frolll the media of exchange. Meulen is critical of the gold standard because of the instahility created hy fluctuations in the value of gold. C- K note that "Meulen cla imed that an a hst ract unit of account is nonetheless superior to a laissez-faire gold standard hccanse only the former can perfectly register changes in the value of gold" (1987. pro 581-82). This suggests that Meulen expected his system to produce sOllle sort of price stability since nominal prices only register real values when the price level is constant. In discussing Meulen's system C-K also observed that Instead of being a claim to a certain weight of gold, bank notes would be H claim for their "face worth" of gold according to the market conditions of the metal. Under the gold standard, notes are claims to a varying worth of gold and. hence. {/ varying worth of commodities. (p. 583, italics added)
This also creates the impression that under the Meulen proposal bank notes would have a stable purchasing power. However, this is clea rly not what Mculen means by "face worth." In fi'ee Banking Meulen defines the unit of account: "lIence, since no known commodity continuously evokes the same intensity of feeling, the monetary unit must not be a commodity . . . but a claim upon a certain satisfaction . . . Accordingly we select the feeling for a certain standard commodity, and call it one unit" (1934, p. 250). Later Meulen states: "Whether we build up our scale of prices by calling the price of our standard commodity I bank-note pound, unit. macute, fecling, or spasm, the result is the same: the unit of price, bcing a simple unit of fecling, remains absolutely invariable" (p. 273). Meulcn's unit of account
49
Free Banking J] NOTES, COMMENTS, REPLIES
:
III
(which now seems similar to utility) is not defined clearly enough to make any judgments regarding its stability. One might argue that Meulen intended the unit to have a fixed purchasing power and that the clumsy wording merely reflected the terminology of the day. In numerous other passages, however, Meulen makes it clear that price level stability was not his intention: Value is to-day conceived of in terms of only the contemporary feelings in respect of a standard commodity, and, since every known commodity fluctuates in human estimation, our "standard" is no standard at all. Under the invariable price unit system, the stability of the value unit would be unaffected by any excessive issue of paper exchange medium in terms of the unit. Commodity prices would rise on account of the excessive issue of purchasing power as compared with the production of commodities; but the unit would remain the expression of the normal desire and sacrifice feelings in respect of gold [sic] on a certain day in the past. The purchasing power of a given weight of gold at the time of the introduction of the unit system was, for example, I; after the inflation of general prices, the purchasing power of the same weight of gold would be, perhaps, 3. This would indicale that the normal desire for gold, expressed in purchasing power, had increased threefold since the time of the establishment of the unit system. All other prices would similarly have risen, indicating simply that the supply of comm:>dities had not kept pace with the demand. It may be objected that in this case human desire for commodities had not risen, that it was merely the volume of exchange medium which had been inflated. I reply that the volume of exchange medium would never had been introduced into the channels of exchange if the desire for commodities had not suddenly increased. Somewhere there must have been an individual, or individuals who were desirous of consuming inordinately-whose desire to consume exceeded considerably their contemporary power of production. (p. 258)
The proposition that the price level could triple with no diminution in the utility value of the unit of account, although theoretically defensible, would seem somewhat implausible. And the statement regarding the purchasing power of gold increasing threefold seems inconsistent with the statement that "all other prices would similarly have risen." There is further evidence that Meulen did not intend a policy of price level stability in the footnote on page 262 where he provides a lengthy critique to Fisher's compensated dollar plan to stabilize the price level. Surprisingly, his critique is not based on the inability of Fisher's plan to stabilize the price level, but rather on the fact that, if successful, it would amount to a countercyclical monetary policy! If Meulen had a coherent plan to stabilize the "worth" of the invariable unit of account, it would be interesting to see how the plan was to be implemented. His explanation was perhaps the most disappointing part of his proposal: All the present denominations of coins could be retained, the public could still reckon its purchases in pounds, shilling, and pence, the only difference being that gold coins would be gradually withdrawn from circulation, their place being taken by the pound and ten-shilling notes, and that the pound note would no longer indicate a certain weight, but a certain wurth of gold at the market price of the metal, whatever that price might be. (p. 291)
50
Free Banking II 112
.
MONEY. CRFIlIT. ANIlIIANK1N
In this statement Meulen appears to be suggesting that the ahility of the pound note to buy a certain "worth" of gold distinguishes this regime from other monetary systems. But this criterion could be met by any monetary system: even throughout the hyperinflation of 1923 one (paper) German mark could he exchanged for gold "at the market price of the metal." Thus the "hacking" Melllen seems to provide for the new pound notes would have no operational significance. What then is to prevent overissue of banknotes and a depreciation in the unit of account? Meulen's discussion of the French paper "Assignats" is hardly reassuring: The notes thcmselves depreciated ill worth and caused prices of commoditics to bc artificially enhanced only when the quantity issued began to exceed all reason . . . k't it he notcd that the Assignats were not issued as a true credit medium should he. namely. in advances to capable producers. but were issucd to pay government dehts. thus necessitating no balancing production of commodities. (p. 276)
The proposition that the French inflation could have been avoided had Assignats been issued only to "capable producers" seems very dubious. While an application of the "real bills doctrine" need not lead to price level indeterminacy when the unit of account is linked to a commodity. neither the French system. nor Meulen's proposal would have done so. 1n contrast, Fama (1980), Black ( 1981), Hall (1982), and Greenfield and Yeager (1983) all proposed systems whereby price level indeterminacy would be avoided either by linking the unit of account to one or more commodities, or through control of the stock of fiat currency. C-K conclude their discussion of Meulen by asserting that his system would produce exchange media which fluctuated in value in terms of the "abstract" numeraire. However, as the previously cited quotation indicates (p. 291), Meulen intended prices to be quoted in pounds, and pound notes to continue to circulate (albeit privately issued). The evidence C-K cite from page 293 (p. 586, C-K) refers to the risk of the value of notes fluctuating in terms of a given weight of gold. But the numeraire was defined in terms of a given worth. not weight. of gold. Part of the confusion is due to the use of the word "abstract" when describing the unit of account. To an economist writing at the turn of the century, any unit of account which was not linked to a fixed weight of gold (or silver) would appear abstract. C-K also claim that Meulen believed "confidence in the system will grow to the point at which bank notes would not be backed by any gold at alI" (1987. p. 585). Meulen would probably agree with C-K's charactcrization of his vicws, although a close inspection of the text leaves one unclear as to exactly what is meant by "backed." Meulen states that eventually notes would not be "rcdeemable in gold at the bank of issue" and that the "holder of these notes must apply to the goldsmith when he desires gold"( 1934, p. 309). This evolution does not imply any change in the backing but rather in the location at which notes could be redeemed. At no lime would the notes be redeemable for anything but their"face
Free Banking II
51 NOTES, COMMENTS, REPI.IFS
:
113
worth" of gold (p, 291), Thus even before the system evolved to its purer form there would be no "backing" of notes in any meaningful sense,
2, Arthur Kitson In contrast to Meulen, there is abundant evidence that Kitson intended his proposal to result in price level stability, I n The Money Prohlem (1903) Kitson asserted that the adoption of an "invariable ideal unit of purchasing power" would elimillate illterest and make general changes in the price level impossible (1<)03, pp, 15354,224), The lIllit wOlild represellt ":1 eertaill fraetioll of all the exchangeable wealth at a given time" (p, 217), Although a sympathetic readillg might lead one to suppose Kitson intended to define the unit in terms of a fixed bundle of tangible assets, the text is vague on this point. In A Fraudulent Standard (1917) Kitson spelled out the details of his proposal, if not with greater clarity, then at least with more specificity, To assure that the unit had stable purchasing power Kitson would define the pound as a given fraction of national wealth: For example suppose we start at noon, as stated, January I, 1<) Ig, Our wealth is then X pounds, Suppose by noon, January 1,1919, our wealth has increased 10 per cent., then our total wealth = (X + , I X) pounds and the pound which equalled IjXth of the national wealth, January I, 19l9-becomes I/(X + ,IX)th of the national wealth in January, 1919, (1<)17, p, 129) Obviously, for this definition to be anything more than a tautology Kitson would have to mean real wealth, However, Kitson never seems to specify his unit in terms of a given basket of commodities, In fact it is l',nclear whether the term "wealth" is to include only physical assets, or financial assets as well. Later he states that "all additions and subtractions from such wealth would be additions or subtractions of so many pounds, that is, so many fractional parts of total wealth, It is not essential to transform the system into numbers" (1917, p, 142), But if price stability is the goal, then it is necessary to specify the unit of account in terms of numbers of commodities, Kitson's explanation in Liberty (J une 15, 1895, pp, 6-8) suggests that he meant the unit to represent a comprehensive bundle of commodities although he continued to deny the need to compute a price index to implement the system, I n the same issue one reviewer argued that Kitson had admitted that his proposal was a commodity standard: "Then Mr. Kitson is, after all, advocating a concrete unit of purchasing power,-viz" a composite unit agreeing in composition with that of the sum total of all wealth" (p, 2), Also in this issue, another reviewer noted that "as for M r. Kitson's claim that the value of the unit need not be calculated, but will take care of itself, nothing could be more absurd" (p, 4), At times Kitson himself seems confused about the issue, Regarding Fisher's proposal to back the dollar by gold of fixed purChasing power (rather than fixed weight), Kitson writes:
52
Free Banking II 114
MONEY. CREDIT. AND RANKING
The suggestion is ingenious. hut more feasible in theory than in practice. It would he analogous to the legal establishment of 1I mercury-tube as our yard-stick. exposed to atmospheric innuences which would involve an alteration in every yard measure and every foot-rule in the Kingdom whenever there was a change in temperature! (1917. p. 86)
Obviously Kitson does not understand what Fisher was proposing since elsewhere Kitson uses length as a metaphor for purchasing power. Fisher's proposal would eliminate changes in the "length" of the dollar. What is more puzzling is how Kitson could be so confused since he makes the same criticism of a gold standard based on a fixed weight of gold. There are also numerous instances where Kitson belittles the fear that free issue of paper money might be inflationary. Were the unit of account clearly defined (as in the BFH system), then Kitson's view would bejustifiable. Many of the examples he cites with approval, however. such as the greenbacks issued during the Civil War, do not support his assertion (1903, p. 228). Most of Kitson's (and Meulen's) references to price level instability refer to the problem of deflation. As with the silver proponents of the late nineteenth century, the actual method to be employed set~ms relatively unimportant in comparison to the objective of expanding the money supply. C-K seem to accept Kitson's assertions regarding the invariable unit at face value. They state that Kitson's unit "can be defined by setting the value of any commodity on a given day equal to 'one' and pricing all commodities in terms thereof" (1987, p. 579). Yet this definition is either incorrect or redundant. The unit can be defined as the value of X ounces of gold on a given day, but to insure price stability it must also be defined in terms of a given basket of commodities (as in the Greenfield-Yeager system.) Otherwise, linking of the unit of account to the value of gold on a given day merely provides the unit with the appearance of having intrinsic value.
3. Prc'clIr.mrs (?lthe Modern /JFII 71leori.l't.l' The fact tllllt hoth Kitsoll and Meulen repelltedly define the unit of account in tcrllls of itself, indicates that these quotations represent something llIore than lin accidentlll application of circular reasoning. Their proposal does bear a superficilll resemblance to the system proposed by Greenfield and Yeager, and Yeager even discussed the possibility of indirect convertibility of currency into gold. 1 The crucial difference, however, is that under the Greenfield-Yeager system, each dollar's worth of currency would be redeemable for gold of fixed purchasing power (in terms of the standard bundle of commodities). In this respect the Greenfield-Yeager proposal is closer to the Williams-Fisher compensated dollar plan. . Williams (1892) proposed a plan whereby the gold weight of the pound would be adjusted daily in proportion to changes in the price of a specified bundle of 'See McCallum (1985). p. 37. footnote.
53
Free Banking II NOTES. COMMENTS. REPLIES
:
115
commodities. Fisher's (1920) plan differed slightly in that the adjustments would be made bimonthly, and on the basis of a more comprehensive price index (the Wholesale Price Index as measured by the Bureau of Labor Statistics).2 Black's (I9H I) proposal is almost identical to Fisher's compensated dollar plan. Under the Black proposal, "when the price level starts to rise, the price of gold is dropped; and when the pric~ level starts to fall, the price of gold is raised. The only delays in this process should be due to the time it takes to measure the level of the price index" (p. IIH). Fisher had recognized that the existence of an information lag in measuring the price level could lead to destabilizing speCUlation in the gold market when there was an expectation of an imminent change in the price index. He therefore proposed what he termed a "brassage charge"(that is, spread or turnaround fee) of I percent on all gold transactions with the central bank. Fisher noted that the spread should be set at a high enough level to insure that it was larger than any probable change in the gold weight of the dollar (which is the inverse of the change in the price index). Although this type of spread would eliminate the problem of speCUlation, it would simultaneously eliminate the ability of private individuals to arbitrage small gaps between the official price of gold and its free market price. In that case there would be no obvious mechanism by which changes in the price of gold could generate corresponding changes in the price level. Instead it appears that the policy would become operational only through changes in the currency stock. Under Fisher's proposal gold coins would be withdrawn from circulation and replaced with gold notes. Under the "rules of the game" governments maintained a fixed currency I gold stock ratio. If the price index announcement showed a I percent decrease, then the government would increase in the official price of gold by I percent, which would increase the (nominal) monetary gold stock by I percent, and thus the government would increase the stock of currency by I percent. It is the change in the currency stock, not the official gold price, which causes the price level to rise. Because the spread must be large enough to prr!vent speculation (and arbitrage) Fisher might just have well proposed a system with a 1,000 percent spread. This would have more clearly highlighted the similarity of Fisher's plan to modern money rules such as McCallum's (1984) proposal to adjust the monetary base inversely to announced shifts in velocity, Under Fisher's plan the currency stock would show a discrete jump when the price index was announced, yet there would be no discontinuolls changes in either the market price of gold or the exchange rate in response to the anticipated portion of the price level announcement. (Note: If foreign cOllntries were on a IRooke (1824) prop~sed ~ plan similar t~ the ~illiams-Fisher coml?ensated dollar pia!!. Rooke suggested that the offtclal pnce of gold be adjusted Inversely to changes In the average nomlllal wag.: rate. lie argued that this would prevent major price level changes while still allowing some price level Iluctuations duc to "non-monetary" factors. Surprisingly. Rookc's plan is in some ways actually more "modern" than the Williams-Fisher plan in thllt it would be accommodative towllrd supply shocks.
54
Free Banking II III>
:
MONEY. (,REflIT. ANflllANKING
gold standard. then the exchange rate would be linked through arbitrage to the market price of gold.) Thus, the system would not be vulnerable to speculative attacks. In a letter to McCallum (1985, p. 37) Yeager indicated that under the Greenfield and Yeager proposal "competition would probably lead institutions to cash checks (if their holders so desired) and redeem currencies and deposits in whatever quantities of a convenient redemption medium (such as gold or actively traded securities) equalled in total value as many standard commodity bundles as the number of units of account denominating the obligations to be redeemed." McCallum noted that "the Greenfield and Yeager elaim that their system 'is not a variant of the often-pnllJosed composite-commodity or commodity-reserve money' system (1982, p. 303) would seem highly misleading" (1985. p. 37). It might also be noted that th,~ir plan is susceptible to the same speculation problem that Fisher was concerned about. By linking the price of gold to a price index containing only the prices of actively traded commodities, Williams avoided the problem of discrete price index changes generated by an information lag. Hall's (1982) proposal to link the unit of account to a specified quantity of four commodities (aluminum. copper, plywood. and ammonium nitrate) sharcs this advantage.) Of course, hoth proposals are susceptible to price level instability if the relative price of'the specified commodity bundle is not stable.
4. Concluding Remarks The work of Kitson and Meulen was further from the ideas of the modern HFH proponents than one might infer from the C- K paper. It is true that they did anticipate some of the tenets of the new monetary economics, particularly in the area of free banking and deregulation. And it might be argued that the inability of Kitson and Meulen to construct a stable unit of account is a relatively minor defect, were it not for the fact that both the BFH forerunners and the modern BFII proponents place great importance on this issue. Yeager (1983) hegins his discussion of the new monetary economics with the assertion that the instability of our unit of account is the central defect of our current system. Black (1981 l. Fama (1983), Hall (1983), and Greenfield and Yeager (1983) have all proposed systems which would result in approximate price stability. Given the importance many NM E theorists place on the stahility of the unit or account, it is surprising that C-K did not give more attention to the WilliamsFisher system. C-K lumped the compensated dollar plans in with Jevons' tahular standard as mere "indexation schemes" (1987, p. 571) a !though, unlike the Jevons plan, they are clearly policies designed to prevent price level fluctuations, not to index contracts c.gainst such fluctuations. Although the unit of account in the compensated dollar plan is linked to a quantity of gold of fixed purchasing power, at a more fundamental level it is the underlying basket of commodities hy lllall chose this commodity bundle because its price closely tracked the cost of living.
55
Free Banking II NOTES. COMMENTS. REPLIES
:
117
which the purchasing power is measured that constitutes the unit of account. Gold is merely the medium by which the purchasing power of notes is guaranteed. 4 Greenfield and Yeager went far beyond Williams and Fisher in their proposal for a laissez-faire monetary system. It is not clear, however, that the macroeconomic implications of their proposal are much different from the compensated dollar plan.
I.ITFRATIJRE CITED
Black, Fischer. "Ballking and Interest Rates in a World without Money: The Effects of Uncontrolled Banking." JUI/Tl/al uf Bankillg Research I (Autumn 1970),1;-20. _ _ . "A Gold Standard with Double Feedback and Ncar Zero Reserves." lJnpublislll:d, MIT, Dt:cember 191;1. Cowen, Tylcr, and Randall KrosL.ller. "The Developmcnt of the Nt:w Monetary Economics." journal of Pulitiml EconolllY 95 Ount: 191;7),567-90. Fama, Eugene. "Banking in the Theory of Finance." Journal l!( Monetary Econolllics 6 (January 1980),39-57. _ _ . "Financiallntcrmcdiation and Price Level Control." jou/'1lal o( Alul/eltll')' t:cu/lOll/in 12 (July 191;3),7'-28. . . Fi,ha, Irving. Siabilizing the Dullar: A Pla/llV Stabilize the General Price Leve/ll'itliolll Fixing Individual Prices. New York: Macmillan, 1920. Greenfield, Robert 1.., and Leland B. Yeager. "A Lai,sez-Faire Approach to Monetary Stability." Juumal uf Aluney, Credil, and Bal/king 15 (August 191;3), 302-15. lIall, Robert. "Explorations in the Gold Standard and Related Policies for Stabilizing the Dollar." In/lljla/ioll: Callses al/(I EYfects, edited by R. Iiall, pp. 111-22. Chicago: lJ niversity of Chicago Press, 1992. ___ . "Optimal Fiduciary Monetary Systems." journal (July 1993), 33-50 .
0/ Alunetary
Ecunomics 12
.Ievons, William Stanley. Aloney and Ihe Mechanism o/Exchallge. New York: Appleton, 1876. Kitson, Arthur. The Money Problem. London: Grant Richards, 1903. (Originally published as A Scielllijlc Sulutiut! IV Ihe Money Ques{ioll. Boston: An:na, 1895). _ _ . A Fraudulell{ Stalldllrd: All Er{Josllfl'{!(111t' Fftlllllu/l'1ll Owrtlcll'ro/Our
wry SWllclarti. London: King,
/110111'-
II) 17.
Uberty, edited by Benjamin Tucker. Westport, Conn.: C,reenwood Reprint, 1970. McCallum, Bennett T. "Monetarist Rules in the Light of Recent Experit:llce." !l1Ill'rimll Economic Review 74 (May 191;4), 3gl;-91. _ _ ."Bank Deregulation, Accounting Systems of Exchange, and the U nit of Account: A Critical Review." In 'i1ze" New MOlll.'IiIry E·COIIOlllic.I', " Fism/h·.I·ue.l· alltl Ullelllployment, edited by K. Brunner and A. Meltzer, pp .. 13-45. Amsterdam: North-llolland, 1985. Meulen, Henry. Free Banking: An OLI/lineo/a Pulil'y(!(individua/i.\'/II. I.ondon: Macmillan, 1934. 'Black seems to have ret:ognized this point since his system calls for "ncar tern reserves."
56
Free Banking II IIX
MONEY. CREDIT. AND BANKING
Rooke. John. IlIquiry into the Principles of National Wealth. Edinburgh, 1824. Wnllacc. Ncil. "A Lcgal Restrictions Theory of thc Dcmand for 'Moncy' and thc Roll' of Monctary Policy." Fcderal Rcscrvc Bank of Minncapolis Quarterly ReFit'\\' (Winter
1983),1-7.
Williams, Ancllrin. "A 'Fixcd Valuc of Bullion' Standard·· A Proposal for I'rcvcnling (Jcncral Jo"illctlllltions ofTradc." HnJllo/II;c .IO/l/'Iw/2 (.Jllnc 1892). 2XOX9. Ycagcr, Lcland B. "Stable Moncy and Free-Market Currcncics." C(J/o .Iolll"l/a/ 3 (Spring
1983),305-26.
[4] Excerpt from Donald E. Moggridge (ed.), Perspectives on the History of Economic Thought, Volume Four,
20-39
2 Laissez-Faire Monetary Thought in Jacksonian America Lawrence H. White and George A. Selgin
In recent years there has been a noteworthy growth ofliterature on laissezlaire systems of money and banking.! The traditional view among economists, of course, has long been that the supply of money (including bank liabilities) will not behave properly without government control. But the contrary view that unregulated banking would work well is not new. Its early proponents included Adam Smith (1981, [1776], p. 329) and J. B. Say (1971 [I803], p. 271). Those who espoused the laissez-Iaire approach to money were in the intellectual minority even during the heyday of classical liberalism. Yet their ideas had an important influence on monetary legislation, as evidenced particularly in the United States by the Independent Treasury System and by the proliferation of state 'free banking' statutes. Early writers in the tradition of laissez-Iaire monetary thought have recently begun to receive some attention from historians of economic thought, their views being now of obvious relevance to modern concerns. The important American proponents of unregulated money and banking in the first half of the nineteenth century, however, have never had their contributions critically surveyed. 2 This paper attempts to fill the gap. The next section sets the historical stage. Because of their small number, we then treat the leading laissez-Iaire monetary theorists individually. In treating individual writers we focus on four aspects of their thought: I. 2. 3. 4.
the application of free-trade and spontaneous-order principles to money and banking; the theory of restraints against over-issue in a free banking system; the theory of business cycles or panics; and proposed reform measures.
As many writers at that time did, we neglect questions relating to the deposit-taking and loan-making aspects of banking in order to focus on the more controversial questions regarding banknotes. Monetary debates in the early Republic The policy ideas on money and banking that dominated debate during the early decades of the Republic were, for various reasons, quite unsympath20
58
Free Banking II
Laissez-Faire Monetary Thought
21
etic towards laissez-faire in money and banking. Advocates of strong central government, such as Alexander Hamilton, favoured mercantilist policies generally. But even among anti-mercantilist statesmen and intellectuals the area of money-issuing was considered an exception to the doctrines of free trade and open competition. Thus Pelatiah Webster (Krooss, 1977, pp. 221-9), a free-trade Whig, defended the exclusivity of the charter given to the Bank of North America in 17S1 by the Continental Congress. 3 In the debate over the chartering and attempted re-chartering of the First Bank of the United States (1791-ISII), Thomas Jefferson (Krooss, 1977, pp. 273-7) and his followers such as John Taylor (1969, pp. 253342) and James Madison (Krooss, 1977, pp. 262-70) advocated hard money and strict construction of the United States Constitution, which does not explicitly grant the Federal government any monetary powers beyond coinage. Pitted against them were Hamilton (Krooss, 1977, pp. 27S-306) and other broad-constructionist proponents of a national bank. The concept of a laissez-faire monetary system was not introduced into the debate. The opponents of the exclusively chartered national bank, following Taylor and Jefferson, doubted the desirability of any noteissuing banks (as opposed to mere 'discount and deposit' banks). Although the Jeffersonians attacked the Bank as a monopoly, their view that any federal chartering of corporations was unconstitutional, mixed with their suspicion of note-issuing banks, prevented them from considering a system of laissez-faire based on federal charters available to all. The Hamiltonian advocates of bank notes, on the other hand, saw note issue as a government prerogative rather than as a common right. The debate in ISI6, when the Second Bank of the United States secured a twenty-year charter, was similarly framed. The strict-constructionist, anti-monopoly and anti-banknote strain of thought showed new life in the IS30s, being spread particularly by William Gouge (IS33). Hard-money views dominated Jackson's party at the ideological level, though 'bank Democrats' more often held office. The national bank, of course, had its advocates in the Whigs. The conflict between hard money and national bank dominated the debate over Jackson's veto of the attempt to recharter the Second Bank in IS32, as it had dominated the earlier debates. A third approach to the currency question was available, however, and came to play an important role in the debates of the IS30s and thereafter. The laissez-faire or 'free banking' approach, opposed to exclusive noteissuing privileges but not to banknotes per se, was first articulated in the United States in IS27 in a pamphlet by the Reverend John McVickar. It found its most consistent policy formulation in the writings of editorialist
Free Banking II
22
Perspectives on the History of Economic Thought Vol. IV
William Leggett between 1834 and 1837. In the next few years laisse=~faire banking was promoted by historian and journalist Richard Hildreth, who provided the best theoretical underpinning for its advocacy, and by economist Henry C. Carey. Hildreth (\ 840, p. \09) aptly characterized free banking opinion as one of the 'three opposite and hostile systems of opinion prevail[ing] at the present time in the United States, on the subject of banks and banking. One party is opposed to all banks whatever; a second party is in favor of the existing system of a monopoly of banking privileges; while a third pal ty desires to throw open in the business of banking, like all other mercantile business, to free competition.' The free banking writers of this era shared with the hard-money school the fundamental tenet of Jeffersonian political economy that narrow limitation of government intervention was desirable. Both camps vigorously criticized the legislative chartering of banks with special privileges. But the laissez~aire camp diverged in recognizing note-issuing banks as natural creatures of commerce, rather than as products of government intervention. In Leggett's case a gradual evolution from hard-money to strictly laissez-faire policy recommendations can be seen.
Early writers: Butler and McVickar Apart from the First and Second Bank of the United States, the chartering of banks of issue was the prerogative of the state governments. An early application of laissez-faire arguments to banking in America came in a brief pamphlet addressed to a member of the New York State legislature. New York's 'restraining act' of 1804 had outlawed non-chartered banking partnerships. The legislature in 1818 was considering a ban on nonchartered private banking by individuals as well. 4 Benjamin Franklin Butler, a private banker writing under the pseudonym of 'Marcus' (1818), protested that banking should be left as open as every other business, being 'the true and legitimate off-spring of commerce', and being able to trade its products only with members of a discerning public who voluntarily accepted them. The objections to private banking, Butler argued, came from established chartered banks 'desirous of monopolizing an employment which should ever be left open to the free exercise of all'. Prohibition of private banking by individuals 'would shackle the excursions, restrain the liberty, and abridge the rights of the citizen'. Butler was not entirely consistent with his professed principles, however. He disingenuously endorsed the legislature's restriction against private banking by partnerships, who were potential competitors with his own business. A more important theoretical brief for freedom in banking came from the pen of a Columbia University professor, the Reverend John
59
60
Free Banking II
Laissez-Faire Monetary Thought
23
McVickar. His anonymous pamphlet Hints on Banking appeared in 1827, the same year that the first major work of the Free Banking School in the British monetary controversies was published (see White, 1984, p. 62). McVickar (1827, pp. 5-7) clearly enunciated the laissez-faire idea that banking, like any other industry, would regulate itself if left free: 'the business of Credit, like every other business for which there is a demand in society, has its natural limits, and left to itself will regulate itself. If it seems a relatively troublesome industry, that is because it has not been free: 'all the evils of banking, beyond those which exist in other modes of business, flow from needless or unwise legislation.' Mc Vickar (1827, pp. 13-16) charged that the state legislature's process of bestowing monopoly banking charters on politically favoured applicants was rife with corruption, an extremely popular theme among both laissez-faire and antibanking writers. Furthermore - and here again the argument was to be taken up by many others - chartering conferred a false and uniform respectability on all bankers which enabled the 'unprincipled and designing' among them to defraud the public. The power of granting special charters was an atavistic remnant of mercantilism, 'one, and almost the last of a long list of legislative powers, which one by one have been dropped into the lap of society, that real nursing mother of the rights of man, and there entrusted to the rights of individuals', because 'economical science' had shown that society is governed by laws other than statute laws. Free banking was thus presented as an integral part of the classical liberal programme for policy reform. McVickar's account of the self-regulating character of banking consisted of a loosely expressed version of the needs of trade doctrine, which held that the quantity of monetary bank liabilities in circulation was and ought to be demand-determined. There is nothing fallacious in this doctrine provided that the 'needs of trade' refers to the demand to hold bankissued money (rather than the demand for loanable funds), and provided that the purchasing power of money is determined exogenously to the banking system, as it is in a small open economy under a specie standard. Then the nominal quantity of money can and classically should vary endogenously. The needs-of-trade doctrine simply extends to specieredeemable money the basic idea behind the price-specie-flow mechanism, that the quantity of specie in circulation conforms to the quantity demanded at a given level of purchasing power.s McVickar (1827, p. 10) explicitly analogized the demand-side determination of the quantity of a purely metallic currency in a small open economy to the similar determination of the quantity of a specie-redeemable currency: ;If [banks] issue a paper exchangeable on demand, like a metallic currency, it accommodates itself freely to every fluctuation from without, and the [individual] bank is
Free Banking II 24
Perspectives on the History of Economic Thought Vol. IV
governed imperatively in its issues, by the demand for specie that is immediately made upon it, whenever its paper is in excess.' McVickar's statement, however, appeared in the context of an argument (1827, pp. 7-16) concerning 'the business of credit'. Under laissezfaire, credit 'left to itself will regulate itself, - will contract or expand with the varying demands of trade, and will be liable to no other fluctuations than those which arise out of trade itself. Credit would be 'furnished to society at the cheapest rate, and in precisely that quantity which society demands'. By 'credit' McVickar seems to have meant the quantity of bank loans rather than the quantity of monetary bank liabilities. His reference to 'the cheapest rate', i.e. the lowest interest rate, suggests the furnishing of loanable funds rather than money balances. Because he blurred the distinction between money and credit, McVickar did not provide a clear account of the determination of the money stock. 6 He nonetheless showed some understanding of the principle that the circulation of any single bank is regulated by adverse clearings, i.e. that any bank expanding faster than its rivals would, assuming no change in note-holding demands, be forced to slow down by loss of reserves to the other banks as they accumulated and redeemed a greater number of its notes. McVickar (1827, p. 18) stated that if one bank contracts, every other bank must follow 'proportionately, under the penalty of becoming a debtor bank, and thereby losing a portion of the specie out of its vaults'. This statement is correct only under the assumption that the demand to hold notes has contracted generally. McVickar (1827, pp. 24-33, 41) suggested that the bank chartering system made business cycles more severe because chartering reduced the number of independent decision-makers in the money market. The surplus profits in closed-entry banking, furthermore, tempted 'the needy and the speculating' to obtain charters through lobbying. This led to highly speculative banks whose actions increased the severity of fluctuations. In a less clear argument, McVickar attributed the speculativeness of chartered banks to their having excess capital, which they dumped on the market. McVickar's reform proposal (1827, pp. 3~1) was animated by laissezfaire principles, but stopped short of full deregulation because of a fear of fraudulent banknotes. Were banking abuses limited to the commercial realm, 'the remedy would be as simple as it would be efficacious, viz. to cast off all restrictions, and to leave the business of banking to be regulated by the necessary laws of credit'. But notes pass into common circulation, substituting for coin. On this ground McVickar, like contemporary Currency School writers in England, justified 'the interference of the legislature, who, as the guardians of the coin of the country, acquire the right to regulate its substitutes'. McVickar accordingly proposed free entry into banking under a general statute, but with the restrictions (I)
61
62
Free
Banking
II
Laissez-Faire Monetary Thought 25 that 90 per cent of a bank's capital be invested in government bonds pledged for the redemption of notes, (2) that the volume of notes not exceed this 90 per cent of capital, and (3) that $5 be the minimum banknote denomination. The bond-collateral requirement was to be a popular feature of proposals by American writers otherwise professing laissez-faire principles. It gave them a concrete answer to the fear of fraudulent banknotes, allowing them to argue more persuasively that the other aspects of banking (particularly entry) could be deregulated without danger. Collateral requirements became a central part of the 'free banking' legislation to be enacted by New York State in 1838 and by seventeen other states prior to 1860. The idea of compulsory bond-collateralization of notes was also advocated in the 1820s by several monetary reformers in England, including both Henry Parnell (1827, pp. 140--4) of the Free Banking School and his intellectual opponent J. R. McCulloch (1826, p. 280) of the Currency School. For Parnell, McVickar and later American free banking advocates such as Richard Hildreth, this restrictive measure became a token to be paid in order to secure the goal of free entry. From McCulloch, a devoted Ricardian, we can trace the bond-collateral idea back to David Ricardo (1951, pp. 72-3), who made it part of his Proposals for an Economical and Secure Currency in 1816. Ricardo argued that common folk who accept banknotes in everyday transaction are unqualified to judge the trustworthiness of the issuing banks, and suffer when the banks fail, so that for consumer protection the banks should be required to deposit with the government 'funded property or other government security in some proportion to the amount of their issues'. 7 Soon after he had made his case for freer banking, McVickar (1830) curiously enough came out in favour of re-chartering the Bank of the United States. In an anonymous 1839 article he defended both the principles behind the New York State Banking Act of 1838 and the chartering of a national bank as a regulator of the various state banking systems. Two years later he published a formal plan for a national bank. 8 Far from affirming the self-regulating character of a free banking system, he now argued (1841, p. 2) that a national bank was necessary to end 'the mad experiment of attempting to steer the ship without a rudder, and to regulate, without a regulator, the vast and complex machinery of currency and exchanges' (emphasis in the original). The Ricardian element in his thinking had apparently crowded out the Smithian element, for McVickar (1841, pp. 16-17) proposed to give a monopoly over note-issue to a nonintermediating national bank of the sort Ricardo (1951, pp. 276-97) had advocated in his Plan for the Establishment of a National Bank (posthumously published in 1824). McVickar did not explicitly cite Ricardo, but
Free Banking II 26
Perspectives on the History 01 Economic Thought Vol. IV
he did cite the parliamentary testimony of Ricardo's followers, the Currency School opponents of free banking.
William Leggett Andrew Jackson's message to the Senate in July 1832, vetoing the recharter of the Second Bank of the United States (Kroos, 1977, pp. 81632), criticized the· Bank primarily on strict-constructionist grounds. His opposition was seemingly only to its form and not to its substance, as Jackson held out the possibility that he could countenance a government bank of the right sort.9 Yet Jackson concluded by denouncing the Bank as a privileged monopoly that infringed upon the just principle of equal treatment for all. This last theme - that legislative chartering of corporations (and particularly of banks) violated equal rights - was developed into a thoroughgoing programme for monetary reform by William Leggett. Editor of the New York Evening Post for 1834-6, and then author of his own New York newspapers the Examiner (daily) and the Plaindealer (weekly) in 1836-7, Leggett was the intellectual leader of New York's Equal Rights Party, a splinter group of radical young democrats more commonly known as the Loco-Focos. Leggett (1984, p. 81) was unmistakably explicit in his desire to apply the laissez-Iaire principle to banking, expressing his hope 'to see the day when banking, like any other mercantile business will be left to regulate itself; when the principles of free trade will be perceived to have as much relation to currency as to commerce: when the maxim of let us alone will be acknowledged to be better, infinitely better, than all this political quackery of ignorant legislators, instigated by the grasping, monopolizing spirit of rapacious capitalists'. Leggett (1984, pp. 104, 130-1, 146, 164, 179-82) motivated the case for laissez-Iaire in currency in several ways. From political philosophy he derived the equal natural right of every citizen to open a bank and to issue any sort of promissory note that others would accept. On aesthetic grounds he appealed to the naturalness of a simple and parsimonious role for government. Most importantly for our purposes, he drew from his understanding of classical economics the concept of a spontaneous and self-regulating order: 'Enterprise would build up, and competition would regulate, a better system of banks than legislation ever can devise.' Because he wrote a series of editorial essays rather than any extended monograph, Leggett never exposited at length his theoretical conception of the self-regulating processes at work in a free banking system. His various remarks (1984, pp. 118-19, 145-6, 174-5, 186-8) none the less sketch a definite picture of the system's operations. Under laissez~faire, in his vision, there would exist a plurality of well-capitalized banks. These
63
64
Free Banking II Laissez-Faire Monetary Thought
27
banks would command 'the utmost public confidence', because competition for public patronage would force them to offer solid guarantees of security for their liabilities. The guarantees could take the form of pledged collateral property. Banknotes 'would, of course, be redeemable in specie', but the banks, immunized against panics by their solvency guarantees, would find small fractional reserves adequate. Competition would reduce the spread between loan and deposit interest rates to the smallest level compatible with the banks earning a normal rate of return on their capital. As Leggett (1984, p. 119, 130-31, 169) correctly insisted, these features were not merely figments of his imagination but could be observed in the contemporary Scottish banking system. In referring to the Scottish system, and in his theoretical remarks on banking, Leggett showed a distinct familiarity with the writings of the Free Banking School in the British monetary controversies of the day, though he did not explicitly refer to their works.1O Leggett overlooked the Scottish experience, however, in suggesting that under laissez:faire the use of banknotes would dominate specie only in business-to-business transactions. A closer reading of Adam Smith (1981, p. 322), whom he was fond of citing, would have supplied him with evidence to the contrary. The most important feature of free banking for Leggett, as for McVickar, was its automatic regulation of the currency supply. In an early essay Leggett (1984, p. 64) enunciated a normative version of the needs-oftrade doctrine. He was at best somewhat ambiguous on the source and measure of demand for bank-issued money, however, and at worst shaded off into the real-bills doctrine that Adam Smith (1981, pp. 304-8) had propounded and that John Fullarton (1844) of the British Banking School would come to endorse. The real-bills doctrine, interpreted as the belief that an excess supply of bank-issued money cannot be created by discounting 'real bills' (trade credit obligations in the form of commercial paper), is a fallacious version of the needs-of-trade doctrine in which the demand to hold bank-issued money is identified with the volume of 'real bills' offered to the banks. 11 A tinge of the Smithian real-bills doctrine may also be seen in Leggett's (1984, p. 157) later positive needs-of-trade argument that the currency would be self-regulating under free banking because, with hard money being used by the general public, 'banking will naturally confine itself to those operations which constitute its only legitimate field - the mere exchange of bank credit for mercantile credit, to the extent of actual commercial transactions'. Leggett's (1984, pp. 82, 118) more emphatic statements, however, indicated a correct understanding of the adverse clearing and specie-flow mechanisms that prevent a sustained over-issue. Competition would lead banks to accept rivals' notes, and then to redeem them (presumably
Free Banking II 28
Perspectives on the History of Economic Thought Vol. IV
through a note-exchange system, although Leggett does not say so), confronting any over-expansive single bank with adverse clearings: 'The natural rivalry of trade would cause [a bank] to return the notes of other institutions for specie, whenever they accumulated beyond a certain point, and this would prevent overissue.'12 The system as a whole would be checked by external drain if the actual quantity of currency exceeded the quantity demanded. Bankers knowing this would try to proportion their issues to the demand, taking the level of business activity as an indicator: when bankers are left to manage their own business, each for himself, they would watch the course of trade, and limit their discounts accordingly; because if they extended them beyond the measure of the legitimate business of the country, they could be sure that their notes would return upon them in demand for the precious metals, thus forcing them to part with their profits, in order to purchase silver and gold to answer such demand.
In examining the existing chartered banking system, on the other hand, Leggett (1984, p. 97) observed 'the banks ... striving, with all their might, each emulating the other, to force their issues into circulation, and flood the land with their wretched substitute for money'. An obvious question is why competition and prudence did not effectively constrain the chartered banks from over-issuing. Leggett (1984, p. 151) claimed that competition among the chartered banks was severely restricted, though in fact it seems to have been vigorous in the major cities (Schwartz, 1947). He also argued (1984, p. 162, 182-3) that chartered banks were able to over-issue 'by reason of the false character which their exclusive privileges give to the notes they issue'. The imprimatur of the state government upon the issuer, and the receivability of the notes by the government in taxes and other transaction!', was apparently held to make the public less discriminating in its acceptance of notes. This would indeed slow the return of excess notes. If the public does not at all discriminate among banknotes according to bank of issue, then an excess supply of notes created by any issuer can be felt only as an excess supply of notes in general. The issuer at fault suffers no systematic adverse clearings against other banks. It only shares in the (relatively slowly acting) external drain that affects every bank in proportion to its circulation (Selgin, 1988, pp. 42-7). Leggett's business-cycle theory strongly resembles the monetary malinvestment theory put forward by Robert Mushet (1826, pp. 152-8), Henry Parnell (1833, pp. 6-13), and other members of the British Free Banking School, and by some writers ofthe Currency School. In Leggett's (1984, pp. 65-70,97-100, 113-14, 139) version, the cycle begins with an over-expansion of bank loans and note-issues that creates an investment boom.The boom raises prices generally and with a lag causes specie to
65
66
Free Banking 1J Laissez-Faire Monetary Thought 29 leave the country. The loss of specie reserves forces the banks to stop expanding and even to start contracting. The contraction of loans leaves investment projects suddenly unsustainable and thereby instigates bankruptcies, panic and a crash. Leggett placed much more emphasis than the British writers on the malinvestment of capital and misdirection of labour during the boom period. The following passage (1984, p. 98) usefully encapsulates his account: What has been, what ever must be, the consequence of such a sudden and prodigious inflation of currency? Business stimulated to the most unhealthy activity; a vast amount of over production in the mechanick arts; a vast amount of speculation in property of every land and name, at fictitious values; and finally, a vast and terrifick crash, when the treacherous and unsubstantial basis crumbles beneath the stupendous fabrick of credit, and the structure falls to the ground, burying in its ruins thousands who exulted in the fancied security of their elevation. The banks, forced to contract by dwindling gold reserves, could no longer finance 'the projects which would not have been undertaken but for the temptation they held forth'. As an example of misdirection of labour, Leggett (1984, p. 100) observed that fields lay untilled 'because the agricultural population has been drawn off to construct railroads and canals, or layout sites for cities'. Later monetary business-cycle theorists, such as Hayek (1935) and Lucas (1975), have tried to account for malinvestments in long-term or capital-intensive projects of this sort. 13 Consistent with his view that freer competition would allow the adverse-clearing mechanism to restrain over-issues more effectively, Leggett (1984, p. 118) came to view completely free banking as the remedy for monetary disturbances. In his earliest editorials, however, he showed more ofa hard-money outlook. 14 He proposed (1984, pp. 71-3) a ban on small denomination banknotes, a measure Adam Smith (1981, p. 323) had endorsed for Scotland. Later, Leggett (1984, p. 82) urged this only as a transitional reform, and finally (1984, p. 152) he repudiated any attempt to institute a metallic currency 'by the force of artitrary government edicts'. Similarly, he first (1984, pp. 81-2) recommended free entry with a bond collateral system as a reform measure appropriate to 'the present temper of the times', which were not yet ready for full-blown laissez-faire. Later (1984, pp. 118, 146--8, 156--7, 174-5, 186), however, he insisted that competition by itself would induce banks to provide sufficient security to their liability-holders. His final proposal (1984, pp. 145-54) was simply that the state legislature repeal the law that restrained free entry into banking, replacing the chartering system by a law making incorporation freely available to all businesses. The federal government, in order to avoid unjust favouritism, was totally to disassociate itself from the bank-
Free Banking Il
30
Perspectives on the History of Economic Thought Vol. IV
ing system. The only statutes applying to banks, he believed, should be those concerning fraud, breach of contract, and incorporation, that applied to business in all fields. In his arguments for 'separation of bank and state' Leggett provided a laissez-faire rationale for the Independent Treasury System which removed the federal government's money balances from the commercial banks. When President Van Buren in 1837 declared his support for such an idea, the mainstream Democratic Governor of New York, William Marcy (quoted by Sharp, 1970, p. 302), despaired of this seeming endorsement of ideas from the radical wing of his party: 'Is it reasonable to expect that the democrats of the state will range themselves under the banners of Ming, Leggett ... and others of better repute of Washington[?]' 15 As the acknowledged intellectual leader of the New York Loco-Focos (Byrdsall, 1967, pp. 15, 22-7), Leggett was also credited by his admirers with an important influence on the 'free banking' law enacted by New York State in 1838 (Bryant, 1839, p. 23; 'General Banking Law', p. 428). Fritz Redlich (1947, pp. 188-90), who provides the most thorough secondary account of the intellectual and political origins of the 'free banking' law, effectively concurs in concluding that 'Free banking was brought into being by the Loco-Focos'. Working against the mainstream Democrats who were allied with the chartered banks, the Loco-Focos generated a popular movement for free entry into banking to which Governor Marcy and the legislature (dominated by Whigs elected with the Loco-Focos's help) capitulated. Leggett's argument that free competition would itself suppress unreliable banknotes did not prevail, however. Instead, McVickar's bond-collateralization idea was adopted as a safety device.
Henry C. Carey and Condy Raguet After its political success (qualified by the compulsory bond-collateral provision) in New York, the argument for free banking was addressed to the remaining states. The economist Henry C. Carey, a prolific author well known in his day, published two·lengthy pamphlets (1838; 1840) favourable towards free banking. Carey's chief concern seems to have been the promotion of economic growth. 16 His earlier pamphlet offered little in the way of theory, but did make two noteworthy points in its discussion of Scottish banking. First, Carey (1838, p. 81) argued that laissez:faire in credit markets promoted economic growth. Scotland's rapid growth was attributed to the 'comparative absence of restraints upon the employment of capital. Were all restrictions abolished, her growth would be still more rapid.' Second, Carey considered unlimited liability for bank shareholders to be an important restriction on Scottish banking. 11 The lack of freedom to form limited liability corporations compelled risk-averse individuals to
67
68
Free Banking II
Laissez-Faire Monetary Thought
31
become depositors rather than shareholders. He believed that New England had a superior banking system because the availability of limited liability there reduced 'friction', by which he meant impediments to enterprise. Carey's second pamphlet repeated much of the detailed institutional discussion of the first, and added a more developed argument for free trade in banking. He began (1840, pp. 9-11) with a sweeping attack against restrictions on the application of capital and on freedom of trade. He concluded (1840, p. 75) with the argument that perfect freedom of trade in money is even more important than in other goods. Interference with trade produces irregularity in supply and price. Of all goods, money enters into the most contracts for delivery, so that irregularity of its·supply and relative price causes the greatest inconvenience. Under laissez-faire, the currency would be self-regulating: 'by establishing perfect freedom of trade, we should permit the supply and demand to regulate each other, thus giving a safe and steady currency'. Carey did not spell out any self-regulating mechanisms that would prevent over-issue by competing banks; nor did he develop a clear business-cycle theory. But he did argue (1840, pp. 7-11, 65) that 'unsteadiness is produced by restriction, causing capital to accumulate while the owners are seeking their means of investing it'. In contrast to restrictions that caused idle funds to be left temporarily on deposit, freedom to employ capital (particularly to form adequately capitalized banks) promoted soundness and steadiness. He cited as evidence (1840, pp. 41, 50) the relative mildness of business-cycles in Scotland, where banks could more freely be formed, in contrast to England. Carey (1840, pp. 16--17,51-61) saw the unlimited liability statutes of bank shareholders in both countries, however, as a r~striction ~hat discouraged equity-holding and was therefore destabilizing. The recent introduction of unlimited liability to Rhode Island, Carey (1840, p. 61) claimed, had led at once to an increase in the holding of currency and deposits and a reduction in economic stability. In the way of reform Carey (1840, pp. 55-7) explicitly rejected both higher tariffs (later in his career he switched sides on this question) and hard-money remedies. Higher tariffs would create idle capital and hence instability, while the substitution of specie for banknotes and checks similarly entailed 'a diminution of the facilities of trade ... Restrictions cannot give steadiness.' Repeal of usury laws would have a limited, though positive, effect. In order to employ the capitals of small savers, freedom to form limited-liability banks was paramount. Like Leggett, then, Carey (1840, pp. 69-72) favoured the enactment of a general law of incorporation to replace the legislative chartering of banks. He also rejected the bond-collateral requirement of the New York State law,
Free Banking II 32
Perspectives on the History of Economic Thought Vol. IV
though on grounds that it would discourage employment of capital in banks and thereby produce unsteadiness. He offered his own model statute allowing free banking with either limited or unlimited liability, requiring only that the balance sheets of limited-liability banks be made public. Carey characterized his proposal as one to 'ABOLISH ALL RESTRICTIONS, AND ESTABLISH PERFECT FREEDOM OF TRADE'. Yet Carey's understanding of the ideal of free trade in money was different from Leggett's. Carey (1840, pp. 63-4) condemned the Independent Treasury system as a source of instability. Later in his career, Carey's enthusiasm for stimulation of enterprise led him to support the issue of greenbacks together with free banking (Carey, I 865a, I 865b, 1866) as means to an 'adequate' supply of money. Together with his eventually ardent support for protectionism, these views indicate (if there is any consistency to be found) that Carey's support for free banking was not (like Leggett's) based on a normative commitment to laissez-faire, but stemmed instead from his belief that free banking would promote economic growth. Like Carey, Condy Raguet was an influential author who has been cited by historians of thought (Redlich, 1947, p. 202; Dorfman, 1966, p. 659) as a supporter of 'free banking'. Raguet favoured a national bank, however, and opposed Jackson's veto of the re-charter and removal of the federal deposits from the Bank of the United States. IS Raguet (1839) did support the New York 'free banking' law, but principally because of its scheme for collateralizing banknotes. His book made no argument for the application of the laissez-faire principle to banking.
Richard Hildreth Like Leggett and the early McVickar, Richard Hildreth grounded his argument for open competition in banking clearly on laissez-faire principles. Hildreth (1840, pp. 172, 149) insisted that the principle of free trade was just as applicable to banking as it was to other industries: 'As open competition has been found to be the best and safest regulator of all other kinds of trade, so it will no doubt prove the best and safest regulator of the trade of banking.' To establish open competition meant that '[t]he monopoly of bank charters must be abolished altogether ... Capitalists must be left as much at liberty to invest their money in a bank as in a cotton mill.' As especially relevant evidence of the virtues of open competition, Hildreth (1840, pp. 143-4) pointed to the free market for a product he thought quite similar to banknotes, namely bills of exchange. It would obviously be 'absurd and fatal' to grant a monopoly in that market, yet monopolies have curiously been thought proper in the supply of banknotes. Hildreth (1840, p. 165n) found it particularly anomalous that J. R.
69
70
Free Banking II Laissez-Faire Monetary Thought
33
McCulloch, 'a most uncompromising, and somewhat extravagant advocate of free trade in every thing else', was among the advocates of banknote monopoly. Hildreth's (1840, p. 171) case against monopoly of note-issue recalls both the Hayekian idea that centralized economic control requires an unattainable degree of centralized knowledge, and the public choice idea that government officials have their own interests: 'This idea of creating one great bank to superintend and control all the other banks ... supposes a superior degree of knowledge and of disinterestedness on the part of the men who may happen to be chosen directors of the great bank, quite superhuman.' The mistakes of the Bank of England during the suspension period, and of the Second Bank of the United States early and late in its career, provided evidence of 'how they may be deluded, and how readily they may become the instruments to delude others'. Hildreth (1840, p. 167) dismissed the notion, propounded by the Currency School in England, that a fixed artificial rule could be laid down for dictating to all banks appropriate expansions and contractions of the currency. The proper control of any individual bank's issues instead required practical knowledge gained through experience with actual decentralized conditions. Similar views were being expressed at the time by Free Banking School writers in England (White, 1984, pp. 130-6). A public-choice orientation may clearly be seen in Hildreth's account (1840, pp. 123-4) of the rent-seeking and log-rolling processes associated with the procurement of exclusive charters. In ascribing self-regulating properties to a convertible currency with plural issuers, Hildreth (1840, p. 88) exposited the needs-of-trade doctrine with unusual clarity. Banknotes cannot be kept in circulation unless 'demanded by the business wants of the community', these wants being identified as 'demand in the country for a circulating medium'. Or again (1840, p. 157): 'The amount of notes, payable on demand, that can be kept in circulation, is and must be limited by the demand of the community for a circulating medium.' Variations in the quantity of bank-issued money were the endogenous result of exogenous movements in the price level or real activity. These movements altered the nominal demand for money proportionally, the equilibrium ratio of nominal transactions to money (what we today call 'velocity') being assumed constant: The rise in prices, the activity of trade, the spirit of speculation, create a new use and a new demand for the circulating medium, the amount of which is increased accordingly .... These fluctuations in the amount of the circulation are absolutely necessary to keep up a due proportion between the medium of trade and the amount of trade.
Free Banking II 34
Perspectives on the History of Economic Thought Vol. IV
Hildreth (1840, p. 158) gave an unusually clear explanation of the adverse clearing mechanism that would check over-issues, even noting that increased gross clearing balances would cancel out in the event of a concerted system-wide over-issue: 19 If a number of banks issue a quantity of notes beyond what are needed for the operations of trade, those notes soon come into the possession of other banks, and when they are presented for payment, unless those other banks have also issued an extra quantity of notes, which having fallen into the hands of the first set of banks, will serve for the redemption of theirs, - the issuers will be obliged to redeem their excessive issue in specie, a process which will soon compel them to put a stop to that issue.
The greater the number of issuing banks, Hildreth added, the less likely is a system-wide over-issue: 'With a single great bank, or a small number of banks, an excessive issue may easily happen; but with the increase of banks the improbability increases of any such concert of action as would make it possible.' Thus a free banking system is more strongly selfregulating than a monopoly or restricted-entry system. A system-wide over-expansion in the currency would also be self-correcting in Hildreth's view, but by a different route: the price-specie-flow mechanism. A fall in the purchasing power of domestic currency would cause an external drain of reserves, forcing corrective contraction by the bank(s). Unlike Leggett or the British Currency and Free Banking Schools, Hildreth (1840, p. 198) thought this self-correction would operate so promptly that 'the direct power of banks to raise prices is very limited'. Consequently he rejected the monetary theory of business cycles, developed by those writers, in favour of the British Banking School views that macroeconomic fluctuations grew naturally out of speculative economic activity, and that any cyclical role played by banks derived from variations in their intermediational activity (1840, pp. 159-65). Unlike the Banking School, however, Hildreth derived implications favourable to free banking from this cycle theory.2o He argued that bank loans based on deposits were far more liable to fluctuate in volume than loans based on equity, because the volume of deposits varied cyclically. A greater number of banks, as would exist under free entry, implied to Hildreth an increase both in total banking equity and (the total sum of deposits apparently assumed unchanged) in the ratio of equity to deposit liabilities, and therefore an increase in 'the ratio which the stable portion [ofloans] bears to the unstable portion'. This greater stability in loans would 'give a comparative stability to trade'. With respect to reform measures, Hildreth, in his first book on banking (1837), argued only for open competition and against a national bank. An expanded version was published in 1840, just two years after the passage
71
Free Banking II
72
Laissez-Faire Monetary Thought
35
of the New York State Free Banking Act. He had now become a supporter of the compulsory bond-collateral idea embodied in that act, and even reprinted the act's text in an appendix (1840, pp. 200-9). Hildreth (,1840, pp. 154-5) justified compulsory collateralization in the nowstandard way as an effective means of protecting bank note recipients who were unable to judge every note's solidity.21 Once the notes were thus secured, he insisted, no other legislative interference was warranted. Stockholders and depositors could look out for themselves. The measure 'completely does away with that plausible argument in favor of monopoly' that free entry would mean unsafe notes. Likewise, Hildreth (1840, p. 184) concluded, it removed the safety rationale for banning small notes. Hildreth thus warrants being called a laissez-faire monetary theorist, though the laissez-Iaire part of his message (that compulsory security for notes was a sufficient regulation) was somewhat obscured by the interventionist part (that such a restriction was desirable).
Conclusion 'Free banking' legislation in the New York State sense, providing free entry for new banks but with a bond-collateral restriction on their issue of bank notes, spread to a majority of the states prior to 1860. The National Currency Act of 1863, which provided federal sanction for notes collateralized by federal bonds, may be viewed as 'a sort of national free banking act' (Rockoff, 1974, p. 142). The success of the movement for such legislation had two important effects. First, it defused for the time being any incipient movement for truly laissez-faire banking. Enthusiasm was directed towards winnable legislative battles for freer entry rather than toward developing a more idealistic position. There were no noteworthy American contributions to laissez-Iaire monetary theory in the next few decades after 1840, though briefs on behalf of state 'free banking' legislation continued to appear. Second, the bond collateral restriction of the National Currency Act proved to be far from innocuous. This restriction made the supply of currency notoriously 'inelastic', and thereby contributed importantly (along with other provisions of the Act) to the banking panics of the late nineteenth century (Noyes, 1910; Vera Smith, 1936, pp. 133-4). In having this impact, the regulatory provisions of the National Currency Act themselves set the stage for a post-bellum revival of the laissez-Iaire approach to money and banking.22
Acknowledgement The authors thank the Scaife Foundation for research support. Valuable comments on an earlier version were received from Milton Friedman, Thomas F. Huertas, Bala Subrama-
Free Banking II 36
Perspectives on the History of Economic Thought Vol. IV
nian, Larry Schweikart. Richard H. Timberlake and participants in the Austrian Economics colloquium at New York University.
Notes I.
2.
3. 4. 5.
6. 7.
8.
9. 10.
II. 12.
13.
See for example Black (1970). Rockoff (1974), Klein (1974). Hayek (1978), Fama (1980). Greenfield and Yeager (1983). Rolnick and Weber (1983; 1984; 1986), White (1984), and Selgin ( 1988). Nataf (1984a, b) has recently discussed French writers; White (1984) the British literature; and Cowen and Kroszner (1987) writers who associated lai.vse::-faire with separation of the unit of account from the medium of exchange (anticipating a particular strain of the recent literature). The most important older secondary account of laissezIaire banking thought is Vera Smith (1936). who surveys the monetary regime debates in several nations but neglects the US debate before the Civil War. Redlich (1947) discusses the American opposition to a national bank and the movement for bondcollateralized 'free banking' legislation in New York State, but says little about theoretical economic arguments. Madeline ( 1943) discusses events rather than theories. Mints (1945, pp. 138-41) and Miller (1927. pp. 159-65) discuss views on competition in noteissue only briefly and unappreciatively. Wilhite (1958) discusses Webster and his contemporaries. On the early history of New York State banking legislation, see Hammond (1936, pp. 184-96). An early British monetary writer to make this extension explicitly was Lord King (1804, pp. 116-20). On the needs-of-trade doctrine in the British literature see White (1984, pp. 90, 122-6). The proposition of a demand-determined currency stock was present in Adam Smith (1981, p. 300). The confusion between money and credit, and consequent misunderstanding of the money supply process, persists today. For a critique, see Greenfield and Yeager (1986). On the similarity of McVickar's views to those of McCulloch and Ricardo see Redlich (1947, p. 294, n. 47). Redlich notes that the same combination of Smithian free banking with Ricardian bond-collateral found in McVickar is also found in Parnell (1827). Ricardo and McCulloch both eventually advocated bank note monopoly, however, while Parnell advocated free entry. Dorfman's (1966, pp. 516-22,713-20) secondary account of McVickar's works fails to note the inconsistency in McVickar's views. The same inconsistency exists in the writings of McVickar's student William Beach Lawrence, who proselytised both for 'free banking' and recharter of the Second Bank of the United States (see Dorfman, 1966 pp. 720-31). Madeline (1943, pp. 45-51) discusses Jackson's plans for a new national bank more favourable to his political interests. Explicit evidence that Jacksonians were following the British free banking literature may be found, however, in Senator Thomas Hart Benton's (1854, pp. 188-9,262) citation of Henry Parnell's and others' speeches and writings. Benton himself took a hard-money line, but occasionally (e.g. 1854, pp. 226, 248) made statements-favourable towards Scottish-style free banking. See also Redlich (1947, pp. 295,n. 62) on citation of Parnell by Samuel Young, sponsor of an early New York State free banking bill. For an elaboration of this critique, see White (1984, pp. 120-2). The real-bills doctrine as a guideline for prudent lending is another matter. Adam Smith argued that lending on real bills had both virtues. Elsewhere, Leggett (1984, pp. 146, 187) argued that depreciation of over-issued notes would provide the check against excessive issues by any single bank. This is consistent with the adverse clearing check if depreciation makes it profitable for other banks and brokers to acquire and redeem the notes. There are linkages here: the British Currency and Free Banking School writers influenced both Leggett and later Wicksell, Mises and Hayek. Lucas, in turn, has acknowledged an affinity with Hayek's work.
73
74
Free Banking II
Laissez-Faire Monetary Thought 14. IS. 16.
17. 18. 19.
37
The evolution of Leggett's monetary reform proposals is treated at greater length by White (1986). The same evolution may be seen in the Loco-Foco movement more generally; see Redlich (1947, pp. 189-90), who surprisingly does not cite Leggett. Alexander Ming was a prominent Loco-Foco. Dorfman (1966, pp. 789-805) and Sharkey (1959, pp. 153-71) discuss Carey's thought and influence. Carey's greatest renown in America was later gained, incongruously, as a protectionist. In France, however, he was well known among economists as a free banking advocate (Nataf, 1984b). The same conclusion has been reached recently (on different grounds) by Carr and Mathewson (1988). Dorfman (1966, pp. 363, 376, 607-12, 659) provides an intermittant secondary account of Raguet's evolving banking views. Another spurious 'free banking' advocate was Charles Duncombe (1841). In fact, cancellation would take place only in the mean. Concerted over-issue would still increase the variance of daily clearings against each bank, rendering its reserves inadequate and prompting it to retreat to its original level of issues. See Selgin (1988, pp. 80-
2). 20.
21. 22.
John Stuart Mill (1848, pp. 675-6), James Wilson (1847, pp. 30-35. 97-104) and Thomas Tooke (1844, pp. 44-5) of the Banking School all defended competition in note-issuing, but not because they thought it would moderate business cycles. Tooke had earlier (1840, pp. 202-7) been quite hostile to free banking. On the evolution of Tooke's policy views, see Arnon (1984). Hildreth (1840, p. 155) suggested indirectly that the acceptance of notes was in some sense involuntary. A companion piece, Selgin and White (1990), discusses this post-bellum literature.
References Arnon, A. 1984. 'The transformation of Thomas Tooke's monetary theory reconsidered" History of Political Economy 16 (2) (Summer). pp. 311-26. Bagehot, W. 1873, Lombard Street, Henry S. King. London. Benton, T; H. 1854, Thirt)' Year's View. vol. I. D. Appleton. New York. Black, F. 1970, 'Banking and interest rates in a world without money: the effects of uncontrolled banking', Journal of Bank Research 1(3) (Autumn), pp. 9·-20. Bryant, W. C. 1839, 'William Leggett'. US Magazine and Democratic Rel·iew. 6 (July). pp. 17-28. Byrdsall, L. 1967, A History of the Loco Foco or Equal Rights Party. Burt Franklin. New York. Carey, H. C. 1838. The Credit System in France. Great Britain. and the United States. Carey, Lea, & Blanchard, Philadelphia. Carey, H. C. 1840, An.nt·ers to Questions: What Constitutes Currency? What are the Causes of Unsteadiness of the Currency? and What is the Remedy? Lea & Blanchard. Philadelphia. Carey. H. C. 1865a, The Currency Question. Collins. Philadelphia. Carey, H. C. 1865b, Lellers to the Hon. Schuyler Colfax. Collins. In Carey (1875), Philadelphia. Carey. H.C. 1866, 'The National Bank Amendment Bill'. reprinted from The North American and United States Gazelle, April. In Carey (1875). Carey. H. C. 1875, Mi.\·cellaneous Papers on the National Finances. the Currency. and Other Emnomi£" Subjects. Henry Carey Baird and Co .• Philadelphia. Carr, J. and Mathewson. F. 1988. 'Unlimited liability as a barrier to entry'. Journal of Political Ecomony 96(4) (April). pp. 766-84. Cowan, T. and Kroszner. R. 1987. 'The Development of the New Monetary Economics', Journal of Political Economy, 95(3) (June), pp. 567-90. Dorfman. J. 1966, The Economic Mind in American Civilization. vols I and II. 1606-1865. Augustus, M. Kelley. New York.
Free Banking II 38
Perspectives on the History of Economic Thought Vol. IV
Duncombe. C. 1841. Dllncomhe·.~ Free Banking. Augustus M. Kelley. New York (1969 reprint). Fama. E. 1980. 'Banking in the theory of finance'. Joumal I!l Monetary Ecolllmric.l· 6(1) (January). pp. 39-57. Gouge. W. 1833. A Slrort History I!f' Paper Money and Banking. T. W. Ustick. Philadelphia. Greenfield. R. L. and Yeager. L. B. 1983. 'A laissez-faire approach to monetary stability'. Joumal I!l Money. Credit and Banking 15(3) (August). pp. 302-15. Greenfield. R. L. and Yeager. L. B. 1986. 'Money and credit confused: an appraisal of economic doctrine and Federal Reserve procedure" Soutlrern Economic Journal 53(3). pp. 364-73. Hammond. B. 1936. 'Free banks and corporations'. Journal of Political Economy 44(2) (April). pp. 184--209. Hayek. F. A. 1935. Prices and Production. 2nd edn. Augustus M. Kelley. New York (1967 reprint). Hayek. F. A. 1978. Denationalisation of Money. 2nd edn. Institute of Economic Affairs. London. Hildreth. R. 1837. Tire History l!lBanks. to II'hiclr is Addeda Demonstration of tire Advantages and Neces.l·ity of Free Competition in the Busine.u of Banking. Hilliard. Gray. Boston. Mass. . Hildreth. R. 1840. Banks, Banking. and Paper Currende,~. Greenwood Press. New York (1968 reprint). Joplin. T. 1826. An Essay on tire General Principles and Present Practice of Banking in Engluncl ancl Scotland. 5th edn. Baldwin. Cradock & Joy. London. King. Lord P. 1804. Tlrouglrts on the Restriction of Payments in Specie. 2nd edn. In Earl Fortescue (ed.). A Selection From the Speeclres and Writing.~ of the Late Lord King. Longmans. 1844. London. Klein. B. 1974. 'The competitive supply of money'. Journal of Money. Credit. and Banking 6(4) (November). pp. 423-53. Krooss. H. E. (ed.) 1977. Documentary History of Banking and Currency in the United States. Chelsea House. New York. Leggett. W. 1984. Democratick Editorials: Es,w),s in Jacksonian Political Economy. ed. L. H. White. Liberty Press. Indianapolis. Lucas. Jr. R. E. 1975. 'An equilibrium model of the business cycle" Journal of Political Economy. 83(6) (December). pp. 1113-44. Madeline. Sister M. G. 1943. Monetary and Banking Theories ofJacksonian Democracy. n.p .• Philadelphia. Marcus [pseud .. for Benjamin Franklin Butler] 1818. Remarks on Private Banking. E. & E. Hosford. Albany. McCulloch. J. R. 1826. 'Fluctuations in the supply and value of money" Edinhurgh Reviell' 43(86) (February) pp. 263-98. McVickar. J. 1827. Hints on Banking. in a Letter to a Gentleman in Albany: hy a Nell' Yorker. Vanderpool & Cole. New York. McVickar. J. 1830. Introductory Lecture to a Course of Political Economy. Recent(I' Delil'ered at Columhia College. Nell' York. J. Miller. London. McVickar. J. 1841. A National Bank: its Nece.uity. and Mo.~t Advisable Form. n.p .• reprinted from an article in the Nell' York Reviell'. New York. Mill. J.S. 1848. Principles l!l Political Economy. Augustus M. Kelley. [1973 reprint]. New York. Miller. H. E. 1927. Banking Theories in the United States Be/c!re 1860. Harvard University Press. Cambridge. Mass. Mints. L. 1945. A History I!l Banking Theory. University of Chicago Press. Chicago. Mushet. R. 1826. An Allempt to Explainfrom Facts the Effects of the I.uues of tire Bank of England upon it.l· Oll'n Interests. Puhlic Credit. and Country Banks. Baldwin. Craddock & Joy. London. Nataf. P. 1984a. 'Business cycle theories of mid-19th-century France'. Unpublished ms .• University of Paris.
75
76
Free Banking II Laissez-Faire Monetary Thought
39
Nataf. P. 1984b. 'Competitive banking and the cycle'. Unpublished ms .• University of Paris. Noyes. A. D. 1910. History of the National Bank Currency. Government Printing Office, Washington. Parnell. H. 1827, Observations on Paper Money, Banking and Overtrading, James Ridgeway, London. Parnell, H. 1833. A Plain Statement of the Power 0/ the Bank 0/ England and the Use it has Made of it. 2nd edn. James Ridgeway, London. Raguet. C. 1839. A Treatise on Currency and Banking. Grigg & Elliot, Philadelphia. Redlich. F. 1947. The Molding lJ/' American Banking: Men and Ideas. (Part I. 1781-1840), Hafner. New York. Ricardo. D. 1951. The Works and Correspondence 0/ David Ricardo. vol. 4. Pamphlets and Paper.\' , 1815-1823. ed. P. Sraffa. Cambridge University Press. Cambridge. Rockoff. H. 1974. 'The free banking era: a re-examination'. Journal of Money, Credit, and Banking 6(2) (May). pp. 141-67. Rolnick. A. J. and Weber. W. E. 1983. 'New evidence on the free banking era'. American Economic Review 73(5) (December). pp. 1080-91. Rolnick. A. J. and Weber. W. E. 1984. 'The causes of free bank failures: a detailed examination'. Journal of'Monetar\, Economics 14(3) (October). pp. 267-91. Rolnick. A. J. and Webe'r. W. E. 1986. 'Inherent instability in banking: the free banking experience'. Cato Journal 5(3) (Winter). pp. 877-90. Say. J. B. 1971. A Treatise on Political Economy. Augustus M. Kelley. New York. Schwartz. A. J. 1947. 'The beginning of competitive banking in Philadelphia: 1782-1809'. Journal of Political Economy 55(5) (October). pp. 417-31. Selgin. G. A. 1988. The Theory of Free Banking: Money Supply Under Competitive Note Issue. Rowman & Littlefield. Totowa. NJ. Selgin. G. A. and White. L. H. 1990. 'Laissez faire monetary theorists in late nineteenth century America'. Southern Economic Journal 56(3) (January), pp. 774-87. Sharkey. R. P. 1959. Money, Class, and Party: An Economic Study of Civil War and Reconstruction. Johns Hopkins University Press, Baltimore, MD. Sharp. J. R. 1970. The Jacksonians ver.\'us the Banks: Politics in the States after the Panic 0/ 1837, Columbia University Press. New York. Smith. A. 1981. An Inquiry into the Nature and Causes of the Wealth 0/ Nations eds R. H. Campbell. A. S. Skinner. and W. B. Todd, Liberty Classics. Indianapolis. Smith. V. 1936. The Rationale o/Central Banking, P. S. King, London. Taylor. J. 1969. An Enquiry into the Principles and Policy of the United States (ed. L. Bartiz). Bobbs-Merrill. New York. Tooke. T. 1840. A History of Prices. vol. 3. Longmans. London. Tooke. T. 1844. An Inquiry into the Currency Principle. 2nd edn. Longmans. London. White. L. H. 1984. Free Banking in Britain: Theory, Experience, and Debate, 1800-1845, Cambridge University Press. Cambridge. White. L. H. 1986. 'William Leggett: Jacksonian editorialist as classical liberal political economist'. History 0/ Political Economy, 18(2) (Summer). pp. 307-24. Wilhite. V. G. 1958. Founder.\' o/American Economic Thought and Policy. Bookman Associates. New York. Wilson. J. 1847. Capital, CUrrelll)', and Banking. The Economist. London.
[5] Laissez-Faire Monetary Theorists in Late Nineteenth Century Anlcrica* GEORGE A. SELGIN LA WRENCE II. WHITE Ullil'ersily (!f Georgia Alilel/S, Georgia
I. Introduction
Economists have recently begun to reconsider many forms of regulation of money and hanks from what may be termed a laissez-faire perspective. Some have questioned fundamentally the rationale for any government role.' This development raises an interesting question for historians of thought: what intellectual support did the concept of a free-market monetary system receive in the 19th century, the heyday of classical liberalism and laissez-faire thought'! This paper takes up the question by critically surveying American monetary writings from the last half of the 19th century.l We find that the idea of a laissez-faire monetary policy was never very popular in the United States, though it did have proponents whose work was often insightful. Most of the relevant literature was written in response to consequences wrought hy the monetary legislation of the Civil War. That legislation included not only the National Bank Aets (IX(,)-4), but also an ancillary measure (effective IX(6) placing a prohibitive tax on state hank notes. These laws forced hanks of issue to hecome National Banks, and till" to huy federal ho"tis as collateml for their notes, which had the intended effect of creating an extra demand for the fedel'lll government's wartime debt. In addition, Congress authorized the isslle of legal tender "grcenlmck" currency by the Treasury. A frequent theme in the post-bellum monetary literature was the alleged deficiency or "inelasticity" of the currency supply. Some-the greenbackers and silverites-wanted the government to increase its contributions to the money supply. Others, especially toward the end of the century, argued for the establishment of a centralized agency with special powers of note issue. A handful of writers renewed the radical Jacksonians' campaign for laissez faire in money and banking:' In their view, the defects of existing arrangements could best he solved by deregulation of hanking and by removal of the federal government from the husiness of slipplying money. ·For comments (m an (' ... rlier dran we thnnk Milton Friedman. Thomas F lIucrtas, Bida Suhrnl1lani;lO. Larry It. Timherlakc. 1. Recent works defending or at least ~ymp;1thctical1y cXHmininJ!, the idea of a lais",c7·f:lirc l11onet:lry system include RI.ck IIJI. Fam. 1221. Friedmen and Schwarlll241. Greenfield and Yeeger 12{;1. Hayek 1271. Klein 12RI. Selgin
Schwrikarl end Richerd
1441. Vauhel 15.11. Wallace 1571 and While 11iIl1. 2. A companion piece 1621 deals with the antc-hcl1uT11 American lifer.ature <1nd cites the sC'condary liferature frcatin~ other nations. J. We confine our disc.:lIo;;sion tn writers who :lcceptcct a specic standard;1" the nntmHI found;ItinTl for;, 1;1i"'sl~7.·faire
774
78
Free Banking II LAISSEZ-FAIRE MONETARY THEORISTS
775
II. Mainstream Atlitudcs toward Rcgulation of Moncy and Banking If thl!rl! is a stallllard vi!!w of nin!!tel!nth-cl!nlUry Am!!rican monetary thought, it is that hlissl!zfaire ideas were commonplace. Lloyd Mints 131, 41, widely regarded as the authority in this arl!a, offers the following judgment: During the nineteenth century adherence to the concepts of It/i.\·.w!Z It/ire led many to helieve tlmt the issuance of hllnk notes against fractional reserves was a 'right' of anyone who desired to engage in this uctivity. It was further contended by many thut no regulation or control of the h'lnks was necessary, that free and competitive banking would operate to promote the welfure of the community. While these views were not unllnimously held, it is dellr that they ... were adequale to prevent ratiollal legislation on this subject.
In fact most banking authorities, in the latter as well as in the earlier half of the century, continued to tr!!at money and banking as exceptions to the general principle of laissez faire. This was true even for some of the principle's more uncompromising proponents, including William Uraham Slunner, Francis Walker, and Simon Newcomb. Sumner, who has been described 1461 as "an unequivocal exponent of the doctrines of economic individualism in their most extreme form," held banks 10 be inherently inflationary and in need of regulation to control their issues. For this reason, and despite his view of the Jacksonian Bank War as "one of the greatest struggles between democracy and the money power," Sumner believed that the First and Second Banks of the United States had been needed to restrict the Slate banks 149, 265; 48,631: He also defended other regulations designed to limit the issues of private banks, such as prohibition of small notes. Like many of the Jacksonian Em hunl-lIIollcy school, Sumner believed that prohibiting small notes would, by keeping specie in circlll'llion, somehow compel banks to hold adequate reserves. Apparently he did nut consider that hanks' Ile!!d to huld reserves fur the settlement of interbank clearings would impose an adeqllate check Ull their expansion. Sumner also refused to be persuaded by the example of Scottish free banking 148,56, 1161. lie attributed its success not to the system itself, but tu the peculiar "Illoderation, sagacity, and scicntilic knowledge" of Scollish bank managers. "Without thes!! qualities in the managers," lit: wrotc, free banking along Scollish lines would be "as wild as any scheme of p'lpcr lIIuney." , Views similar to Sumner's were expressed by Francis Walker, whose works on /'o/ili('(// H(,OIlOIllY 1551 and MOlley 1561 were cspecially popular. Though generally opposed to government intervention, Walker rejected the view that the money supply could he left in the hands of private producers 155, 2!!6-7, 461 III He c1aimcd, for cxample, that governmellls had monopolized the proJuction uf coin in order tu protcct the "poor and ecunolllically wellk." lie objccte
available ,lali,lie,. Sec Van Fenslermaker 15t; 521. 5. The Scoui!:lh bank!) were prohibitcd from i!)!)uing nOlcs ulU.lt..!r I pound an.:r 1765. SUllln.:r did nul rl.'gard Ihal kimJ uf IUcu~urc lakl.!'l1 alullt..! a:, suftich!nllo guard again~l ovt..!rcxpan~ion.
79
Free Banking II 776
Gcorge A . .'ie/gill alld Lall"rellce II. While
to circumstances unique to Scotland. These were (I) the ability of their managers to ascertain the value of real estate collateml owned by would-he horrowers (due to Scotland's comprehensive system of land registration); (2) the unlimited liahility of unchartered banks; 0) the fewness and great strength of Scottish banks;' and (4) "the shrewd Scottish sense and the strong Scottish will." /laving thus accounted for the special record of the Scottish banks, Walker inconsistcntly went on to arguc that their record was not special after all. lie endorsed the Currency School writer Lord Overstone's assertion that Scotland under free banking was just as suhject to commercial pressures, economic stagnation, and losses due to insolvency, as England during the same period. 7 Another representative writer of this era, known for the sophi~tication of his monetary theory, was Simon Newcomb. A general supporter of laissez faire (which he termed the "let alone" principle),' Newcomb made an exception for the issue of bank notes 135, 443-561. His view was that note issue is "not performed in the best manner when left entirely to private enterprise. because some of the conditions which insure good performance are wanting." Newcomb refcrred to past "evils suffered by leaving banks free to issue notes at their own pleasure."· /Ie even doubted whether note issue "should be considered a legitimate 1function 1 of any bank" 1.15. 1661. /lowever, since he also disapproved of government-issued currency, Newcomh concluded by advocating a government-regulated private monopoly bank of issue, similar to the Bank of England 135, 5101.'" Although he was less well known than any of the previously mentioned authors. Sidney Shcrwood's reaction 145, 2771 to the idea of an unregulated monetary system is indicative or the mainstream sentimcnt of his em: I think that you would not lind any strong party in the United States at present wishing to go hack Isicl to a system of perfectly free !lanking or hanking without governmental regulation. anymorc th:m you would lind a strong party advoc:'ting our going hack til that condition where privatc tinns were alillwed til issue cllins."
III, Moran, Wilson and Sturtevant The first substantial post-helium work to argue for a laissez-faire monetary policy was Charlcs Moran's MOlley, Currellcies, alld Ballki,,!( 1321. Moran was critical of metallic money as well as government paper money. lie argued that the money supply should instead consist solcly of "hank notes and bank credits, where banking and hank isslIes are left entirely free. controlled Ii. Walker cites the figure nf' t t hanks of issue existing in tR73. Entry intn the nnte-issue husiness. h"",e\w. had heen dnsed ofI since IR44. In that }ear. Scotland had 19 hanks nf issue. Moreovcr. conccntrution of hanking """ nfl
peculiar feature of the Scottish syslcm. It was simply
fI
consequence nf freedom of note issue cnl11hincd wilh unrL'slrit:led
branch hanki"g ~md could he secn in Camuin under similar circumstances. 7. Contrust White 160. 44-4'11. whn aftrihutes the relative cyclical stahility of the Scnftish system tn ilS rcla,i\·cly li1issc7.-I~lirc
character. R. Newcomh did. howevcr. distinguish the "let alnne" principle from what he called thc "keep flut" I'rindrl". The former opposed govcnlmcnf interference wilh privnte enterprise: the latter opposed J!(lvcnll11cnt cntcrpriq· n.r . government production of guods and services) not conncc-tcd with pmlCClin!! the rights of citizens. Nrwcol11h did nlll accept the "keep out" principle as a gcnrml nile.
"r
'l. If Newcomh "'as referring tn the ante-helium "free hanking" sy"en". he had failed to appreciate 'he role regulatin" in afTecting their performance. Sce RocknfT 1401. Rotnick and Weher 141; 42: 4~1 and White 1611. to. ror New"omh's criticisms of the Legal Tender Acts sec A (',.i/;cal To."m,;/la/;"" 1~41. II. Shcrn'(lOd's slalcmrnt recalls n more rcccnt stalement made hy Friedman 12.'. 2201: .. J\1I(1win~ hank!' III i"'''II(, ... haml·lo-h:md CUrT'Clll'y . . . has lillie or nn support arnon~ cconnmistc;, hunker.. , or the p"hlil'."
Free Banking II
80
LAISSEZ-FAIRE MONETARY THEORISTS
777
by nOlhing but the enlightened self-interest of bankers and their dealers." Such .111 arrangcment would eliminate monetary crises which were due exclusively to "crroneous legislation in regard tu banks and currcncics" 132,4, 151." Moran was a proponent of anti-Bullionist and Banking School views." His theory of the self-regulating nature of bank money resemhled thc real-hills doctrine, although he did not insist that banks hase their issues on short-term commercial loans. His view seems to have been that, so long as hank money was issued in loans for which interest was paid, a Fullartonian "law of Ihe rellux" would guard against overexpansion. That is, any exces~. issue of notes would be quickly returned through loan repayments." Signilic'llltly, Moran explicitly denied that convertibility into specie was needed to limit bank issues 132, 261. Note issues occur only "after industry has produced something useful that is required to be forwarded from the producers to the consuillers." The purchases of consumers .. furnish the means of returning the credits to the banks in liquidation of the loans which caused thcir issue, rendering it wholly unnecessary to redeem bank issues in coin." Indeed, Mown held that guarantees of convertibility "should never be imposed or assumed." He thought that such guarantees would eventually be voluntarily ahandoned in a deregulated system 132, 471. In saying all of this, Moran was true to his anti-Bullionist convictions. I' I-!owever, those convictions-which led him to argue that the Bank of England could not overissue, even during the Restriction, and evcn though it had lIIonopoly privileges-were inconsistent with Moran's insistence on the need liJr unrestricted competition in banking. The contradiction between Moran's anti-Bullionisfll and his defemie of free hanking W,IS 1Il0st apparcnt in his positive appraisal of the Scottish systems 132, 14-151. Muran pointcd to Scotland's relative immunity (during the free banking era) from monetary crises that aJ'ili,ted England throughout the same period. lie noted, furthermore, that thc criscs "extcndcd th<:m,clves to Scotland" after Peel's Acts of 1844 and 1845. Yet, Moran's anti-Bullionist perspective cannot account lilr why the restricted English banking system should have been more prone to IIHlnetary crises. Nor does it jibe with his observation that the Bank of England's powers of issue w<:r<: "grcatly incr<:ased by its monopoly of the issue of notes in England within 65 miles of I.onllon." Finally, Moran recognized that the issues of the Scottish banks werc regul,tted by "the exchange (/1/(/ redemplio/l of their notes twicc a week in Edinburgh, and twice a week in Glasgow" (<:lllphasis added). Thus when discussing the actual operation of an IInreguhlted syst.:m, Moran IIIlCOtlSCillusly abandoncd his anti-Bullionism, with its Fullartonian "law of the reflux." Instcad, he rd<:rred to adverse clearings-which depend on competition and conwrtibilily-as the mechanism actually limiting issues. This tacit theory-switching enabled Moran to argue lilr unregulated banking despite his endorsement of anti-Bullionist and Banking Schools doctrines which treat competitilltl as immaterial to preventing over-issue. Later writers who were sympllthetic til thc 12. Moran rc<.:ognizcd thul crisis IIlight also lll'isl! frolll real JblUl'banccs, su~h as wars alld (.;mp failurc~. II[ from fustcn.:d by cxccs~ivc cOldiJClh..:C 132. 4~-491, Nevel d .dcss, crisi:, would only involve munetary Ji:-.lulbalKcs illlhL' prcscn..:c of Icgall'cslril:li\m~ on muncy and banks. for furtlll:r cvidCIH.:c of Moran\ theory of cri:-.b sec "Palik's i.IIlU Pulitks l ' 1331. 13. Mnri.lll's argulilents on this senl'!.; WI.!I\: Hot particularly novel ano WUlJlJ nol witilslanJ ('fitiL'~tI anal),!)>':s like those ufVillcrl54, 145fr.1 and Minis 1311. 14. AI 0111': poillt Moran even implied Ihat the n:gulalcd nOle issucs of thl.! Nutiou:..1 Banks wcr~ gO"CnlI.!O by the IIccJ, uf lraJc 1.l2, 551f.1. Fur Fullartuu', thwry. ,cc While 160, 126-2KI. 15. MlI~1 wrikl's, illdudillg IlIcmbers uf the Bmlking Sl.:hool, illsbtl.!o 011 Ihe IIL'~J for l:\)Ilv~rtihilily eVL'n IlhHlgh 111L'if I'l.!al-bilb Ih~uriL's implies lilill convel'tibility Wi.lS IlOl essential 10 prcvelll ovcris~IIL'. AUOlhcr Allh:riC<J1I prllpOlll.!lI1 of LUII..,istcnt allti-Bulliullislll was Slephcn Cl)lwcli 1161. ~p~cUlalivl! Illaiinvcstlncnt
Free Banking II 77R
(I('o/,Xe A. S('iXill alld Lal\'/'('I1("('
H. Whil('
theories of the Banking School managed with little difficulty to incorporatc them into proposals calling lilr increascd ccntrali7.ation of currcncy supply. Two othcr carly post-bcllum writers, Gcorgc Wilson and Julian M. Sturtcvant. also argucd for governmcnt's withdrawal from the bitnking husiness. and did so beliJre thc call liJr hanking reform hec,lIll(, widespreml. They did not, however. olrer any detailed analytical support for their views. Wilson was a pamphlcteer who was mainly interestcd in opposing all manner (If government-supplied l11onics. Besides attacking the Legal Tender acts responsiblc lill" the issuc of the greenhacks. he opposed the government's coinagc monopoly and the "communistic" National Uanking system 164, 46-91. He insisted that banking was "no mo~e the business of a govcrnment than any other calling." Sturtcvant cxpressed his views more soberly, in an economics texthook. Observing tlmt commodity money had come into usc without any assistance from government. he argued that bank credit would also dcvelop in soundest fashion if left free to evolve on its own 147,971: Ir an individual or a privale co-paJ1nership can procure so mueh credil in a communilY Ihal Iheir notes payaole on demand will circulale as a medium or exchange. we know no reason why Ihe law would interrere helween them and the pUhlic. Each man may be safely lefl to lake care or himselr. Although he recogni7.ed that the volume of commercial liabilities was capahle of influencing prices, Stur1evant denied tlmt this implied the need for a central bank. lie likewise rejected other means for artificially regulating or limiting bank issues, such as the hond-deposit systcm which, he noted, required maintenanec of a large govcrnment debt. Sturtevant argucd that all the government need do, in addition to permitting "the IHltural and spontaneous developllleni or credit," is to atteml to "its own proper function of rigidly enforcing all contracts ;rccording to the true intent and meaning thereof" 147, 1001.
IV. Edward Atkinson Though most of his writings on the money qucstion appcarcd after thosc of Moran, Wilson. and Sturtcvant. Edward Atkinson's first criticisms of government involvement in money and banking wcre puhlished in the mid-IR70's. lie therefore deserves to be viewed as one of the earlier post-bellum advocates of laisse7. faire in money and banking. A classical liberal who hased his views on thc writings of Adam Smith and Frederic Bastiat, Atkinson had personal contacls in Washington rivalling those of economist Ilenry Carey (who ironically, given his earlier slIrport for free hanking. hecame onc of Atkinson's leading opponcnts on the IlH)ney and tariff isslleS)."· Atkinson's views on monetary rolicy reflect his understanding of money and hanks as products of spontaneous evolution. In a discussion that compares favorahly with Menger's cl:!'sic alwlysis I JOI, he descrihed thc origin of commodity money. Drawing on Ridgeway l:llj I. i\lkin'0I1 cxplained that gold (or, in more primitive societies, less valuahle precious metals) was adopted :1$ a medium of exchmlge and unit of account "without process of law. rule, or statule" III. Y,OI. Ilis view that natural developments had encouraged the usc of a convenient and reliahle stand:ml lE"d him to regard Icgal tcnder legislation as pointless except as a device for governments to dl'fratld
Ie). Sl'e WilliulllSOIl
1('.'1
f(lr (In intcllc('ltHlI hioJ!rllphy or Alkinson.
81
82
Free Banking II LAISSEZ-FAIRE MONETARY THEORISTS
779
the public by forcing the use of bad (overvalued) money in placl of good money III, 344-51." In other words, legal tender laws caused Gresham's Law to take effect. Elsewhere Atkinson gave a good account of the outgrowth of modern banking from deposit-taking by goldsmiths 121." Atkinson's critical stance toward legal lender legislation caused him 10 oppose both greenb,lcks and bimetallism. The Legal Tender Acts had been passed to confer a "Ii)rced Imll1" to the government to pay for the war. Whether or not this had been justilied during the war, there was no longer, in Atkinson's opinion, any justification for the Acts. He therefore campaigned for rapid resumption of specie payments, and argued that surplus revenues should be used to retire ,IllY greenbacks left in circulation after resllmption was achieved 14; 51. Atkinson's opposition to bimetallism was not so uncompromising: consistent with his laissez-faire views, he did not oppose silver money per se, but only laws aimed at fixing a mte of exchange between gold und silver currency. In letters to Grover Cleveland and to William Jennings Bryan, he recomlllelllied a compromise: the issue of silver purchase certificates that would exchange against gold al a mte established by the free murket.'· In Atkinson's view, an ideal system would have its currency supplied exclusively by private banks. Atkinson took exception to those hard-money writers who had shown "bitter prejudice against banks and bankers" and contended that banks should be "left as free from legal restrictions as possible" 19,4, 141. Like many previous proponents of unregulated banking he alluded to two (not necessarily consistent) reasons tor believing that unregulated banks would not overissue. In an carly work, Atkinson argued that the banks' need to hold reserves lor maintaining convertibility restricted their issues appropriately III. Later, however, he upheld the real-bills doctrine, c1aililing Ihat bank issues would automatically adjust to agree with the amount of real property "heing moved or assigned at any given time": If the system uf banking is sound and bankers arc prudent, the sum of bank-nutes, bank-dcp()~its, and uther fimlls by which titles arc transferred to property on its way to conSUlllers, can never exceed the nominal value of thc commodities: hence money is said to be plenty or otherwise, whcnthe 4uantity of commodities is abundant or otherwise 13, 13,231. Atkinson was apparently oblivious to the problem of nominal indeterminacy involved in this Banking School argument-a problem transparent in his own version of it. lie had lost sight of the critical role of convertibility and a limited reserve supply in limiting bank issues. Although his later theory of unregulated bank issues was flawed, Atkinson's criticisms of existing regulations were generally cogent. He complained that the government-bond-holding reljuirements imposed upon issuing banks had wilhdrawn loanable funds from commercial uses 19, 51. lie also pointed oul that Ihey were a cause of the perverse elasticity of the money supply that contribllled 10 financial panics. 20 Specifically, he noted Ihat banks' inability 10 increas\! profitably their note issues at limes of high demand tor currency (n:lative to lotal money demand) led to withdrawals of high-powered money from bank reserves. This in turn caused an und\!sired contraction of tolal money supply 17, 5631. At other limes a decline in the demand li)1'
17. Sec ab,) ALkill>on', IM94 leuer 10 H. D. Macleod. "uolcd in Wiliialll..1Illb3, 303nl. 1M. Compare l'uwcll13bl and Richards 13MI. 19. See Williall,"un Ib3, t411. Fur a cmuplcle 1i,1 of Alkinson's sub,Lanlial wrilillg' un bimelallblll consul! Ihe bibliugraphy in Ihis work.
2(). For Atkin~on'~ ~cneraJ views on IIAe causes of business cycles
1101·
s~c
"Cause alld KCllh!c..Iy I(}r
Businc~s Ikpn.'s~il)n"
Free Banking II 7XO
Georg(' A. S('/gill alld L(/lI'rell(,(, H. While
currcnl'y could have the opposite erfect. ellusing a return of high-powered money to hank reserves and :m unwarranted expansion of total money supply. St:ltutory reserve requirements aggmvated thcse elleets: lJllller Ihe present l'omlitions of compulsory reserve on deposits ami investmenl of capilal in honds. the hank in!! comnnll1ity is kgally lilrhidden either to extend Mlpport 10 IIll'rl'iHlnls or ii' noll' cirl'ulution al Ihe time when hoth arc most nee,ied. On Ihe other hand. hanks nHly he. :md often ure. oppressed hy the accunllllillion of governmenl noles which eunnot he used excepl in unwholesome speculation at :1 period when there is little or no call for slllull nole circulalion II). 121. Besides favoring the repe:ll of hond-eolhlteml and st:ltutory reserve requirements, Atkinson argued for a discontinuution of the prohihitive 10% tax on state-bank note issues, which :liso interfered with desired adjustments in the currency component of the money stock IXI. In
V. William Urollgh The panic of IRI)3 brought a great outpouring of monetary puhlications. a few of which viewed the panic as further evidence of the undesirability of government interference with mllney and hanks. In this group the contributions of William Brough-an oilman hy profession ~I--·stand Ollt. In Tire Nall/l'lIl I.CIII' (!f MOlley 1141 Brough discussed the spontaneous evolution 1'1' money amI argued for free choice in currency. lie pursued the same themes in a hlter work, 0l't'll II/illl.\" (/1/(/ Free /lallkillR
1151.
Like Atkinson, Brough opposed himetallism, which he pointed out would lead tn an alternating monometallism favoring the overvalued metal 114. 2{) .. 57, 13fi-91. lit' too slI!!!!estcd a compromise to the silverites: not himetallism or monometallism hut "free metallism." This arrangement would require the elimination of allleg:ll fender laws and fixed exchange rates hetwecn gold and silver. It would make the choice of a particular metallic standard a mailer of CIlIl\'Cnience and elliciency. instead of 3n unplanned consequence of Gresham's Law. What the sihwites 21. Hroup.h wus the lir ... Prt'sidcnt of the Oil Producer" .. At;sflci:lIioll.
83
84
Free Banking II LAISSEZ-FAIRE MONETARY THEORISTS
7!! I
should have opposed, Brough argued, was not the gold standard per .I·e hut laws that prev.:nted free choice in currency. Brough believed tlUII the quantity of money should he regulated by market forces 114, 951. Although in his earlier book he brielly invoked the real-bills doctrine in defense of his position, i( is clear from his later work that he recognized the imporlltnce of convertibility in limiting bank issues 114, 65; IS, 43, 60, 64-701. 22 Under a system of competitive note issue, a strict cnlilrcement of convertibility would lead 10 a demand-elastic nominal money supply, with minimal lIuctuations in the value of gold liS, 57-85J. The importance of competitive note issue to Brough is evident from his choice of the Scottish and Canadian Banking systems as representatives of his ideal 114,45-91. He emphasized their relative freedom from b:lIlk failures and suspensions, dlil:iently low ratio of gold to total money supply, uniform gold redeemability with par cirl'ulation of currency, and regional uniformity of interest rates. Brough denounced Peel's Acts of 1!!44 ami 1!!45, which closed olf entry into Scottish nOle issue, calling them "a brihe" to the existing banks liS, 471. lie insisted that the only shortcomings of the C:IIl:ldian system stemmed "from government interlerence," including the issue of legal-tender government notes 114, Hili. /laving praised Canada's and Scotland's systems, Brough went on to condemn England's liS, 145J. lie regretted the fact that the English, having "generally conceded that trade must be krt absolutely untranllllelled," failed to appreciate "that an equal freedom is essential in money and banking." The monopoly privileges of the Bank of England led to the centralization of the gold supply, which was a less st:lble armngement than the free hanking outcome in whicll "each hilllk would keep its own reserve." 2.1 Brough drew on evidence from other banking systems to criticize the National Banking System of the United States. He denied, lirst of all, that state banking had failed because of ov.:riy light regulation. Instead, he attributed its failure to "the vicious charters under which State Banks wen: urganized" liS, 53-4J. 24 The National Banks were also burdened by unwarnlllted hlWS, including especially the bond-collateral requirements which intertered with note issue. Brough, like Atkinson (and many other less radical authors) noted that these requirements prevented b:lflk nlltc issucs from accommodating seasonal changes in the demand lilr curr.:ncy. As a result, hanks lost high-powered lIIoney to the public in the autumn months, only to have it return in the summer. The banks of the west and midwest would send ~ull1l11ertill1e deposits of high-powered money to their correspondents in New York and Chicago, who paid the highest interest for it J14, IO!! I, but this simply added to the correspondents' vulnerability when the dt:mand I'm currency rose again.~~ In Brough's view this situatioll was largely to blame tilr the severe panic of I!!Y]. Bad as it was, however, the panic might have been even worse had it not been for "the action of the people, who promptly took into their own hands the supplying of a medium of cxdmnge, ignoriug the laws that make such action a penal ulfence." Brough mentioned in partinllar the use of payroll checks of well-knuwn linlls, clearinghouse certilicates, and round-denolliination
21. Morwver, ill his taler buuk llruugh cxplicilly cnllorsell Ihe lIulliun Repllrl anll ii' r"bUII"t III' Ih" r""t-bill, duclrinc 115. t431. 2]. The :-.illW.: view had bel.!li cxpn'!~M:d by WaltC:f BugdHlt in Lom/Junl Street 112. 66~·tIKI. 24. Agaill M!e the work!) dleu in nuh.: ') ahove. 15. Uruugh Jill lIut seC rC!!louit.'tiulI.!lo a~i.liIlSI bralll.:h banking a~ .. factur I:ollirihulilig Iu the vulu\,.'Ci.lhilily uf Ihi:'\ arrangclllcnt. He argul!d hJ the cuntrary that branch bUllking YluulJ have the umk:,irallle ClJlbeliuellcc uf pnulHlting gn:alcr centralizatiun uf ballking 114, 162-1661. Uis upillivns uilihis mallfo!'C are f.:k-arly iIlCOII~i'!'h:nl wilh his uthcfwbe Irl!c UIOUkCI '!'Ialll.:c.
Free Banking II 7H2
85
Gl'(//'IW A. Sl'/gin find Lawrel1ce fl. Whill'
cashier's chccks ,IS cmcrgcncy currency 115. 15R-601.'· Brough also noted the signilicant fact that there wus no spcciul run on gold during thc punic: in fact, papcr currcncics sold at ,I premium relative to gold. Thercli)re. no ccntrul hank wus needed to supply mlditional reserves (even if \lIll~ could he relied upon to do so). Rnther free hanking, perhaps in l·onjllnl·tion with thl' lise of c1euringhousc ccrtilicutcs, wOlild havc prcvented the currcncy shortuge without cncollmgin!! h,mks to hold "Iiuhilities which ,Ire not volunturily assumcd" 115, 142, 1601. In other words. u free hunking relimn would ,Ivoid creation of a privileged issuer cupuhle of urhitrarily Cl"eatin!! hi!!h-powercd moncy.21
VI. ,':lllleS UilwOI·th and Austin Willard Wright
Wc h,lve secn thut the cuse for laisscz faire in money and hunking could he huscd on cthicul or politicul considcrations. Thc cthical upprouch was a mujor thcmc in Jmncs Dilworth's /;i'c'(' !JlIllking, A Nall//'(// Rig/II 121 I. Dilworth characterized regulated and ccntralized hanking systems. such as England's. as "a fruitful sourcc of injusticc, 11 limitation of frccdom to opportunity. and ... potcnt in dcvcloping poverty ,1I1d crimc" 121.6.5711"1In placc of corrupt and patcrnalistic rcgulations Dilworth dcsircd an arrangcmcnt where unyonc could cngagc in hanking and note issuc: II' A is imlchted to B tothc ml10unt of twcnty·live dollurs. he should n"t be restricted in his ri!!ht to plly this indehtedness by l'reuting u new one. lie should he permitted. if II is willing to IiIke such payment. to make live or more notcs pltyahlc on denmnd. It is no onc's husiness hut that Ilf A and II 121. 62-JI. Nor would such a policy lead to thc circulution of inferior eurrencics: '11ll'Sl' notes Ill' !I. 's wOllld nnt he Ihlls cirelll:ttl'd unlcs~ tltnse wltn tonk 11t,·1It Itad laitlt in lit" inle~rity and sound husincss IlIctllllds of !I.. Wltcn a plItenml gOVl'ntmcnt pmltihits the use Ilf suclt of the fcur Ihat snm~ innocent or iJ,!llnnll11 holder of sHch nnll" will p(l~'\ihly Iw a Inser , , , Ihal I!0vl'rnlilelit siands IIpon " pilf in inlclk'l'llIal dl'vdopmcl1l wilh tlmt parl'ot whll rl'ht'cs to takc tltl' nccl'ssllry hcnllh rcstorin!! lirug intn Itis housc hcc:tllsc of his fcar tlml snl1l(' "I'
l"lIITl'I1Cy. hCC."HIISl'
his innoccnt anli i!!,u)flml childrcn will get a pUrl of it anli poison thelllselves. Though hc relicd hcavily on cthical argumcnts, Dilworlh also bclieved that governml'nt interferenl'c with moncy and t,ilnks was cconomically unjustificd. "Why," he askcd. "should thc Governmcnt control the qua lily and quantity of currcncy-issue any morc than it should control thc qUlllily lllld quantily of cotton produccd?" For il to pcrform this task properly would require it to employ as administrators "angels that havc not yct heen sent to earth." Rlilher than lIltempl the impossihlc. "thc Govcrnment should retirc from Ihe h,mking business cntirely" 121, R.'. 911. Likc Brough. Dilworth 'lfgucd that Ihc elimimllion or governmcnt intl'rli.'rence woultl 0pl'n the W:ly to a free markel silvcr coil1:lge to satisfy proponents of himetallism. lie also helicveclthnt frce hanking would solve the prohlem of currency shortage that plagucd the Nlilintwi Bank, ami helpcd inspirc hoth thc silvcr and grccnhack movcments 121, -'6, 1211. lie dl'snihl'd instance, where. in Ihc prescnee of restrict ions on note issue. small checks had to he cmployed liS CIl1TCIll·Y. The checks. he said. would circulatc for several months. after which they "came hack tn the 2(1. For a simihlr mmlysis or thc Pimk of IH().' rrom :1 ("rl'c hnnkinp: pcrspc(':livc scc dl':uinghollsct; in this :md olhl'r crise... S('C' Timhcrl:lkc 1501. 27. For;t list of BrouJ!h's reform ~lIJ!J:!csti(lns. "'C(' {}P(·" AUnt,\ ""rr'(' nt1lrki",~ 115.
""d
\\':1I"Ill'f I~XI. 2J-3~1
On
Ihl' Inll'
fllr
Free Banking II
86
LAISSEZ-FAIRE MONETARY TIIEORISTS
7K3
bank on which they were drawn, soiled and covered with endorsements." This was proof that neither bond-secured currency nor an infusion of high-powered money was required to prevent contractions due to increased currency-demands 121, IOK-131. Moreover, bond-secured currency rcquin:d the maintenance of a substantial government debl. Dilworth found this llrr
Austin Willard Wright was similarly convinced of the wrong-headcdness of govenlillent illvulvclncllt in (Ihllley and banking. III a shun article entitlcd "Unwarranted Gowrnmellt Intt:rkrenCl.:" Wright claililcd that monetary problems Ol:curred ollly as a result of "obstructive gov.:rnnlcntal intcrkn:m:e attempting to dday or defeat the natural order of things." Legal tender legislalion was unwarranted hecause reliable monies had comc into usc, not "al Ihe behest of gIlV':IIIII1':lll, ur because \If stalulory enactllll:nts, hut solely because of recognized ability to pcrfornl th.: functions of nHlIIcy." Even in the case of thc gold dolhlr "constituted authority simply recognized what had already been accepted and adopted by the people." Wrighl thought it "reIllarkable" that people should think government capahle of regulating the value of llIulley and qualilied to remedy existing arrangements: Everybody is cognizanl of lite facl Ihat Ihe present undesirable slale "fthings in rdalion 10 Illulley is it direct cOllsequencc of govcnuilcill inierkrelll:e. and as a reilledy it is proposed Ihal we be
1M. Thi~. lIf L'UUf!:aC, wa:, au cXi.Jggcralioli. Elimination of the blllllJ:, digiblc for us\.! a:-. i.:ollalc.:ral would (i.UlU eventually Jid) IlICiJll dilililliJlioll l)f natioIHl1 hank noles. Howcwr. it Jid nul pn:Vl:lIl 1i.llillilUI bJ.lIl..s fWIll ~Oillinuillg III fUlh:til)JI as Jc,;po~il banl-.:-,. It i~ IrIJl.!, though. th...tl thdr ;,Ibility 10 fUllction thell Jqh:nd:J IIIl the ilh,:rcaSl'u usc of (.:hCl'k .. (whil.,:h ",crt.! nul yd wiJdy lhL'J in many pans ur the country) amI until!.! cxbtclIL'l! of ::.OIIiC llUhiJc :o.IHUU..: of pilp~r l'UITl.'III.:y.
2Y. C""'p"re
R,,~k,,1f
140 I.
30. Thi:-, vi~w Wit!) also shared by Lysandt:r SPOllllc.;(. Iknry Mudl.'lI. lIugo BilgriJllI, and Alnat J. N\I..-l. illL:a:-, an.: Ji~l,;u:-,st:J by COWL:II and Kro:-,ncr 11(.)1. It is ~ppilrcntly shan.:d tuJay by Nl.'il W'lllal.'c 1571.
who~c
Free Banking II
suhjecled 10 some more of the same kind of uninlelligent governmenlalism. In order Ihat a real cllre of our linan('ial ills may come ahoul it is ahsolutcly necessary thaI the government cease wholly from interference and go completely out of the hanking husiness. together with illl Ihal is implied in Ihut conneclion 165. 3-51. Wright. like Dilworth. thought that regulation. inlldditioll to disrupting the 110 I'll W I alld heneficial operation of hanks. encoumged dishonesty and corruption. His view bore sOl11e resel11hlance to the llIodern "Public Choke" l110del of government policy. although stated l110rc emphatically: The poliliciun und spoilsman are everywhere presenl. lIml these arc the men who will have 10 do wilh our filUmciul policy so hlllg us the people 1Iliow it to remain a puhlic function. The pofitician is ever a coward; a mere plan'-hunting demagogue. without convictions in rclation to any suhject: he is only lilled with desire for ofTice ... The spoifsmlln is all thllt the politiciun is, mill in addition is wholly dishon(·st. But if these men, who. owing to our own crilllimil indifTerence :md ne)!lcl'f of political duty, lire our law-givers, were ever so wise and well disposed, they l'ollid not cure ollr financial ills. !I. system of sound IiIHlnce ClIIlnot he created, it IIlllst cOllie uhout in the onfy possihle way und thut is the natural way 165, 71.
VII. Free Banking vs. Asset Currency Wl~ have seen tlwt nlllny advocates of laissez faire in l110ney and hanking invoked the real-hills doctrine (or something like it) in defense of their views. During the mid-IX90s an "asset currency" movement urose that ulso made uppelils to the real-hills doctrine. Participants in the ;Issetcurrency movement-including the respected authorities Chlll'les COl1llllt und Ilorace White -did not. however, view deregulation as 1111 adequate means tilr llIonetary reforl11." Thus although the call tilr "asset currency" initially merely indicated opposition to hond-colhltcral requirements, it cal11e increasingly to he linked to propoSllls for a centrali7.ed agency of sOl11e kind that would he uniquely exel11pt from the note-issue restrictions imposed on the Illitional hanks. This was partly a consequence of the politic'al failure of earlier proposals elilling for deregulation of note issue togcther with interstate hnlllch hanking. Branching was intended to ensurl' the active redemption of f'r'l'l'ly-issued notes. mill therehy to gual'll against thl'ir overissul'. Stymied hy thl' IIlIit ·hankin!! tilrces, retimllers turned to the alternative idea of IHlving a ccntral agency issue assct l'urrc'Ill'Y in lieu of commercial hanks doing so themsclves I.W. (,3-1251. This central agenl'y would supply "emergency currency" hy rediscounting private hllllks' holdings of short-term conllllercial paper. Asset currency was viewed, in the new reform proposals. not as something that would Illllllrally arise through deregulation, hut as an ol/erl/o/il'(' to dcregulation which actually involved a more centrali7.ed system of 1ll!l!1ey supply. In thc development from the Baltimore Plan of 11194, to the Fowler Proposal of 1900. tothe Aldrich-Vreeland hill of 19011, and finally to the Fedenll Reserve Act of 19IJ, thcre is a clear trend away from deregulation and towards centrali7.ed issut,. One way in which the asset-l'llrrency writers enwuraged this trend was hy their faillll'l' tn distinguish centnlli7.ed asset-currency systems from competitive ones. Unlike advocates of free hanking, they did not hesita'~ to include the tIlonopoli7.ed systems of France, Belgiulll, nnd Germany in their list of aml'I:~ements worth emulating 120. 202; 17, III. When they cited the
.'1. Sec Comm' /171. The ~lI'gl1l11cnt" of thc~C' ~Hllho.. " \H'n." c"l'ntu:tlly ittlopll'd hy 111l'l11hl'r~ of Ihl' "()l1alil:tliH' cr~dil control. who were in!'lnllllcntal in pronlOling the FClil-nll Rl'!'crvc 1\(', 12()). Ff1r:I nith:al :lI1nly . . io.; PI' l.iHlJ!hlin's vicw'\ sec (jjr1on :mc.l Rnprr /251.
School" of
87
Free Banking /I
88
l.AISSEZ-FAIRE MONETARY THEORISTS
785
competitive Scottish and Canadian systems they neglected to mel,1 ion any important dillcrcnces in historical performance and principles of operation between centralized and competitive systems. This approach reflected the movement's reliance on the Banking School doctrine, which did not fully appreciate the role of interbank clearings among co-equal rival issuers in promptly clll!cking any bank's overexpansion.-" They did not see that a privileged central issuer of bank notes would not have its over-issues checked liS rapidly or as reliably as a competitive one, bec.luse its liabilities would be a high-powered money held as reserves by ordinary b.mks. When over-issued, instead of being sent back through the clearing system for redemption in specie, its liabilities would be held as excess reserves and would thus prompt an expansion of the entire system. Although much of their theoretical analysis was also weakly grounded the proponents of free banking and a laissez faire monetary policy were not rebutted by the proponents of asset currency (or by anyone else, for that matter). Ironically, while most of their policy recommendations were being brushed aside, their least tenable theoretical arguments were actually shared by their opponents. Just as "free banking" laws (with compulsory bond collateraliz'ltion of notes) took the wind out of the pre-Civil War movement for laissez faire banking by co-opting its call tilr fn:edom of entry, the asset-currency movement neutralized the post-Civil War case for free banking by co-opting its l3anking School rhetoric. The real-bills doctrine, ill-suited to building a sound case for free banking, was adapted to provide an unsound case for central banking. 32. Thll~ CUIliUlI c1aimcu th"'l allY bank ovcrbsuing notes (prcsulII
References I. Alf..in . . oll. Euwanl. All ArgwJI£'1Il for the Com/iliolU.ll R£'IJt!u/ (!flhe L£'NUI1t-lUlt'f Act. Host on: A. Williams and (''''''paIlY, 1~7.j. 2. - - - -, .. All Easy Lesson in Money and Banking." Ai/tllIl;t" MU/ilhiy, Augusl 1~7.j, 195··201>. , W}WI h a LJullk! What Sen'ict' LJOt'S £I/Jallk l)t'rj
J. I~M2.
"Pay",elll uf Ihe L.egat Tender Deb!." B,.",I",."I'>. July I~I;X, .jW-·MI. "The Surplus Rcvl.!nuc," Tht' fopular ScieJlc#! MOII/hly, June I~KH. 145··9, FiliwlCe tJlkllJtml.ing. Rcprinted from the Shoe tliU/ Lemller R('/wrlt'r. lkccmbcr HiYO. 1551· 55, "The Unil Vatue ill AU Trade." t;lIgill"rillg Mug"zill", Augu'l 1~93, 555-67. "1),) We Need Siale Ilank CurrencyT Ellgilleerillg Ma!:uzille, July tg93, 497-505. lJ .. - - - - . Tht' /Jallking Pri"niJ/e, New Y~)rk: Rdorm Club Sl)UIlU Curn:lIl'y CUlIIllIiltce, IH95. 10. - - - - . "Cau!)c and RCJlIcdy lor Businc!)s Oepression." I:.'''gillt'erillg Magazille, July 1~96, 611-20, It. - - - , "The Philosuphy uf Money." nit! Mo"i.I·I, April 1~96, 337-50. t2. Bagehol, Waller. LOII/b"rd Street. London: Henry S. King, IM73. 13. Btack, Fischer, "Banking and Inle(esl Rales ill a World Wilhoul Money: The Elfeels of Unconll'lllle
u"d h," Bu"ki"g. New York: G.P. PUlnal1l's Suns, tK%. 16. Cutwell, Slepheli. The Wuys ",UJ M""".I· ufPuymem. Phitadetphia: J.Il. l.ippinwlI Cumpany, t~59. 17. Cunant, Charles A., "Banking Upon Uusille~~ As!)cls," SOlIll,I ('Uff('II('\I, I>c~cmha J, IXY7. 1M. - - - , "The Funclions of Centralized Hanking." /Junker's Magazillt:~ (ktuhcr 1~14. IY, Cowen. Tyler and Ranoy Kroszner. ""The Ocvdoplllcllt of (he Ncw Monctary b:unomics," }ounwl ufPolili('ul Ecollomy, June I'.IH7, 567-90. 4. 5. 6. 7. g.
---, ,-,. -----, --- . ---, -----,
89
Free Banking II ?Xc,
IIl1d
Gcorge A, Selgill
(I/ld 1,(/\,,/,('/lCI'
H, While
2(), 1)c;IIl. \VilliHIII n. "NHlural nunk Currellcy illHI N;llioll:ll Unn!..: Currcncy:' in "rtll'l;od Ilrohll'm,\" in IIrmVl1g CfI"elln', ediled ny Waller Henry Mull. New York: Mat'll1illan, 1')07. 21. Dilworth . .failles. /:,.rr Ranking, n Nnwrot Righ,. New York: Conlincntal Puhlishing COll1pHny. IXl)7.
"Bclll~ing inlhc Theory of Fin:lIll"C," .Jollrllal (!fMfllr(',,,,.y 1~·('()Iro",if·.'\. Jilllu:lry IC)XO. J9 ~.7. 2.'. Friedman. Millon . .'~'s.\'IlyJ in "o.~i';l'l· Hco1l(lmit's. Chicago: University of Chicngo. 195.'. 24. - - - and Anna Jacohson Schwart7.. "Has Government Any Role in Money'!" Journal ({ MOl/flary ,,"'('o1lomi(',\', January 19H6, 37-62. 2~. nirttlll. LlInec Hllli Don Roper. "1. l.t1l1fCIlCC l.illIghlin HII(llhc Quantity Theory of Mont~y." Joun1al (~rr(/Ji';(,(ll F,-nllomy, Deccmhl'r Iq7K. 599-625. 2(). Grcenfickl. Rohert L. and Leland B. Yeager, "A l..(lissc7. Fairc Approach to MonctOJry Stahility." '/0/1,.,,111 01 '.tOll(,Y. Cn'(H, alld lIanki1ll-:. Al1~II.s1 19K.'. 302--15. . 27. Hayck. F. A. /)t'IWlio1llllisaliof1 (!fMol/('Y. 2nd cd. London: Institutc of Economk Alrairs. 197X. 2M. Klrin. Benjamin, "The Competilive Supply or Money." Journal (~{ MOlley Crl'flillllld f1(",~illg. NovelllhL'r 1(J74.
22. F:mw. ElIJ,!cnc.
42.1-5.1.
2(j. Lnughlill. J. Luurence. !lankin,:: Ut/(Jrm. Chk,Igo: National Citi7.ens· Letlglle, 1')12 . .10. Menger, Carl, "On Ihe Origin of Money." /itr"'''lIIi(,jollrl/al, June IH92, 2.19-55 . .11, Minis. Lloyd. A /lislory ({B(mk;"x .,.h('ory, Chie:tgo: University ofChit:Hgo Press. 1945. 32. Moran. Charlcs. Mmu'y, Currnrdrs, and Rankin!:. New York: S. W. Green. IR75. n ---, "Panics and Politics." The Nalioll, Octoner Hi, IHH4, 324-5. -'4. Newcomb. Simoll. A Crilical E.mmi,wlioll C!.fOllr ':;mll/cial Poli(... IJllr;"K ,he SOlllh(,1"" Reh('/t;olJ. New York: D. Appieton and CO"'P'IIlY, 1965. 35. - - - . Prindl'les '?( roli'icall~(,OIwmy. New York: Harper allo Brother.s. IKX6. 3/i. Powell. Ellis T. Tire El'OllIliOIl of Ihe MOlle,' Markel, 1385-1915. New York: A. M. Kelley, 1%(1. 37. Prm'l,('dings f!{,hr Am('r;nUl flollkt'r... A.'I,'iocialion. New York: Amcric;:m Bankers A.'~(lCiilli()II. IXXX . .1X. Richards, R. I). 71",/,'orll' Hi."lory of'!/allkillg ill lillg/wlt/. New York: A. M. Kelley, 1%5. 39. Ridgeway. WilliHIl1. 71,(' 0";1(;" of Metallic C",.r('1wy and Weixhl Standards. Cambritlge: The University Prc~s. IH92.
40. Rockoff. lIugh. "The force Ranking Era: A Re-cxamination." JO/lnllll (If Alon('.", Cr('di, alld l1allAillg. May 1974. 141-67.
41. Rolnick. Arthur J, und WHITen E. Wehcr. "Ncw Evidence on thc Free B.mkinl! Era." A",rr;nlll I:'cflllomic R""irl\', Decell1ner 19HJ, IOXO-'1I. 42. - - - . "The CHlISCS of Frce IhUlk Failures: A Dclaikd E:~;:lInimlli()n." .Iounlfll (II M(lI/"ta1"\' I:·nl/l"mio. Octoher 19R4, 2117-91.
43. - - - . "Inherent Instahility in Banking: The rree Banking Experience." Cmo JOllrnal. Winler 1(}X6. X77-90. 44. Selgin, Gcorge. Tlte 711eory (!f Free Rallki,,!:: MOlle.v SIII'ply Unti('r Compelilil'e Nol(' 1,\".\"11(', Totow(I. N.J.: Rowman illld I.itllelield. )f)HH. 45. Sherwood. Sidm·y. 'l"h"/li,\'lor", tll1d Theon' of Mon('.\'. Phil.lde!phi,,: J. B. Lippincoll. IX9."\. 46. Stem. Bernhard.l. "Sul1Iner. \Villi:"" ("imh:nn," in TIl(' 1~'I1("ydol'('di(/ (!{ ,",' So";al S·";('",·('S. l'dilt'd hy hlwin A. R. Sl'li~n"U\, Vol. 14, New Y(lrk: Macillillan, 1'1.\'1. .J7. SllIrtl'v;lI1l. Julinl1 M./~·('(I"(J",i,·.'I. orlhrSd,·w·(·,{W('fll,h. New YOlk: (j. P. Putnil1ll\ SOli'" IXXI . •1X. Stllnner. Willi~"" (inlhlllll. A lIi""ol'Y '~fAm('ri("tm C/l,.n·I1(\'. New York: lIenry 11011 and ('OlllP;WY. IK7'1 49. . A"dl"'1I' .ladson. rev. c.·d. Ho"ton: II(luJ,!hloTl, Mimi" .lIId COI1lPiIl1Y. IWN. 50. TimhcrlHke .Ir .. Riclwnl .1. "The ('entr(ll Bmlking Role of Cle(lrinp.-ifollse As"ociatiol1"." ./,,/Om" (~f Mo,,('\,. C"'di,, (//1(/ R(/Ilki/l~, Fenmary I'IH4, 1-15. 51. Van fenslcrmaker . .I. The Del'('lopm('nl f!( American Commercial BmlkinK. 17112-18.37. Kent. Ohio: Kenl Stale lJnivef'ity Press, 1'165. 52. _ _ _ • "Thc SI'ltistirs of A,"cric~n Cornmercial Ranking. I7R2-IRIR." Journal (!( f:crm(lmic lI;'filol'.\'. SCpICfllher 1965,400·-41.1. 5:l. VcHlhcl. Rol
("lifO
j(lllnml. \,\Iinll'f 19H(,.
92742.
54. Viner. Jacnh. SHulif's in ,h(' Tit,.",.." (~ll"J(',."tllimUlI Tmdf'. New Y('rlo:: Ji;Irprr .lnd Brother". 19.",4. 55. Walker. Fr~n(.'i.'\ A. I'n/i,h'al J-..'cmrnmy. New York: Henry }lolt and Company, IKR.'. 56, - - ' - . Monry. New York: Henry Holt and Company, I'IRI. 57. \Vctllacc. Neil ... A Lc~~1 Restrictions Thenry of the Demand ror 'Mollcy' and the Role of f\·lonctary "CIlley" Federal Reserve Rank of Mil1lleClpnli~ Quorlrrly RCl'i!'I1'. Winter 19R.l. 1-7. ~R. W;.uner. John de \Vitt. "The Currency famine of 1R9J." Sound C/lrrell(\,. Fchnmry O~(JS ..n7 ·Sr,. 5q. \Vhite, Eugenc NcI<.;on. Thl' RrRlllalimr a"d Rflnrm (~( ,hi' Am(,,.j(,(H1 Ranking Sy.\,,'m. 1J.?88 ICJ2fJ. P,il1("('lon. N.J.: Princeton University Press. 19M."'.
90
Free Banking II LAISSEZ-FAlKE MONETARY THEORISTS
60. Whil~. Law[\!ncc H. Free BUllkin}.: ill Britain: Theory. t.:xperiem'l\ hri\Jgc University Pn:s~. IlJH4.
Qli(J
7R7
Dehate, IHOO-IH45. Camhrillg\!: Cam-
61. - - - - , "Regulatury Sources of Instability in IliUlking," CI.l/o Jounud, Winter IY86, 891-7. 62. - - - and Geurge A. Sclgin. "Laissez Fain: Monetary Thought in bcksonian America," in Perspecli\J('s 0/1 the lIi.ltory of t:cU/wmic Thought, ediled by Donald E. Mllggridge. Uplcadon, U.K.: Edward Elgar, tilrthcllming. 63. Willian1Son, lIarold Francis. t:dward Atki"",,": The Biugraphy of all Aml'ricClll UberaII1i27-/yli5. Boslnn: Old Cnrner Book Slllrc, 1934. 64. Wibon, George. NUliolllll Bunkill}.: Examined. Kansas City, 1880. 65. Wright, Au~tin Willard, "Unwarranted Governnlcnt InrcrfcrclU.:c." Soulld Cu,.,.e",')', Scplt!lllbcr IH96. Reprinted from Hlec/,.i" f:IlMineeriIiN. July IM96.
Part II The 'Free Banking' Era in the United States
[6] Excerpt from Fritz Redlich, The Molding American Banking: Men and Ideas, 187-204,292-7 Chapter VII FREE BANKING ' I
The idea of Free Banking' was neither of American vintage nor did it become a drivLng force in this country before the middle of the 1830's. Its roots, however, can be traced back at least to the eighteenth century and its history must be kept in mind if one wishes to understand the development which the concept underwent in America. The thought that banking (with note issue as its characteristic feature) might be free, that is, "entire liberty be allowed to everyone to take up the trade of banking'! had already been expressed by Sir James Steuart. But the idea was born even before the 1760's when Steuart published his book, for prior to this time it had been embodied in the Scotch banking system. Nevertheless the concept of Free Banking, that is de facto freedom of bank note issue, never gained much popularity or importance in the old world, outside of Scotland. Only after having been transplanted to America did it find conditions favorable for growth. In this country the imported basic notion, combined with other ideas of various origins and adapted to the existing external world, came to serve as the proper solution to the difficult banking problems of the 1830's. While that process was under way, the sources of the basic idea were generally unknown, as can be inferred from a Michigan court decision which declared in 1835 that Free Banking was unheard of. As a matter of fact the earlier history of the concept has remained almost forgotten ever Since, so that a modern author could erroneously characterize Free Banking as a truly Jacksonian idea. What then, were the social conditions in which the concept of Free Bsnking could mature in America? Free Banking was an alluring idea to the pioneers of the 1830's who generally were debtors and who could achieve success only with easy credit. Indebted pioneers (or pioneering
have-nots) were by necessity equalitarians. Although politically influential, these hard working farmers and small folk in frontier towns were, of course, inarticulate. Their feelings and wishes had to be formulated and expressed for them by their representatives. These men in turn could not but look to Europe for suitable ideas, since intellectually at that time America was still a European province. Thus the desires of our pioneers had to be met with the help of imported concepts framed into rules of behaviour and into laws. The popular drive toward Free Banking crystallized but slowly and stemmed from another movement very different in character which has been described by Bray Hammond. When the fate of the Second Bank of the United States was sealed, the founding of new banks appeared not only desirable to debtors and have-nots, but also sate beyond doubt to investors and would-be capitalists. Consequently the State of New York, then leading in many respects, was flooded with applications for bank charters; and the incorporators could proceed without fear of failure backed as they were by a public eager to buy bank stock. This state of the public mind was, of course, sensed by the legialators, who saw chances for filling their own pocketbooks and for promoting their political ambitions. General corruption followed in the wake of this movement. s To render banking free, that is, the issue of bank notes free, seemed to be the only way out. In 1835 a bill was introduced in the New York Assembly which aimed at the repeal of the so-called restraining acts passed earlier in the nineteenth century and directed against private, that is, unincorporated, banking. However, this bill of 1835 was not the first in which Free Banking was proposed in America. A bill to that effect had already been passed by one branch of the New York legislature about 1810, While
Free Banking II
94
THE MOLDING OF AMERICAN BANKIBG
188
the Maryland legislature in 1831 debated a Free Banking bill entitled "An act to regulate private banking."· However, in spite of these precedents, most of the New York legislators in 1835 were still strongly in favor of individually chartered banks. Many of these men may have thought along the lines of an Ohio legislator who reported in 1838 or 1839: Free Banking "is the wildest scheme of Utopian speculation prevalent on this subject. It is founded on the fallacious supposition that the business of banking should be conducted solely with a view to private gain and in total disregard of the public interests."s Thus the first attempt of the 1830's at providing Free Banking in New York proved abortive. The early sponsors of that measure needed more than inarticulate popular approval. They needed the backing of a strongly organized political movement such as came to be provided by the Equal Rights Party.s This much can be said about the early history of Free Banking in America down to the time when the realization of the underlying idea became a possibility. II
It has been mentioned that the idea of Free Banking pure and simple was expressed in economic literature at least as early as the 1160's. A few years later no less a person than Adam Smith, the admirer and advocate of Scotch banking, took up the notion and elaborated it as follows: "If banks are restrained from issuing and circulating bank notes or notes payable to the bearer for less than a certain sum; and if they are subjected to the obligation of an immediate and unconditional payment of such bank notes, as soon as presented, their trade may, with safety for the public be rendered in all other respects perfectly free. " This conviction of Adam Sm1 th probably stemmed from his experience with Scotch banking. Vice!!!!!, in the process of interaction between ideas and the external world, once stated in Smith's influential book, the suggestion became one of the ideological starting pOints of the development toward Free Bank~ in America. The Bank War of the 1830's, on the other hand, set the stage on which the idea could be realized in this country. The fact that Scotch banking not only had persisted up to
that moment, but that the idea of Free Banking had been transplanted also into England, was of no small importance. The Scotch banking system and English country bankers and joint stock banks, permitted outside of London under the law of 1825, could serve as models for the American exponents of this type of enterprise. The existence of those models and the knowledge of their practices inspired American economists and statesmen? while they were developing their fundamental notions. However, when they set up a Free BRnking system of their own it proved to be very different from both the Scotch and the English prototypes, regardless of the identity of the underlying idea in Great Britain and America. It is the combination of the basic with certain accessory ideas which is specifically American, and which was due to the particular economic and sooial environment of this country. A discussion of the historical setting may well precede the detailed description of the ideological development; and thereby a previous thread ie picked up. As has been mentioned before, the realization of Free Banking became possible in the v&ke of the Bank War. When in the fateful decade of the 1830's President Jackson waged war against the "monster," in the first instance (whatever his ultimate intentions may have been}S he intended to kindle hostility only against the Second Bank of the United States. However, the emotions whioh he excited and the reasoning of his.followers turned against all chartered banks; and some of theBe men now reached the conclusion, not fostered by the President himself, that banking, or in other worda the issue of notes, should be open for all. In the language ot the time, they felt that individuals should be entitled "to. multiply indefinitely monied oorporations." Thus the old European idea of Free Banking suddenly became a driving foroe in America. It is one of the ironies of history that the scheme, although fitting perfectly into the intellectual and social milieu of Jackson's disoiples and backers, was not put into practioe by faithful Jackson men. Against the res1stance and votes of the Demoorats Pree Banking v&1 brought into being by the Loco-Pooos, the radical wing which had broken aV&y from that party and had become independent trom and even hostile to the mother organization. The Democrats themselves were unable to
Free Banking II
95
FREE BAIIK=I:::.NG=--_ _ _ _ _ _ _ _ _ _ _ _ _..!1~89~ drav the consequences from the1r 1deolog1es., for 1n the fight aga1nst the Bank of the Un1ted States Jackson1an democracy had become al11ed v1th the state chartered banks. These nov clamored for protect10n and the Democrat1c party could not abandon off1c1ally 1te old al11es. Byrdsall, the f1rst secretary of the Loco-Focos, takes pa1ns to make clear that bankers vere the lead1ng figures 8 1n the Democrat1c party mach1ne dur1ng the turbulent meet1ng of October 29, 18'5 vh1ch gave the Equal R1ghts Party 1ts n1ckname. Th1s vas the sett1ng 1n vh1ch Free Bank1ng could become an actua11ty. Pub11c op1n1on vh1ch found express10n 1n the b1tter d1scuss10ns betveen the exponents and enem1es of the proposed meaaure came to prov1de back1ng for the movement, the second element needed for 1ts success. It 1s notevorthy that both fr1ends and foes of the plan, Loco-Focos and Democrats, respect1vely. 8~00d on the same 1deolog1cal ground. the natural r1ghuph110sophy of the e1ghteenth century as, 1nc1dentally, vas to be expected because of the1r common ancestry. In becoming the possess10n of the common man. hovever, the noble 1deas of the Lockes. Humes, and Rousseaus had become debased and vere passed about 11ke small co1n 1n the market. The arguments.of the Equal R1ghts men. the fr1ends of Free Banking, ran about as follovsl The sovere1gnty of the people has passed from the leg1slature 1nto the hands of chartered 1nst1tut10ns; such corporat10ns possess monopo11es wh1ch are offens1ve to freedom, contrary to the gen1us and sp1r1t of democrat1c 1nst1tut10ns. and subvers1ve to the great and fundamental pr1nc1ple. of equal r1ghts and pr1v1leges asserted 1n the charter of our l1bert1es. "The r1ghtful pover of all leg1s1ation 1s to declare and enforce only our natural rights and dut1es and to take none of them •••• The 1dea 1s quite unfounded that on enter1ng 1nto soc1ety we g1 ve up any natural r1gh t. " To th1 s the conservative fr1ends of chartered banks ansvered 1n s1milar language. Governor Marcy of Nev York (1786-1857). then st11l the1r spokesman, expla1ned that the cla1m to 1ssue paper currency stood upon a foundation qu1te d1fferent from pr-operty r1ghts. The restr1ct10ns 1mposed by all governments on c01nage had never been cons1dered the
1nvas10n of a property r1ght and consequently the r1ght to 1ssue notes could not be cons1dered as a common r1ght v1thheld by leg1s1atures and to be restored. 'O In the struggle" of the Equal R1ghts men aga1nst monopo11es and spec1f1cally aga1nst chartered banks, as represent1ng bank1ng monopo11es, the1r 1deas developed. In 18'5 they resolved: "That ve are opposed to all bank charters granted by 1nd1v1dual states, because ve be11eve them founded on and as g1v1ng an 1mpulse to pr1nc1ples of speculat10n and gamb11ng, at var v1th good morals and just and equal government, and calculated to bu11d up and strengthen 1n our country the od10us distr1but10n of wealth and pover aga1nst mer1t and equal r1ghts."u And 1n the Declarat10n of Pr1nc1ples of 18,6 one may read that the party holds: "unqual1f1ed and uncomprom1s1ng host111ty to bank notes and paper money as a c1rculat1ng med1um, because gold and s11ver 1s the only safe and const1tut10nal currency. " They therefore demanded that the state governments no longer author1ze the 1ssue of bank notes "1n open v101at10n of the const1tut10n of the Un1ted States." Hovever, 1n 18,6 the development of the party program took a turn tovard Free Bank1ng vhen Edvard Curt1s (1801-1856) vas nom1nated cand1date for Congress. Curt1s, a lavyer, vas later Collector of the Port of Nev York, and the supposed author of a ser1es of ant1-Bank essays. In a letter to be read if h1s name ever came up as a cand1date for any office, he tr1ed to teach the extrem1sts a lesson 1n banking. "It 1s my opin1on that the r1ghts of the people to compete v1th the 1ncorporated banks 1n deal1ng 1n money and cred1t ••• ought to be restored. The repeal of ex1st1ng restraints 1n th1s respect 1s a measure vhich the advocates of equal rights may veIl 1ns1st upon.... A reformation of the bank1hg system [that is to say: not 1ts abo11shmentl] is becoming a prom1nent point among the lessons of the polit1cal reform of the day." It 1s not surpr1s1ng that at f1rst th1s letter produced dissat1sfaction among the Equal R1ghts men, but after some add1t10nal correspondence the cand1date vas confirmed vithout recant1ng h1s op1n1ons on banking. In consequence the party abstained from instructing 1ts prospective representatives to vage var on the eXisting banks. They were asked only 1f they vould advocate the
96 190
Free Banking II THE MOLDING OF AMERICAN BANKING
exclusion of all bank notes of $10 and under that amount from circulation as currency. Thus an important step toward Free Banking was taken, and there can be no doubt that Curtis had rightly gauged public opinion. First, there was a general craving for credit and, secondly, many people did not want to destroy the banks. Besides the credit to be had,they hoped for a share in the golden harvest which they expected in the form of profit and from speculation in bank stock. Under the pressure of the popular movement resistance now crumbled, as can be seen best in the messages of Governor William L. Marcy.'3 In his annual message of 18'7 Marcy ascribed the bad times to the facts that banks possessed the exclusive privilege of furnishing the circulating medium and that "the business of loaning money [was] embarrassed by restraints imposed on other associations and on individuals." That is, of course, the cheap money argument typical of depressions. He, therefore, considered the so-called Restraining Act "so far at least as it denies to individuals and associations the right of receiving deposits and making discounts as unquestionably injurious." In this respect he suggested modifications of the law, whereas, with regard to the issue of bank notes he still wanted to sustain all prohibitions. This program was in line with the demands of the Loco-Focos, as expressed in the State Convention at Utica in September 18,6,'4 and it was actually embodied in a law in 18'7. However, after a short time the Governor was forced to go even further. In his annual message of 18,8 he recommended reserve requirements for notes and deposits alike, an exceedingly sound and modern proposal. At the same time, since the November elections of 18'7 had resulted in a landslide for the Whigs backed by the Loco-Focos, he could not but drop his resistance to a general banking law. 'S More than this, he had now to favor exactly the system which he had fought the year before. Thus he proposed to "discontinue the present mode of granting charters and to open the business of banking to a full and free competition, under such general restrlctlons and regulations as are necessary to insure to the public a large and sound currency."'S Thls could be done either by passlng a general banklng law or by an entire repeal of the restralning laws
which forbade the business of banklng to unauthorized lndlviduals and associations. Allaying doubts as to the constltutlonal competency of the Legislature, the Governor expressed his convictlon that lt had the power to pass a general banklng law, conferrlng corporate powers. But he added that according to the spirit of the constitutlon of the State of New York a twothirds majorlty would be necessary. Thls message is interesting not only for what lt sald, but also for its terminology, whlch shows the influence of the Equal Rlghts Party. The Governor lntroduced the paragraph in question as follows: "Monopolies are undoubtedly incompatible with that equality of civll right which is the great object of a free Government to secure to all its citizens." He then proceeded to state that banks, althouSb not strictly monopolies, possessed privileges withheld from individuals and associations and that they therefore shared in the odiousness with which monopolies were justly regarded. 17 After the public has accepted Free Banking in principle, the only objection which could be raised was that suoh a course would impair the soundness of the circulating medium. This argument weighed heavily indeed, and not everybody was willing to brush it aside as lightheartedly as Samuel Young whose activ1ties will be discussed shortly. The latter did not fear that Free Banking would produce redundancy of currency and that spurious bills would be forced into Circulation, for, according to his optimistic views, reflecting typical ly eighteenth-century philosophy, "the great mass of mankind are indUstrious, eoonOmical, and attentive to their own 1nterests and know how to manage their own affairs in their own way much better than legislation can guide them."l. Most people, however, thought otherwise; and the Free Banking concept pure and simple had to be combined with another idea, 1n order that safety might be guaranteed to the note holder. The outcome of this process marks the change from the general-asset banknote to the government-bond-baoked bank-note, and here entered the th1rd element of the development, the creat1ve m1nd. III
One cannot stress too strongly that the government-bond-backed bank-note was
"Free Banking II .FREE BANKING neither a higb capitalistic device nor was it rooted in classical economic theory. It was a posthumous child of Mercantilism, adopted and developed by one of the great classical economists, as will presently be shown. The government-bond-backed banknote originated when an old scheme was revamped to secure the notes at free banks. The connection at banking with public debt, that is, public creditors obtaining the privilege of banking, is an old device which even antedated Mercantilism and which vas generally known in America. This plan, as practiced for instance by the Bank of England, had come into this country when the first Bank of the United States was chartered with only one-fourth of its capital to be paid in specie, and the rest in stock ot the United States. Nothing vas needed now except the reshaping of the measure in such a way that an unlimited multitude of banks, instead of one privileged bank, would become tied up with public credit, this time not with a view to the public creditors, but for the benefit of the note-holders, that is, the creditors of the banks. This reshaping was done in America by John MacVickar,18 who at an early date elaborated the Free Bank1ng idea while combining it in one plan with the idea of the government-bond-becked bank-note. However, one should not overestima.te his achievement, since the development of the latter idea was not his work and since the same combination of ideas had also been arrived at about the same time in England. The Free Banking idea and the idea of the government-bond-backed bank-note alike are of royal ancestry. The former, as mentioned above, was promoted by Adam Smith; the latter was enunciated by David Ricardo, who undoubtedly derived his inspirations from the note issue privilege of the Bank of England and who may have known that similar plans were "favorsbly entertained by Mr. Pitt who was obliged to relinquish only an account of the difficulties occasioned by the War. "20 In his Proposals for ~ Economical ~ ~ Currency, published in 1816, Ricardo wrote: 21 "What objection can there be against requiring of those who take upon themselves the office of furnishing the public with a circulating medium to deposit with government an adequate security for the due performance of their engagements.
97 191
••• those whose habits and pursuits are little suited to explore the mechanism of trade are obliged to make use of money and are in no way qualified to ascertain the solidity of the different banks; accordingly we find that men living on limited incomes, women, labourers, and mechanics of all descriptions are often severe sufferers by the failures of country banks. • •• Against this inconvenience the public should be protected by requiring of every country bank to deposit with government or with commissioners appointed for that purpose funded property or other government security in some proportion to the amount of their issues. ,,22 It is this Ricardian idea which MacVickar combined with the one promoted by !ldam Smith. MacVickar (1787-1868) came from a well-to-do New York family. His father, aship-ovning merchant and immigrant from Ireland, was one of tha first in New York to engage in the China trade. A director of the Bank of New York, he was also connected with several insurance compenies. 29 John, the son, studied at Columbia College in New York, became an Episcope.l clergyman, and in 1817, professor of moral philo·sophy a t his alma. ma ter. In this capeci ty he became one of the early American teachers of political economy, then considered a branch of moral philosophy, as it had been in England at the time of Adam Smith. MacVicker is interesting to us as the author ot an anonymous pamphlet: ~ 2!! BBnkill8 .!!!.! ~ to .! Gentleman in Albany; ~ .! New Yorker, dated February 17, 1827, and published in New York in the same year. Written to an influential member of the New York legislature, this letter, accord1ng to a leading New York banker, actually contr1buted to the subsequent developmant. 24 It contains in fect the first proposal of that Free Banking system which was to be established a decade later. MacVickar I s plan is understandable es the outgrowth of his "Weltanschauung" end his theoretical conceptions of banking, both of which he preaches in the flowery language of his generation. "The principles of banking are much less mysterious in their nature than is generally supposed, and may be mastered by any man of conunon understanding. Everyth1ng depends on looking at it in its original simpliCity, since the mysteries of its operations are almost entirely the fruit of legislative interference
Free Banking II
98
THE MOLDING OF AMERICAN BANKING
192
lend their own credit or they lend their capital. In the former case the businsssmovements of Nature, are clear, simple, man is enabled to anticipate his funds by and harmonious. II "One has to look at bank- discounting commercial paper which has oring as one of the natural and obvious pro- iginated in actual business transactions, visions of trade; one of the simple, but "by the real exchanges of society." What beautiful, conbrivances, by which men, takes place here is just a swapping of united together by the bonds of mutual credit. The bank lends the businessman the confidence, economize the precious metals credit of its name, as MacVickar has it. and anticipate future funds by taking However, when banks discount accommodation promises in lieu of immediate payment. paper for investment purposes (that is, This l.S the essence of the whole matter.,,:i!a bills which have not originated from preSuch opinions on the principles of vious business transactions) the bank then banking are put into a larger framework. lends its capital to the businessman: it Nature is uniform, instinct is always wise, "creates him funds." The former is a legitand self interest left free to competition imate function of banking; the latter "is will reach the mark of public good with a an irregular operation, the business of ths precision beyond all the wisdom of the money lender." Thus the decisivs point is wisest law givers. MacVickar's "fundareached: If the banks should bs confined to mental position" is that evils flow from lending credit they could not go beyond the needless and unwise regulation. Let law real needs of commerce. The banks would be then secure the integrity of contracts safe and the currency secure. Ths interest [which is imperiled by the fluctuations in of the banker and that of the public wmuld the value of money), and credit will take be one and the same. The discounts would care of itself; or in other words, credit, rise and fall with the pulse of trade. But like every other business left to itself, when the banks are able to lend out their will regulate itself. Legislatures are capital, such harmony does not exist. They unable to do it; they have granted to inlend their capital when they have a surplus, dividuals and corporations monopolies and not when commerce requires it. Here lay, privileges that belonged to the community, for MacVickar, the root of the evil, for the and they should learn from experience. bankor needed capital only to support hie The thing for them to do is to give up credit. That the exclusive use of capital credit into the hands of those whom alone to this end would be best attained when it it concerns and who will certainly care was "vested" and "from under his [the banksufficiently for their interests. This er's) control," is a strange argument from will be done anyway, if not by our own such a stout believer in natural economics.29 generation then by our children, since the This flaw in MacVicker's reasoning, prejudices against freedom are destined to however, can b9 explained easily ae a rebe worn away soon. 27 lapse into Mercantilist thought. Sir James What would be the consequence of Steuart in his once famous discuseione of acting in the proposed manner? Error and banking had expreasly maintained that the fraud in banking would be limited within capital of a bank served only ae collateral the narrowest possible bounds, as soon as security for the notes issued and its purcredit is put under the sharp sighted con- pose was to establish for the bank in questrol of self-interest. Credit would then tion credit and confidence with the public.'" be able to obtain power only through the Thus it was Marcantll1sb tradition from voluntary confidence of society, which which MacVicker derived this suggeetion,Sl would rarely be bestowed except upon inand in this connection one needs to study tegrity and wealth. Thus society would be another ideological line which tied in here. furnished with credit at the cheapest rate This line, as far as America is concerned, and in preCisely the quantities which it started from that feature of the Bank of demands. 2s England's charter which gave ita "proprietors" banking privileges in consideration of The preceding quotations picture clearly MacVickar's philosophy, and his their advancing funds toward the carrying theories on banking must be viewed against on of the war with France. Or, in other this background. He distinguished between words, banking privileges were given betwo forms of bank credit. Banks either cause the recipients thereof invested in
which in Credit has obscured those natural processes which, like all the other 25 •••
99
Free Banking II FREE BANKING government ob11gat10ns, the cap1tal wh1ch they ra1sed for the projected Bank of England. Such 1nvestment 1mp11ed that the bank had no 11quid cap1tal for 1ts bank1ng bus1ness proper, but 1n l1ne w1th the theory, then develop1ng and just ment10ned, cap1tal was not necessary for banking. Follow1ng the po11cy of the Bank of England some of the early Amer1can money banks, as descr1bed prev10usly, 1nvested considerable parts of the1r capital stock 1n government ob11gat10ns, and such and other s1milar 1nvestments of the cap1tals of banks wsre common by 1816. The Amer1can country banks, on the other hand, the successors at Mercantilist "bank1ng on private cred1t," spread allover the country the 1dea that, 1n banking, capital was not needed at all. ss If cap1tal was not necessary for banking, what was? In the eyes of many banking theor1sts of the eighteenth and early n1neteenth centur1es, 1t was "cred1t" wh1ch was essent1al for banking. This 11ne of r~ason1ng also came down from the Mercant111st era, exactly from the late seventeenth century. Clapham has made the statements. that the Bank of England, when founded, was cons1dered as drawing purchas1ng power out of that "fund of cred1t" to wh1ch "almost mag1cal qual1t1es" were ass1gned by contemporaries. Th1s tradit10n was developed 1n th1s country and, in the words at another modern author,sS by 1800 people believed that there was "someth1ng calLed credit" "somehow and somewhere 1n . the corporate personality of a bank" on wh1ch to 1ssue a c1rculat1ng med1um. Robert Morr1s asserted that "paper money was the ch11d of credit" and aga1n that the pub11c would depos1t money 1n the bank to be established "ava1ling themselves of 1 ts cred1 t." 3a In the newspaper language of that t1me th1s fundamental idea was e~ pressed, in one instance as follows: "Nothing is necessary to make th1s representative of money [bank paper] supply the place of specie, but the credit of that office or company who de11vers it, wh1ch credit cons1sts 1n 1ts always be1ng ready to turn it into spec1e whenever requ1red. ,...,. Consequently when Rhode Island legislators 1n 1809 tried to remedy an unsound banking s1tuation they d1d so by 1mproving the cred1t basis of the banks of the state. 38 In th1s setting the dist1nct10n between bank1ng on capital and banking on
19'
credit was developed. 39 By 1829 a wr1ter 1n the Free Trade Advocate 40 (probably Candy Raguet h1mself) dist1nguished between banks of d1scount and banks of c1rculat10n accord1ng to whether they lent the1r capital or their cred1t. But the different1ation was already contained !n ~ 1n the Report of 1820 of the Secretary of the Treasury. If the banks followed a certain policy "and if they should reta1n an absolute control over one half of their cap1tal and the whole of the cred1t wh1ch they employ by discount1ng to that amount nothing but transadtion paper payable at short dates •.• the rema1n1ng part of the1r capital might be advanced upon long credit to manufacturers and even agr1culturists without the danger of being under the necessity of call1ng upon such debtors .•. if emergenc1es .•• occur." It has already been described that Isaac Bronson (however he himself may have acquired 1t) made th1s idea the cornerstone of his bus1ness policy and hoy he promoted it in public: banks should bank on their credit only and use their capital for safe long-term 1nvestments, thereby securing their cred1tors.·' It was from Bronson that MacVickar took the 1dea which d1d not fit at all into his ovn general philosophy and which he expressed as stated above. This fact 1s knovn through the fortunate discovery by Abraham H. Venit· a of a letter wr1tten by MacV1ckar to Isaac Bronson. This letter shoWS that MacVickar was not only dependent on Bronson, but that he was conscious of that dependence. But when 1t came to mot1vating h1s suggest10ns based on Bronson's notions, MecV1ckar reasoned as Ricardo had done: If the evils of present banking could be confined to the commerc1al classes all CfJuld be left to be "regulated by the necessary laws of Credit." But bank money is not so confined: it 1s issued 1n small denominations and thereby passes 1nto the hands of people vho do not consider notes as promises to pay, but as an "equivalent for value." Thereby the notes come to be a substitute for coin and so or1ginate nev evils. Therefore and therefore alone, legislature acquires the r1ght to interfere 1n order to guard the 1nterests of the many and 19norant.· 3 Thus MacVickar arr1ved at h1s proposals, vh1ch played such an 1mportant part in the development of American bank1ng.
100 194
Free Banking 11 THE MOlDING OF AMERICAN BANKING
Not all of the proposals are interesting, but two are essential: 1. Banking is to be a free trade, in that it may be freely entered by individuals and associations under the provisions of a general statute. 2. The amount of bank capital is to be freely fixed by such individuals or associations, but only one-tenth to be invested at the discretion of the banks, the remaining nine-tenths in government stock. From the latter the banks in question would receive interest, while the principal would remain in pledge for the redemption of their notes.·· While there is documentary proof for the dependence of MacVickar on Isaac Bronson,·5 his dependence on Ricardo is evidenced by the eimilarity of the arguments used by both men. This theory is all the more probable since MacVickar, as an economist, was thoroughly familiar with the diecussions on monetary questions which were taking place in England in the 1820's and especially with the writings of the Ricardian John Ramsay McCulloch. It was McCulloch's opinion that banks should be prevented from issuing notes unless they had previously given security.·e According to him it was not enough that all notes were made payable on demand by law. Compliance with this wise rule should be enforced by compelling bankers to give security for their issues and by prohibiting them from issuing notes before satisfying the government of their ability to honor them. Such security was to be proportioned to the issues and not to the capital. (This proposal, of course, was modern and fundamentally different from MacVickar's antiquated Mercantilist thought.) Incidentally by the same measure McCulloch wanted to force the banks to have at their disposal resources over and above the capital required for conducting their business.· 7 MacVicker's ideas were known to at least a small circle in Albany in 1827. About two years later they were more widely spread since, as already described,·e Joshua Forman inclUded them in his legislative proposals which in turn led to the New York Safety Fund Act of 1829. It is not certain but highly probable, that the incentive to combine in one proposal the two sets of ideas came from James Alexander
Hamilton. In a letter to Martin Van Buren, supposed to have been written in December 1828, he suggeeted this combination to the Governor of New York. To be eure, in this connection, Hamilton did not refer to MacVickar, but to "Bronson," 1.e., Isaac Bronson, the New York banker, and as mentioned above, the original source of a good part of MacVickar's ideas. He thus described Bronson's plan: The capital ought, before the bank can commenoe its operationa, to be inTested in the United Statee or State Stooks - and these securities ought to be placed be70nd the control of the bank; that the7 ....7 be a fund in reserve, and. ne'Yer to be used or to be .1'811able for IIl\T bankins operation; and the IIlIIOUDt of the note. to be isoued neTer to exceed the IIlIIOUDt of this fund; and as a .an of praventins IIl\T excess let it be provided that after the bank ehall have SiTeD seovit7 for ite notes, the7 ehall be countera1gned and stamped b7 the proper officer or cDlllllissionsr within whoae control the stock ie placed. [Ricardo's ideal] ••• B7 thio prevision the capital of the bank vould be eecurad to the public aa a fund to be applied in P8lID8nt of the note. in Circulation, and these could not at IIl\T t1llle exceed the IIlIIOUDt thereof. It "",,14 alao be incapable of beins loaned to dealera [bank cuetomera], and consequently not liable to be lo.t; but above all the bank vould thus be compelled to loan it. credit.... [Follova the .ussestion that bank. lean on .hort term paper on1;r. ] ••
A few months later Bronson himself published his plan anonymously in the article "General Propositions ••. " in the In!. Trade Advocate of July 11, 1829,50 an article which at the same time contains an elaborate theoretical justification. The plan amounts to a suggestion to provide security against the failure of banks by "requiring their capital to be paid in full and to be permanently loaned in mortgage security or vested in stock, prohibiting by proper penalties the employment of any part of it in banking operations and limiting the issues ••• to the amount of capital."5. It will be seen at a glance that Bronson's plan contains only one aspect of MacVickar's scheme. He is not concerned with the question of Free Banking. Having cleared up this link between
Free Banking II FREE BANKING MacV1ckar's (or Bronson's) and Forman's reason1ng, a compar1son of the latter's proposal w1th that of MacV1ckar may follow: W1th MacV1ckar the not10n of a bank note backed by spec1al secur1ty resulted from cons1stent reasoning based upon definite opinions on banking. In Forman's scheme, the same device stood disconnected from the main train of thought. It was added to the ma1n ideas at the end of h1s memorandum "to render the system perfect," and was coupled w1th the suggestion to confine the banks to the discounting of short term bus1ness paper, i.e., Bronson's pet proposal. Actually, Forman d1d not believe in the possib11ity of realizing these ideas for the time being, as he himself says. He recommended the plan 1n the same way as MacV1ckar had, but although he even elaborated t'he arguments, 1n the new context their cogency was lost. Forman reasoned: banks should loan their credit and not their capital; they should be compelled to invest the latter "in safe publ1c stocks or put it out on bonds and mortgages ••• and not to employ it in their ord1nary bus1ness." To emphasi.ze the d1fference between the two authors note should be made that while MacVickar had suggested the investment of bank capital in government stock alone, and thereby had based h1s proposal on precedents (Bank of England) and recent Eng11sh ideas, Forman was th1nking of all sorts of safe investments, including mortgages. This latter idea was traditionally American, and a step back toward Mercantil1st "bank1ng on real and personal estates" as it had been practiced by the colonial land banks. 53 This feature of Forman's "exposition" was undoubtedly due to Bronson's and/or James A. Ham11ton's 1nfluence. s4 Free Banking, of course, was not mentioned by Forman for his was fundamentally a scheme not compatible with Free Bank1ng. Thus only one of MacVickar's two basic ideas can be found in Forman's plan. This adaptation, however, was important and helped prepare the way for the later realization of MacVickar's scheme. For thus at least one aspect of it became widely known 1n Albany, being published 1n the Assembly Journal as a suggestion by Forman. It is strange irony that Forman, through h1s action,'himself belped pave the way for a plan destined to supplant his own scheme after the lapse of a decade; although he was undoubtedly forced to take such action.
101 195
MacV1ckar's and Forman's proposals of 1827 and 1829, respectively, were followed in 1829 by those of Eleazar Lord who aimed in the same direction. The latter (1788-1871) had originally studied for the m1nistry, but had to give up that vocation because of serious eye trouble. He became a very successful businessman in New York City, where he was founder and first president of the Manhattan Fire Insurance Company and repeatedly president of the New York and Erie Railroad, wh1ch he guided during d1fficult years. Lord published in 1829 a pamphlet Credit, Currency and ~1Qg in which he made suggestions very sim11ar to those of MacVickar. However, there is a decisive difference in their arguing. Attention has already been drawn to inconsistenc1es in MacVickar's and Forman's reasoning. Regardless of their belief in ~-~, they arrived at the propos1tion that the freedom of the banking business should be restricted, at least as to the use of 1ts capital. Lord, however, set the same proposal into a modern framework. 8ays Lord: 55"The object of a 11 conventionalss regulations of the currency should be to secure an adherence to the principles which naturally govern it." 80 far Lord is in agreement with MacVickar. But after this statement he goes on: "for though the operation of these principles cannot be superseded or h1ndered permanently, yet they may. be temporarily violated and reSisted." On the basis of such opinions it is consistent to force the businessman to invest his capital in a way prescribed by law. The redemption of bank notes in coin is, according to Lord, not a sufficient check; it does not make the currency definitely safe and un1form. There is too much temptation to.profit from the fluctuations due to expansion and contract10n. The system therefore must be changed in such a way that no banker 1s able to make excessive issues. As to the rest, Lord thinks along the lines of MacVickar. Excessive issues are due to the fact that banks loan their capital and credit alike. Thus Lord with consistent logic arrived, as MacVickar had inconsistently,at the conclusion that the capital of the banks should be 1nvested in "permanent securities," whereas their credit should be "employed in the operations of discounting to an extent not exceeding the amount of capital invested."se Thereby the banking system would become selfregulating; all banks would secure their
102 196
Free Banking II TIlE MOLDING OF AMERICAN BANKING
notes in the same way, all bank paper would be uniform in value, and the temptation to overissue notes would disappear. Once the security of the notes to be issued was provided for, there could obviously be no objection to the establishment of as many banks as individuals were willing to set up, provided they possessed the requisite capital. Or, in other words, there was no longer any reason to resist Free Banking, just as there was no reason to object to the multiplication of other enterprise. The more, the better! A comparison of the three proposals of MacVickar, Forman, and Lord is interesting: MacVickar starts with the intention of making Free Banking possible, foresees difficulties, and suggests a scheme to overcome them so that competition may be introduced into the field of banking. Forman does not think of Free Banking at all and connects MacVickar's safety valve (the government-bond-backed bank-note) with a machine (Safety Fund System), which already possessed a safety valve of its own. Lord, finally, elaborates a scheme which is supposed to do away with expansions and contractions of bank currency; and it seems so propitious that he finally does not see any reason why, after its introduction, Free Banking should not be permitted. He starts where MacVickar ends and ends where MacVickar had taken his departure. Did Lord know of MacVickar's ideas and especially had he read the Hints 2li Banking? It is not impossible that the two men had social contacts. Lord was a leading New York businessman, as was MacVickar's fa~er. Lord, as proved by numerous publications, was interested in· economic problems; MacVickar taught economics. The first edition of Lord's treatise has a chapter VIII (which was not reprinted in the second edition of 1834) entitled: "Popular Errors respecting Currency and Banking," and pages 120-121 in that chapter could be a criticism of MacVickar's sanguine expectation that banking once left alone would be self-regulating. However, nothing of this can be proved; and both these men may have started from the same set of experiences and may have arrived at the same conclusions independently. Born in 1181 and 1188, respectively, they belonged to the same
generation; and this alone may explain their common ideas. 57 IV So much about the emergence of the ideas essential for American Free Banking. The legislative actions of the 1830's already mentioned should by now be understandable. It will be remembered that Governor Marcy in his annual message of 1838 suggested the enactment of Free Banking in the State of New York; but although the soil was well prepared the measure would hardly have become an actuality but for the crisis of 1831 and the collapse of American banking which followed in its wake. Free Banking was a piece of reform legislation caused by a national calamity and as such parallels the Louisiana Banking Act of 1842. se Samuel Young (1189-1850) was the champion of the measure in the upper house of the New York legislature. He had been a leading Democrat in the Assembly for many years and had taken part in the Bank War by a series of articles in the Saratoga Sentinel in 1831-32, written under the pseudonym of Umbra. 59 In these articles he attacked Albert Gallatin's Just published Considerations 2li ~ Currency and Banking System of the ~ States and the McDuffie Report. In 1835, after having become state senator, he promoted the repeal of the restraining acts and thereby may have attracted the attention of the young Equal Rights Party, then still called Anti-Monopoly Party. In 1836 the party sent him a copy of the Declaration of Principles and offered him the nomination as governor of New York if he would sign that document. In a very long and wordi letter Young declined the honor because he considered it wrong to provoke a split in the party; and in addition he took issue with the radical views on monetary questions expressed in the Principles. He thought it impossible to exclude paper circulation altogether and instead proposed as his program the repeal of the usury and of the restraining acts, a proper limit of bank issues, and the exclusion of all notes of less than twenty dollars. He attacked "odious and detestable monopolies," which laid a tax on the population; but internal improvements seemed even worse to him,
103
Free Banking II FREE BANKING because they inflicted slavery and taxation on a future generation. eo In 1837 and 1838, when the question of Free Banking came up in the legislature, Young fostered in the Senate the General Banking Law, as the pending reform legislation was originally called. e1 He was then the chairman of a committee, consisting of one senator from each senate district, to report "on sundry petitions for the passage of a law creating a. general system of banking." The two decisive reports, from which the following is quoted, were probably his work. He saw three dangers in the existing banking system; viz, demoralization of the public through the process of getting charters, unmitigated inequity in the distribution of stock, and monopoly. Therefore he recommended the repeal of the restraining acts, a measure which would have led automat1cally to free banking. In doing so he referred to the Scotch banking system with which he had already been acquainted when he wrote the newspaper articles of 1831-32. Sir Henry Parnell's Scotch System of Banking e2 was, according to his own statement, his source of knowledge. In thoae articles he had praised that system because the Scotch banks by their "active compe.t1tion" had maintained the value of their notes at par when the Bank of England suspended. In his report of 1837 he went into more detail. es In Scotland, he explained, note issue was regulated by competition alone; consequently, supply and demand, and the spirit of free competition among the banks supplied the country with a circulation not disgraced by any failures. The ideal Scotch set-up was then compared with the Safety Fund System of New York. This latter system Young criticized for its inabiiity to free the state, as Van Buren had put it, from the two crying sins committed by banks, namely fraud ~nd failures. Lack of success in this respect was supposed to be due to the fact that under the Safety Fund System the banks were "connected together by the bonds of fraternal feeling, mutual sympathy, and identity of interest .... And to the same extent that ~his unity of feeling and action exists, is that free competition neutralized which is the only safe and salutary regulator of all pecuniary dealings."64 The interest attached to this assertion needs no comment but the fact
197
should be stressed that Young shared his Weltanschauung with Forman, the creator of the hated Safety Fund System. Young stated, in words almost identical with those of Forman, what was the generally accepted 'truth' in those days, "The monetary laws, the laws of trade, and indeed all the laws which appertain to national, civic, and social intercommunication among men are as determinated and fixed in their general results as the laws of light, heat, and gravvitation. If the various elements which enter into problems in political economy were fully and justly appreciated the result in all cases could be as accurately ascertained as in mathematical sciences .... And if the legitimate principles of government and the effect of the exercise of its powera upon the affairs of life were accurately understood by all, there would for the same reason be no parties in politics. "65 Young added to his report of 1837 a bill fOl' a Free Banking act. Impressed, as he was, with the Scotch system of jointstock banking his bill did not embody the ideas of MacVickar and Lord, but went back to the Scotch model and to the English proposals described above,66 The underlying idea of the Young bill was to permit banking to anyone willing to protect the creditors of his bank by pledging some property in addition to the paid-in capital. Or in the language of the bill,67 Section 1.
Any person or aSBociation of per-
Bona may establish banka of diacount, depoe 1t
and circulation upon the terms and conditione and subject to the liab1l1ties hereinafter prescribed. Section'. Three trustees ehall be appointed by the person or persona establishing Buch bank, who ehall not be interested therein ... to whom each person Or shareholder establishing Buch bank shall execute a mortgage of unincumbered real estate situated in thia state, or mako a conditionel transfer of personal property, which shall In the opinion of the said trustees be of the value of at least one and a half the amoWlt of auch shareholder' a interest in the capital stock of such bank,as aeouri ty ... for the payment of all d ebte and the redemption of all notes, bills and evidences of debt to be issued by such banking association. [Section? tredlt10nally limit. the circulation to one and a half times the bank IS capital.)
104
Free Banking 11
198
THE MOLDING OF AMERICAN BANKING
Section 16. In case any circumstances shall render a resort to the secur1 ty held by the tnlatesa ... necessary to insure the payment of the debts of such association .. . such trustees ....y make appl1cat1on to the court of chancery .•. j and the sald oourt my thereupon gno,nt an order requ1r1ne the pres1dent ••• to ahow cause why an attachment shouJ.d not issue, attach1ng the property of such assoc1at1on and transf err1ng the same to the trustees, with power to collect the dues and to dispose of the property of suoh assoc1at1on, t"8ether with the secur1t1es previously pledged to such trustees ... or of 80 DDlch of the said property and securities held as my be necessary for aueh purpose; ...
Commenting on this bill, Young stressed as its main advantages the compulsory paidup capital, publicity of the names of the shareholders (the respective sections are not quoted), and their personal responsibility as established in sections 3 and 16 .• 8 As a matter of fact, these sections contain some sort of compromise, for Young's ideal was the unlimited personal liability of the shareholders 8s as established in the Scotch Joint stock banks, whereas in this bill an additional, but only limited, guaranty was made available over and above the shareholder's interest in the bank. Young seems to have derived his inspiration mainly from McCulloch's article in the Edinburgh Review mentioned above. 7o The similarities are striking, ao that one can go as far as to say that the basic ideas of the article were shaped into a bill by the New York senator. The purpose of the bill according to Young, was to correct the ruinous contractions and expansions of the circulating medium; and this was also McCulloch's intention. Young wanted to reach this goal by excluding small notes from circulation and by opening banking to competition, in the "wholesome spirit" of which he and his generation believed with religious fervor. The author of the Review article, however, trusted less to free competition than to the deposit of securities. By compelling bankers to increase the deposit of securities as soon as they intended to add to their circulation McCulloch believed that excessive issues could be stopped. This latter idea was not embodied in the Young bill,
Since, as has been mentioned, free competition was supposed to prevent overissues. However, the main feature of section three of the bill, the deposit of security pver and above the paid-1!P. capital was apparently derived from the Review article which proposed that banks which wanted to issue small notes were "to hold a supplemental capitel as a security over and above the capital that was required for the active conduct of the business." This measure was considered justifiable since "supplemental capital would not be unproductive," but would draw interest from the government stock in which it was to be invested. 7 ' The proposal to use real estate as security, incorporated in the Young bill, was of course traditionally American. However, the bill was against the trend in too many respects to become the basis of the Free Banking Act of 1838.
v Legislative action which finally resulted in the famous New York Free Banking Act of 1838 began in that year with two bills, the one originating in the Assembly, the other in the Senate. The Senate bill was not only brought in by Guilian. Verplanck (1786-1870), but was also supposed to have been his work. 72 Verplanck was familiar with banking problems since he had played a certain part in the Bank War. As Democratic chairman of the Ways and Means Committee of the House of Representatives in Washington he had written the well known report that, in the opinion of the House, the government deposits were safe in the custody of the Bank of the United States. Thereby he had taken a stand against President Jackson, and was forced to withdraw from the Democratic party.73 Joining the Whigs he ran unsuccessfully for Mayor of New York, but in 1837-1841 was a member of the New York Senate. In contrast to the Young bill Verplanck's bill embodied the essence of MacVickar-Lord thinking; consequently it came much nearer to the Free Banking Act, as finally passed, than Young's bill of the preceding year. According to section I, any number of persons may form a joint stock company for the purpose of discounting notes, receiving deposits, dealing in exchange, making loans on real and personal securities,
Free Banking II FREE BANKING 1ssu1ng notes, and for some other transact10ns hab1tually performed by banks, Such jo1nt stock compan1es, by their art1cles of assoc1at1on, were entitled to 11m1t the1r 11abi11ty to the jo1nt funds of the company, a clause wh1ch was necessitated by the fact that the common law Joint stock company 1s characterized by the un11m1ted liabi11ty of the partners. For constitutional reasons the bill d1d not establish corporations, but joint stock compan1es instead; however, in sections 19 and 20 the word corporation crept in. Joint stock companies, as envisaged by this bill, could issue notes of denom1nations h1gher than fifty dollars without special 11m1tat10ns (section 10); but notes of smaller denominations could be issued only if they were secured by bonds. This clause will be recognized as being in line with Ricardo's original idea. However, section 20 which established this rule was actually dangerous and would have invited orgies of wildcat banking. Any of the following securities could have been pledged: public stocks of the United States, the State of New York, or of any of the states of the Union, or of any of the cities of the State of New York. All these securities could be deposited at 90 per cent of the market value of such stock in the city of New York. Unsatisfactory as this was, mortgages at least were not included in the securities envisaged by Mr. Verplanck. Sections 23 and 24 are of part1cular interest. The bank commiSSioners, who were to hold the stock as security for the redemption of the notes to be issued, would also have to procure such notes and the plates from which they were to be printed, to keep them, and to issue such bills to the companies pledging stock under the law. This clause deserves some comment. MacVickar was mainly interested in stating principles and was less concerned about putting these principles into practice. Lord, however, thought of details; and it occurred to him that the public would have to be secured against fraudulent issue of notes over and above the limitations set by law. For this purpose he proposed that the notes be stamped by purlic authority and that the circulation of unstamped notes be prohibited. This idea had already been set forth in R1cardo's pamphlet quoted before. 74 In the second place, however, Lord went a step further:
105 199
It would be a great achievement, he felt, not only to have the notes stamped by the government, but to have them all supplied from the same plates under public author1ty, with convenient blanks left for the 1nsert10n of the names of the respective banks and for other details. Th1s feature, with its advantage of mak1ng bank notes uniform, was first embodied in the Verplanck bill. The day after the Senate bill had been entered, a similar bill was introduced in the Assembly by one W. G. Patterson. This bill was said to have been drawn by John Canf1eld Spencer (1788-1855), a lawyer and politic1an of many years experience in bank matters. As a member of Congress in 1817-1819 he was on the committee which examined the affairs of the Second Bank of the United States and the Comm1ttee report was his work. His interest 1n bank1ng pers1sted, and fifteen years later, when Jackson attacked the Bank, he was among its defenders. Since the 1820's he had been constantly active in state politics. In 1827, for instance, he was on the board appointed to revise the statutes of the state. To prevent abuses by moneyed corporations this board proposed certain regulations which were enacted into law in December of the same year. 7!5
Although the Spencer bill was, like the Verplanck bill, based on the MacVickarLord train of reasoning, there were some important differences. Section I of the Spencer bill established the principle that any person or association of persons might create banks of discount, deposit, and circulation. Here, as in the Senate bill, the creation of corporations was carefully avoided for constitutional reasons. However, section 13 of the Spencer bill contained an idea alien to the Verplanck bill: It made shareholders or partners of banking enterprises to be created,jointly and severally liable for the payment of all notes put into circulation. Such unlimited liability, to be sure, could be avoided if and when any banking association, for the purpose of securing the bill-holders, invested any portion of the capital (not exceeding Qne-half of it) in stocks of the United States, the State of New York, or of any other state designated by the comptroller (Section 14). This clause meant upholding the connection with older Mercantilist tradition through the link between capital and security to be deposited for the benefit of the noteholder.
106 200
Free Banking 11 THE MOLDING OF AMERICAN BANKING
Here, as in the Verplanck bill, mortgages were not included in the eligible securities. The sound sections 23 and 24 of the Verplanck bill have very weak counterparts in the sections 9 and 15 of the Spencer bill, which provided that all notes should be countersigned by some state officials. Both bills were unsatisfactory with regard to the security to be pledged. This latter shortcoming was clearly seen by a politician who had already been instrumental in setting up the New York Safety Fund Act, Abijah Mann, Jr. (1793-1868). In both cases he played the same part. Entering the stage at a strategic moment, he did not contribute any basic ideas; but his critical mind possessed the ability to gauge necessary tactical moves and to sense politically possible combinations. From 1828 to 1830 he was a member of the New York StRte Assembly; for the next two years postmaster, appointed by President Jackson; a Democratic member of Congress, 1833-1837; again in 1838 a member of the State Assembly. Here he served on two committees dealing with banking matters, having been prepared for this work by his membership in the congressional committee which examined the Bank of the United States. He tells us in his own words how he came to act in the decisive moment in 1838. 7e Beginning with his experiences in Congress during the Bank War, he wrote: "I was appointed a member of the committee sent by the House of Representatives to examine that institution in the panic session of 1834, when the Bank, in violation of its charter, refused to submit to examination, and its directors refused to testify, through fear that they might criminate themselves. 77 In the course of performing that duty I came into much free discussion with the late Colonel Benton, and also with Mr. Biddle, 77a who were both accomplished and intelligent on banking and finances generally. The English joint stock system, as it was called, was then recently in limited operation with individual banking, with occasional disasters. I was told that the individual banker in England, when he organized his business, always invested the largest share of his capital in the securities of the Government, reserving sufficient for his current business, as he soon learned by experience
that, with a small cash capital and his bills receivable, he could always protect himself under ordinary circumstances, and 1n extraordinary circumstances he could always take up money with those securitias in tha market at home or on the continent on terms not ruinous to himself." Mann did not understand that this policy of the English banks was possible only because the Bank of England had already become a central bank, the American counterpart of which his party was just then destroying by refusing to recharter the Second Bank of the United States. Mann continues by describing how the Verplanck and Spencer bills were introduced in the New York legislature and then goes on: "Both were speCially referred to the same committee of the whole in the Assembly, and in due course came under discussion and consideration, for about a week, with a great variety of views and opinions upon the details. I did not participate in it until the close, when I pointed out some of the defects in the practical pperation of the material provisions of both bills, until I think there were not ten men in the house who would have voted for either bill. I then stated to them, in conclusion, that if they were determined to pass a general banking law, which I did not think was necessary, I thought one could be framed which could be endured, and which would, perhaps, essentially improve and simplify the wbole system, and I mentioned the principle of taking the securities by State stocks of the State of New York, into the hands of the financial officer of the State to redeem the currency, and limiting the issues of the notes to the amount of securities, so that every bank note should be simply a certificate in effect, in the hands of the holder, that he was entitled to a share in the bond equal to the amount of the note; so that a bank note, of a broken bank, should be a little better after broken than before, because it should draw interest after demand and refusal of payment. The suggestions seemed to infuse new hopes into the drooping minds of many of the Whig paper system men, who cOmposed the whole body, except twenty-three democrats, who were of the hard money persuasion of Mr. Benton. Mr. David B. Ogden, who was a leading and able member, immediately moved to refer both bills to a select committee
107
Free Banking II
201
FREE BANKING of nine, to report a Dew bill, which was adopted; and on that same Saturday, with hasty zeal, the committee was appointed and notified to appear at a meeting that evening .... The committee, under the advice of Mr. Ogden, immediately desired me to draw up the bill I had suggested in the conclusion of the debate •••• " "I drew the bill with your [1.e., Flagg's1 78 valuable suggestions, except those sections to provide for the formation of associations. It ~as my intention to have placed the business of banking in the hands Of large capitalists, as in England, or in the hands of corporations." However, Mann was in doubt whether corporations for the purpose of banking were permissible under the New York constitution and therefore did not provide for them in the first draft of his bill. David B. Ogden discovered the omission. 78 He made it clear to Mann that he desired to have a large bank in New York which could act as a subst1tute for the defunct Bank of the United States. It can not be ascertained whether what later became the Bank of Commerce was projected at that time and whether Ogden was thinking of the project, even though he was not among its founders. Mann and Ogden then made a deal. Although the former was of the opinion that under the New York constitution banking corporations could be authorized only by a "two-thirds vote on each singly and separate17," he was willing to drop his scruples if Ogden would accept the 100 per cent backing of notes by securities. The outcome was, in the words of Millard Fillmore, "a species of corporation" depositing securities to the full amount of their issues. 80 The associations thus created could contract, sue, and be sued in the name of their pZ'esident; the shares were transferrable, and the shareholders not individually liable for the debts of the association. As to the rest, Mann built his bill with stones taken from both the Verplank and Spencer bills. His section 1 corresponds to sections 2, and 24 of the Verplanck bill which thus came into the final Act. Sections 2, }, and 9 embody the principles both of Free Banking and of bond security to be pledged for the benefit of the note-holder. The security was to consist "of certificates of public stock
issued by the proper authority of this state, or either of the several states, or by the Congress of the United States, at the par value thereof." This last clause was copied from the Spencer bill, but the superfluous tying together of bank capital and security for notes issued was dropped. With this exclusion the Mercantilist element in the underlying set of ideas was also abandoned, and the result resembles Ricardo rather than MacVickar and Lord. S1 There was certainly no telling improvement over the two earlier bills, and Mann's claim indicates a braggart's gross overestimation of his own importance. Neither of the three bills envisaged the use of mortgages among the securities to be deposited. This proposal, Mercantilist in its origin S2 and in line with traditional American banking thought, was contributed by Charles Brook Hoard (1805-1886), a young local politician and a postmaster of Jackson's and Van Buren's appointment. Through an amendment it came into the engrossed bill (as section 7) against Mann's better vision. Much wiser was a second amendment proposed by Daniel Dewey Barnard (1797-1861). A well educated and widely traveled lawyer, he later became American minister to Prussia. Possibly under European influence he suggested a 20 per cent specie reserve against the banks' notes, in addition to the deposited bonds. Accordingly, section " of the Act prescribed a 12 1/2 per cent specie reserve against circulation ( a clause which was dropped again in 1840).88 VI
Thus the New York Free Banking Act came into being, as the result of a long ideological development. so Not too successful at first, it became the object of discussion and the subject of numerous amendments. The basic ideas, however, survived unaltered for many years to follOW; viz., that banking should be a free trade, but that at the same time no bank should be allowed to supply a circulating medium which was not secured outside of the bank and independent of it. sS Free Banking once put into practice was one of those institutions which spread 11ke wildfire, although with a lag of time dur1ng which t.he Act was made workable by
108 202
Free Banking /1 THE MOLDING OF AMERICAN BANKING
various amendments. In New York itself no less than 134 free banks were chartered between April 1838 and the end of 1839; but most of these were unsound and went down quickly enough. It was a genuine achievement, therefore, to develop the first successful business organization on the basis of the new ideas, the Bank of Commerce, which for some time towered above all the other banks in the city of New York. This bank was created by a number of merchants, banker~ capitalists, and jurists, with Chief Justice James Kent drawing up the charter; but the leaders in this group were the partners Samuel Ward and James Gore King of the banking house of Prime, Ward and King. Ward became the first president and until he retired shaped the policy of the bank during its early years; whereas King was among the signers of the articles of association and sold a substantial portion of the original stock to the public. The bank found in George Curtis an exceedingly competent cashier and to him as the actual manager much of the success was due. B8 In the 1840's the idea of Free Banking won minds allover the United States (although not to the exclusion of all other notions on banking). Percolating westward, particularly after a period of slow growth, it suddenly became the banking idea of the day. Free Banking represented so perfectly the underlying spirit of the period that everybody seemed to have been just waiting for the formula. Among those who contributed to the victory of the Free Banking idea were Henry Charles Carey, who advocated the principle of free banking,B7 Condy Raguet, and Richard Hildreth. The writings of the latter two men familiarized the American leaders in this field with the new system of banking. Raguet repeatedly discussed it in his Financial Register, going so far as to declare New York Free Banking "the greatest improvement of banking which has yet been introduced in our country." In his Treatise Qll Currency and Banking of 1840, he reprinted the New York Free Banking Act, and devoted a whole chapter to commenting thereon. He also reproduced the Report of the Comptroller of 1840 on the "Operations under the Act to Authorize the Business of Banking" together with amendments to that act. Already in the
Treatise he expressed his expectation that the "legislatures of the different states ... may ... adopt the New York sys tem," whiCh was identical to a suggestion that they do SO.B8
At an earlier stage, on the other hand, Hildreth had propagated the idea of free competition in banking, for in 1837 he published his History of Banks t.9. J!h!£h is added ~ Demonstration of the Advantages and Necessity of Free Competition in the Business of Banking. Bs Here he pleaded that "the monopoly of bank charters must be abolished altogether and not in one or two states only but in all." "Capitalists must be left as much liberty to invest their money in a bank as in a cetton mill." Hildreth expected that free competition introduced into the field of banking would result in the "Laws of Trade" controlling the currency. Thus Hildreth promoted the idea of free banking. His hazy and radical ideas could not provide the basis of an act; but once Free Banking had become a reality in New York, Hildreth logically Joined its defenders. He became convinced that the principle of "requiring security from banks" (not originally embodied in his conception) was "the oull legislative measure with respect to the currency which the government could reasonably and advantageouslyadopt."so Consequently in his Banks, Banking and Paper Curreucies S1 of 1840 (in which his book of 1837 is substantially reprinted) he embodied in his earlier reasoning those essential features of the New York act, thus becoming one of the promoters of the New York brand of Free Banking. Like Condy Raguet, he felt impelled to reprint the whole act. S2 Thus promoted by leading thinkers on banking and backed by public opinion, Free Banking was adopted by one state after another, by New Jersey in 1850, OhiO, Illinois, and Massachusetts in 1851, Indiana and Tennessee in 1852, Louisiana in 1853, Wisconsin in 1854, MiSSissippi in 1856,' Michigan (for the second time) in 1857, Iowa and Minnesota in 1858, and finally by Pennsylvania and Kansas in 1861. s3 All the state laws in question were more or less copies of the New York law, except for Louisiana which engrafted certain features of the Banking Act of 1842 onto the New York ideas. s4 The fact that Free Banking in its original form was abused in many
Free Banking II
109 20,
PREE BAlIXlNG
states. espec1ally 1n the ~ddle West. does not detract from 1ts 1mportance nor d1d 1t destroy the pers1st1ng 1nfluence of 1ts underly1ng 1deas. One 1nterest1ng deta11 remains to be considered: v1z. the fact that the Pree Banking idea started to migrate from New York even before 1t had been put 1nto practice 1n that state. Actual17 Mich1gan took over the princ1ple as early as 18'7 when the final New York act had not even been drafted. The explanat10n 1s s1mple. The Michigan settlers in the early 18,o's came MOst17 from the western countiea of New York and were imbued w1th the sp1r1t preva11ing 1n that state;that 1s. th&7 were against monopo11es. but in favor of banking. aa descr1bed prev10usly. Such notions migrated West in the minds of the pioneers of Michigan. and when a situation developed quite similar to that 1n the mother state. they looked for suggest10ns to that more developed commonwealth. Thus 1n 18,6 the Michigan leg1slators copied the New York Safety Pund Act; and when the debates on Pree Bank1ng were scarcely under way 1n Albany. they eagerly embraced the bas1c ideas of the new scheme. as Just as had been the case 1n New York. so in Michigan the leg1s1ature was swamped in 18'7 w1th applicat10ns for bank charters and could see no way out except by embod71ng the pr1nc1ple of free bank1ng in an act. Be At that time 1n New York only the Young b111 was drafted. and 1ts 1nfluence can be traced in the Michigan act. The Young b1l1 as well as the MiChigan Pree Bank1ng Act of 18'7 (the f1rst Pree Bank1ng law paseed 1n Amer1ca) estab11shed 1n sect10n one the Pree Bank1ng pr1nc1ple. The Young b1l1 prov1des 1n sect10n , that each shareholder shall execute a mortgage on un1ncumbered real estate or make cond1t10nal transfer of personal property. which 1n the op1nion of a trustee 1s of the value of at least one and a half of the amount of such shareholder's 1nterest. Th1s secur1ty 1s to be pledged for the payment d all debts am. the redempt10n of notes and s1m11ar ev1dences of debt. Sect10n 2 of the Mich1gan Act closely corresponds to the above-quoted section, of the Young bill. reading as follows: "The president and directors of such assoc1at10n shall before commenc1ng operat10ns furn1sh good and suf-
f1c1ent securit1es to the aud1tor general, for and 1n behalf of the people of the state. to be held as collateral secur1t1es, when all other liab111t1es of such assoc1at10n shall fa11, or prove to be 1nsuff1c1ent for the purpose 1ntended, ••• and for the redempt10n of ali notes, which secur1t1es shall be in the full amount wh1ch such associat10n shall at any t1me have 1n c1rculat10n or be 1ndebted, and shall cons1st e1ther of bonds or mortgages upon real estate w1thin the state. or 1n bonds [that 1s. personal bonds] executed by reaident free holders of the state, and shall forthw1th lodge 1n the hands of the bank commissioner." In both the New York bill and the Michigan act. just as 1n the older Safety Pund acts, no dist1nct10n 1s made between note-holders and other cred1tors. In all these proposals in 11ne w1th Mercant111st tradit10n. security 1s supposed to consist of mortgages on real estate and personal property or personal bonds executed by the free holders. respect1vely. One essent1al d1fference between the New York b111 and the Mich1gan act should be stressed; and 1n this respect M1chigan took a forward step. Young wanted to have secur1ty g1ven over and above the 1nvested cap1tal, and there was to be no connect10n between secur1t1es pledged and outstand1ng 11ab111t1es. In the Mich1gan act the connection between capital and secur1ty to be pledged was severed and replaced by a 11nk between secur1ty and 11abi11t1es. Nevertheless Young's idea of a security over and above the 1nvested cap1tal was also 1ncorporated 1n the Michigan act, 1n that a director was 11able to th~ extent of h1s 1nd1v1dual property and a stockholder to the amount of stock held by h1m. It is not surprising that this sort of land-bank-open-to-all ended in d1saster, all the more so because bus1ness could be started when on17 per cent of the capital had been pa1d up, whereas the Young bill required the full paying1n of cap1tal. In add1t10n, 1n M1ch1gan real estate mortgages were taken "at true cash value" dur1ng boom times; while certa1n restr1ct1ng clauses were c1rcumvented, exam1nat1on evaded, and the 11ab111ty clauses rendered 1neffect1ve by court decis1ons. It 1s d1ff1cult to trace w1th certa1nty the MiChigan p10neers who f1rst put the idea of Pree Bank1ng on a statute book.
,0
110 204
Free Banking II THE MOLDING OF AMERICAN BANKING
In January 1831 Edwin H. Lothrop, member of the Committee on Banks and Incorporations, a leading farmer end Democrat, reported in the House of Representatives the "bill to organize and regulate banking associations." In the Senate Edward D. Ellis moved "that the Committee on Incorporations be instructed to inquire into the expediency of providing by law for a uniform and equal system for the regulation of banking associations and that they be instructed to report by bill or otherwise." A few days later a "bill to regulate banking associations" was brought in by Ellis himself, a newspaper editor, who also reported the bill. The biographies of these men mention no connection with the Michigan Free Banking Act of 1831, perhaps because the fate of this law was such that its
authorship gave no reason for local or family pride. On the other hand, this s1lence of the biographies cannot be taken as proof either for or against the leadership of the two men in this matter. S ? So far the westward migration of the Free Bank1ng idea has ·been traced from New York to Michigan. When it came to the enactment of the idea 1n New York in 1838 the already existing Michigan Free Bank1ng Act was known in New York,8s and it is not impossible that the direct relationship between notes and securities pledged (characteristic of the New York act) was adopted, at least partly, under Michigan influence. If this was the case, then the idea or1ginated in New York, journeyed to Michigan and then back home, improved in this one important point.
Free Banking II
111
THE MOLDING OF AMERICAN BANKING
292
7.
Chapter VII FREE BANKING l. This chapter is the revised version of the essay "Free Banking: The History.of an Idea and its Exponents," first published in this author's Essays in American Economic History. Recently, Vera Smith (.Ql1. cit., passim) has treated the concept of free banking and its history in the various European states. 2. Inquiry into the Principles of Political Economy (London, 1767). Quoted from the reprint of Book III, published under the title Principles of Banks (London, 1810), 237. 3. Proposals as to the distribution of bank stocks were made by the Bank Commissioners of the State of New York in a report of 1/27, 1837. 25th QQns!:., 2nd ~., ~ Executive Document 12, 234-5. 4. Publicola, Vindication of ~ Currency ,of, the ~ of New ~ and .!!. ~ of ~ Report ••• addressed 1Q The Hon. Isaac Pierson (1818). Page 24 contains the following passage: To diminish the evils of the preseht (1810) banking system competition among the existing banks should be promoted or every citizen should be authorized "to bank, either solely or with associates, under suitable regulations and restrictions. The latter course is the most simple •••• It can be done in such a manner as to render the public perfectly secure •••• A bill to this effect was actually passed at the last session of the legislature through one branch of that body .... " The Maryland bill is reprinted in Condy ~aguet's Financial Register, II, 398400. 5. U.S. 26th Congr., 1st ~., ~~ tive Document 11£, 1087. 6. Hammond, Jabez D" ~ Hi-story of !:21illcal Parties in the State of New York (COoperstown~1846):-U:- 447T In'1837 the New York Bank Commissioners were
8.
9.
10.
11.
still against Free Banking (the report quoted a.bove). Albert Gallatin, in contrast, was a friend of free banking as early as 1834. Adams,Henry, The Life of Albert Gallatin (Philadelphia, 1880), 651-.-The Scotch system of banking was known allover America at that time. Thomas Hart Benton referred to it in Congress as early as 1831, and probably under his influence the matter was dis CUD sed in Missouri in the winter of 1834-35. It was a Southerner, Hugh Swinton Legare, who in a speech in Congress in 1837, remarked: "Everybody has heard or ought to have heard of the Scotch system of banking;" Writings (Charleston, 1846), I, 302, 303. The system is also referred to in the Report of the New York Bank Commissioners of January 27, 1837, 25th Congr., 2d~., ~ Executive Document 12, 236, 237. Helderman, .Ql1. Cit., 42-43, 53, 63-64 describes the various futile attempts to introduce into America during the 1830's and 1840's its characteristic feature, stockholder's l1abil1 ty. Of the publications dealing with Scotch banking Albert Gallatin's Considerations of 1831 may be mentioned. Its Appendix contains among "Notes and Statements" one on the subject extracted from the Report of ~ ~ Committee of ~ ~ of ~ .Q.!! Promissory ~ of Scotland and ~ (May 26, 1826) • Benton, Thomas Hsrt, Thirty Years' View, I, 158. c. J. Ingersoll to Biddle, Feb. 2, 1832; McGrane, B1ddle Correspondence, 172. ------ -----George D. Strong, president of the Commercial Bank, and Isaac L. Varian, "a bank director," later state senator and mayor of New York. At that meeting the candidacy of G1deon Lee (1778-1841) for Congress was opposed because he was president of the Leather Manufacturers' Bank. Byrdsall, F., The His tory of the Loco-Foco £!: Equal Rights Party (New· York, 1842), 21, 24, 26. See also Hammond, Bray, "Free Banks and Corporations: The New York Free Banking Act of 1838," The Journal of Political Economy, 44 (April 1936), passim. Byrdsall (.Ql1. cit., 15, 27, 39, 57) reprints party documents. Marcy in his annual message of 1837, State of New York, Messages ~ the Governors, III, 629. For the following: Hammond, Jabez, .Ql1. Cit., II, 478 ff •. and Byrdsall, .Ql1. cit., 27, 39, 41, 81 ff., 88.
112
Free Banking II FOOTNOTES
12. This program may have been inherited from the Workingmen's Party, launched In 1828 In New York, Phlladelphla and Boston. Myers,!h! Endlng of Hereditary American Fortunes, 47. 13. lew York, Messages from the Governors, III, 611 fe, 657 f£, 695 ff. 14. Byrdsall, ~. £!!., 73. 15. The introductlon of a general banking law was facilitated by the fact that general incorporation laws had already preceded this piece of legislation, and that since the Safety Fund Act of 1828 the public no longer looked at each bank as a particular enterprise but as belonging to a class of institutions. Hammond, Bray, ~. Cit., 190. 16. ~~, (1838), 9. 17. One should compare this phraseology with that used by Senator Young, whose role In the development will be examined shortly. Said Young in a committee report of 1837: "The monopoly characteristic of the [present banking] system, the exclusion of competltion, the creatlon and continuance by law of a privileged order who have [sicl] the sole power of selling or lending their promises to pay in the shape of notes ••• are features utterly at war with equal rights and free government." 18. New York ~, !12.£. 55.... 1837, 7. 19. Dorfman, Joseph and Rex Guy Tugwell, "The Reverend John MacVickar," Columbia University Quarterly, XXIII, No.4, Dec. 1931. Seligman, Edwin R. A., "The Early Teaching of Economics in the United States," in Economlc Essays, .Q£!!tributed in honor of John Bates Clark (New York:-1927). - - - - - - - - 20. Edlnburgh ~ (Feb. 1826), XLIII, 297. 21. Pages 35-6. 22. Professor Herbert Somerton Foxwell of Cambridge, England, wrote on one of Ricardo's pamphlets which is now in the Kress Collection of the Baker.Library at Harvard: "The ablest of all Ricardo's writings. It is the foundation of most of the modern currency systems." Ricardo's tract is quoted by Cleaveland, ~. cit., Introduction, lill. Thi proves that the essay was known in America by the middle of the nineteenth century. SCOVille, The Old Merchants of New York - -(2nd serie;;: 281-291. 24. Williams, John Earl, "A New York View of Finance and Banking," in Old and ~ (November, 1873). However, Williams goes too far in his statement. Condy
293
Raguet acknowledged MacVickar's importance in reprinting the Hlnts In 1838 when the Free Banklng legislation was under discussion. Flnanclal Reglster, II, Nov. 21, 1838. 25. MacVickar always writes Credlt with.a capital. 26. MacVickar, ~. £!!., 5, 6. 27. ~., 7, 13-14. 28. ~., 15. 29. Ibld., 28-31. 30. ~Clples.!lf Banks, 188, 189. 31. MacVickar hlmself was silent In this respect, but Eleazar Lord and a New York pamphleteer knew, as the latter has it, that "this is no new expedient," for the Bank of England dld not employ its capltal tn making discounts or In any of its ordinary business, slnce it was invested in government stock. Letters to the Honourable ~ Woodbury (New Yor~1837), 20. (Venit,~. cit., 213, footnote, identlfies RogerM. Sherman as the author of this pamphlet whlch was heretofore ascribed to Isaac H. Bronson.) James Alexander Hamilton (Reminiscences, 82 ff.) stressed this pOint. See also Gallatln, Consideratlons in Writings, III, 273, 319. 32. See pages I} and 49 ff. 33. This author suspects that the idea that capital was not necessary In banking wa.s perhaps an awkward expression of a conclusion drawn from the dlscovery that in payments credit could serve as money. By identlfying money with capltal, such a mlsconception could easl1y ari~e. 34. Q2. clt., I, 24. 35. Stokes, ~. clt., 267. 36. Carey, ~, 37; Sparks, Dlplomatic Correspondence, VII, 448. See also Sulllvan, Path to ~, 41, 51. "Paper owes its currency to the credlt of the emitter." "The notes of private bankers are Issued upon the credit of thelr reputation and prlvate estates." 37. Provldence Gazette, February 20, 1784, quoted by Stokes, ~, £!!., 266. 38. Stokes, ~. clt., 281. 39. Banklng on credlt, of course, was a vel1ed expression of what was de ~ manufacturlng money or what is called today the creatlon of purchasing power. Opdyke In hls Report on the Currency of 1858 distlnguishes between a moderate and an Immoderate Infuslon "of the fic.tlt1ous or credit element" Into the currency (page 14). 40. II, 1 ff. 41. As late as 1841 the New York Bank
Free Banking II
113
THE MOLDING OF AMERICAN BANKING
42. 43. 44. 45. 46.
47.
Commissioners in their ~ Report for 1840 (page 10) commented on Free Banking in terms of Bronson's philosophy. Qp. cit., 203, 204. Hints, 38. Ibld' J 38. See above, page 193. As early as 1819 an American author writing under the pseudonym of Scaevola suggested that banks should give security with the courts for their notes. In case of non-payment the note holder sbould be entitled to sue on the bond. Amos Kendall in an article of September 1819 recommended that proposal. See William Stickney's AutobiographY of Amos Kendall, 225. Later, substantially the ~ea was expressed by Isaac Bronson in Free Trade Advocate, July 11, 1829. The-P;;sage is quoted by Condy Raguet in ! Treatise on Currency ~ Banking, 2nd ed., 204,footnote. In the 1850's it was realized in Massachusetts that the New York Free Banking Act of 1838 embodied ideas of Ricardo and McCulloch (U.S. 36th ~., 1st ~., ~ Executive Document !!.2, 74) • All these quotations from McCulloch were reproduced by an American author of the 1850's: Cleaveland, £P. cit., Introducduc.tion, li ff. A whole collection of similar quotations from McCulloch can also be found in Raguet, Treatise, 203, 204, footnote. It will probably be impossible to prove or disprove whether MacVickar knew the following English sources which contain ideas similar to his. One is an anonymous article in the Edinburgh ~ (February, 1826, Vol. 43) entitled: "Thoughts on Banking." The author (J. R. McCulloch, as a matter of fact) besides discussing the introduction of Joint stock banking and the suppression of small bank notes, suggested as alternative to the latter measure "to allow private banking companies to issue notes as at present [i.e., small notes), but to oblige such as chose to avail themselves of that power to deposit securitie. for their payment in the hand of the government" (page 280). When he discussed his proposals in detail on pages 291-2 he made it clear that he thought primarily of security for small notes in order to protect the public. Only at the last moment he took a step further, broadening the scope of his proposal and making it apply "as well
48. 49. 50. 51.
52. 53.
for large as for small notes." The justification of the suggestion in the Review article is permeated by a theory completely different from that of MacVickar. Sir Henry Brook Parnell took up the ideas of the Edinburgh ~ article in his Observations of ~ Money, Banking and Overtradlng ••• (London, 1827), a book which later influenced American thought. Ricardo's reflections quoted above, were reprinted in the Review and from there were taken over by Parnell, who wanted to place English banking "under the control of continual and efficient competition." Since MacVicksr' s letter is dated February, 1827, he can scarcely have been influenced by Parnell, although here is the same combination of Adam Smith's and Ricardo's thoughts which we find with MacVickar. The original ideas of both ftdam Smith and Ricardo were conceived with reference to small notes. In their development, however, this connection was dropped. This was done in McCulloch's anonymous ~ article during the discussion; while MacVickar and Parnell no longer use that same starting pOint. MacVickar, following Adam Smith, wanted to forbid completely small notes below $5, but at the same time he adopted Ricardo's suggestions worked out with regard to such small notes. Slightly later, Lord ftlthorp in a speech of 1833 suggested that the government should grant charters to nonissuing joint stock banks within the sixty-five miles if they deposited the paid-up capital in government securities or other equally good funds (Gilbart, History, 63). This English author discusses the problem of bankers depositing securities for their issues with a view to England (~., 90). See page 91. J.A. Hamilton, Qp. cit., 83, 84. See footnote 46. ~ ~ Advocate, II, 26. One will realize the difference between the plan as presented by Hamilton and' by the author himself. However, there is a discrepancy in Hamilton's letter. ftctually Bronson, as described above (see page 47) invested the capital of his own bank in mortgages. ~ York Assembly ~ (1828), 182 ff. MacVickar wanted to leave one-tenth of the capital to the discretion of the banks; Forman nothing.
114
Free Banking II FOOTNOTES
54. .Q.E. cit., 86. 55· All quotations are from the second edition, entitled~, Currency ~ Banking, the above is to be found on page 62. 56. Ibid., 84, 90. 57. As to the theory of generations, see the author's History of American Business Leaders, Vol. I, Ch. I. 58. Helderman, £E. cit., 9. 59. Republished anonymously in 1832 as a pamphlet entitled: Considerations 2a the Bank of the United States, in which its ;ep;;-g~n~to the c~utlOn ••• (Albany, 1832) • 60. Hammond, Jabez, £E. £1!., II, 447-8; Byrdsall, £E. cit., 61 ff; Jenkins, John S., History of Political ~ in the ~ of New 12!:!i (2nd ed., Auburn, 1849) gives many dats on Young, easily to be found through the unusually good index. 61. For the following: State of New York, Assembly and ~ Journals, 1837 and 1838. ~ Documents 22, 1897~ Report of the select committee ••• on sundry petitions for the passage of a law creating a general system of private banking, and Document 68, 1838, Report of the select committee on so much of the Governor's message as relates to the repeal of the restraining laws and free competition in the business of banking. Less important are Assembly Documents }22, 21§, (1837). 62. He means the book quoted before in footnote 47, the exact title of which is: Observations on ~ money, banking and overtrading, including those ~ of evidence ~ ~ the committee of the ~ of ~ .'!!!l!£!:!. explain the ~ System of Banking. (London, 1827). Parnell's book was widely known at that time. Thomas Hart Benton quoted him repeatedly (e.g., £E. cit., 188, 202) and so did Nicholas Biddle and Samuel Young, the latter in his articles cited in footnote 59. In Senate Document 22, 13, 14. This argument is repeated on page 18. Hon. Samuel Young, Oration delivered at the Democratic Repu~Celebration~f the 64th Anniversary of the Independence of the ~ States, July i, 1840. (New York, 1840), 10, 11. On page 14 is an interesting elaboration of Jacksonian ideas. 66. See McCulloch, esp~cially his anonymous Edinburgh ~ article. 67. N.Y. Senate, Document 22, (1837), 21-24.
295
68. In his Early American Land Companies. Shaw Livermore gives the background of this proposal treating the trend toward stockholders'liability in the period 1800-1825 (261 ff.). 69. Gilbart's contemporaneous History (78, 79) contains an interesting description of the advantages of unlimited liability and especially its influence on the behavior of the shareholders. 70. See footnote 47. The article was quoted in Parnell's book. 71. ~, £E. cit., 293. 72. This bill and those to be discussed presently are to be found as Nos. 65, 66, 215, and 308 in a volume without title page: Legislative ~ 1838, owned by the New York State Library, Albany, New York. 73. Daly, Charles P., ~ Q. Verplanck, His Ancestry, Life, and Character (New York, 1870). Bryant, William Cullen, A Discourse of the Life, Character, a.nd Writings of Guilian Crommelin Verplanck (New York~1870T-:--Hart, C. 'H., 1> Discourse on the Life and Services of G. C. Ver";l'an~ (New-Y;;-;k-;-l870). -- ---74. See above 191. 75. Paine, Willis S., £E. Cit., 25. 76. The following is taken from a letter of Mann to A. C. Flagg, reprinted in Flagg, £E. cit., 40-42. Joseph Sabin in his Dictionary of ]looks relating to America quotes a pamphlet of Abijah Mann, Correspondence ~ A. Mann, Jr. and Hon. A. C. Flagg ••• (Brooklyn, 1868), but no copy could be' located. Mann in his vanity must have told people wild stories about his share in the Free Banking Act. Bonnefoux, once a Free Banker himself, relates: "Mr. A. Mann, Jr •••• vas, I believe, the gentleman who drew up the seven or eight sections which formed the groundwork of the Act to authorize the business of banking, passed April 18, 1838. The original idea, Mr. Mann told me, sprung from the example given by Mr. Stephen Girard WhO·, when he established his bank in Philadelphia, voluntarily hypothecated his immense property for the due payment of his circulating notes." Bonnefoux, Vindication, 11. 77. A vivid description of this attempted examination is given by Benton, £E. cit •• 458 ff. Se~ also u.S. £22 Congr., 1st ~., ~ Report 481, passim. 77a.On April 3, 1834 Biddle wrote to J. G. Watmough, "Tell me something about this Mr. Mann of New York," (Biddle Letter~).
------
Free Banking II
115
THE MOLDING OF AMERICAN BANKING
78. Flagg as Comptroller of the State of New York had to carry the law into execution. His interesting report with some documents added is in U.S. 25th Congr., 2!:!! ~., House Executive Document ££r, 2Pl ff. 79. David B. Ogden (1775-1849) vas a famous New York lawyer and scion of a socially prominent family of ironmasters. He played no important part in public life, but in 1838 served a term in the Assembly. 80. Report of 1849, quoted from the reprint in U.S. 31st ~., 1st ~., ~ Executive Document 68, 128-9. The cases arising out of the Free Banking Act are discussed by Bray Hammond, passim, and Livermore, ~. cit., 287 ff. Some of these pertain to the legal character of Free Banks. 81. It was one of the doctrines of Mercantilist banking theory that capital functioned exclusively as security for the note-holder. Thence, as mentioned before, MacVickar took his suggestion that bank capital should be "vested" and "from the banker's control." Based on the same Mercantilist idea American legislatures had developed the principle of limiting the issue of notes to twice or three times the capital. MacVickar and Lord went a step rurther: they proposed the limitation of note issues to the amount of capital; and since both suggested at the same time its investment in securities, notes and securities were tied together indirectly. The amount of the notes was "never to exceed the IlIIIOunt of their pledged stock," says MacVickar (page 38}. and according to Lord (page 84) the bs nks had to use their credit alone in dIscounting "to an extent not exceeding the amount of capital invested." Ricardo, who had freed himself from Mercantilist tradition, could be direct and logical from the beginning in suggesting the deposit of securities, "In some proportion to the amount of their issues." (See page 191). 82. In consequence thereof, to the extent that note issues were based on mortgages Free Banks in fact practiced Mercantilist banking on private credit. Individuals transferred mortgages to banking associations upon the condition that they received accommodation loans immediately. They melted down real property into symbolical money, as Sir James Steuart would have expressed it. (~Report of the New ~ Bank COmmissioners, for 1840, 9.)
83.
84.
85.
86.
87. 88.
~
of 1840, chapter 363, section 6. The amendments are reprinted in bill ~, 10 ff. Horace Greeley in his Recollections mentions Willis Hall as the man to whom the state is especially indebted for the Free Banking Act. (Pages 125-26.) Dixon Ryan FOX, in The Decline of Aristocracy in the Politics of New~, Studies in H1story, EconomiCS, and Public Law, No. 198 (New York, 1919), 4-5, misquotes the passage. A thorough investigation by the present suthor as well as by the reference librar1ans of the New York Public and the Albany State Libraries brought nothing to light to corroborate this statement. Willis Hall (1801-1868) was a Whig politician, an Assembly man in 1838, Attorney General of the state in 1839. He wrote a few political pamphlets and after 1848 was close to the Free Soilers. A. C. Flagg (~. cit., 43) criticized his wild and vague banking theories. If there is anything at all to Greeley's statement, Hall must have played his part behind the stage and may thereby indeed have had a meritorious influence on the development. This, however, is mere guess work. An Act to s.uthorize the business of banking, passed 4/18, 1838, ~ of the ~ of New York, passed at the 61st session (Albany, 183~ Most important are the sections 1, 2, 3, 4, 7, and 11. The history of the Act has been written pyMillard Fillmore in the document quoted above and by Helderman, ~. cit., 21 ff. Important are the ~ Reports of the New York ~ Commissioners, especially those of 1840 and 1841. O.S. 26th ~., 1st ~., ~ Execillve Document ~, 115 ff. and 2nd ~., ~ Executive Document 111, 122 ff. For the beginnings see also Ogden, Remarks, 28 ff. ~y Trust Company of New York, One ~ ~ of Banking ~ (pr~ vstely printed, 1939), 3-10. Curtis had been ca.hier of tHe Exchange Bank in PrOVidence, had played a part in public affair·s there, and was destined to become one of the leading figures in the young New York Clearing House. The first attempt to base a Free Bank in the state of New York on exeeptionally good security was made by L. Bonnefoux, ~. ill., passim. Principles of Political Economy, Second Part (Philadelphia, 1838) 257 ff. Financial Regisber, II, 10; Treatise, 201 ff, 243 ff, 308 ff, 321 ff. The
116
Free Banking II FOOTNOTES
last quotation may be found on page 202. 89. See especially, pages 124 ff, 131, 132, 136. 90. Hildreth, R., ~ ~ to his Excellency ~ ~ .Q!l Banking and the ~ £y (Boston, 1840), 14. 91. See especially, pages 147 ff., 153 ff.. , 200 fr. 92. See also the articles on "Free Banking" in ~ ~ XII, 610 fr., XIII, (1852) 127 ff., XIV, (1853) 28 fr., 151 ff. The case for and against Free Banking is summed up in this series, the author being opposed to it. 93. Kentucky and Virginia adopted the bondsecured bank-note rather than general incorporation in 1851 and 1852, respectively, being preceded in this respect by Michigan which in 1849 incorporated this feature into bank charters. In Maryland e Free Banking bill was defeated in 1852. Helderman, £Q. cit., 149 and 97. In 1857 it was suggested by the Superintendent of the Banking Department of the State of New York to' extend the principle of bond security for banknotes to the still existing Safety Fund banks. ~' MagaZine, XII, 615 ff. 94. The Louisiana Free Banking Act stipulated a thirty-three and one-third per cent specie reserve against deposits and a restriction of at least two-thirds of all loans to ninety-day paper, thereby becoming the most modern banking act of that time. The progress is here due, as it is so often, to a combination of two different sets of idea~. 95. An Act to organize and regulate banking aSSOCiations, No. XLVII, Acts of the Legisla ture of the ~ta te of MiChigan, passed at the annual session of 1837 (DetrOit, 1837), 76 ff. 96. Felch, £Q. cit., 114. The act was praised in 1837 by a contemporary author for "divesting the state of its undue control of banking." Rafinesque,.2l2..!. cit., 37. 97. ~higan Senate Journal, (1837), 45, 61, 147; ~ of the Michigan ~ of Representatives, (1837), 144. Biographies of Ellis and Lothrop in Michigan Biographies (Lansing, 1924), I, 272 and II, 36, respectively. 98. Hammond, Bray, £Q. Cit., 196.
297
Free Banking II
117
REFERENCES Benton, Thomas Hart. Speech on the Resolutions offered by Mr. Clay ... relative to the Removal of the Public Deposites from the Bank of the United States. (Delivered In the Senate, Jan. 2, 3, 6, 7, 1834.) Washington, 1834. Benton, Thomas Hart. Thirty Years' View or A History of the Working of the American Government for Thirty Years, from 1820 to 1880. New York, 1854. Benton, Thomas Hart. "On Banks and Currency" In Bankers' Magazine, Vol. XII (1857/58). Biddle, Nicholas. Oration Delivered before the Pennsylvania State Society of Cincinnati, on the Fourth of July, MDCCCXI. Philadelphia, 1811. Biddle, Nicholas, compiler. Commercial Regulations of the Foreign Countries with which the United States have Commercial Intercourse.
Washington, 1819. Biddle, Nicholas. Address delivered before the Philadelphia Society for Promoting Agriculture, at its Annual Meeting of the 18th of January,
1822. Philadelphia, 1822. Biddle, Nicholas, Mathew carey, John Sergeant ... [and others, inclUding Richard Peters, Jr.). Address to Those on Whom Heaven has Bestowed the Goods of Fortune and What is More Valuable, Hearts to Make Proper Use of Them for the Public Benefit. Philadelphia, 1824. Biddle, Nicholas. Eulogium on Thomas Jefferson, delivered before the American Philosophical Society on the Eleventh Day of April, 1827.
Philadelphia, 1827. Biddle, Nicholas. Documents from ... , President of the Bank of the United States to Hon. G.M. Dallas. 22d Congr., 1st sess., Senate Doc. 98 (1832). Biddle, Nicholas. Account of the Proceedings on Laying the Corner Stone of the Girard College for Orphans, on the Fourth of July 18:53: Together with the Address, Pronounced on that Occasion at the Request of the Building Committee. And a Description of the Plan of the College, by the Architect. Philadelphia, 1833. Biddle, Nicholas. An Address delivered before the Alumni Association . ... September 30, 1838.
Princeton, 1835. Biddle, Nicholas. Two Letters Addressed to the Han. J. Quincy Adams concerning a History of the Recharter of the Bank of the United States.
London, 1837. [Four further letters to the same were published in the newspapers and are to be found in the Financial Register, Vol. I, pp. 396, 342,400; Vol. II, p. 391.) Biddle, Nicholas. Reply to the Report of a Committee of the Bank of the United States of Pennsylvania. Paris, 1841. [Biddle, Nicholas.) Questions Submitted to the PreSident of the Bank of the United States by Mr. Cambreleng with his Answers Thereto.
[n.p., n.d.) Biddle, Nicholas. The Correspondence of Nicholas Biddle dealing with National Affairs, 1807-1844,
ed. by Reginald C. McGrane. Boston, 1919.
[Bronson, Isa.a.c.) An Appeal to the Public on the Conduct of the Banks in the City of New York, by a citizen. New York, 1815. [Bronson, Isa.a.c.) "General Propositions Explanatory of the Elementary Principles of Banking" In The Free Trade Advocate, ed. by Condy Raguet, Vol. II (1829). Flagg, A.C. Banks and Banking in the State of New York from the Adoption of the Constitution in 1777 to 1864. Brooklyn, 1868. Gallatin, Albert. The Writings of .... , edited by Henry Adams. Philadelphia, 1879. [Ga.llatin, Albert.) "Banks and Currency" In Amer. Quart. Rev., Vol. VIII (1830). Ga.llatln, Albert. ConSiderations on the Currency and Banking System of the United States. Philadelphia, 1831. Gallatin, Albert. Suggestions on the Banks and Currency of the several United States . ... New York,l84l. Hamilton, James A. Reminiscences; or Men and Evente at Home and Abroad during Three Quarter of a Century. New York,
1869. Hammond, Bray. "Long and Short Term Credit in Early American Banking," in Quart. Jour. Econ., Vol. XLIX (1934-35). Hammond, Bray, "Free Banks and Corporations: the New York Free Banking Act of 1838," in Jour. Pol. Econ., Vol. XLIV (1936). Hildreth, Richard. History of Banks: to which is Added a Demonstration of the Advantages and Necessity of Free Competition in the Business of Banking. Boston, 1837. Hildreth, Richard. Banks, Banking, and Paper Currencies. Boston, 1840. Hildreth, Richard. A Letter to his Excellency Marcus Morton on Banking and the Currency.
Boston, 1840. Jackson, Andrew. Messages. Boston, 1837. Jackson, Andrew, "Letters and Papers of ... " in Builetin of the New York Public Library, Vol. IV (1900). Jackson, Andrew. Correspondence of ... , edited by John Spencer Bassett. Washington, 1926. Lord, Eleazar. Principles of Currency and Banking. New York, 1829. Lord, Eleazar. On Credit, Currency and Banking. [2nd edition of Principles of Currency and Banking) New York, 1834. Lord, Eleazar. A Letter on National Currency addressed to the Secretary of the Treasury. New York, 1861. Lord, Eleazar. Six Letters on the Necessity and Practicability of a National Currency and the Principles and Measures Essential to it. New York,1862. [McVickar, John.) Hints on Banking in a Letter to a Gentleman in Albany, by a New Yorker. New York,1827. [McVickar, John.) "Importance of a National Bank," in The New York Rev., Vol. VIII (1841).
118
Free Banking II REFERENCES
Mann, Abijah. Correspondence between A. Mann Jr. and Hon. A.C. Flagg in Relation to the General Bank Law of New York. Brooklyn, 1868. Parnell, Henry Brooke, 1st Baron Congleton. Observations on Paper Money, Banking, and Overtrading . .. which explain the Scotch System of Banking. 2nd edition, London, 1829. Raguet, Condy. An Inquiry into the Causes of the Present State of the Circulatory Medium of the United States. Philadelphia, 1816.
Raguet, Condy. "Report on the Renewal of Bank Charters made to the Senate of Pennsylvania by ... Chairman on the 16th of January 1821," in Examiner, Vol. II (1834/36). Raguet, Condy. "Report of the Committee of the Senate of Pennsylvania, appOinted to enquire into the extent and causes of the present general distress," in Financial Register, Vol. II (1838). Raguet, Condy. A Treatise on Currency and Banking. London, 1839. 2nd edition, Philadelphia, 1840. Ricardo, David. Proposals for an Economical and Secure Currency; with Observations .... London, 1816.
Ricardo, David. Plan for the Establishment of a National Bank. London, 1824. Smith, Adam. An Inquiry into the Nature and Causes of the Wealth of Nations-with a Life of the Author and Introductory Discourse by J.R. McCulloch. Edinburgh, 1828. Steuart, Sir James. Principles of Banks and Banking of Money, as Coin and Paper ... London, 1810. (Being Book III of his Inquiry into the Principles of Political Economy, London, 1767.) Van Buren, Martin. The Autobiography of ... Washington, 1920. Venit, Abraham H. "Isa.a.c Bronson: His Banking Theory and the Financial Controversies of the Jacksonian Period," in Jour. of Econ. Hist .. , Vol. V (1946). [Young, Samuel]. Considerations on the Bank of the United States in which its Repugnance to the Constitution Albany, 1832. Young, Samuel. Oration Delivered at the Democratic Republican Celebration of the 64th Anniversary of the Independence of the United States. July 4, 1640. New York, 1840.
[7] HUGH ROCKOFFo
The Free Banking Era A Reexamination INTRODUCTION
During the two decades preceding the Civil War the regulation of banking in the United States was left to the states. Various schemes were tried, but the most famous and most controversial was free banking. The term free banking meant something very specific at this time: it meant a banking system with free entry and a bond-secured note issue. Free entry provided that any potential banker who could raise a certain minimum of capital could start a bank wherever he chose. Under the older system of chartered banking, the potential banker had to secure a special grant from the state legislature. The bond security provision of the law worked in the following way. Banks were allowed to issue paper currency redeemable in gold or silver under free banking as well as under other regulatory systems. But under free banking, designated government bonds had to be deposited with a state authority as security for all circulating notes issued by a bank. The bank, so long as it remained solvent, was entitled to the interest on the bonds. But should it fail to honor its notes, the state would sell the securities and reimburse note holders out of the proceeds. Free banking was intimately related to wildcat banking, the formation 'I am indebted to the members of my Ph.D. thesis committee, Robert W. Fogel, Donald McCloskey, and Richard Zecher for potent doses of help and encouragement. I have also benefitted greatly from criticisms made by Stanley Engerman and Roger Hinderliter. They are not, of course, responsible for any remaining errors of fact, interpretation, or conclusion. This caveat applies with special force to the first three named, as this paper has changed somewhat since they were most intimately connected with it. HUGH ROCKOFF
is assistant professor of economics at Rutgers University.
120
Free Banking II 142
MONEY. CREDIT. AND RANKING
of banks with note issues of far greater volume than they could hope to continuously redeem. While something similar to wildcat banking could and did happen under the older system of chartered banking, it was less likely because it required the initial cooperation of the legislature. Judged by its ability to survive, the free banking law proved reasonably successful. Free banking laws were first passed in Michigan in 1837 and in New York and Georgia in 1838, an understandable response to the destruction of the Second Bank of the United States, and the resulting increase in the importance of state laws in regulating the banks. On the eve of the Civil War over half the states, including the most populous, had free banking laws. As late as 1858 free banking laws were passed in Iowa and Minnesota, and Pennsylvania adopted free banking in IR60. Michigan's legislature passed a free banking law in 1857, even though an earlier law produced disaster. Thus, the National Banking Act passed during the Civil War, a sort of national free banking act, was the continuation of a movement that was well underway and perhaps accelerating in the decade preceeding the Civil War. But traditionally, historians have judged the experiment, at least in the western states, a failure. However, despite the attention paid to American monetary history during the past decade, this episode has not been reexamined. The purpose of the present study is to begin such a reexamination in the light of the extant quantitative data, frail as it is, and modern monetary theory. Our primary effo.rt will be aimed at filling three omissions in traditional interpretations. The first omission is an explanation of the diversity of experiences under superficially similar free banking laws. The most striking contrast is between the free banking laws passed in Michigan and New York in the late 1830s. These laws appear very similar on the surface and their historical roots are entwined. Yet free banking in Michigan was a disaster giving rise to some of the most famous stories of wildcat banking, while the New York law (after some initial rough sledding) was the basis for one of the most successful banking systems of the ante-bellum period. No serious evaluation of free banking can be completed without an explanation of why some free banking laws exploded in "hyper-inflations" while others did not. Some writers have hinted that the source of instability was the security provisioll liS, p. 164; 19, pp. 55-56J. However, these suggesliolls do 1101 constitute an analytically complete statement. This issue is treated in section
2. A second important omission from traditional interpretations, at least from a modern point of view, is that they make no attempt to quantify the damage produced by wildcat banking. This issue is treated in sections 3 and 4. A third omission from traditional interpretations is the failure to ask whether free banking could have produced any benefits for the
Free Banking II
121 HUGH ROCKOFF
:
143
states which adopted it. We discuss one possibility, a more efficient allocation of bank capital, in section 5. 1 Before plunging into these issues however, it will be useful to examine the behavior of a bank-issued currency.
I.
THE ECONOMICS OF A BANK-IsSUED CURRENCY
Dllfing the free banking era, as we have noted, paper currency was issued by banks. The specie price of bank notes varied depending on the location and condition of the issuing bank. This state of affairs has attracted considerable attention because it appears to differ so greatly from our present system. The market's response, generally, was a system of private brokers who exchanged bank notes for specie, discounting the notes as a charge for their services. The discounts they charged were published in "bank note reporters." These were periodicals which listed each bank by state and city, and the discount (as a percentage of the face value) on its notes currently prevailing in the financial center where the reporter was published. The reporter also described any counterfeits. The specie paying banks would typically be listed at a small discount from par. Another group of banks was the "broken" banks, and the notes of these would often not be acceptable at any price. Some observers have deduced that the ante-bellum banking system was chaotic from a count of failed banks and counterfeits in bank note reporters [26, pp. 177-78]. However, any deduction about the state of the currency must be made carefully because the reporters generally listed counterfeits and bank failures even if the notes had been removed from circulation years before. The reason for this was that someone who had managed to save a stock of worthless notes might reintroduce them at some later date. A detailed comparison of a pair of bank note reporters by the same publisher issued in December of 1846 and November of 1843 revealed that slightly more than ninety percent of the banks listed as broken, failed, closed, or fraudulent in the 1846 reporter had also been listed in the earlier issue. In this particular case the danger in using bank note reporters for making inferences concerning the state of the ante-bellum banking system is similar to making inferences about the state of the banking system in the 1940s from a list of failed banks going back to the early part of the depression. Table 1 displays the nationwide pattern of discounts for selected dates. It shows that most banks were at small discounts from par, discounts that decreased over time, perhaps as a result of improvements in transportation and communication. 'The classic histories of nineteenth century banking [13,26], generally eschew 'luantitalive They tend 10 concentrate on the safety of bank notes and deposits L2S].
mea~urement.
Free Banking II
122 144
MONEY. CREDIT. AND BANKING TABLE I Discoums ON BANK NOTES AT PHILADELPHIA, SELECTED DATES, 1845-58 (Percents)" November
November
November
November
November
November
1845
1850
1855
1856
1857
1858
State
New England New York New Jersey Pennsylvania Delaware Maryland Vir~il1ia
North Carolina South Carolina Georgia Alabama Mississippi t Louisiana Tennessee Kentucky Missouri Ohio Indiana Illinois Michigan Wisconsint
1/2 3/4 3/4 0 0 1/4 i
1-1/2 1-1/2 2 5-1/2 80 2 2-1/2 1-1/2 1-3/4 2 2 70
65 ?
3/8 3/4 0 0 0 1/2 3/4 1-1/4 1 I
3-1/2 75 1 2-3/4 1-1/2 1-1/2 1-1/2 1-1/2 75 3 ?
1/4 1/2 1/4
1/4 1/2 1/4
0 0
0 0
1/2 I 1-3/4 I I
1/2 I 1-1/4 3/4 I
.5
.5
? 1/4 2-1/2 I I I 5 2 1-1/2 2
? 1-1/4 5 I I I 5 2 1-1/2 2
3/4
3/8 1/2 1/4
I
3/4 0 0
(J
0
.5
1/2 5/R
10 10 10 10 5
I
?
.5 20 10 10 10 10 20 10 20
1 3-1/2 ? 3/4 1-1/2 3/4 I 3/4 2 1-1/2 1-1/2 1-1/2
·Sources: Various iS5ue~ of Van Courl',~ Cou"terfeit Detrctor and Banle Note Lbt. tThe ? for Mio:;sissippi and Wisconsin indicates th;Jt these notes were of doubtful value, and only purchao:;ed under "recial circumstances. The tahTe gives the mooal discount.
An implication of this pattern of discounts is that it is unlikely that there were large fluctuations in the rate of exchange among regions during the free banking era except during the suspensions of specie payments in the early I 840s and late 1850s. This would seem to eliminate the possibility of regional devaluations curing incipient unemployment regularly. However, further research on this point is warranted. The devaluations that could take place during suspensions (note the large discounts in November, 1857) might have been sufficient to have reduced unemployment when the competitive position of particular regions deteriorated. Whatever the benefits from flexible exchange rates, a heterogeneous currency did make exchange less efficient. Each time a transaction took place the seller had to make some judgment about the quality of the particular set of bank notes being offered. Clearly the process of making this judgment used real resources, particularly labor, that would not have been used if the currency were homogeneous. Moreover, because of the inconvenience of the paper currency, people were led to use larger amounts of specie than otherwise. In 1859 the ratio of specie outside the treasury to the total money supply was .41. Twenty-five years later, under the National Banking system, the ratio was only .15 [12, pp. 7, 224-25; 32, p. 648]. Nevertheless, it seems unlikely that the heterogeneous nature of the
Free Banking II
123 HUGH ROCKOFF
:
145
currency was a major brake on economic growth, for in many crucial respects the system was little different from that which prevails today. Locally we use demand deposits. But these are not generally acceptable as a means of payment. Each time we wish to make a purchase by check from a businessman we force him to make some judgment about the quality of the money we are offering. Instead of having to worry about different kinds of bank notes a merchant today must worry about different kinds of deposits which could be as numerous as his customers. Counterfeiting currency is now rare, but forged checks and insufficient balances are a constant irritation. Yet no one today would argue that the heterogeneity of our deposit money is a serious impediment to the growth of national income. This analogy can be extended. Since deposit money is not generally acceptable it is more convenient when traveling to use currency or traveller's checks. During the free banking era people used bank notes locally but switched to specie (or possibly the notes of some well-known bank) when traveling. Today without much complaint we pay a small charge for travelers' checks. During the free banking era one had to pay a similar charge to a broker who exchanged "foreign" money for the local currency. We do not wish to argue that the system worked perfectly. A national currency might have been preferable. But in the light of the close similarity between the workings of the ante-bellum system, properly understood, and our present-day system, the inefficiency of a heterogeneous currency should not be exaggerated. It should also be remembered that free banking did not add to the heterogeneity of the currency except as it led to an increase in the number of banks. To the contrary, free banking greatly reduced economic heterogeneity by standardizing the assets that banks held against notes. In New York the particular kind of security backing the note was stamped on it. It was even possible under free banking for all of the free banks to use the same type of note in order to inhibit counterfeiting.
II. THE CONNECTION BETWEEN FREE BANKING AND WILDCAT BANKING 2 Wildcat banking was made possible by the free entry provision of the free banking law. But whethe~ it was profitable depended on the design of the bond security provision. Let us consider several cases. First, suppose the market value of the eligible bonds was less than the "legal value" (the value of the notes that could be issued on a bond). Under this condition it obviously paid to set up a zero reserve bank providing the notes could be placed in circulation at close to par before the enterprise went bankruot. 2This section was stimulated by [5, pp. 86-95}.
124
Free Banking II 146
:
MONEY, CREDIT, AND BANKING
The "income statement" of a wildcat bank of this sort with a circulation of $100 might look like this: Market value of notes issued -Market value of bonds (or mortgages) deposited Net Income
$100 90
$ 10
This simple model explains several important cases of wildcat banking. In Michigan (in 1837) mortgages on land were eligible and they were accepted at par [2 I, p. RO]. It was thus possible to create a mortgage on a worthless piece of property, have it certified as being valuable by some friends, and then transfer it to a wildcat bank in exchange for a mass of bank notes. This seems to be the process by which many of the Michigan wildcats issued currency. Others were simply frauds which operated in violation of the free banking law. It should be noted, however, that there was an important difference between Michigan's experience with wildcat banking (the nation's first) and later episodes: Michigan's occurred during a legal suspension of specie redemption. Thus, the situation in Michigan was unique, a group of men could issue bank notes with practically no cost to themselves and unchecked by the need to redeem the notes in specie. However, while the free banks were not required to redeem their notes in specie they were required to have 30 percent of their authorized capital on hand in the form of specie when they began operation. The often quoted complaint of the Michigan bank commissioners, "gold and silver flew about the country with the celerity of magic; its sound was heard in the depths of the forest, yet like the wind one knew not whence it came or whither it was going," [36, p. 1129] referred to attempts by the wildcat bankers to evade this requirement. 3 It is interesting to note that some 20 years after her first disastrous experiment with free banking Michigan passed a second free banking law. The eligibility requirements were far more strictly drawn than for the first [23, p. 366]. The monetary statistics of the resulting banking system clearly show that there was no wildcat banking. Michigan's two free banking laws were not part of a controlled experiment. Nevertheless, the contrasting experiences under the two laws appears to be strong evidence that the source of wildcat banking was to be found in the provisions of the free banking law rather than in something else, for example, in the conditions of frontier life. Indiana's experience with wildcat banking was perhaps another example of the simple model. According to the free banking law, bonds were to be accepted only at the minimum of the market and par values, so that there does not seem to be room for a banker to have received notes J
It has been quoted by [14, p. 28J among many others.
125
Free Banking II HUGH ROCKOFF
:
147
worth more than the bonds deposited as security [18, p. 24]. Sut according to one authority the auditor may have valued Indiana bonds at par [2, p. 172]. This could have been done because the auditor was not aware of the significance of the distinction between the par and market values. In a "currency reform" which took place in 1855, the security provision of the law was considerably rewritten and the requirement was changed so that banks received at most $91 dollars worth of notes for every $100 dollars of bonds deposited [18, p. 34]. No wildcat banking took place after this change in the law. The clearest case of the simple model occurred in Minnesota [25]. The initial version of Minnesota's free banking law was passed in 1858. In this version only bonds of the United States, Minnesota, and other states approved by the banking authority were eligible. Soon after passage, however, this part of the law was amended to extend eligibility to the Minnesota State Railroad Bonds. These were to be issued on behalf of certain railroads in exchange for mortgage bonds. When the railroads applied for the bonds due them under the law, the governor refused to issue the bonds unless a first lien on the assets of the railroads was given to the state. The railroads refused, and took their case to the Supreme Court of Minnesota. The ruling was in favor of the railroads. The nature of the ruling along with threats of repudiation made in certain newspapers led to a rapid depreciation of the bonds in the market. This need not have caused wildcat banking, because under the free banking law the banking authority was empowered to reduce the price at which he accepted depreciated bonds. Under the circumstances, however, the banking au thori ty was reluctant to take this step because it might appear to be a repudiation of the state's debt. The railroad bonds, virtually worthless in the market, were accepted at 95 percent of par. The result was wildcat banking. There is certainly one and perhaps several episodes of wildcat banking which do not seem to fit the simple model we have been using. However, a natural extension of it seems sufficient to cover these cases. Suppose that a wildcat bank could expect to survive for as long as, say, one year. Then, even if the value of the notes issued was no greater than the market value of the bonds deposited, the interest earned on the bonds might be sufficient to induce wildcat banking. The "income statement" of such a bank with a circulation of $100 might be as follows: Market value of notes issued + interest on bonds + surplus returned to shareholders
$100 6 4 $110
- Market value of bonds deposited
$104 Net income
$ 6
126
Free Banking II 148
:
MONEY. CREDIT. AND BANKING
New Jersey's episode seems to fit this model. The free banking law of New Jersey passed in 1850 provided that bonds of the United States, New Jersey, or Massachusetts could be deposited. The value of the notes that could be issued was set equal to the par price of the eligible bond with the additional provision that the market price be at or above par. There were no signs of wildcat banking during the first year of operation. However, in IR51 the free banking law was amended to permit banks to deposit bonds isslIed by New York, Ohio, Pennsylvania, and Kentucky. These were all somewhat closer to par although above it. Finally, the law was altered in 1852 to permit banks to deposit Virginia's bonds [10, pp. 52-64]. The latter amendment soon provided a bitter lesson in the danger a state faced if it linked its currency to the debt of another state. About one year earlier Virginia had passed a law which permitted her government to issue debt from time to time. In subsequent years Virginia greatly increased her debt. The largest addition was made in 1853 when some $4.6 million was issued, surely one of the largest deficits of any state in the ante-bellum period [34, p. 554]. Part of this debt was absorbed by New Jersey's free banks during a rapid expansion of the free banking system. Wildcat banking in New Jersey was similar in style to that in the western states [8, pp. 76-77]. It appears that wildcat banking could have been prevented if the laws had required a free bank that failed to pay damages to note holders in addition to the face value of the note, and if this protection had been assured by demanding that the value of the bonds deposited exceeded the value of the notes issued by an amount sufficient to pay slIch damages. A damage rate which suggests itself on a priori grounds would be the normal rate of interest, since this is what the note holder could have earned had he invested in some safe alternative. So far we have discussed wildcat banking in a very different way from what is usual. We have emphasized the bond security provision and state debt policies. We have not even mentioned the remote areas in which wildcat banks were located, or the various tricks used to keep notes in circulation. Of course wildcat banks used these tactics. The more notes they could get into circulation, the greater their profit. However. it is not usually recognized that in at least two ways the bond security provision contributed to the tendency of banks to locate in hard-to-reach places. First, under free banking, the state banking authority was empowered to sell all of the securities of the bank if a single note holder was refused specie. Thus, the banker who did not want to honor demands for specie had few alternatives to locating his bank as far as possible from the principle area of circulation. A second factor was the knowledge each note holder possessed that should a bank fail all of the notes would be redeemed on a pro rata basis out of the proceeds from the sale of the deposited bonds, and usually in the banking authority's office in the state capital
127
Free Banking II HUGH ROCKOFF
:
149
or in the state's major commercial center. Thus each note holder separately had an incentive to let someone else bear the costs of seeking out the wildcat banker, and, being refused specie, of informing the banking authority. The framers of the free banking laws were apparently walking a tightrope. If they made the gap between the market price of bonds and the "legal price" small or negative they risked, as we have seen, wildcat banking. If they made the gap slightly larger they would produce a "sound" free banking system as was the case in New York, Ohio, and Louisiana. If they made the gap still larger the free banking law could prove abortive. To take one example of the latter phenomena, Massachusett's free banking law was passed in 1851 but no banks were organized under it until 1859. Massachusetts limited the bond holdings of free banks to bonds of states in New England (which had small debts), and to bonds issued by towns in Massachusetts. The state securities sold at substantial premiums in the 1850s. Yet the par price of the bonds was the legal maximum. Moreover, the bond had to be made equal to one yielding 6 percent, so its par value, for the purposes of the law, was reduced by one-sixth if it yielded 5 percent. As the decade drew to a close a number of free banks were started on the basis of securities issued by towns in Massachusetts. These securities were, presumably, somewhat cheaper since they were more liable to default.
III.
LOSSES ON BANK NOTES
A bank failure was, in the first instance, a transfer from the note holder to the wildcat banker that left the net wealth of the community unchanged. But the uncertainty and inconvenience caused by wildcat banking could have produced decreases in total real income. In this section we examine the redistribution of wealth, while the efficiency effects are dealt with in section 4. Table 2 presents an estimate of the losses ultimately suffered by holders of free bank notes through the year 1860. It includes losses due to wildcat banking as well as losses due to ordinary mismanagement and bad luck. For the most part the estimates were compiled from standard secondary sources. When, as sometimes happened, somewhat arbitrary adjustments had to be made, a procedure which tended to bias the estimate upward was followed. In interpreting Table 2 it is useful to distinguish between the states which had "sound" free banking laws, that is, with adequate bond security provisions, and those which did not. In the former, with the exception of New York, losses were mild. In New York most of the losses came in the early years of the law when the security provision allowed the bonds of states besides New York to be used.
128
Free Banking II
150
MONEY. CREDIT. AND BANKING TABLE 2 LoSSF.s SUFI'F.RF.D BY HOI.DF.RS OF FRF.F. BANK NOTF_~ FROM TIIF. FIRST YEAR OF FREE BANKING THROUGH 1860*
State
Vermont Massachusetts Connecticut New York New Jersey Pennsylvania Ohio Indiana Illinois Michigan
First Yea,
1851 1851 1852 1838 1850 1860 1851 1852 1851 1837
Loss (dollars)
24,500 0 0
394,700 6,000 0 77,600
227,900 21,300 1,000,000
State
Michigan Wisconsin Minnesota Iowa Georgia Florida Tennessee Alabama Louisiana
First
LO!li~
Year
(dollars)
1857 1852 1858 1858 1838 1853 1852 1849 1853
-t
Total
1,851,900
0
96,900 3.01K) 0
0
·Each estimare is rounded off tJ hundred •. The dates refer to the years in which the free banking laws were p'"ssed and are laken from lhe Slalule. of lhe various slale •• In some stales additional losses occurred during Ihe Civil War due. generally, to the depreciation of southern bonds. For a list of leading references and assumptions used in estimating lhe losse•• see Ihe Appendi •. t-signifie. Ihal hllie or no banking was done under lhe fr.. banking law.
The experience under wildcat banking was quite varied. In Michigan the notes in many cases, although not all, became worthless. If the estimate in Table 2 were correct, the total volume of wildcat money would have amounted to about II percent of the annual income of Michigan in 1840 [9, p. 98]. However, while the bank commissioners refer to $1,000,000 as a low estimate, it is likely that they had in mind the face value of the wildcat issues. Even from the first, many of these notes may have borne heavy discounts for it appears that the public caught on rather quickly to the condition of the wildcat banks. The bank commissioners had already closed a number of banks and had officially reported their findings in March of 1838. Disenchantment was sufficiently widespread by April of 1838 to induce the legislature to suspend the free banking law [22, pp. 246-47]. Thus, the true condition of the wildcats was surely common knowledge sometime early in 1838. Since few of the wildcats were organized until after the general suspension of specie payments in June of 1837, the life span of the typical Michigan wildcat was about six months at the most. The episodes of wildcat banking which occurred in the 1850s were apparently not as costly to the note holder. In Indiana, for example, it appears that losses were frequently less than 5 percent. Moreover, many of the notes were probably accepted initially at some discount from their face value. It is even conceivable that some note holders made money from the wildcat bankers by taking the notes from the banker at a larger discount than prevailed when the notes were finally redeemed by the state. Seventeen free banks, a substantial number for a frontier state such as Indiana, survived to the Civil War. Today, after nearly three decades with almost no bank failures, we might
Free Banking II
129 HUGH ROCKOFF
:
151
regard the failures that occurred in Indiana as catastrophic, but contemporaries took a more tolerant view. To illustrate this we quote at length the state auditor's view of the damage: The experiment of free banking in Indiana, disastrous as it has been in some particulars, has demonstrated most conclusively the safety and wisdom of the system. The original bill was crude and imperfect, admitting of such construction as held out to irresponsible men inducement and facilities for embarking largely in the business of banking, without the ability to sustain themselves in a period of revulsion. That revulsion came. . . and yet the loss to which the billholder was necessarily subjected. in many cases, did not exceed five percent, and in no case exceeded twenty percent of the amount in his hands. [37, pp. 183-84.]
The other episodes of wildcat banking were more similar to Indiana's experience than to Michigan's. Few reliable estimates comparable to those in Table 2 exist for the non-free banking sectors of the. banking system, or for deposits.4 The upshot is that our estimate must be judged as it stands. Nevertheless, it seems to be a rather small number. It means that by 1860 note holders had probably lost less through the failure of free banks, including the wildcats, than they stood to lose in that year from a 2 percent inflation [12, p. 225]. Undoubtedly, this estimate could be considerably refined. But it seems unlikely that the unearthing of new data will require a substantial upward revision.
IV.
TOWARD A MEASURE OF THE EFFECT OF WILDCAT BANKING ON THE EFFICIENCY OF EXCHANGE
We would like to know whether in addition to the redistribution of wealth to which Table 2 is addressed, there was also a decrease in the income of the community as a whole. Our analysis follows Bailey's examination of the cost of anticipated inflation [1]. To explicate the argument we will use Figure 1. Here the cost of holding bank money in cents per dollar per year is measured along the vertical axis, while the amount held is measured along the horizontal axis. The vertical distance between the demand curve, DD and the horizontal axis is a measure of the marginal productivity of money. The area under the DD curve is thus the total value of monetary services. This relationship permits a deduction of the efficiency costs of wildcat banking. Suppose that the cost of holding money given a "sound" banking system would be Ca' This cost might be simply the interest on U.S. treasury bonds, a safe alternative to money. Under 4 An estimate by Jay Cooke placed losses on all bank nOles al $50 million per year [24 vol. I, p. 327]; also quoted in 6, p. 21].
r
130
Free Banking II 152
MONEY, CREDIT, AND BANKING
Cost of
Holding
o
o Real Cash Balances
FIG. I. The Demand for Money
wildcat banking, costs would be higher, say C b , because from time to time a money holder would expect to discover that part of the money he held had been issued by wildcat banks. For example, if the yield on U.S. bonds were 5 percent, and if a money holder expected that over the course of a year 5 dollars out of a 100 would turn out to be worthless because they were issued by wildcat banks, then the total cost of holding money would be 10 percent per year. Returning to Figure I, people would wish to hold only M b of real money balances under a regime of wildcat banking. The cross hatched area in Figure I, the expected loss rate (C b - C a) multiplied by the total amount of money held, M b' is the amount of wealth that people anticipate will be lost due to wildcat banking. We can consider the cross hatched area a pure transfer from note holders to bankers leaving the wealth of society as a whole unchanged. However, this transfer produces a loss in efficiency. The value of the services produced by the money which is not held under wildcat banking but which would be held under a "sound" banking system is the entire area between the DD curve and the horizontal axis bounded by M band M a' The social cost of these services is the product in alternative uses of the resources used in producing this amount of money.
131
Free Banking II HUGH ROCKOFF
:
IS3
For simplicity, assume that the social cost is simply the amount of specie held by banks multiplied by the interest rate-hence it is represented by the line C xr where r is the reserve ratio. A more complete analysis would " notice of the use of other forms of physical capital (bank have to take buildings and furnishings) and labor in producing money. The efficiency cost of wildcat banking, then, is the stippled area in Figure 1, the total value of the services of money that society loses, less the resources saved when the production of monetary services is reduced. It is important to emphasize that the stippled area is a measure of the amount of resources saved by the use of money which would otherwise be used in precisely those activities which historians have designated as having been effected by the quality of the currency in the free banking era. The labor used by merchants in carefully examining the currency offered to them is the most frequent example. Another is the inconvenience experienced by travellers who had to convert their home bank notes into notes circulating in the region they were visiting. Perhaps most important is the return from productive capital which is foregone because people must hold larger balances of specie. To directly apply the logic embodied in Figure 1 we would need to know the stock of bank money, the expected loss from wildcat banking, and the elasticity of the demand for money with respect to the cost of holding it. Unfortunately, direct estimates of the latter two are out of the question. Forming an estimate of the expected loss from the data fragments we possess would involve making extremely arbitrary assumptions about how expectations were formed. Moreover, the elasticity of demand cannot be estimated because regional interest rates have not been developed for the ante-bellum period. However, we were able to estimate a rough substitute for the true demand function which allows one to obtain an idea, albeit an imprecise idea, of the losses from wildcat banking. This substitute is a cross-state demand function which includes a dummy variable that takes the value 1 in states which had experiences with wildcat banking and 0 in states which did not. The coefficient on this variable will be a product of the elasticity of demand and the cost of holding money, provided that we have taken account of the other important variables. The actual regression that was run was an ordinary least squares regression of the following form (I)
where M W
the stock of bank money per capita, wealth per capita,
132
Free Banking II
154
MONEY. CREDIT. AND BANKING
U FJ
E
n, ~J' ~2' ~3' and ~4
urbanization, a dummy variable which state experienced wildcat not, a dummy variable which state had a "sound" free it did not, a random error term,
takes the value 1 if the banking and 0 if it did takes the value 1 if the banking system and 0 if
the coefficients to be estimated.
The use of wealth was suggested by demand for money studies based on modern data which indicate that wealth or income is an important variable. Urbanization was considered a good proxy for the demands for money generated by commercial and industrial activity. However, this interpretation is open to challenge, urbanization may be a proxy for variables effecting supply. For example, a greater density of banks may induce people to hold more money. Or it may even be a proxy for our unobserved regional interest rates. But even if one of these alternative interpretations apply we can still use the reduced form equation to estimate the monetary deficit in the wildcat banking states. Experimentation with a number of other variables failed to reveal a strong relationship with per capita money holdings. The F2 dummy was added so that we could compare wildcat banking directly with chartered banking. The equation was estimated for the year 1860. This was the only year in the ante-bellum period that met the twin requirements that it be a census year so that wealth estimates would be available, and that it follow the major wildcat banking episodes. In fact, the latter requirement was only partially met for Wisconsin and Illinois. Money was defined as the sum of deposits and currency (net of currency held by banks) for the year 1860. The monetary data was taken from the summaries given in the Report of the Comptroller of the Currency for 1876. Since the Report generally gives the data for the end of the preceding year, the comptroller's series were backdated one year. After deleting those states for which the data was obviously incomplete, we had a sample of 27 observations. The nominal money holdings were then deflated by the number of free people [32, pp. 12-13; 33, p. 7]. The wealth index was constructed by summing the state by state estimates of real and personal property given in the census of 1860 [30, p. 319]. This index was then deflated in the same way as the money index. The urbanization variable is simply the percentage of the total population living in communities with population greater than 2,000 [31, pp. (1-30)-(1-37)]. Two specifications of the basic equation are presented below. One uses a broad definition of the states in which free banking produced wildcat
Free Banking II
133 HUGH ROCKOFF
:
155
banking, viz. Florida, Tennessee, Indiana, Illinois, Michigan, and New Jersey, (variable F I ) and a broad definition of the states with sound free banking systems, viz. New York, Ohio, Louisiana, and Virginia, which adopted a conservative variant of free banking (variable F1 ). A second specification used a narrow definition of wildcat banking including only those states which experienced intense episodes, viz. Indiana and Michigan (variable Fi), and a narrow and more proper definition of those states with sound free banking systems, excluding Virginia (variable These regressions are reported below along with the t statistics in parentheses and the R2 statistics corrected for degrees of freedom.
Fn.
M =
-S.56 + .017W + 54.02U - 6.55F. + 10.32 F2 (-1.97) (4.79) (5.64) (--I.S6) (2.56)
= .72 M = -9.47 + .01S W + 55.91 U - 7.S1 Fi + IO.SSF!
(2)
R2
(- 2.14)
R2
(4.62)
(5.63)
(-1.67)
(4.S0)
(3)
= .69
Other definitions of F. and Fl produce similar results. If we use the coefficient on F. or Fi to estimate the loss from wildcat banking we get a rather surprising result. Taking a linear approximation, the stippled area in Figure 1 is given by the following formula,
where L stands for the loss from wildcat banking. If we assume, for example, that the risk free rate of interest, c", was 10 percent, that the expected loss from wildcat banking, c b - c", was 10 percent and that the reserve ratio r was also 10 percent-three assumptions which bias the estimate upwards-and use the coefficient on F. as an estimate of Mb - Ma' then we get the following results:
L
= 6.55 [.10(1 - .10) + .5(.10)] = $.92
If we use the coefficient
(5)
Ft the result is L
= $1.09
(6)
Thus, the residual effect of wildcat banking was to lower income per capita in the effected regions by about $1.00. This finding, however, can be subjected to a number of substantial
134
Free Banking 11 156
:
MONEY. CREDIT. AND BANKING
criticisms. For example, wildcat banking may have affected the reserve ratios of the banks-banks may have been forced to hold greater reserves because people were more likely to return their notes if signs of trouble developed. In terms of Figure I wildcat banking may have shifted the Cnxrcurve upward. Thus, a calculation of the loss that included the cost of holding greater bank reserves might lead to a different conclusion. And, indeed, there did exist substantial interregional differences in bank reserve ratios of a sort that would lead one to believe that western banks had to hold large specie reserves because the public was worried about wildcat hanking. Introducing banking reserve ratios explicitly into the analysis, however, involves two vexatious problems. First, the effect of other economic variables on bank reserve ratios, such as interest rates or the distance between banks, must be taken into account. Second, and more important, the relevant ratios are not known. Gold and silver are clearly what count for the economy as a whole, since the economy must give up real resources to secure these and only these reserves. But when we examine separate states, notes or other assets issued in other states have the same property. In general we do not have information on the net balance of bank notes or other assets between anyone state and the rest of the economy. For these reasons we have not been able to isolate the effect of wildcat banking on the reserve ratios of banks in the effected regions. Perhaps the most serious objection to the analysis is that it involves comparing one ante-bellum banking system with another. The real point at issue, it might be argued, is whether the heterogeneity of the currency led to losses in comparison with the centrally-directed monetary system that existed before 1840 and the federally regulated system that existed after the Civil War. We can still use Figure I. But, now the horizontal axis records bank money balances for the economy as a whole. The cross-hatched area would represent the expected loss due to the discounting of bank notes used in interregional trade. However, this area could no longer be interpreted as a pure transfer, since the discounts would in part reflect the cost of the resources used in handling uncurrent money. The situation is similar to the analysis of a tax on an ordinary commodity when the proceeds of the tax are spent on socially wasteful activities. Stanley Engerman has attempted to measure the effect of the destruction of the Bank of the United States on the economy by determining what national income would have been in the 1850s had the economy been able to operate on the same specie-money ratio that prevailed in the 1830s, while maintaining the stock of money that existed in the 1850s [II J. U si ng this framework, Engerman showed that the cost of destroying the Bank of the United States, orfrom our point of view the cost of the late ante-bellum banking system, was small, about .15 of I percent of national income on an annual basis. It is true that per capita holdings of bank money
135
Free Banking II
157
HUGH ROCKOFF
TABLE 3 BANK PROFITS IN NEW YORK BOSTON. AND PHILADELPIIIA.
1849-59·:r DIVIUENOS AS A FRAcrlON O~ Ncr WOK"H
DI YIOENDS AS A FucnON Of THE PAR VALUE Of' CAPITAL
Year
1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859
New York
Boston
Philadelphia
New York
Boston
Philadelphia
8.79 8.70 9.38 9.03 8.85 8.87 9.08 8.61 7.73 6.91 7.56
8.06 8.36 7.82 7.78 8.08 8.65 7.98 7.81 7.73 7.43 7.31
9.79 10.60 10.30 10.27 11.13 11.40
7.63 7.33 7.51 7.69 7.84 7.86 8.()8 7.72 6.91 6.20 6.78
7.33 7.33 6.88 6.87 7.32 7.71 7.09 7.10 7.()0 6.76 6.70
8.14 9.04 8.61 8.70 9.30 9.54 9.52 9.()() 6.111 6.74 6.74
II.!K)
10.26 7.12 8.03 8.03
·The first three columns are unweigtued averages for all bunks in the cily. The second columns are the corresponding column to the left multiplied by the aggregate ralia of par capital to par capital plus surplus. tSources: I. Dividends. Boston; Joseph G. Marton. A Century 0/ FilWnce (New York: Greenwood Press. 1969) pp. 99-101. lOB-II. New York. 1849-51: Hunt', Merchant,' Magaline. vol. XXIIl. p. 89; and vol. XXVII. p. 92; 1852-59: Banker's Magazine, vol. XIV. pp. 556-57. Philadelphia, 1849-55: Banker's Muga1.ine, vol. X, p. 978; 1856-59: Communicatiun. (J{ the Auditur General 0/ Penn5ylvania Rdafive 10 Banks tmd Sewings Institutions, 1856, 1857, 1858, and 1859, passim. 2. Cup;tal and Surplus. Boston and New York; Annual Report of the Set'rttary of fhe Treasury on tile Condition of lhe SUIre lJanks (in Iht: Uouse Congressional Documents), !H49-18.59. Philadelphia, annual Report of the Sec:rdury 0/ .he Tr
increased greatly in the post-bellum era. But it seems reasonable to conjecture that a large part of these increased holdings could be attributed to the decline in interest rates, the increase in wealth, and the increase in urbanization.
V.
THE EFFECTS OF FREE ENTRY ON THE ALLOCATION OF BANK CAPITAL 5
A fundamental question concerning free entry is whether it improved the allocation of bank capital. However, there is no easy way of determining whether a particular allocation is more or less efficient than some other. Economic theory suggests that the rate of profit on capital is a good index of the efficiency of allocation so long as there is no significant divergence between the social and the private rates of return. We have, therefore. assembled some of the surviving data on the rates of return to capital in banking that allow a comparison between free banking and non-free banking states. We begin by comparing bank profits in the three major ante-bellum financial centers. Table 3 gives the average dividends as a percentage of par capital and as a percentage of par capital plus surplus for dividend-paying banks. The outstanding feature of this table is the consistently high profits earned by the Philadelphia banks. The simplest explanation for the high profit rate of Philadelphia's banks is the existence of monopolies fostered SThis section can be usefully compared with [29], which discusses post-bellum banking market structure.
136
Free &nking II 158
MONEY, CREDIT, AND BANKING
TABLE 4
CIlY,
COMPARISONS OF BANKS IN NEW YORK
New York City
Ralio ~O Dividends to market price of bank stockt 8.25 Capital to population 40 Earning assets to total assets .79 Equity to total liabilities .33 Notcs to dcposits .17
BOSTON, AND PHILADELPHIA,
1850-60.
BO~lon
60
%
~O
60
%
~O
7.25 60
-1.6 4.1
7.61 174
6.68 231
-1.6 2.8
8.04 107
.80 .40 .10
.1 1.9 -5.3
.87 .54 .45
.87 .52 .39
0.0 - .4 - 1.4
.76 .34 .36
Philadelphia 60
%
7.21 -1.4 27 -13.8 .86 .35 .24
1.2 .3 -4.1
·S""rer" n.lnnce Sheel., U.S. ('ongre ••• 1I0use l>OC:umenl 122. 32nd Congre... lSI Se.. ion. pp. 7R. 79. "R-127. 1%-20.1. U.S. Congre.. Ho"se Documenl 77. 361h Congr.... 2nd Se •• ion. pp. 9R. I~O. 159. 171-1R4. Stock price.
are from the Banlc:tr',.. MORad"t, I"Opulation from the Cen~u~e~1 and the dividend~ from the ~otJrce~ Ii!illcd in Table
J.
tThe dividrnd price rnli(l!ll nre fur IK~I Rnd IK~9. l'he percent !lyrnhol re'en to the annuAl perer-nlnge rnle of chAnge.
by the unwillingness of the legislature to charter new banks. Contemporary observers were certainly aware of the lack of bank capital in Philadelphia. Consider the following item from the Philadelphia North American which was reprinted in the Banker's Magazine: The Southwark [bank] was again refused an increase of capital at last session of the legislature, though, as will be seen [from its dividend rate] it is one of the best banks in the city, and were the legal sanctions granted, might easily obtain subscriptions to any desired amount of new capital. [3, p. 996.]
It is conceivable that Philadelphia's banks held riskier portfolio's than did Boston's, but this argument is less plausible for New York. Some evidence is presented in Table 4. The dividend price ratios for several years are given. They do not show a significantly higher rate for Philadelphia in comparison with that which would be expected for New York if its banks held riskier portfolios. However, the dividend price ratios are lower for Boston, indicating that these banks may have had a more conservative investment policy. Perhaps, the key factor here was the high capitalization rates in Boston. 6 The dramatic fall in bank capital per capita in Philadelphia-the product of a stable amount of capital and rapidly growing population-and the otherwise similar nature of its portfolio tend to confirm the diagnosis based on contemporary observations. The problem of allocating bank capital was more difficult in the areas of new settlement. For here the rate of growth of population, and changes in the distribution of population were more intense, so that errors in the allocation of bank capital were more costly. There is abundant qualitative evidence that allocation by state governments on the frontier was unsatisfactory. 6
An econometric study of bank portfolio behavior [16] produced similar results.
137
Free Banking II HUGH ROCKOFF
:
159
In Indiana the state bank, an often praised monopoly partly owned by the state, was set up before the railroads were built. The railroads, of course, completely altered the importance of various communities making the allocation of capital among the branches of the state bank uneconomic. People felt that it was time for a "new shuffle and deal." The upshot was that the legislature allowed the charter of the first state bank to expire and adopted free banking in its place [20, pp. 124-25]. In Tennessee, a state owned bank was formed as a relief measure dming the depression of the late HOOs. At its very inception the decision on where to locate branches involved the directors in a statewide controversy. While commercial demands for banking facilities were taken into consideration in the location of the branches, there was continual dissatisfaction. The location of the branches was a subject of debate each time the legislature met and in each gubernatorial campaign during the life of the bank [7, pp. 96-97]. In Missouri the chief source of controversy was the contention by St. Louis businessmen that the state bank did not allocate sufficient capital to satisfy the rapidly growing needs of St. Louis. In the late 1850s the legislature answered this criticism by chartering a new set of privately owned banks. While this law was not, strictly speaking, a free banking law, it did include a bond security provision [4, pp. 241-48, 253-56]. Free banking in Ohio was not preceded by a state owned bank, but rather by a system of chartered banks with standard provisions. Nonetheless, the belief was widespread in 1850 that Ohio was suffering from a lack of banks. Her newspapers pointed out that Ohio was third in population but far from third in bank capital. Papers in Cincinnati and Cleveland complained that these rapidly growing cities were being denied bank capital under the charter system. These arguments helped overcome the opposition and secure passage of the free banking law in 1851 [17, p. 208]. Examples of this sort, evidence of a widespread belief that free banking would improve the allocation of bank capital on the frontier, do not prove that in fact that free banking could or did improve the allocation of bank capital. But together they seem to establish a presumption that something was wrong with the older system of chartered banking. 7 There exists little quantitative evidence by which one can gauge the impact or potential impact of free banking legislation on western banking systems. The best evidence is for Ohio, where we can examine bank profits before and after the free banking law of 1851. This information is presented in Table 5, along with bank profits in certain other states for comparison. It does appear that in Ohio profits were high before free banking and were lowered as a result of it. While the low initial profits of the free banks could be attributed to a "start up" period, they could not be attributed 7 An alternative interpretation of these complaints is that they represent the traditional plea of the farmer for "easy" credit.
138
Free Banking II 160
MONEY, CREDIT, AND BANKING TABLE 5 BANK PROFITS IN OHIO BEFORE AND AFTER FREE BANKINO'
Banks
I.
2. 3.
4. 5. 6. 7.
R.
Ohio, free banks Ohio, state banks Ohio, independent banks Ohio, old banks State Blink of Indiana Hank of Kentucky Banks of New York City New England Municipal Bonds
1850
1851
15.0 13.6 12.1 10.0 10.5 9.2 5.1
15.3 14.1 12.4 9.4 9.0 9.6 5.1
t 1852
18S3
8.2 13.7 10.1 9.4 13.0
9.6 10.0 8.9
5.0
5.0
·Sollrce~: Dividtnds wtre obtRined from the !ources listed below the appropriate balance sheet information where necessitry was obtained from the annual reports on the condition of the stale banks made by the secretary of the treasury and printed as House documents. (1-4) Charles Clifford Hunlinron. A Hislory of Ba.kl'R a.d C,m,.cy I. Ohio B,fort Ihe Ch'U War (Columbus, Ohio: Hoer Prinling Co., 1964). pp. 21 .213.278-79, and 293-94; (5) William F. Harding, "n.e Slale Bank of Indiana,"' lournal of Polilical "co"omy. 3 (December. 1895), p. 23; (6) Gen. Basil W. Duke. Hblory of Ihe Ba.k of Kenlucky (Louisville: John P. Morlon & Company. 1895). p. 140; (7) Table 3; (8) Sidney Homer. A Hblory of ,.It,.sl RaltS (New Brunswick. New Jersey: Rutgers University Press, 1963), p. 287. t Prorits are defined as the ratio (as 8 percentage) of dividends to nominal capital. These rates are similar to more sophisticated ratios. The old banks were individually chartered. The state banks were similar except that each contributed to a common "safety fund" for the relief of note holders of failed banks. The independent banks were bond secured but entry was limited.
to the restrictions imposed by the bond security system since the independent banks also faced these restrictions. The preceding discussion relies on intraregional comparisons. To gauge the impact of free banking it is also worthwhile examining measures of efficiency available for all states. One measure is the density of incorporated banks with respect to popUlation. This is a rough index of the number of banks a potential customer faces when he enters the market as a depositor or borrower. Table 6 gives the number of incorporated banks per hundred thousand in 1840, 1850, and 1860. The most striking feature of the table is the smaller ratio of banks to population in the free banking states in 1860, the opposite of what one might expect from stories about wildcat banking, although the high ratio for Wisconsin in 1860 is an exception. A second supplement.try measure which suggests itself is the number of unincorporated banks. If the state authorities were restrictive in the issue of charters, or if incorporated banks were taxed heavily, then we would expect to see private banks developing as substitutes. We know little of the activities of private banks in the ante-bellum period. They were not limited liability institutions, nor could they issue bank notes. But it seems reasonable to suppose that they could offer some competition to the incorporated banks in the issue of deposits and the making of loans. Table 7 presents data on the private banks from the Banker's Almanac; 1859 and 1860 are among the best years for our purposes because the list of private bankers was gradually lengthened as users of the Almanac noted omissions. The higher ratio for the free banking states tends to contradict the hypothesis that free banking reduced the private banking sector, confirming the results from Table 6. However, it appears in both
139
Free Banking II 161
HUGH ROCKOFF TABLE 6 INCORPORATED BANKS PER 100,000 INHABITANTS BY STATE, 1840, 1850, and 1860· BANKS PER
100,000 IN 1840
Sial.
BANKSPEk
100,000 IN 1850
BANKS PER
tOO,OOOIN 1860
Sial •• wilh Opcraliv. Fr•• Baw... Sy.lcm. in 1860t
New York Louisiana Ohio Indiana 1I1inois Wisconsin Minnesota Avg.
3.91 13.35 2.43 1.90 1.89 3.23
6.52
4.45
4.12
5.60
2.93 1.42
7.89 1.84 2.35 2.89 5.47 14.18 1.74 5.19
All Olh.r Sial ••
Maine New Hampshire Massachusens Vermont Rhode Island Connecticut New Jersey Pennsylvania Delaware Maryland District of Columbia Virginia North Carolina South Carolina Georgia Florida Alabama Mississippi Arkansas Kentucky Tennessee Michigan Iowa Missouri Kansas Nebraska Avg.
9.36 9.12 15.45 5.82 56.88 10.00 6.97 2.85 4.47 13.64 2.18 1.33 2.36 4.20 9.26 1.18
5.49 6.92 13.16 9.87 46.62 9.97 5.31 2.29 6.52 3.94
lUI 15.95 14.88 12.70 51.43 16.09 7.44
2.32 2.07 2.09 1.99
4.12 3.12 2.84 1.70 1.43 .83
.26 .16
10.20 2.18 2.77 1.89
1.93 2.20
.78
.88
8.23
6.26
1.26
3.06
10.71 4.51
3.71 2.97 .27 1.93 3.56 1.87 3.45 7.82
·Sources: Banks: U.S. Complrolh:r ol.he Currency. Rtpo" 1876. pp. xcvi-ccxi. Populalion: U.S. Bureau of .he Census, Hislolk.1 S,.,;',ic. uflh. Ulli,." 5,., .., Culuni.1 Ti"'t"ClI1I57(W•• hinlllon, D.C., 19611), pp. 12-1). tin SQme: of these slale. ulher
kind~
of banks were important.
cases that the free banking effect is obscured by regional factors, making a simple point-in-time comparison insufficient. The potential gains from an improved allocation of bank capital were small when viewed in relation to the economy as a whole. The par value of bank capital in 1859 was $422 million [35, p. xcv]. Thus, if banks earned a monopoly profit of, say, 2 percent on the average, the total transfer would amount to an annual flow of $8.44 million, and this is an overestimate-many states had free banking laws. A calculation of
140
Free Banking II 162
MONEY, CREDIT, AND BANKING TABLE 7 INCIDENCE OF PRIVATE BANKS BY STATE,
No. OF STATE
1859 AND 1860*
RATIO OF
No. OF
PRIVATE
PRIVATE TO
PRIVATe
RATIO OF PRIVATF. TO
BANKS 1859
INC. BANkS 1859
BANKS 1860
INC. BANKS 1860
Slal.s wilh Operative Fre. Banking Systems in 1860
New York State New York City Louisiana Ohio Indiana Illinois Wisconsin Minnesota Avg.t
35 79 14 143 46 124 37 33
.14 1.46 1.17 2.70 1.24 2.58 .38 16.50 3.27
33 78 10 148 36 136 17 25
.04
3 n.a. 15 n.a. 4 I n.a. 82 2 II 6 18 5 3 10 9 19 12 2 31 9 50 76 32 7 4
.13 1.42
.77 2.85 .97 1.84 .16 10.00 2.27
All Olher Slates
Maine New Hampshire Massachusetts Vermont Rhode Island Connecticut New Jersey Pennsylvania Delaware Maryland District of Columbia Virginia North Carolina South Carolina Georgia Florida Alabama Mississippi Arkansas Kentucky Tennessee Michigan Iowa Missouri Kansas Nebraska Avg.
3 n.a. 18 n.a. 7 I n.a. 86 2 18 8 20 6 2 13 7 14 15 I 33 18 58 100 31 4 7
n.a. .10 n.a. .08 .01 n.a.
.99 .17 .56 .32 .21 .10 .46 n.a. 2.33 n.a. n.a. .89 .46 19.33 .84 4.0 3.5 1.91
.04 n.a. .08 n.a.
.04 .01 n.a. .91 .17 .35 .28 .17 .15 .34 4.50 2.38 n.B. n.a. .69 .26 12.5 6.33 1.07 4.67 2.67 1.88
'Source" Prival. Bank. 1859. 1860: Merc\~nt', and Banker', Reg"t.,. 1859, pp. 26-40; 1860, pp. 28-42. Incorporaled Banks, Stal ... 1859. 1860: U.S. Com~roller of Ihe Currenc~. Rtport, '876, ~.•cvl-cui. Incorporaled Bank•• N.w York Cily. 1859. 1860: U.S. Congr.... ouse Documenl 112.3 th Congr••• , 2nd .. ion (1859) pp. 116-17; U.S. Congress, House Documeol49. 361h Congre ... 151 Session (1860) pp. 113-14. t In some of Ihese slales olher kind. of banks w.re importanl.
Free Banking II
141 H UGH ROCKOFF
:
163
the pure efficiency loss might well produce an even smaller flow. A flow of $8.44 million in 1859 would have been only 8.28 per capita [32, p. 7]. Thus, the main import of this section is f07 an understanding of the development of the banking system rather than the economy as a whole.
v (. CONCLUSIONS The following facts and conclusions can be drawn from our reexamination of the free banking era: 1. Free banking laws were passed in eighteen of the thirty-two states, with Michigan passing two laws, one in 1837 which was subsequently repealed, and a second law in 1857. In nine states, little or no banking was done under the law, or it was given only a brief trial before the Civil War. In three states, New York, Ohio and Louisiana, some of the "soundest" banking of the era was accomplished under free banking laws. (n Tennessee and later in some of the states which initially experienced wildcat banking the system was a more modest success. Six states, Michigan (after 1837), Indiana, Illinois, Wisconsin, Minnesota, and New Jersey, experienced wildcat banking. In one of the latter, Minnesota, this was clearly due to efforts by the state to force its bonds to a higher price than they were currently bringing in the market. In three states, Wisconsin, Illinois, and New Jersey, wildcat banking can be traced to the linking of the supply of currency with the debt of another state. In only one of the 19 free banking experiments, Michigan (after 1837), did wildcat banking result from a system of currency backed by privately issued securities. 2. Wildcat bank notes lost most of their value only in the first episode in Michigan. In other cases the losses were much less. More typical of free banking was the average loss of 15¢ on the dollar for failed banks in New York. 3. Simple regressions suggest that episodes of wildcat banking had a mild long-run impact on the services people derived from holding money in the effected regions. 4. The evidence for or against the proposition that free banking improved the allocation of bank capital is too slender to support firm conclusions. However, it does appear that New York City benefitted from free banking in its competition with Philadelphia for financial leadership, and that in the West free banking was, or at least appeared to be, a way of solving the vexatious problem of how to allocate bank capital in a region of new and rapid settlement.
142
Free Banking 11
164
MONEY, CREDIT, AND BANKING
LITERATURE CITED
I. BAILEY, MARTIN J. "The Welfare Cost of Inflationary Finance." Joumal of
Political Economy, 64 (April 1956),93-110. 2. BAKER, HENRY F. "An Historical Sketch of Banking in the State of Indiana." Banker's Magazine, 7, New Series (September 1857), 161-79. 3. Banker's Magazine, 8, New Series (June 1859). 4. CARLE, JOHN RAY. The Bank of the State of Missouri, Studies in History, Economics and Public Law, Vol. 102. New York: Columbia University, 1923. 5. CAGAN, PHil.!". Determinants and Effects of Changes in the Stock of Money, 1875-1960. New York: Columbia University Press, 1965. 6. _ _ . "The First Fifty Years of the National Banking Act-An Historical Appraisal," in Banking and Monetary Studies, edited by Deane Carson. Homewood, Illinois: Richard D. Irwin, Inc., 1963. 7. CAMPRELL, CLAUDE A. The Development of Banking in Tennessee, Nashville, 1932. 8. DILl.IST1N, Wn..LlAM H. Bank Note Reporters and Counterfeit Detectors, 1826-66. Numismatic Notes and Monographs, 114. New York: American Numismatic Society, 1949. 9. EASTERLIN, RICHARD A. "Interregional Differences in Per Capita Income, Population and Total Income, 1840-1950," in Trends in the American Economy in the Nineteenth Century (The Conference on Research in Income and Wealth. Studies in Income and Wealth, vol. 24). Princeton: Princeton University Press, 1960. 10. ELMER, LUCIUS Q. C. A Digest of the Laws of New Jersey, 1709-1861. 3rd ed. Bridgeton, New Jersey: Elmer and Nixon, 1861. 11. ENGERMAN, STANLEY L. "A Note on the Economic Consequences of the Second Bank of the United States." Jountal of Political Economy, 78 (July / August 1970) 725-28. 12. FRIEDMAN, MILTON, and ANNA JACOBSON SCHWARTZ. Monetary Statistics of the United States: Estimates, Sources, Methods. New York: National Bureau of Economic Research, 1970. 13. HAMMOND, BRAY. Banks and Politics in America: From the Revolution to the Civil War. Princeton: Princeton University Press, 1957. 14. HELDERMAN, LEONARD C. National and State Banks: A Study of Their Origins. Boston: Houghton Mifflin Company, 1931. 15. HEPRURN. A. BARTON. A History of Currency in the United States. New York: The MacMillan Company, 1915. 16. HINDERLITER, ROOER and HUGH ROCKOFF. "The Management of Reserves by Ante-Bellum Banks in Eastern Financial Centers." Explorations in Economic History, II (Fall 1973), 37-54. 17. HUNTINGTON, CHARLES CLIFFORD. A History of Banking and Currellcy In Ohio Before the Civil War. Columbus, Ohio: Heer Printing Co., 1964. 18. INI)IANA. Laws of Indiana (1855). 19. LASIlON, OSCAR. "The Early Ways and Crazy Days of Banking." Banker's Magazine, 154 (spring 1971) 49-58.
Free Banking II
143 HUGH ROCKOFF
:
165
20. MCCULLOCH, HUGH. Men and Measures of Half a Century. New York: Charles Scribner's Sons, 1889. 21. MICHIGAN. Laws of Michigan (1837). 22. _ _ . Laws of Michigan (1838).
23. _ _ . Laws of Michigan (1857). 24. OBERHOLTZER, E. P. Jay Cooke, Financier of the Civil War. Philadelphia: George W. Jacobs & Co., 1907. 25. PATCHIN, SYDNEY A. "The Development of Banking in Minnesota." Minnesota History Bulletin, 2 (August 1917), 111-68. 26. REDLICII, FRITZ. The Molding of American Banking. New York: The Johnson Reprint Company, 1968. 27. ROIIERTSON, Ross M. History of the American Economy. 2nd ed. New York: Harcourt, Brace and World Inc., 1964. 28. SYI.LA, RICIIARD. "American Banking and Growth in the Nineteenth Century: A Partial View of the Terrain." Explorations in Economic History, 9 (winter 1971-72), 197-227. 29. _ _ . "Federal Policy, Banking Market Structure, and Capital Mobilization in the United States, 1863-1913." Journal of Economic History, 29 (December 1969),657-86. 30. U.S. BUREAU OF THE CENSUS. Eighth Census of the United States, 1860, Mortality
and Miscellaneous Statistics. 3 I. _ _ . Eighteenth Census of the United States, 1960, vol. I. 32. _ _ . Historical Statistics of the United States, Colonial Times to 1957. Washington D.C.: Government Printing Office, 1960. 33. _ _ . Ninth Census of the United States, 1870, vol. I. 34. _ _ . Tenth Census of the United States, 1880, vol. VII. "History of the State Debts," pp. 523-645. 35. U.S. COMl'"rROLLER Of' Currency, 1876.
THE
CURRENCY. Annual Report of the Comptroller of the
36. U.S. CONGRESS; HOUSE. House Document 172. 26th Cong., 1st Sess., 1840. 37. _ _ . House Ex. Document 102. 34th Cong., 1st Sess., 1856.
ApPENDIX
The following list gives the leading references and assumptions used in estimating the losses in Table 2. In a number of cases, other sources had to be consulted. I. Vermont. We have discovered only one free bank which failed: the South Royalton. Its notes were secured by mortgages and Virginia bonds. A. W. Kenney, "The Banks," in History of Royaltoll Vermont, ed. by Evelyn M. Wood Lovejoy (Burlington, Vermont: Free Press Printing Co. 1911), pp. 502-506. The mortgages may have produced considerable losses. Total losses were estimated as an arbitrary 20 percent of the South Royalton's circulation of $122,570 in August 1856. U.S. Congress, House Doc. 87, 34th Congress, 3rd Session (1857), p. 27.
144
Free Banking II 166
MONEY. CREDIT. AND BANKING
2. Massachusetts. U.S. Congress, House Ex. Doc. 25, 37th Congo 3rd Sess. (1862), p. 41. 3. Connecticut. None of Connecticut's free banks failed during the life of the free banking law which was repealed in 1855. For a list of the Connecticut free banks see Forrest Morgan, ed. in chief, Connecticut as a Colony and as a State, orOne of the Original Thirteen (Hartford Connecticut The Publishing Society of Connecticut, 1904), vol. 3, p. 21 J. 4. New York. Carroll Root, "New York Bank Currency," Sound Currency 2 (February, 1895), p. 19. 5. New Jersey. Losses were compiled from the summary sheets to the annual statements of the banks appearing in the legislative documents of New Jersey. fl. Pellll.~ylvml;a. John Tom Holdsworth, Financing an Empire: History of Banking in Pennsylvania (Philadelphia: S. J. Clarke Publishing Co., 1938), pp. 583-585. 7. Ohio. According to Huntington only 7 bond security banks had failed through 1857. Charles Clifford Huntington, A History of Banking and Currency in Ohio Before tlte Civil War (Columbus, Ohio: F. J. Heer Printing Co., 1964), p. 249. Assuming that each was a free bank and had the average circulation of a bond secured bank in 1857, and that the rate of loss was, say, 10%, we get an estimate of 68,000. In addition I have added a ten percent loss for two banks existing in 1858 and 1860 but which failed to report in February 1861. 8. Indiana. U.S. Congress, House Ex. Doc. 102, 34th Congo 1st Sess. (1856), pp. 181, 182, 185, and similar reports on other dates. 9. Illinois. George William Dowrie, Tlte Development of Banking in Illinois, 1817-1863, (Urbana, Illinois: University of Illinois Press, 1913), pp. 152-153 and U.S. Congress, House Doc. 76, 26th Cong., 2nd Sess. (1861), pp. 220-221. 10.-11. Michigan. U.S. Congress, House Doc. 172, 26th Congo 1st Sess. (1839), p. 1129. It was deduced from other evidence that there were no failures under the law of 1857 through 1860. 12. Wisconsin. Leonard Bayliss Krueger, History of Bankillg in Wisconsir1, Studies in the Social Sciences and History, No. 18 (Madison, Wisconsin: the University of Wisconsin, 1933), p. 69. 13. Minnesota. Discounts: Sidney Patchin, "The Development of Banking in Minnesota," Minnesota History Bulletin 2 (August, 1917), p. 160. Circulation: U.S. Congress, House Doc. 49, 36th Congo 1st. Sess. (1860), p.296. 14. Iowa. Howard H. Preston, History of Banking In Iowa (Iowa City: State Historical Society of Iowa, 1922), p. 75. 15. Georgia. Bank note reporters published in the 1840's listed two free banks in Georgia: the Ruckersville Banking Company and the Exchange Bank. See, for example, Bicknell's Counterfeit Detector and Bank Note
145
Free Banking II HUGH ROCKOFF
:
167
List, November 1, 1843, p. 27. The former bank apparently redeemed all of its notes; Laws of Georgia, 1853-1854, p. 192. For the Exchange Bank we have used its maximum reported circulation; U.S. Congress, House Doc. 226, 29th Congo 1st. Sess. (1846), p. 680. 16. Florida. J. E. Dovell, History of Banking in Florida, 1828-1954, (Orlando, Florida: Florida Bankers Association, 1955) pp. 44-46, and passim. This estimate is uncertain. 17. Tennessee. Claude A. Campbell, The Development of Banking in Tennessee (Nashville, 1932) pp. 150-151. U.S. Congress, House Doc. 49, 36th Cong., 1st Sess. (1860), p. 169. 18. Alabama. Theodore William Mathews, Statutory Protection of Bank Creditors Prior to the Civil War (unpublished Master's thesis, University of Chicago, 1930), p. 242. 19. Louisiana. George D. Green, Finance and Economic Development in the Old South (Stanford, California: Stanford University Press, 1972), p.23.
[8] Journal of Monetary Economics 12 (1983) 127-158. North-Holland
ON THE ECONOMICS OF PRIVATE MONEY Robert G. KING* Ulliversity of ROc/lest"', /
The application of laissez-faire to the monetary system is a key ingredient of important proposals for monetary reform, both new and old, that seek to discipline government creation of money through competition. This paper provides an overview of issues to be encountered in the construction of theoretical models of private monetary economies, including consideration of market imperfections that have been hypothesized to arise in these systems. Further, some descriptive history of the less regulated monetary systems of 19th century N.:w York is undertaken, with an emphasis on discussing the nature of market generated banking ami monetary institutions in light of these critiques.
1. Introduction The evolution of the monetary standard of most developed economies has been intertwined with a history of government intervention into the business of money and banking. In most economies, governments have long been the sole legal issuer of circulating notes and sole producer of coin. In addition, governments have imposed substantial regulations and taxes on private issuers of demand deposits and other short-term liabilities. Hayek (1976) argues that the present monetary standard- government notes not convertible into any basket of goods - is not an accident given the absence of competition, obtained by the exclusion of private note issuers and the regulation of substitutes for circulating notes. Rather, from his perspective, it is the absence of competition that permits public notes to be unbacked and for a systematic depreciation to occur. As an alternative, Hayek supports unfettered private money and banking, with individuals free to issue notes and other claims in any chosen manner and quantity. Although the government would not be barred from the marketplace for money, it would be - in Hayek's view - subjected to the discipline of competition. Recently, Hall (1981) has argued that a national government has an important role to play in the selection of the monetary standard and the ·Valuable comments on a much earlier paper with this title were provided by Rooert Bano, Gary Gorton, and Charles Plosser. They are not to oe held responsible for any remaining confusion. Support from the NSF is gratefully ack nowledged. 0304-3923/83/$3.00 @ 1983, Elsevier Science Puolishers B.V. (North-lioliand)
Free Banking II 128
R.(;. KiIlK. 011 til(' ('('OliO/nics o(pril'ote mOlley
control of its market value, but that the detailed organization of money and banking should be left to the private sector. Proposals for monetary reform such as Hayek's and Hall's are attractive to economists who believe in laissez-faire. Many more support the less dramatic deregulations of banking that are now in process in the United States. But such major interventions such as Hayek's and Hall's - and probably our ongoing reforms - cannot be evaluated with any currently extant economic model. At present, economists know little about the operation of fully private monetary systems or about the implications of efficacy of key interventions into such systems. This paper seeks to provide an overview of the economics of private monetary systems. Section 2 is a forward-looking description of the issues to be encountered in theoretical analyses of private monetary economies, which draws on a rising tide of research in this area.! The central focus of future research in this area must be the types of monetary institutions that the private market will deliver. In addition, the discussion in section 2 concerns particular forms of market failure that have been suggested as arising in private monetary systems. Section 3 is a backward-looking description of a less regulated actual monetary system,' specifically that which prevailed in New York during the period prior to the Civil War, which is sometimes called the 'Free Banking' interval. The goal of this historical discussion is to highlight the qualitative features of this interval, while abstaining from any formal econometric history. In keeping with the focus of section 2, particular concern will be directed to the nature of private monetary institutions during this interval. On topics such as these, it is well to state one's hunches (or biases) explicitly. Throughout, the discussion stresses the costs of obtaining information about individual actions and the incentive problems that arise from incomplete information about individual actions. From my perspective, these will prove central to our ultimate understanding of major aspects of monetary and banking institutions: (i) the simultaneous use - as media of exchange of circulating notes issued by third parties and personal evidences of indebtedness, such as checks and individual notes, and (ii) the internal organization of the banking institution, including the nature of individual responsibilities and the structure of assets issued and held. Further, imperfect information and monitoring form the chief potential basis for government intervention, since in a world of costly and heterogeneous information, private market outcomes may not be socially optimal. 'The hasic reference is Klein (1974). who studies an environment in which there are l'sscntially many unconvcrtihlc monetary asscts. Tauh (19R2) analY7.es the competitive supply of nloncy in an ovcrlapping gcncrations model. stressing the time inconsistency of the optimal plan formcd hy a competitive supplier. hut excludes reputational equilibria of the variety discllSsed hy Klein. Banking panics are discu.~sed in recent work hy Diamond-Dybvig (1982) and (jorton (1980).
147
148
Free Banking II R.O. KinK, On the economics oj"priPlIle money
129
At an earlier stage of research, it seemed possible to focus simply on the first aspect, developing individual demands for alternate media of exchange as determined principally by the relative costs of demonstrating an ability to pay for goods received. In this world, one could treat financial institutions including banks as passively supplying those types of issues that individuals require. But, two factors worked against this view, which has been adopted by Fama (1980) in his important work on banking and exchange systems. First, it seemed difficult to reconcile this view with certain features of the structure of relatively unregulated note and deposit-issuing institutions. Second, historical information suggested that this view would rule out the principal basis for public regulation of banks, a priori. That is, in New York and many other states, public regulation occurred because of a widespread view that private arrangements to secure notes and other assets were not in the public interest. In particular, bank failures were popularly viewed as arising from the fact that banks were set up in a manner that permitted 'fraudulent' behavior by their operating officers and that such failures were frequently alleged to have important external effects on the banking system and on trade. 2. Private monetary and banking systems In this section, we imagine an economy in which there is a governmentally determined monetary unit but there is no other intervention of any sort. In particular, the government has specified that the monetary unit is a dollar, which corresponds to a specific amount of a commodity (gold). Further, the economy has adopted this monetary unit as its means of stating prices of goods and services (as numeraire) and as unit of account. The government does not, however, necessarily produce any special claims denominated in dollars nor does it take any action to affect the purchasing power of the dollar. All details of organization of money and banking are under the control of the private sector. The purpose of this section is to discuss factors affecting the types of monetary institutions brought about by private market forces and to outline and assess critiques of these institutions that have served to rationalize public interventions. Over a wide variety of countries and time intervals, several major features of monetary and banking systems recur and, consequently, should be the object of a theory of private monetary and banking arrangements. First, commodity money has been an element of most systems since early times. Second, circulating bearer notes and token coins have always existed, albeit in a broad range of forms. In U.S. history, note is:;ue includes private noninterest-bearing notes payable on demand; private interest-bearing notes payable at fixed term; government non-interest-bearing notes; and
Free Banking II
IJO
R.(i. 1\ ;11,1(.0'1 1/'" (,",,",JI1/ic'S
or ;'r;,'",1' mlllley
governmcnt interest-bearing notes on a demand and term basis. Third. to a greater or lesser extent, variolls type of debt have been lIsed in exchange, ranging from 'checks' against demand deposit accounts to bills of exchange drawn by a particular individual with payment at a specified date. The discussion below seeks to provide some tentative explanations of these institutions. In order to cover a wide range of topics likely to be rclevant in future research on private monetary economics, the discussion employs the non-formal, though abstract approach previously used by Friedman in his investigations of commodity standards (1951) and monetary control (1959). Clearly, this is not a substitute for formal models and the discussion below makes references to more completely worked out theories in some situations.
2.1. Monetary models Received theories of monetary economies typically stress three c1ements. 2 First, trade takes place between individuals because (i) it is efficient for production to be organized in a specialized, decentralized manner, and (ii) individuals' preferences are not similarly specialized. Second, indirect exchange is efficient relative to barter because (i) individuals' preferences are not symmetric - implying an absence of exact coincidence of desired trading patterns - and (ii) the process of exchange is itself costly, in terms of time or costs of physically exchanging goods. Third, in order for indirect exchange to be compatible with individual incentives - that is, to insure the feasibility of trades - it is essential that there be a binding mechanism for enforcing the required wealth transfers. Although such thcoretical analysis provides basic insights into the demand for means of payment, it is silent on the factors determining the extent to which particular institutions will fill this need. For example, wealth payments might take the form of binding individual promises, transfers on the accounts of a centralized record-keeping system, or a commodity money. In discussing various monetary institutions, the following sections focus on demand influences, assuming that supply is forthcoming according to a constant returns-to-scale technology and that there are no important information or enforcement problems in the supply of means of payment. Critiques of privatc monctary systems discussed in section 2.4 below focus on departures from these assumpt ions.
2.2. Tile commodil y mOlley-·credit economy The natural starting point is an economy with two mechanisms for executing payments, a commodity money and a well-functioning credit 2Thc Iheoretical lillc or Ihoughl slll1lllwri7ed hcre slem>. al least rnlm Jevons (19111 ,lnd involves importanl recenl conlrihlliiolls hy Ostroy (In.', alld Towllselld (1980).
149
150
Free Banking II KG. King, On lilt! ecunumics uj"privlIle money
131
system, each of which is provided by a group of specialized competitive producers. The commodity money is assumed to be produced from fine gold at a neglible cost. It is packaged in forms which make exchanges easy to execute, since it is presumed to possess the classical properties of a monetary metal (divisibility, malleability, etc). As coins are circulated, however, some depreciation does occur which is assumed to be due to general wear and tear (rather than clipping, shaving, etc.) Exchanges using commodity money as a means of payment may be rapidly undertaken, as it is assumed that there are relatively small costs of ascertaining the value of coinage. 3 Further, the identity of the trading partner is irrelevant in an exchange where commodity money is the means of payment. The conventional private and social cost of this mechanism is the cost of forgone use of monetary gold (the interest rate) and depreciation. Exchanges in which credit serves as the means of payment, by contrast, require that larger amounts of information be accumulated, in order to assure that only feasible exchanges take place. These information requirements dictate that credit transactions will principally be made through third parties - financial intermediaries - and also are central to the nature of individual-intermediary relationships. First, an initial 'fixed' cost of obtaining and maintaining information about an individual's wealth, etc., implies that it is efficient for each individual to select only one 'bank' and that no individual transactors will choose to extend credit directly to a typical trading partner. Second, immediate messages about credit ~ransactions are necessary so as to secure the approval of the intermediary, which avoids the possibility for fraud. Black (1970) and Fama (1980) discuss the supply side of the credit system. Black's competitive intermediaries ofTer 'accounts' to individuals, with specified interest rates for positive and negative balances, with the latter subject to a maximum negative value that depends on the amount of an individual's fixed property against which the intermediary has claim. Fama (1980) emphasizes that a substantial portion of wealth transfers take place within such an 'accounting system' and stresses that competitive, unregulated financial intermediaries will price transactions services (messages, recordkeeping, etc.) and portfolio management separately. The use of the accounting system in exchanges is limited by the transactions costs it imposes (time costs of issuing and processing messages and the accounting costs of record keeping). Conventional transactions demands for commodity money and for the output of the accounting system arise from cost minimization by individual traders. "Brunncr ,Iml Mellzcr (1971) idcnlify Ihc dlicf rolc of 'Illuncy' as an ccononlill'r on informalion (;ull~diun alld pro...:cssillg.
COSily
151
Free Banking II 132
2.3. Allc/'Ilalive
u.(;.
meal1s
4
I\ill~.
Oil/I,,' ('CO/WillieS "{pri"lI/C'
IJIIJlJry
payment
Private incentives to economize on commodity money and the informational costs of the accounting system would create a demand for alternative means of payment. Bank checks and bearer notes historically havc served as intermediate mechanisms between the credit system and commodity money. Circulating bearer notes are promises to pay amounts of the unit of account to the 'bearer', by some individual. Use of such circulating notes as a means of payment implies that the accepting party in an exchange would not have to expend any resources learning about the credit-worthiness of his trading partner but, rather, only about the note issuer. Notes will circulate if the cost of information about note issuers is small relative to the costs of commodity money in at least some transactions. In the present context, it is easiest to think about interest-bearing circulating notes being issued by the 'financial intermediaries' discussed above, but there is nothing that requires this coincidence of note issuing and credit managing. Bank checks are a short-term credit agreement between buyer and seller of a product, which are ultimately executed through the accounting system. This mechanism reduces the information requirements at the point of exchange (by not requiring an immediate message) but increases the information that an individual trader requires about the trading partner. In many economies, including the pre-Civil War New York experience to he considered below, all of the ahove monetary institutions operated sill1ultaneously. Payments were partly made through commodity money, hearer notes, hank checks and accounting transfers. Traditionally, this coexistence has been viewed as renecting the heterogeneity of exchanges and individuals with respect to transactions costs, which presumably have a predominant informational component, with each means of payment being efficient in some but not all exchanges.
2.4. Critiqlles of prilJate monetary systems There is no shortage of critical comment on various aspects of private monetary systems, with the common thread being. a stress on supply-side elements not explicitly considered above.
2.4.1. Natural monopoly themes Various modern authors argue that note issue is a natural monopoly. One conventional view 4 stresses the scale economies applicable to a note-issuer ·Parl or Illy graduale cducalioll alld reilcralcd hy several colleagues hut ror which no citation could he round.
152
Free Banking II R.G. Killg, 011 tile ecollomics of priVllle mOlley
133
facing a stochastic private demand for redemption, who holds gold and interest-bearing assets as backing, with the latter subject to some transactions or liquidation costs. S Internal increasing returns to scale occur if demands for individual redemption exhibit sufficient cross-sectional independence and if the transactions/liquidation costs are specific to the firm. Another critique relies on a sort of external economies of scale. Tobin (1980) provides a statement of this view, asserting that the use of 'a particular money by one individual increases its value to other actual or potential users,.6 It seems that such external economies would most plausibly arise if users of circulating notes faced costs of ascertaining the value of particular notes. For example, an increase in the number of users of a particular note could lower the probability of meeting an individual uninformed about the value of one's note and, hence, the expected cost of trades. Either of these two sources of increasing returns to scale should make the relevant industry - note issue or, perhaps, other elements of banking exhibit a tendency toward concentration.
2.4.2. Privute note isslie and cOllnterfeitillg Cagan (1963, pp. 19-21), argues that a uniform national currency has substantial benefits over the heterogeneous issues of individual banks, principally because the latter leads to ready counterfeiting, in discussing the U.S. free banking period. He asserts that counterfeiting made it desirable for individuals to 'avoid using notes or to incur the expense of hank note detectors (weekly publications of the period, that printed up-to-date lists of depreciated notes and counterfeits for ready reference); no gains ofTset these costs, which therefore burdened the economy as a whole'.? Further, Cagan suggests that the U.S. could not have undertaken its major post-civil war expansion of economic activity without a uniform national currency. The necessity of detecting counterfeits clearly interferes with a principal 5MiIler and Orr (\966) provide a 'transactions demand' model of firm cash management which possesses the increasing returns to scale property. bThis assertion is abstracted from a larger discussion in which Tobin (1980, pp. 86-81) argues that 'social institutions like money are public goods'. In addition to the dement discussed in the text, he notes that 'there is a strong fiat element in the designation, whether by formal government mandate or by informal social consensus, of any commodity as numeraire and means of payment'. The present discussion has fully abstracted from any concern with the selection of the numeraire or governmentally selected units of a standard money. Consequently, we remain silent on Tobin's second criticism, except to note that even if the numeruire is designated by the government - as in Hall - that this does not determine the means of payment. Tobin recognizes this point, but notes that it is 'hard to imagine and, I suspect even harder to illustrate historically, a unit of account disembodied from a gelienilly accepted means of payment'. 'Cagan (1963, p. 20). The description of bank note detectors, in pan:ntheses, is inserled from ellrlier in Ihe same parll!!raph.
Free Banking II R.G. Killg, 011 tile ecollomics of private mOlley
134
function of commodity money and circulating notes, the execution of rapid exchanges with small information requirements. One position is that the public sector has superior capabilities for preventing counterfeits of commodity, token, or note issucs, but it is unclear why such a comparative advantagc requires public note issue rather than public provision of enforcement. By contrast, Cagan appears to argue that it is the heterogeneity of private currency which raises enforcement costs, whieh is a variant on one of the natural monopoly themes discussed above.
2.4.3.
II/(~[ficie'llcy
oj' ill I cre'st -heo,.ill};
1101 es
Samuelson (1969) and Fama (1981) focus on a potential gap between private and social costs of circulating notes. Samuelson takes as given a Zero (nominal) own rate of return on notes, the social cost, and explains why laissez-faire will not be optimal in this circumstance, if the opportunity cost of money - given by the nominal interest rate - is not zero. Fama describes features of individual exchange - computation and information costs - which may lead individuals to prefer a money with a zero interest rate in terms of the numeraire. In Fama's view, the gap between the private issuer's cost of producing notes and the return on the assets collected is pure economic profit, a logical government confiscation in a world of taxation with positive deadweight losses. With a zero rate of return on circulating notes, it is necesary to reconsider the competitive supply of circulating notes. As Stigler (1968) and others have stressed, price is not the only dimension along which firms can compete. Note issuers would have an incentive to take actions increasing the attractiveness of their issues to holders, increasing the 'quality' of note issue by expenditures directed to aiding acceptance over a wide geographical area, reducing the likelihood of counterfeiting, etc. It is possible, but not necessary, that competition in this non-price dimension can make government intervention undesirable. If government docs not tax or confiscate such pure profits and private banks do not compete away the profit, then private note issue should be an extraordinarily profitable business.
2.4.4. Exle,.,llI/ ejTects (!f hallk failu,.e However, U.S. monetary history - of the past and present century convinces one that the chief government interventions in both the business of note issue and banking have not been rationalized on these grounds. Rather, these interventions have taken place because of a general perception that there were important external innuences of events in the monetary sector, specifically the failure of note issuers and deposit institutions to make good Oil promises. III A I'l'Og,.(/1/I Jill' MOllela,.y SIaM/it)', Friedman makes a
153
154
Free Banking II RG. King, On the economics of private money
135
particularly clear statement of this position, during his rationalization of public intervention into the monetary sector. The introduction of fiduciary elements would not require government intervention if such promises to pay were always fulfilled, or, alternatively, if the community were willing to carry to the extreme the doctrine of caveat emptor. But the first is not likely to occur, and the second neither is likely to occur nor is it c1car that it would be desirable if it did. What is involved is essentially the enforcement of contracts, if the failure of an issuer to fulfill his promise is in good faith, or the prevention of fraud, essentially of counterfeiting, if it is not. Both are functions that most liberals would wish the state to undertake. It so happens that the contracts in question arc peculiarly difficult to enforce and fraud peculiarly difficult to prevent. The very performance of its central function requires money to be generally acceptable and to pass from hand to hand. As a result, individuals may be led to enter into contracts with persons far removed in space and acquaintance, and a long period may elapse between the issue of a promise and the demand for its fulfillment. In fraud as in other activities, opportunities for profit are not likely to go unexploited. A fiduciary currency ostensibly convertible into the monetary commodity is therefore likely to be overissued from time to time and convertibility is likely to become impossible. Historically, this is what happened under so-called 'free banking' in the United States and under similar circumstances in other countries. Moreover, the pervasive character of the monetary nexus means that the failure of an issuer to fulfill his promises to pay has important effects on persons other than either the issuer or those who entered into a contract with him in the first instance or those who hold his promises. One failure triggers others, and may give rise to widespread effects. These third-party effects give spccial urgency to the prevention of fraud in respect of promises to pay a monetary commodity and the enforcement of such contracts.' Two related questions arise when one views the failure of banks - issuers of circulating notes, demand deposits and other claims -- as the principle criticism of laissez-faire in money and banking. First, why do banks fail? Second, what (if any) elTects does the failure of a bank have on third parties in the economy? The discussion starts by considering 'theories' of bank failure and then turns to consideration of potential external consequences. (a) Theories of bank fai/llre
The present discussion outlines alternative stories about the origins of bank failure, some of which are specific to monetary institutions and others which are more general.
155
Free Banking II 136
R.C;. K ;IIK. 011 /"1'
('Co/WIII;C.'
of pr;flll/€'
/IIOII('Y
IIf eil"cl//al ;IlK Illlles. In the passage quoted above, Friedman descrihes p~(lperties of circulating notes that are asserted to make fraudulent manipulation by note issuers particularly important: Notes pass hand-tohand; travel to individuals far removed from the issuer in space and the date of issue in time; and are only infrequently redeemed. The crucial general element appears to be that holders of circulating notes are unlikely to closely monitor the activities of a note issuer, because notes represent a small fraction of an individual's wealth and are held only for a brief period. Thus, it is possible for individual note producers to 'overissue', printing more notes than can be redeemed b r securities held and, consequently, inflicting capital losses on noteholders if simultaneous redemption occurs. Underlying this story is the fact that monitoring by one noteholder would be sufficient to protect all. Yet, because any specific individual would be unable to claim the benefits created for other noteholders, there would be no incentive to undertake this activity. Private agreements on the capital structure of a note-issuing firm could, in principle, resolve this problem. For example, consider a 'bank' issuing a thousand one dollar circulating notes and a one thousand dollar bond, with the latter specified to be the junior claimant in the case of bank failure. Plausibly, monitoring that is privately inefficient for a noteholder will be privately efficient for the bondholder. 8 Of course, for this monitoring process to be maintained, it is essential that the owner-entrepreneur and the bondholder be unable to collude, for example by the bondholder simultaneously being a large borrower from the note issuing firm.
OI)(,I";SSI/£,
F"al/d, !llill/I"e, and agency relations. The business of financial intermediation necessarily involves an agency relation. 9 That is, some individuals (variously described as depositors, savers, shareholders, etc.) employ others (operators of financial intermediaries) to manage wealth, yielding some elements of control in exchange for a pattern of returns superior to that which they could create for themselves. There are inherent conflicts in the relationship between depositors (principals) and operators of financial intermediaries (agents). One view - widely expressed in the regulatory community - attributes bank failure to fraudulent behavior by bank operators (agents), but does not involve any essential difference between classes of bank liabilities (in contrast to the above discussion of 'overissue' of circulating notes). Klein (1974) argues that the capital value of a reputation for issuing sound money provides an important counterweight to the short-run gains from fraud. However, he also notes that reputational forces are likely to moderate "To cover the allendant costs of monitoring note issue, the bondholder would have to receive a higher gross return. ·'Sec. for example, Fama and Jensen (1982).
156
Free Banking II R.G. King, On the economics of pril1ate money
137
rather than eliminate the possibility of bank failure due to fraud, either through 'overissue' or other channels.
Demand liabilities and fluctuations in asset vailles. The traditional theory of bank failure - utilized by Friedman and Schwartz (1963, pp. 108-109 and, especially, pp. 351-356) among others - takes as given the structure of bank assets and liabilities. The bank is presumed to isslie demand liabilities (notes and deposits) and to hold loans, securities, etc that have fluctuating returns. In this sort of setup, failures occur when fluctuations in asset values take place. Thus, for example, Friedman and Schwartz (1963, pp. 354-355) attribute the failures of the 1930s to reductions in the value of marketable assets and increased defaults on loans, occasioned by the general decline of the period. lo Of course, this value of this sort of explanation is limited because it docs not explain why the market generates the specific asset and liability structure of banks. In particular, even if the demand characteristic of the liability is taken as given - viewed as to be explained by its role in the exchange process, for example - there remains the question of why competition docs not ensure that banks hold only securities whose returns imply a negligible risk of failure. (b) Balik lui/lire und third parties Bank failures have long been asserted to have impacts on other individuals than those directly involved (claimants, owners and operators, and those with liabilities to the bank). The traditional concern over bank failures involves implications for monetary quantities and consequent impacts on real variables such as output, employment, etc. through a Keynesian-type demand mechanism. Friedman and Schwartz (1963, pp. 351-353) argue that the bank failures of the 1930s were principally important for this reason. In Keynesian-type models, it appears that all sorts of interventions into the banking sector may potentially be rationalized, on the basis of an external dTect imbedded in the connection between the banking sector and the rest of the economy. Alternatively, there may be relevant external effects of bank failure within the banking sector itself. Various interactions have been suggested historically, II but the present discussion focuses on externalities arising from 'OFriedman and SchwartL (1963, p. 355) also discuss the imp0rlam:c uf a vicious dr.:le during the interval: 'banks had to drop their assets on the market, which inevitably furced iI dcdine in the value of these assets and the remaining assets they held.' "One other common assertion is th,lt the inlen:unnccled nature uf hurrowing and knding arrangements itself produces an externality. lIowever. in the absellce Ilf imperfecl information. this appears to be simply one ,)1' many unccrl'lintics buill inlo any linancial arrangcllIcnl.
Free Banking II
IJX
R.G.
Kill~,
all ,111' ecmrnll1ics fl( prim//' mm!e),
informational events, which perhaps corresponds to Friedman's suggestion that 'one failure triggers another and may lead to widespread effects'. Informational externalities are the concern of a developing literature on rational expectations economies in which markets are incomplete. I 2 In uncertain environments with heterogeneous information, individuals who are relatively uninformed may have dramatic shifts in their beliefs induced by changes in publicly available information. Such revisions in expectations occur if the new information is correlated with an underlying variable that inOuences individual's welfare but is not directly observed. Further, revisions in expectations may occasion important changes in the behavior of individuals. Thus, the observed outcome of one particular banking arrangement perhaps, the revelation of 'overissue' or other fraudulent activity - could lead to actions by depositors that would occasion the disruption of activities of additional banks, which would not. otherwise have taken place. Gorton (1980) discusses how crisis episodes in banking typically followed the failure of a large financial or non-financial corporation, which he interprets as causing shifts in the beliefs that depositors held about default probabilities. \3 Gorton (1980) suggests that suspension of convertibility is a privately efficient response to shifts in depositors (or noteholders) beliefs, providing time for an individual bank to demonstrate so that its operations are linancially sound. However, suspension also involves positive costs to individual transactors, so that the negative externality of information about bank failure may be moderated but not eliminated by this tactic. Efficient private contractural arrangements between banks and depositors would not typically take into account the informational by-products of particular outcomes. Thus, informational externalities could form the basis for a desirable government intervention into the contractural process. Alternatively, organizations of private banks might be able to internalize these informational externalities, by various systems. 14 The standard theory of bank failure - utilized by Friedman and Schwartz (1963), for example - predicts that failures occur and claimants suffer losses when major declines in asset values take place, lowering the value of the bank's portfolio below the par value of its claims. Since dramatic shifts in asset prices are correlated with informational events in models without 12Thc conccpt of an informational externality operating through the price systcm in a rational expectations framework is discussed by Grossman (1977). In the present discussion, the externality operates through quantity outcomes, but the general concept is the same. 1-'The existence or inrormational externalities seems to be the central element in Gorton's proposed explanation or banking panics, delined as large scale changes in the volume of bank notes/deposits. However, as Gorton's analysis combines (i) bank deposits that are demand liabilities, (ii) imperrect, heterogeneous inrormation, and (iii) interactions between the linancial structure and real productive activities or the banking sector, it is not clear how crucial each of these is to the large, discontinuous changes in note or deposit issue. 14Gorton (1980) argues that bank clearinghouses partly Iilled this role.
157
158
Free Banking 11 R.G. King, On the ecollomics of private money
139
externalities, it will be a challenging empirical task to determine the importance of informational externalities in this context. 3. Banking. in 19th century New York The banking industry of New York prior to the Civil War - more precisely, prior to the National Currency Act of 1863 - provides valuable information lS on the operation of a monetary system that is relatively unrestricted by modern standards. During this interval, banking was regulated at a state level, so that there is much cross-sectional variation in state experience that is also relevant. However, to keep the topic manageable and to permit greater detail, the present discussion focuses on New York alone. Based principally on geographical grounds, this decision is not without content, for economic historians 16 have traditionally viewed New York as having among the soundest banking systems of the era. On the other hand, New York has a valuable internal diversity, since two distinct systems of bank regulation existed side by side. In particular, during this interval, banking in New York suffered from a number of perceived deficiencies that attracted public attention and legislative response. First, issuers of circulating notes - banks and other companies - at times failed, so that noteholders could only obtain a fraction of the face value of their notes. Second, notes typically sold at a discount in regions far from a bank's physical location and, in particular, those of country banks sold at a discount in the financial centers of New York and Albany. In the period, the state attempted two different schemes to remedy the first deficiency - systems of bank insurance and registered notes secured by government bonds - and promulgated various regulations in an attempt to secure redemption of notes over wide geographical areas. I 7 The principal public discussion concerned bank notes during this interval, "The statistics presented in this section and the bulk of other information on bank failures are drawn from reports of the relevant bank regulatory agencies (see footnote 25 below) published in the Assembly Docllmellts of New York State during this interval. Detailed information on sources and the statistics on which the summary tables of the text are based is available as unpublished appendices from the author. 16The historical view of banking before the Civil War has evolved with the prevailing economic views of the day. In the late nineteenth century, for examplt:, the New York banking system was widely recalled as highly successful and, to an extent, as the foundation for national reforms. However, after the 1930s, tht:re was a tendt:ncy to view all pre-Civil War banking experience as calamitous. The recent rt:visionist view of this inh:rval was initiated by Rockoff (1975), in an important study which emphasized the overall strength of pre-Civil War systems and t:xplored cross-state differences in experience with 'free banking'. Recent studies by Rolnick and Weber (1982a, b) also grapple with these differences. 171n 1839 and early 1840, discounts were particularly large (ranging as high as six percent on some country bank notes) during a period of gem:ral economic decline and uf substantial bank failure. In May 1840, all banks were re(juired to maintain an agent and an oflice in New York or Albany which would accept notes at a discount of nut IIIme than one-Imlf percen\. This redemption margin was later lowered tu one-quarler percen\.
Free Banking II 140
KG. K illg, 0'1 the economics of private money
although other creditors lost heavily during bank failures as well. During debates on bank reform in 1837, for example, Albert Gallatin defended severe restrictions on the issue of paper bank notes but opposed regulations on bank deposit and discount business. Gallatin particularly emphasized that depositors needed no special protection, pointing out that the choice of a deposit bank was one about which an informed jUdgment was possible and that banks would take actions to earn the confidence of depositors. ls This public vicw is of empirical importance, for it partly determines the data readily available to us. The. reports of the bank regulatory agencies, for example, provide no information on depositholder losses during bank failures. 3.1. The chartered bankillg system From the adoption of the Constitution in 1789 to the enactment of Free Banking legislation in 1839, New York operated variants of a chartered banking system. Unde'r this legal setup, an entrant into the banking business had to obtain a charter from the state legislature, which circumscribed the operations of the institution. For example, the charter under which the Bank of New York began business in 1791 specified that the bank could have a maximum capital of $1 million and could contract debts up to three times the amount of its capital. The charter, however, did not specify how the total debts of the bank were to be divided between circulating notes and other obligations. 19 The process of charter-granting was widespread in the early U.S., not simply a feature of the banking sector. For example, canals and bridges were built and operated by private companies under government charter. The 'franchising' of banking by the New York state government had two predictable consequences. First, it was necessary to rule out entry by nonchartered institutions, which was initially achieved by the incorporation requirement. Later, as individuals sought to organize banks as unincorporated associations, more direct restrictions were promulgated, including the Restraining Act of 1804 which forbade banking by unincorporated businesses. Second, the political sector captured a portion of the monopoly rents in ways that ranged from bank provision of public works to payoffs of legislators. Prior to the expiration of the (national) charter of the Bank of the United Slates ill 1811, ten banks were chartered in New York. During 1811-1829, thirty-two more chartered banks were added. '·See Chaddock (1910) ror a discussion or Gallatin's remarks in the Albany Argus, a contemporary newspaper. '"Knox (1910, p.393) provides an excellent summary discussion or the early stage or the chartered 1l1Inking system, which draws heavily on the reports or the New York State banking c(ll11l11is~iol1crs and comptrollers during the rree banking interval.
159
160
Free Banking II R.G. Ki/lg, Oil/he ecollomics of privu/e
IIIUIII1)'
141
The first bank failure occurred in 1819 and was followed by four more during the subsequent decade. These failures apparently involved substantial losses to noteholders, as banking failure was a central concern of banking legislation in 1829. 20 The legislature also was apparently influenced by the 1826 indictment of several New York insurance executives for fraud and the illegal issue of circulating notes - engraved and otherwise composed so as to resemble bank notes - in the wake of the failure of their corporations. 21
3.2. Chartered batiks and the safety fund system The safety fund system represented a form of insurance for the creditors of an individual bank, much like later institutions such as the Federal Deposit Insurance Corporation. Each chartered bank annually contributed one-half of one percent of its capital - the stockholders' initial subscription - until the bank's safety fund balance had reached three percent of its capital. Income from the safety fund's investments was distributed in proportion to the bank's cumulated contribution, after deduction for certain regulatory expenses. Under the safety fund plan, if a bank failed its circulating notes were to be redeemed from the safety fund if the bank's resources were insufficient to pay :its creditors. The further contributions to finance these payouts - to return the fund to three percent of total capital -- were again assessed proportionately, with each bank contributing one-half percent per year until the necessary level of aggregate capital was reached. This insurance scheme involved more detailed restrictions on the activities of chartered banks. Notes could be issued only to twice capital and loans were not to exceed two and one-half times capital. Further, a panel of bank commissioners was appointed to monitor the activities of the incorporated banks and bank reporting/auditing standards were raised. (However, there were no penalties set for issues of notes or extensions of loans in excess of the charter allottments nor for bank officials' deception of bank commissioners.) Finally, the Safety Fund Act also specified that suspension of specie payments was one ground for insolvency. Importantly, however, the Safety Fund Act did not specify any important restrictions on the process of note issue, deposit creation or portfolio management of the chartered banks. For example. safety fund banks continued to have substantial discretion about the physical characteristics of notes, denominations, etc. 2I1How~ver. information on both bank activities - note circulation, deposits, etc. - and bank failufl!s is sketchy during this early period. Root (1895), who is the source of the above information on bank failures, is not encouraging about the availability of much additional data on this period. 21 Puinc (1914) summ,tri7cs this cpisodc.
Free Banking II 142
RG.
Ki/lg,
0"
tile el,'O/lOmics o( pril)(/te lIumey
3.3. The free hallking system, /838-/963 In reaction to perceived deficiences of the chartered banking system and in line with a general Jacksonian tide of reducing government barriers to entry in markets, New York initiated free banking in 1838. Under this system, individuals and associations could enter the banking business freely and, in particular, participate in a state administered plan for the issue of registered and bond-secured circulating notes, which represented two important departures from the safety fund system. Banking associations were required to have a minimum capital of $100,000 and offered limited liability to their stockholders, unless the association's rules specified otherwise. The free banking plan, like so many other institutions of the period, was adapted from British counterparts: the English joint stock banks 22 and the Scottish free banks.23 However, the government occupied a far more important position in the New York free banking system, as it defined the class of eligible securities, served as custodian of the securities; printed and registered notes, and served a pivotal role in the liquidation of insolvent and failed banks. The free bank had some options in its selection of the assets that would serve as security for its notes. First, one class of notes - bearing the label 'secured by public stocks' - was issued on deposit of bonds of New York, other states, or the U.S. government that bore at least five percent interest. Second, another class of notes - bearing the label 'secured by public stocks and real estate' - was issued on deposit of a mix of government bonds and mortgages on real estate. The mortgages had to bear at least six percent per annum and to be for no more than fifty percent of the value of the real estate. I n addition, mortgages could not constitute more than one half of the lotal security for this class of circulating notes. 24 The bank regulator provided the bank with notes - bearing its name, etc. - in exchange for mortgages and government debt. However, if changes in asset value took place, there was no continuous rebalancing requirement so that the amount of security varied over time. Initially, the Comptroller's office and later the state banking department had plates made for the production of each free bank's notes and supervised/recorded the production of notes. 25 22See, e.g., Gilbart (1871). 2JThe Ranker's magazine of this period contains many discussions of the Scottish free banking system, including extracts from Gilbart (1871). A modern analysis of this system is provided by White (1981). 24lnitially, the comptroller of the state of New York permitted free hanks to use six percent securities from other states in lieu of live percent securities from New York. In March of 1839, however. the comptroller began to accept U.S. and New York securities at par and those of other states at market. This administrative action was incorporated into a general revision of the free hanking act in 1840. During 1839, the comptroller reported accepting the bonds of other states at market prices varying from 70 to 95 percent of par value. B(;enerally, in the discussion helow, the term 'bank regulatory authority' is used to descrihe the relevant state agency for the class of banks and period in question. For free hanks, this was
161
162
Free Banking 11 RG. ·Killg,
011 the ecollomics ()/'private mOlley
143
Further, free banks were required to convert notes into gold on demand and had to hold a minimum amount of specie reserve against circulating notes -- an average of twelve and one half percent over each two week period - according to the act of 1838. (This reserve requirement was removed within two years.) The Free Banking Act of 1838 also specified that if a free bank refused to redeem its notes in specie for a specified period (ten days after receipt of a formal complaint and issue of a formal warning) then the bank regulator would sell off the deposited securities and redeem circulating notes with the proceeds of the sale, net of collection and auction costs. Other creditors of the bank had only secondary claims on the proceeds of such a sale. Finally, the Free Banking Act of I S3S ruled out any issue of non-registered circulating notes by free banks, specifically prohibiting the issue of any instruments of amount less than one thousand dollars.
3.4. A broad overview During the free banking era, New York's banking sector - with its dual structure of incorporated banks 26 in addition to individuals and associations operating as free banks - displayed rapid growth, even after the initial surge of free bank entry in 1838-IS39. Over the twenty year interval IS41 through 1861, the banking sector's volume of outstanding circulating notes doubled, increasing from $14.5 million in 1841 to $28.0 million in 1861. Even more dramatically, deposits at New York banks increased by more than six times as New York City became the financial center of the U.S. economy, rising from $17.2 million in 1841 to $111.9 million in 1861. 27 The bulk of these deposit balances bore interest. 28 There was also substantial growth in the total number of banking institutions, which more than doubled bel ween I R3R and 1863, rising from 133 to 301. Nationally, the United States was on the gold standard during this period, with substantial exports of specie over the 1839-1860 interval. 29 The U.S. treasury of the period also began some tentative moves toward a 'countercyclical' monetary policy, according to Timberlake (197S, pp. 73-85), conducting open market purchases of government securities during business downturns in the latter twenty years of the period. the comptroller of the state from IH3K to IK50 and the banking superintendent from 1851 onwards. For chartered banks, this was the bank wmmissiuners from IH29-1H45, the comptroller from 1846-1850 and the banking superintendent from 1851 onwards. 26The vast majurity of incorporated banks were invulved in the Safety Fund sys\em, with the exception of five chartered banks whose perpetual authorization apparently let them avoid the system. 27lnterbank balances - from banks in New Yurk and other states - are excluded from the totals discussed. 2"See Myers (\ 970, p. 124). 29See Timberlake (1978, pp. 70-71).
Free Banking II 144
R.G. Kin/<, On tile rcO/wlllics o!private money
Over this interval, the U.S. economy went through several severe business fluctuations. The annual statistics on national domestic trade compiled by Smith and Cole (1935) in their study of business fluctuations during 17901860 show a sustained period of diminished activity from 1840-1843; a subsequent expansion followed by sharp declines in 1848 and 1849 with a trough in I HSO; a recovery followed by a sharp decline in 1861. Working with NBER business annals, Mitchell (1930, p. 397) also reports severe business downturns beginning in late 1839, late 1847, 1857 and late 1861. The behavior of commodity prices over the two intervals gives a sense of the long-run price level stability provided by the gold standard, even in the face of major gold discoveries [see fig. I(a)]. Discount rates on high grade three-to-six month paper in New York City are highly volatile [as shown in fig. I(b)]. For example, during the peaks of financial distress in October-November of 1839 and September-October of 1857, discount rates reached thirty percent for a brief period. During the sharp economic downturn in late 1847, discount rates rise from about seven percent to about eighteen percent.
3.5. Free bank failures, 1838-1863 Bank failures play a prominent role in most contemporary and modern discussions of the period of state regulated banking prior to the Civil War. Common elements of these stories are the bank operator's abuse of a 'fiduciary' position: notes are issued that are not entered on the books of the bank; high quality assets are replaced by low quality assets in the vaults of the bank; loans are made to uncreditworthy individuals, etc. Upon discovery of these actions, the bank's creditors -- including stockholders, depositors and noteholders - have assets that are worth only a fraction of the amount of originally invested or the face value of claims. During the free banking era in New York, bank failures did occur and noteholders and other creditors took losses. However, the circulating notes of free banks organized under the statute of 1838 were backed by a specific class of securities that were held by a third-party, the state regulatory agency. Consequently, losses on circulating notes occurred only when (i) the market value of the securities fell below the value of notes outstanding, and (ii) the bank could not redeem its notes for specie, as a result of the general condition of its balance sheet. At present, it is not possible to identify particular causes of bank failure, partly because bank regulators saw themselves principally as custodians of securities against bank notes and did not - at least in the case of free banks - discuss the reports of bank receivers that could afford a more detailed picture of bank actions and the bank balance sheet. Table 1 provides some summary statistics on losses to free banks. The first
163
164
Free Banking 11 R.G. King, On the economics of private money
145
w·~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 11f.;3
1048
1838
1863
1858
TIllE
..
~~~~~--~~~~--~~~~--~~~~.-~~~~--~~~~--~~~~----~--,
,,
Fig. I. lOll \\,il
'"''
I~~
LIt:tI
It."
,....
II~'
IX.1X IX(I~. II>I Discoullt rilte, III Ncw YOfJ. City, 1SIIIII',',,: Smith OIlld ('ole I 1'1.1)1.1
iXJX iH62.
165
Free Banking II 14(,
lUi.
/( ;II!:. 011
,I", I'I'''"''IIIi<"S
or I";f''''('
llfOlfl'.\'
Table I Statistics on free bank failures, 1839-1863.'
Average loss rate (percent)
Fraction notes in failed banks (percent)
2.187 3,991
4,06 0,00
9,73 0,00
11/43 11/44 11/45 11/46
3.363 5.038 5.544 6,235
0.18 0,00 0,00 0.00
1.06 0.00 0.11 0,00
17R.2 40.4 0,0 0.0
11/47 12/48 h 12/49 12/50
9.310 9,265 10,191 13,198
0.36 0,00 0,00 0,00
1.91 0.43 0.00 0,00
9.4 12,2 20,2 6,3
98,9 81.0 260,2 216,8
12/51 h 12/52h 12/53 9/54
14.452 17,004 21,631 21,968
0.07 0,06 0,09 0.03
0.68 0.48 1.20 0.99
IO/55-9j56 10/56-9/57 10/57-9/58 10/58--9/59
0.0 6.8 0.0 0.0
48.8 48S.8 31.0 40.4
9/55 9/56 9/57 9/58
23,308 26,469 22,018 21,761
0.00 0.03 0.00 0.00
0.21 1.85 0.14 0.19
10/59-9/60 10/60-9/61 10/61-9/62 10/62--9/63
0.0 3.6 0.0 0.0
0.0 702.9 41.5 0.0
9/59 9/60 9/61 9/62
23,142 26,617 23,730 33,319
0.00 0.00 0.00 0.00
0.00 0.00 0.17 0,00
Losses ($1.000)
Failed bank circulation ($1.000)
(Date)
3/399/40 10/40-9/41 10/41-9/42 10/42·-9/43
19.6 185.2 88,9 (J,O
273.5 1,383,1 212,8 0,0
12/41 1/43
10/4-'·9/44 10/45-9/46 10/46--9/47
6,0 0,0 0,0 0.0
35.S 0.0 6,0 0,0
10/479/48 10/489/49 10j499/50 IOj50-9/51
33,9 0.0 0,0 0,0
10/51-9/52 10/52-9/53 10/53-9/54 IO/54·9j55
10/44-9/45
Total note circulation ($1,000)
'Sour('t': Compiled from New York state regulatory agency reports. Detailed appendices available from the author on request. hThis number is a linear interpolation from circulation figures at intervening intervals.
is the average percentage loss, which is the ratio of the total losses suffered by noteholders in a given interval to the total stock of circulating notes. The total loss is calculated from detailed information on free bank failures during this interval (available in unpublished appendices). This loss ratio is not the preferred economic measure of losses suffered by noteholders, sinee the redemption of bank notes by the state regulatory authority occurred at variOlls intervals after the initial suspension of bank operations. A preferable measure would utilize a market quote from a bank note broker, but the relevant data has not been investigated. 30 A second measure of the magnitude of bank failure is obtained by dividing .1°Dillistin (1949) reports the location and availability of bank note reporters one market quote - during this period.
which give
166
Free Banking 11 KG. Killg, Oil/he e('()tu/lllics (!f'privule
lilli/ley
147
the total circulation of failed free banks by the stock of circulating notes. In constructing the list of failed free banks, only those that were reported as failed were included. The more comprehensive list compiled by Rolnick and Weber (I 982a) includes many banks that were described as voluntarily closing by the bank regulatory authorities. Consequently, use of that alternative measure would yield it much larger fraction, given the substantial rate of turnover in the banking business during this period. During the initial few years of the free banking system, bank failures were clearly an important phenomenon, irrespective of the measure employed. This is fortuitous, as accurate figures on note circulation appear to be impossible to obtain for the first two intervals. During October 1841 through September 1842 (the first ye&r for which the circulation figures are reliable), eight banks failed, resulting in an aggregate loss rate of about four percent. Nearly ten percent of the total stock of circulating notes were claims on failed banks. The Free Banking Act of 1838 specified that state government debt from any state was acceptable as security for circulating notes. During the early years of the free banking system, dramatic changes in the prices of state debt occurred as some state governments failed to meet their interest payments. 3t Root (1895, p.20) summarizes the sales of securities deposited by failed banks over the 1838-1848 interval. Most of the losses derived from declines in the value of debt of the southern and western states - in the view of the that failed to meet their interest payments or considered 1840s repudiation of state debt. For example, the state debt of Indiana and Illinois sold for less than half its nominal value when auctioned ofT to pay noteholders. The experience of 1838-1840 motivated New York to a number of changes in the free banking law, including a limitation on the types of government debt that could be used as security for circulating notes (initially, free banks were restricted to New York debt in May 1840, although this was broadened to include u.S. securities in April 1849). The state constitution of 1M46 also altered the free banking law, specifying that (i) the state legislature could not pass any laws legitimizing suspension of convertibility, (ii) making bank shareholders personally liable for issues of circulating notes (a 'double' l'Rolnick and Weber (19H2b) provide a modern study that stresses the existence of two portfolio restrictions, specit: convt:rtibility and bund-security as caust:s uf bank failurt:. Yet, there are natural means for a bank to undo any pure portfolio restriction. Hanks should simply have as owners 01' creditors individuals who wuuld otherwise hold amounts of government deht. As an extn:me example, a bank could back its notes by one hundn:d percent specie reservl!s if it is owned by an individual who would otherwise hold a portfolio of govt:rnment bonds. However, if the securities held by the state banking authurity served as a bond against the bank's redemption of its notes, as appt:al'S mort: likdy to me, then nuh:holder losses would nt:ct:ssarily be tightly linked to the value of the assets hc:ld by the state banking authority. Also, the timing of chartcrt:d bank failurt:s were highly corrc:lah:d thost: of frel! banks (see below), despite the fact that these banks were nut subject to the bond st:curity provision.
Free Banking II 148
KG. King, On tile
economic.~
(!f private money
liability as it was limited to the amount of their share of the capital stock of the bank), and (iii) making circulating notes the primary creditor of all state banks. In the twenty year period extending from October 1842 through September 1862, quite a different picture emerges. The average annual percentage loss was about 0.04 percent, described by a contemporary observer as being far less than the loss arising from wear, tear, and shaving of specie coins. The largest percentage loss - about one third of one percent - occurs during a period of general financial distress and sharp economic decline, 1847. Thus, the picture which emerges from these aggregate statistics is one of relatively secure stable-valued money, when taken in conjunction with the average inflation rate over the period. On average over the twenty year period, the notes of failed banks comprised about one-sixth of one percent of the total volume of circulating notes issued by free banks. Periods of financial distress and general economic depression also corresponds to intervals where there is a relatively high fraction of note issues belonging to failed banks.32
3.6. Chartered ballkfailures, 1838-1863 Consideration of chartered bank failures under the safety fund system must be divided into two periods. In the initial years of its operation, 1829-1838, the resources of the fund grew through bank payments and there were no claims on the fund. 33 Eleven safety fund banks failed, however, during 18401843, a period of generally depressed economic activity that was also marked by many free bank failures. The claims of the creditors of these banks exceeded the safety fund and the ensuing settlement in 1845 effectively precluded any insurance payouts to chartered banks for many years. Thus, during the latter years of the system, chartered banks differed from free banks principally in that they were exempt from the bond security provision and were required to make payments into the safety fund, but they no longer had state insurance of their notes. The insolvency of the safety fund system resulted from a May 1843 court ruling that all creditors - not just noteholders - had claims on the fund. This ruling reversed the practice applied in initial failures prior to 1841 and .I2The IK61 ohservlltilln is heavily innuenced hy the simullaneous closing of several Alhany hanks in May of 1861. In this episode, four banking associations failed - according to the state banking commissioner in his report of 1861 - due to 'gross and criminal mismanagement by the officers ... together with palpable inefficiency on the part of directors'. Together, these failures caused the suspension and liquidation of their correspondent bank, the Bank of the Interior. However, no losses were sustained by noteholders in this episode. Hprior to 1838, the bank commissioners had acted to assist several banks in financial difficulty, making temporary loans from the fund to redeem notes, but had received repayment with interest.
167
168
Free Banking II R.G. King, On the ecollomics of private mOlley
149
was widely viewed as a surprise, contrary to legislative intent, by contemporary observers. 34 Because the safety fund was a creditor of failed banks, contemporary accounts of the regulatory authority are far more detailed than for free banks. Chaddock (1910, pp. 309-325) provides a summary of aspects of bank failures from several annual reports of the regulatory authorities. In four out of eleven cases during this 1840-1843, banks engaged in issues of notes in excess of the amounts placed on the accounts of the bank. For example, in 1841, investigators from the office of the bank commissioners found that the cashier of the Commercial Bank of New York - essentially the chief operating officer of the bank during this period - had put about $56,000 worth of notes into circulation without entering these on the bank's registry. In three failures, the operating officers of the bank substituted high quality for low quality assets, again engaging in an accounting fraud to cover their actions. In many failures, the assets of the bank were relatively worthless upon liquidation because the bank had made large loans to bank directors, officers and to other individuals who are typically described as unscrupulous friends of the bank's officers. It is difficult to construct a measure of noteholder losses in safety fund banks because of the substantial delays that arose in payments, as a result of the financial difficulties that the safety fund system encountered in this period. For most of the noteholders of the three banks that failed prior to November 1841, repayment was relatively prompt. However, in November 1841 through March 1842, six additional bank failures occurred, with total liabilities that far outstripped the resources of the safety fund. In May of 1843, an injunction prevented any further payments from the safety fund, until the creditors of one of the banks failing in November 1841 - both noteholders and others had been fully compensated. In 1845, the legislature authorized an issue of state bonds, secured by the future payments from the safety fund, to pay the creditors of insolvent banks. No further insurance payments from the safety fund were permitted until after the state debt had been fully paid off. Nevertheless, table 2 provides some statistics for 1839-1863 analogous to those presented for free banks in table I. During the first four years of the table, ten banks failed with note issue that represented a substantial fraction of the outstanding notes of all chartered banks. Losses to noteholders are not reported for the tirst four years due to the timing difficulties discussed earlier. Two major regulatory changes for safety fund banks came out of this episode, in legislation of April 1843. First, the safety fund banks were required to have their notes registered and countersigned by the state regulatory authority, although they were not required to deposit securities. '4Knu}( (1910, pp. 410·413) Jisl:lIsst!s this cpisoJc in Jctail.
Free Banking II
169
ItG. K;IIg. 011 11r(, ('collomit's (){,P";I'lIll' """/('Y
150
TlIhle 2 Stlltistics on chnrtcrcd hank lililmes. I!DK-llI63." ---.... _._- .. -<---
Losses ($1,000) 10/38··9(.19 10/39-9/40 IO/40--9f4 I 10/41-9/42
Failed hank circulation ($1,000)
(Date)
Towl notc circulution ($1.0001
0.0 268.9 265.8 1,119.0
1/3K 1/39 1/40 1/41
12,432 19.373 10,630 15,235
Average loss rate (percent)
Fraction notes in failed banks (percent) 0.00 1.38 2.50 7.34
10/42-9/43
0.0
70.0
11/42
9,734
0.00
0.72
10/43-9/44 10/44-9/45 10/45-9/46
0.0 0.0 0.0
0.0 0.0 0.0
11/43 11/44 11/45
13,850 15,115 I 5,lI3 I
0.00 0.00 0.00
0.00 0.00 0.00
10/46-9/47 10/47-9/48 IO/48-9f49 10/49-9/50 10/50-9/51
0.0 0.0 0.0 0.0 0.0
0.0 185.5 0.0 0.0 0.0
11/46 11/47 12/48 b 12/49 12/50
16,033 16,927 14,262 13,972 14,220
0.00 0.00 0.00 0.00 0.00
0.00 1.09 0.00 0.00 0.00
10/51-9/52 10/52-9/53 10/53-9/54
0.0 0.0 0.0
0.0 0.0 0.0
12/51 b 12/52b 12/53
13,484 12,391 10,942
0.00 0.00 0.00
0.00 0.00 0.00
10/54-9/55 10/55-9/56 10/56-9/57 10/57-9/58
o.n n.n
SOli. 5
9/54 9/55 9/56 9/57
9,539 8,032 7,551 5,104
1.56 0.00 0.00 1
1.56 0.00 0.00 9.96
0.0 0.0 0.0
9/5l1 9/59 9/60
4,843 4,829 5,142
0.00 0.00 0.00
0.00 0.00 0.00
1O/5H9f59 10/59-9/60 10/60-9/61
148.5
"
n.u o.n
0.0
n.o n.o
"So/(r('l': Compiled from New York stnte regulatory agency reports. Detailed appendices availahle from the author on request. bThis number is a linear interpolation from circulation ligures at intervening intervals. 'Failure of the Lewis County Bank; losses presumed equal to one hundred percent of circulation. dAs discussed in the text, the magnitude of the losses during 1857 is unclear.
Second. future claims on the safety fund were limited to noteholders and could not be made by other creditors of failed banks. In addition, the state constitution provided the general alterations in the banking business described previously, including the 'double' liability for issues of circulating notes. Five additional failures of safety fund banks occurred during the remainder of the period (one each in 1848 and 1854 and three in 1857). Despite fraudulent embezzlement by the cashier of the Canal Bank of Albany in 1846, a receiver ultimately paid off noteholders at par. In the case of the Lewis County Rank in 1854, a firm failed whose debt and loans were almost the entire assets of the bank, a situation that had been concealed on its
Free Banking II
170
R.G. Killg, On the ecollomics of priVlIIe mOlley
151
books. According to the contemporary account of the bank superintendent, the noteholders were the only creditors and, when the debts proved to be worthless, suffered an extreme loss. II is not clear whether there was any recovery from the bank's owners under the 'double' liability provision. When the safety fund paid off its outstanding state debt in I S66, a small fraction of the notes of the Lewis County Bank were redeemed without interest. Information on the three safety fund bank failures during the general financial crisis of I S57 is sketchy, as the bank superintendent regarded the matter as one principally for receivers due to the condition of the safety fund. Most of the notes that were returned to the bank superintendent voluntarily by the receiver, presumably indicating that the receiver had satisfied these claimants. In table 2, statistics summarize the twenty one years over which chartered banks operated without insurance or bond security. The fraction of notes in failed banks appears slightly smaller than that for free banks, with the exception of the 1857 collapse of three chartered banks having about ten percent of the total outstanding notes.
3.7. The crisis of 1857 During the nineteenth century, the U.S. economy underwent periodic episodes that contemporary observers viewed as crises. Common features of these episodes were sharp increases in interest rates and declines in prices and outputs in many productive sectors. In these episodes, the volume of notes, deposits and loans made by the banking sector typically declined dramatically. Not infrequently, banks temporarily suspended convertibility of notes and deposits into specie during these intervals. The episode in the fall of 1857 is widely viewed as one of the major crises of the nineteenth century, which exhibits the general characteristics discussed above. First, the wholesale price measure plotted in fig. l(a) fell by about twenty-five percent during September 1857 through January 1858. Second, a general decline in economic activity was in progress during 1856-1857. For example, Smith and Cole (1935, p. 106) report that the earnings of American railroads slumped during this two year period, with a dramatic decline in New England railroad earnings in the fall of 1857. Third, discount rates on commercial paper rose from about ten percent in August to thirty-six percent in early October, after which they fcll back to the tcn percent range in late December [see fig. l(b)]. Important movements in asset prices also occurred during this interval. Financial historians 35 and contemporary observers·H , suggest that the l5Dewey (1922. pp. 261 165) and Myers (1970, pp. 126 12XI an: two eXilllIplcs. The quoted description is due to Myers. l"for example, the superintendent of hilnks I,ll' the Slale "I' Ne" YorJ,. in his Annual Report of 1858.
171
Free Banking II R.G. King, On tlte economics of private money
152
Table 3 Banking statistics during the crisis of 1857." Date
Loans
(A) Weekly average data
8/22/57 8/29/57 9/5/57 9/12/57 9/19/57 9/26/57 10/3/57 10/10/57 10/17/57 10/24/57 10/Jl/57 11/7/57 11/14/57 11/21/57 11/211/57 12/5/57 12/12/57 12/19/57 12/26/57
Olt
Specie
Notes
Deposits
8,694,011 8,671,060 8,673,192 8,322,316 8,073,801 7,838,308 7,916,102 7.523,599 8,087,441 6.8114,739 6,334,748 6,434,312 6,258,652 6,283,417 6,520,783 6.555,000 6,348,494 6,309,466 6,352,187
64,241,471 60,860,801 57,260,609 57,334,121 57,851,965 56,918,863 52,798,365 49,745,176 42,696,012 47,1173,901 51,853,159 56,424,807 60,601,556 64,917,932 64,307,380 64.444,375 62,907,999 63,710,106 65,239,374
8,835,575 7,838,308 6,352,187
69,233,090 56,918,863 65,239,374
23,560,317 19,284.596 17,547,777
35,127,336 26,621,031 14,741,211
New York city banks
120,139,582 116,588,919 112,221,365 109,985,573 108,777,421 107,791,433 105,935,499 101,917,570 97,245.R26 95,593.518 95,317,754 95,866,241 95,239,247 95.375.432 94.963.130 96,333.687 96.526,037 97,211,690 97,902,035
10,097,178 9,241,376 10,227,964 12,181,857 13,556,186 13,327,095 11,400,413 11,476,294 7,!!43,231 10.411,643 12,8R3,441 16,492,153 19,451.967 23,167,9110 24,303,144 26,069,833 26,058,877 27,957,326 27,142,098
(8) Quarterly data on New ll>rk city bal1ks
6/6/57 9/26/57 12/26/57 (e) Quarterly dala
6/6/57 9/26/57 12/26/57
115,338,592 107,791,433 97,902,035 /JI1
13,134,714 13,327,095 27,142,098
New York country bal1ks 75,470,240 63,055,341 56,308,030
1.235,720 994,504 2,171,323
'Sources: Compiled from materials in Gibbons (1873) and the 1858 annual report of the New York bank superintendent.
'immediate spark' to the crisis of 1857 was the failure of the Ohio Life Insurance and Trust Company, whose New York agent ran up heavy losses in railroad security speculation. The company - which had functioned essentially as a bank of deposit financing purchase of primary securities (with call loans but did not issue notes -- closed its New York office in late August of 1857 with New York banks holding substantial amounts of its commercial paper. Traditional discussions emphasize that note and depositholders rapidly opted to .redeem their holdings. Table 3 reports weekly average figures on the banks of the New York clearing house reported by Gibbons (1873, pp. 33~337). These indicate a dramatic decline in deposits from $61 million on August 29th - the week the New York office of Ohio Life closed - to $43 million on October 17th. Over the same
172
Free Banking 11 R.G. King, On the economics of private money
153
interval, circulating notes outstanding fell from $8.7 million to about $8.1 million. Suspension occurred on October 13th, which was followed by a rapid decline in notes outstanding from $8.1 million to $6.3 million on November 14th. Deposits, by contrast, climbed back to $61 million over that interval. When resumption of specie convertibility occurred on December 11th, the New York city banks had tripled their specie holdings from $9 million to $26 million and cut loans from $109 million to $97 million. Deposits had increased from $61 to $63 million and circulating notes had fallen from $8.1 million to $6.3 million. Developments in New York banks outside of the city can be viewed only at the quarterly interval [see table 3, parts (8) and (C)]. In general, the country banks exhibit the same pattern of movement as the city banks: loans and notes outstanding both fall, while specie rises. There is, however, a major decline in deposits at country banks that occurs from June to December (about sixty percent), in contrast to the rough constancy of deposits in banks in New York city. Table 4 exhibits the diverse movements in asset prices during the fall of 1857, with weekly quotations on prices of a long-term U.S. bond (bearing a six percent coupon), a state debt instrument (Indiana bonds with a coupon of five percent), and a railroad equity share (New York Central). These prices Table 4 Asset price movements during the crisis of 1857."·
Date 8/21/57 8/28/57 9/4/57 9/11/57 9/18/57 9/25/57 10/2/57 10/9/57 10/16/57 10/23/57 10/30/57 11/6/57 11/13/57 11/20/57 11/27/57 12/4/57 12/11/57 12/18/57
US 6s ( 1867-1868) 116.75 116.75 117.00 117.00 117.50 117.25 117.50 117.50 117.75 114.00 1111.00 112.00 III.()() 112.00 109.00 110.00 112.00
Indiana 5s
NY Central RR stock
82.5 81.5 76.0 80.0 80.0 80.0 75.0 70.0 70.0 65.0 73.0 73.0 75.5 7l!.O 8UI 81.0 82.0 82.5
76.50 74.13 76.13 70.50 69.75 65.50 53.00 53.50 65.75 59.63 (,5.25 70.25 75.50 79.00 74.()0 73.50 72.00 76.25
"Bond prices have par value of one hundred dollars. Stock price is in dollars. ·Sources: Banker's Magazine, various monthly issues in 1857.
Free Banking II 154
ItG. 1\ illl(. 011 1/1(' e("llllmnics
fl." private lIIone.v
all dcclined during thc height of the crisis and subsequently recovered, but thc magnitudc of thc price movemcnts is much larger for state debt and railroad cquity than for thc U.S. government bond. 37 During thc crisis of I R57, sevcn free banks and three chartered banks failcd. In addition, contcmporary journals rcported a number of other banks as suspended (prior to the general suspension), which was widely perccivcd as indicating a possihle failure:1R
3.8. Tire Natiollal CII/."ellcy. Act: Elld of all era The National Cur rency Act of I R63 established a national banking system based on the principles of bond security, government note registration and frce entry that had been the foundation for free banking in the State of New York. Few free banks joined the national system within the first few years of its opcration. 39 'n March of 1R65, however, additional legislation imposed a duty of ten percent on state notes paid out by any national or state bank, with this tax scheduled to take effect in July of 1866. The vast majority of New York banks converted to the national system, although about one sixth of the total remained as state banks with neglible note circulation.
3.9. Evidellce
0/1
critiqlles of p,.ivllte monetary systems
The New York experience during the twenty years before the Civil War provides an important potential laboratory for judging critiques of private monctary systcms. Along with othcr historical experiences, a main lesson is that relatively unregulatcd private systems do not completely fall apart. 40 New York had relatively sound provision of circulating notes, at least, with two differcnt systems of bank regulation. Based on the currcnt investigation, thc following initial assessmcnt of the spccific critiques discussed in section 2.4, is offcrcd.
N £1/111'01 lIIollopoly themes. Thcre appears to be little tendency toward conccntration in note issue or the general business of banking during this interval. Even within the narrow geographical confines of New York city, a large numbcr of note issuing institutions coexisted without any apparent tendcncy toward a monopoly position. '''Note also that U.S. sccurity was at a premium during the CriSIS, so that free b,lI1k circulating notcs backed by these securitics were not subject to loss irrcspective of the condition of the rest of II bank's balance sheet (since securities were accepted at the minimum of par or markct). JKSee, for example, the report of III/lit's Mel'e/wlII Magazine and Financial Chronicle, for October 1857, in which ('ighteen banks were reported as suspended. J"Eighteen banks by the end of 1864, llccording to the bank superindent's reports, out of nearly three hundred. 4°Whitc's (1981) analysis of Scottish free banking is another of the experiences relevant here.
173
Free Banking II
174
R.C. King, 011 tile eCCJnolllics of private mOlley
155
Counterfeiting of private notes. Contemporary discussions such as banking publications of the interval 41 as well as the reports to the legislature of New York by the bank superintendent - do not stress circulation of counterfeits as an important feature of the N.Y. banking experience. Nevertheless, bank note reporters of the interval do include many entries for New York free and chartered banks. As Rockoff (1975) notes, however, such indices are cumulative, including entries of {;ollnterfeits reported earlier but not necessarily circulating at a particular point in time. Any comprehensive evaluation of the importance of counterfeiting would require estimates of counterfeit circulation, which perhaps could be based on fractions of counterfeit circulation in note redemptions to banks during this period. But it is not apparent whether banking records of Ihis interval would permit the construction of such estimates. Inefficiency of interest-bearing circulating notes. In New York, legislation influenced the issue of interest-bearing circulating notes, so Ihat there is not clean evidence on Fama's (1981) contention that only non-interest-bearing notes have survival value. During the early 1800s, interest-bearing circulating notes were issued by banks in amounts as small as five dollars. 42 The practice of issuing post notes - issues payable at a fixed future date, either with or without explicit interest - had originated with the first Bank of the United States and were so called because they were initially used in transactions through the postal system. Post notes were either payable to bearer or to order. In New York, an 1830 law specified that all notes were payable on demand and gave noteholders legal recourse against the issuer if a note was not so honored, which effectively suppressed post nole issues by chartered banks. In the initial two years of free banking, interest-bearing post notes were issued by individual bankers and associations. Court decisions over Ihe interval held this practice in violation of Ihe free banking act 43 and, in March 1840, the legislature added amendments that made issue of post notes a misdemeanor punishable by lines and/or imprisonment. Suppression of post notes was based on a belief Ihat no effective controls could be exerted on such issues and thai 'banks which engage in such enterprises are rotten and sooller or latcr will end in defrauding the community'.44 "Parlicularly, The Ballker's Magazille and Jlwil's Merdullil f\1agazirw am/ ell/lilller";,,/ Chronic/e. Also, for a nol~ engravers view, see Ormsby (1\152). 41See Dewey's (1910, p. 105) discussion of posl nOles and, in particular, ,)f small inlaesl bearing noles in Massachusells in 1836, which he reporls as being issued unly by 'weak' banks. 43See Cleaveland (1857, pp. 113-123) for a dis~ussion of SOill~ uf court ~ases and cAampks of pOSI noles. 44The quotalion is from Ihe opinion or a New Yurk SIal.: judge in a ~ourl case involving pOSI notes, as abslracled in Ckavdaild (Ul57, p. 311l). J.MolI- F
Free Banking II 156
R.G. K ;'I!?, 011 lite
I'Co/WlllieS
of pril'all' mOPIer
Halik failllres and externalities. The losses to noteholders of chartered and free banks during this interval have previously been discussed in detail. In the last twenty years of the operation of each system, losses to noteholders were relatively small. 45 The bond security provision of free banking prevented occurrence of extreme losses due to fraud, which did take place under the bankrupt safety fund system (as in the case of Lewis County Bank in 1854). But, al\ in al\, noteholder losses due to bank failure were not an important chronic element in New York during 1838-1863. However, during the irregular financial crises of this interval - such as that of 1857 - major variations in quantities of bank notes and deposits took place over short intervals, along with failures of free and chartered banks. Not infrequently, these episodes also involved disruptions of the monetary system, including suspensions of specie convertibility of notes and deposits at many banks. The external effects of bank failures and suspension on the monetary system and on trade requires such a detailed evaluation that there is not any point in speCUlating on its ultimate outcome.
4. Summary and conclusions This paper combines consideration of a fully private monetary system, including critiques of its operation, with a qualitative overview of New York State's experience with systems that are relatively unregulated 'by modern standards. But economists have barely begun to scratch the surface of these problems and much work remains to be done. For example, theoretical rcsearch into the nature of banking institutions - including both note issuers and banks of deposit - is necessary to provide explicit frameworks in which some of the critiques of private monetary systems can be fully investigated. In addit ion, substantial amounts of econometric history are necessary - detailed investigations of free banking and banking under the National Banking Act arc particularly relevant - if we are to understand the structure of private note issue and deposit banking under alternative regulatory schemes. Such knowledge i, relevant to the analysis of monetary standards - and proposals for monetary reform - because one important potential reform involves separation of government designation of the standard from the provision of circulating notes. Indeed, Hayek (1976) argucs that unfettered private money and banking is necessary if sustained innatiol1 is to be cl i mi na ted. "In tllhle I, free hllnk noteholdcr losses average less than 0.05 percent pcr yellr. In table 2, s:,fetl' fllnd hank losses average between 0.07 percent and (J.5H perccnt per year, depending on the ,;nknown 10" rate for IRS7.
175
Free Banking 11
176
R.G. King, On the economics of private money
157
References Barro, R.J., 1979, Money and the price level under the gold standard, Economic Journal 89, March, IJ-33. Black, F., 1970, Banking and interest rates in a world without money, Journal of Bank Research, Autumn, 9-20. Brunner, K. and A. Meltzer, 1971, The use of money: Money in the theory of an exchange economy, American Economic Review 61, Dec., 7K4·K05. Cagan, I'., 1963, Till: first fifty years of the national hanking systenl, ('h. 2 in: D. ('arson, cd .. Hanking and llIonetary studies (Irwin, Il
Free Banking II 158
R.G. Kin!!, On till' ecollomics of private mOlley
Painc, W., 1914, Thc laws of the State of New York relating to banks, seventh ed. (Baker Voorhis, New York). Rockoff, H., 1975, The free hanking era: A reconsideration (Arno Press, New York). Rolnick, II. and W. Weber, 19R2a, The free banking era: New evidence on laissez faire banking, unpuhlished working paper, Federal Reserve Bank of Minneapolis, Minneapolis, MN, May. Rolnick, II. amI W. Wehcr, 19R2b, II theory of free bank failures: Tests and implications, unpublished working paper, Federal Reserve Bank of Minneapolis, Minneapolis, MN, Aug. Root, L.c., I R95, New York hank currency, Sound Currency, Feb., 1-24. Samuelson, P., 1969, Non-optimality of money holding under laissez faire, Canadian Journal of Economics, May, 3m-JOB. Smith, W. and 11.. Cole, 1935, Fluctuations in American business, 1790-1860, Harvard University, Camhrilige, MA. Stigler, G.J., 196R, Price and non-price competition, Journal of Political Economy 76, Jan./Feh., 149-154. THUh, 0., 19R2, Private competitive fiat money, unpublished working paper, University of Virginia, Charlollesville, VII, Jan. Tcmin, P., 1969, The Jacksonian economy (Norton, New York). Timberlake, R., 1978, The origins of central banking in the United States (Harvard University Press, Camhridge, MII.). Tobin, J., 1980, Discussion, in: Modeis of monetary economics, Federal Reserve Bank of Minneapolis, Minneapolis, MN. Townsend, R., 1980, Models of money with spatially separated agents, in: N. Wallace, ed., Models of monetary economics, Federal Reserve Bank of Minneapolis, Minneapolis, MN. White, L., 1981, Frec hanking in Britain: Theory and experience, 1800-1850, unpublished working paper, New York University, New York, Nov. .
177
[9] New Evidence on the Free Banking Era By
ARTHUR
J.
ROLNICK AND WARREN
E.
WEBER '"
Oll a reevaluation of the Free Banking Era, which develops and examines far more detailed empirical evidence on this period than has been considered in any previous research, we find that the history of free banking was not as chaotic as many believe. From a casual view of history, the reluctance to allow unfettered competition in banking is understandable. There is a long and costly history in U.S. banking of instability, banking panics, and major disruptions to economic activity. Most date these problems hack to the Free Ranking Era when there were no federal regulations. when entry harriers were low, and when banks were free to issue their own notes lind to compete for deposits and loans. Jlistorillns have found this period chaotic and filled with speculators, wildcat banks, and bank failures. Then came the national banking system (1863-1913), which has been seen as an improvement, although problems still persisted. Under this system, national banks were subject to supervision and regulation by the Comptroller of the Currency, who presumably promoted a much safer and sounder form of banking than the state officials. This new system was still not good enough to prevent the recurrence of major bank panics and was replaced in 1913 by the Federal Reserve System. Not until the nation had both a strong central bank and federal deposit insurance (1933-35), however. did it appear to have solved the inherent instability problem in banking. This historical record appears to have even convinced the staunchest proponents of free enterprise. Milton Friedman (1960). for example, made his position clear when he responded to Gary Becker's (1957) proposlli for laissez-loire banking_ Becker suggested that the United States permit free deposit banking, without any requirements ahout reserves or supervision over assets or liabilities and with a strict caveat emptor policy. Friedman rejected Becker's proposal, arguing that
The argument for free competition in most economic endeavors has much appeal. Under fairly general conditions. economic agents acting in their own self-interest produce an economically efficient outcome, that is, a situation in which no one person can be made better off without someone else being made worse off. Yet despite both the logic of the argument and the ap.parent real world success of free enterprise economies, there has heen one industry, one economic activity, that even ardent proponents of loissezloirc' have heen afraid to leave to the vicissitudes of the free interchange of supply and demand. This is the husiness of hanking, an industry that most helieve is inherently unstahle. This fear and reluctlll1ce to permit free competition in banking is not based on an explicit theoretical foundation that has confronted real world events, but rather on the events themselves. The United States has experienced many periods of major banking panics during which a large number of banks have failed and financial markets have been in considerable disarray. The period considered hy many as the worst is the one period when banks were more or less left alone to pursue their own profit-motivated interests, the period known as the Free Banking Era (1837-63). Its problems are often cited as evidence that banking should be regulated. 'Vice President and Deputy /)irector of Research. Research /)epartment. Federal Reserve Bank of Minneapolis. 250 Marquette Avenue. Minneapolis. MN 554RO. and Professor. Department of Economics. Virginia Polytechnic Institute and State University. Blacksburg. VA 2406 I. respectively. Part of Rolnick's work was done while he was a Visiting Professor at Roston College. and part of Weller's work was done while he was a Visiting Scholar at the Federal Reserve Rank of Minneapolis. We gratefully acknowledge the research assistance of Susan Mendesh. lIerhert Miller. Maureen O'Connor, and Judy Sargent. and the editorial assistance of Kathleen Rolfe. The views expressed are our own and not necessarily those of the Federal Reserve Bank of Minneapolis or the Federal Reserve System.
lalla
179
Free Banking II VOl.. 7J NO.
j
ROLNICK AND WEBER: FREE BANKING ERA
while it had some merit, "it would not ... solve the problem of •inherent instability'" (p. 108, note 10). Clearly, Friedman holds the majority view; much of U.S. bank regulation and supervision is aimed at promoting a safe and sound banking system. Nevertheless, observing instability in banking does not necessarily mean that the instability is inherent. Recent theoretical work suggests, in fact, that government intervention, supposedly aimed at safeguarding the system, may have produced the observed problems. John Kareken and Neil Wallace (1978) find that under certain assumptions there are no bank failures under laissez-Iaire banking. In their model, failures are induced by the nonoptimal pricing of government deposit insurance. Eugene Fama (1980) argues that, under competitive banking, " ... portfolio management activities ... fall under the Modigliani-Miller theorem on the irrelevance of pure financing decisions. It follows that there is no need to control the deposit creation or security purchasing activities of banks to obtain a stable general equilibrium with respect to prices and real activity" (p. 39). J. Huston McCulloch (1981) argues that the maturity transformation activity of banks borrowing short and lending long is not a natural function of financial intermediaries. It is a malfunction that is a by-product of several forms of government intervention that encourages what he calls .. misintermediatioll" (p. 103). Such theories, however, must be able to confront the data. If unfettered banking is optimal, why did it not work during the Free Banking Era? Why did the laissez-Iaire system turn out to be so chaotic and so costly? Why did it lead to wildcat banking and large numbers of bank failures in virtually all states that passed free banking laws? To help find answers to these questions and eventually test alternative theories of banking, we look much more closely at this period than previous researchers. I Using the J Hugh Rockoff (1975) produced the first analysis of Ihe Free Banking Era that tried to support a theory of what went wrong. Examining all states thaI adopted free banking laws, Rockoff argues that the instabililY in free banking systems was caused not by special economic
1081
original state auditor reports for several of the states that adopted free banking laws, we collected individual bank information on the banks that existed under the free banking laws in these states. This information allows us to determine the number of free banks that closed and to estimate what losses, if any, were sustained by the holders of their notes. While our new evidence confirms there were problems with free banking, it challenges the view that free banking led to financial chaos. In the next section of this paper we describe the Free Banking Era and why it has been generally regarded as a failure of laissez-Iaire banking. The second section contains our data on free bank failure rates, on length of time in business, and on note safety and noteholder losses. Using these data, we compare the free banking experiences of the states we consider and contrast our findings with the conventional view of free banking. The final section contains a summary and conclusion. I. The Conventional View of Free Banking
Because of the U.S. public's concern about financial instability and the misuse of financial power, banking has historically been one of the nation's most regulated and supervised industries. These concerns date back to the First (1791-1811) and Second (1816·-36) Banks of the United States. Both banks acled as private as well as public banks, and both reportedly used their financial power to at times rein in any banks they thought were too aggressive in issuing notes and were thus likely to cause currency problems. It was precisely this use of financial power by quasi-private banks, however, that caused Congress to revoke their charters. In both
factors nor by the lack of a central banking authorily, but rather by specific problems wilh the laws themselves and at times a failure to enforce them properly. Specifically, Rockoff argues that the problems occurred when the states allowed free banks to value the notes securing their bonds at par. Here we will not be concerned with testing Rockoff's theory. This is done in our 1982a, b studies.
180 1II11!
Free Banking 11 TilE AMI!RICAN ECONOMIC REVIEW
instances. the puhlic was concerned that these hanks had become too hig and too powerful. Over the years, beginning with the National Banking Act of 1863, the concerns with rinancial instahility and the misuse of rinancial power have produced a variety of laws. regulations, and government agencies aimed at controlling hanks and protecting their depositors. Commercial banks have heen restricted in location. in size, in portfolios. and in the interest they can pay depositors. Interstate IHllIlch hanking has heen explicitly prohihited since the McFadden Act of 1927. and intrastate hranch banking has heen restricted to some degree in almost every state. Bank size has been closely watched, limited first hy branching restrictions and then by the Bank Holding Company Act of 1956 and the Bank Merger Act of 1960. And under the Banking Acts of 1933 and 1935, hanks have been prohihited from underwriting and distrihuting nongovernment securities. owning common stock. and. until recently. paying interest on demand deposits. Not only have bank operations been restricted, they have also been closely scrutinized. No less than three separate federal agencies (as well as state hanking authorities) play a role in regulating, supervising. and examining bank operations. The Comptroller of the Currency oversees the activities of national banks; the Federal Deposit Insurance Corporation (FDIC) oversees the activities of all FDIC-insured banks; and the Federal Reserve System supervises its memher hanks. In order to promote stability and prevent the misuse of financial power, the federal government has clearly played an active role in regulating banking. Only in the period hetween the end of the Second Bank of the United States (11136) and the National Banking Act (11163) have hanks heen subject to almost no federal intervention. During this period, federal government involvement in the regulation of hanking was limited to restrictions on banks that held federal deposits. and even these restrictions were removed by the adoption of the Independent Treasury System in January 11147. So throughout this period, states were essen-
TARJ.E I-STATF.S WITH AND WITHOUT FREE HANKINGl.AWS RY I R60
States with Free Ranking Laws Michigan Georgia New York Alabama New Jersey Illinois Mn!\,":'('hl,s("tl~
Ohio Vermont Connecticut Indiana Tennessee Wisconsin Florida Louisiana Iowa Minnesota Pennsylvania
Year Passed l.aw IR37' I 83Rh IR.1R IR4Q h IR50 IR51 I K~lh IR51" IR51 h IR52 1852 IR52" IR52 1853 h IR53 IR5g h IR58 1R60"
Slates without Free Hanking Laws Arkansas California Delaware Kentuckv Maine Maryland Mis\itosippi Mis,ouri New Hampshire North Carolina Oregon Rhode Island South Carolina TexM Virginia
Sm/T('~:
Rockoff (1975. PI'. 3. 125-30). 'Michigan prohihited free han king after I RJQ and then passed a new free hanking law in I R57. "According to Rockoff. very little free hankin~ was done under the laws in these states.
lially free to design and regulate their own systems, and the system most chose was h:lscd on the free banking acts designed hy New York legislators. The first state free h1mking act that actually became law. however. was Michigan's in 1837. New York and Georgia followed in 1838. (A complete list of the states which adopted free banking and the years they did so is given in Table I.) Two characteristics of the free hanking laws distinguish free banks from traditionally chartered banks. First, the laws "':Ide entry relatively easy. Traditionally. to open a bank. entrepreneurs had to convince state legislators that a new bank was needed and that they were competent bankers. Such legislative charters were not required under free banking laws. Individuals with a certain minimum amount of capital could start a bank whenever and wherever they chose. Second, the free banking laws aHempted to
181
Free Banking II 1'01.. 73 NO.5
ROLNICK AND WEBER: FREE BANKING ERA
protect the noteholders. The laws required that designated state and federal bonds had to be deposited with a state authority as security for all notes. Moreover, banks were required to pay specie for their notes on demand and at par value. Banks received the interest on bonds used as security against their note issue only as long as they honored this specie-on-demand requirement. Failure to pay specie to even a single noteholder meant that the state would close the bank, sell the securities held as collateral, and reimhurse all noteholders. Most free banking laws provided additional protection for noteholders by giving them first lien on the assets of a bank. According to the conventional view, free bank notes did not work very well as a medium of exchange despite this protection. Because there were so many banks of varying reputations issuing notes, bank notes sold at differeut prices in different places, making tran!>actions with sudl notes quite complicated at times. Moreover, noteholders suffered losses throughout this period, especially those who held notes of so-called wildcat banks. These wen: banks that located note redemption offices only in areas where the" wildcats roamed" and then closed, leaving their noteholders with worthless pieces of paper. The literature is filled with references to the dehacle caused by free banking laws and the wildcat banks that appeared and then ljuickly disappeared. John Knox, for example, describes the free banking experience in Indiana as " ... the darkest page ill her financial hi!>tory ... " (1903, p. 70 I). According to Knox, Indiana's free banking law ... was loosely drawn, and opened wide the door for fraud. It was speedily taken advantage of by daring speculators, and banks sprung up like mushrooms everywhere .... Such a flood of paper money, with no substantial backIng could not help proving disastrous. The hills rapidly depreciated, and notes talo.en one aftef'lloon at eighty cellts might he ljuoted the next Illof'lling at si)lty-five, or even lower; thus all values Hlld lill husiness were deranged. [pp.701-02)
1083
Wildcat banking apparently was not confined to Indiana. As Bray Hammond reports, in Michigan, Wisconsin, and Illinois, Speculators bought bonds, issued notes to pay for them, and eluded their debtors by taking to the woods among the wildcats. Notes were issued by banks with no known pillce of business, and no regulllr office hours; and kegs of nails with coin lying on top were moved overnight from "hank" to "bank" to show up as cash reserves just ahead of the bank examiners. [1963, p. 9) Recent work by Hugh Rockoff (1975) adds New Jersey, Minnesota, and even New York to the list of states that suffered at least some degree of wildcat banking. This view of the Free Banking Era has become the conventional view that appears in most standard texts OJl money aJld banking. Dudley Luckett's text is a good example: ... free banking degenerated into socalled wi/deell bUllkillg. Banks of very dubious soundness would be set up in remote and inaccessible places" where only the wildcats throve." Bank notes would then be printed, transported to nearby population centers, and circulated at par. Since the issuing bank was difficult and often dangerous to find, redemption of bank notes was in this manner minimized. These and similar abuses made banking frequently little more than a legal swindle. [1980, p. 242) For students of banking, the implications of this view are very clear. The banking industry, left on its own, is unstable. Unless banks are closdy regulated and supervised, banking will self-destruct, causing chaos in financial markets and substantial losses to bank creditors and ultimately affecting real economic activity. Phillip Cagan goes so far as to assert, "The nation could not so easily have achieved its rapid industrial and COIllllIen:iai e)lpansion during the second half of the ninetecnth ccntury with the fragmented currency system it had during the first half ..... (1903, p. 20).
Free Banking 11
182 10M
TIll, AMERICAN H'ONOMI(' Rf:VTEW
II. The Free nanking Evidence
The conclu~ion that when hanking was left on it~ own it failed has significant implications for reguliltory policy and to a great extent influences policy today. Surprisingly. though. this conclusion is hased on casual empiricism. To hetter understand what went wrong with free han king systems, we have gone hack to the original state iluditor reports to document the financial history of the population of free hanks in four states. We describe in much greater detail than previous research the variety of experiences among states, and we find considerable reason to question the prevailing view of free banking. For our ~tudy we selected four states which altogether represent a wide rilnge of experience with free banking systems and which have state auditor reports available. These were New York, Wisconsin, Indiana, and Minnesota. We chose New York because it ~uppo~edly had the most stable system. Bray Hammond (1957) claims that, because of New York's sound hanking practices, free banking worked well in the state. Rockoff's (1975) work lends some support to Hammond; he finds that New York afler I R40 was relatively free of wildcat hanking. And historians generally regard New York's .~ystcll1 as an exception to the typically chaotic free banking systems of other slates. We chose Wisconsin and Indiana because, while they reportedly had serious prohlems with thcir systems, they also had periods when free banking appeared to work reasonably well. Minnesota, our fourth state, was chosen because its experience is thought to be one of the worst. Rockoff claims that this state's experience is a good example of the way wildcat banking arises and the damage it causes. Due to the apparent diversity of free banking experiences in these four states, a close look at the data provides insights into the generally accepted view that the free banking experience was a failure. The conventional view appears to be bascd on three ideas about the Free Banking Era that are accepted
nrCl-:,lfRfR IOR3
as facts: I) Free bank failures were numerous. 2) Free banks were in husiness for a relatively short period of lime. 3) Free bank notes were nol sllfe. and free bank failures produced subslllnlilliinsses for their noteholders. A large amount of datll on failure wles. on years in business, and on note safely and losses 10 nOleholders was available for our four ~tates (see Tables 2-5). These dala sugge~t that the" facts" underlying the com'entional view of free hanking are nol entirely correct and that the conventional view of free banking is an exaggeration.
A. Failure Rates Our data suggest that the accepled view on free hank failures is overstated. While a huge number of the banks that opened under Ihe frce banking acts closed well before I R(iJ, few of these banks failed in the sense of paying noteholders les~ than the par value of their notes. 2 Specifically, about half of the free banks in the state~ we investigaled c1o~ed, bul less than a third of the hank~ that closed did not redeem their notes at par. Evidence supporting this assertion is presented in Table 2, which summ;ui7.es the exreriellce ill each state. Column I contains thc total numher of free hallks ill each slate. New York with 449 free banks between IR38 and 1863 is our largest system; Minnesota with only 16 free banks between 1858 and 1862 is our smallest. Column 2 reports the number of banks for which we could find redemption information. Column 3 contains the number (and percentage) of free banks that closed. New York had the most closings, but as a percentage of the total number of its free banks it had the least. Column 4 contains the number (and percentage) of closed hanks that redeemed their notes below par and can thus be considered failures.
2We define a free hank failure in this way hecause a major intent of the free hanking laws was to provide a safe currency. The laws made no attempt to insure depositors or shareholders against risk.
Free Banking II
183
ROI.NICK ANV WHHHR: FRHH BANI\IN(; ERA
VOl .. 7.1 NO.5
1U1I5
TA/ll.~ 2-NU~I/IER OF FREE BANKS, FREE BANK CLOSINGS, ANI) FAILURES
IN FOUR STATES
Slate (Free Banking Years) New York
Free Banks
Free Banks wilh Redemption Information
Free Banks that Closed ('if, of Col. I)
(1)
(2)
(3)
441)
445
140
140
104
77
16
16
160 (36) 71) (56) 81) (86) 11 (61) 339 (48)
(11138-63)
Wisconsin (1852-63)
Indiana (11152-63)
Minnesota (111511-62)
Total
701)
678
Free Banks that Failed ('if, of Col. 2) (4) 34 (11) 37 (26) 24 (31)
9 (56) 104 (15)
SOW'"",': New York, Indiana, and Minnesota state auditor reports and Wisconsin state auditor reports as given in U.S. Congress (11138-63).
Table 2 clearly confirms the accepted impression that free banking did not work: a large number of banks closed during the Free Banking Era. Of the 709 free banks in the four states we considered, 339 (48 percent) closed. Over half of all free banks which existed in Wisconsin, Indiana, and Minnesola closed, and the highest closing rate was Indiana's H6 percent. Even in New York, which had the smallest percentage of bank closings, 36 percent of all free banks dosed. However, a somewhat milder picture enll:rges if we consider only those free banks Ihat dosed and redeemed their notes below par, that is, banks which we consider to have faill!d. We find that only 104 (15 percent) of the 67H free banks on which we were able to oblain redemption rate information actually c1o~ed with below par redemption of notes. Thus, only about one out of three free bank c1o~ings resulted in losses to noteholders. Examining the evidence state by Slate, we find only 8 pen.:ent of New York's free banks dused with bdllW par redemption. The beluw par closing rates for Wisconsin and Indiana were virtually e4ual-3 out of 10. Minne~l)til had the highest below par closing (,lte: 56 pcrcent. The individual state results wnfirm the general impression that New York's free banking system worked well and Ihat Minnesota's free banking experience was
among the worst. 3 They also confirm the view that free banking created at least some problems in most states that adopted a free banking law. However, viewing the nearly 50 percent closing rate as a failure rate clearly exaggerates the extent of the problems.
B. Years in Business Another part of the conventional view is that many banks formed under free banking laws were not only unsuccessful, but they were also short-lived. Rockoff, for example, in developing an explanation of wildcat banking under free banking laws, assumes such banks were in business for only a month or two (1975, p. 8). Our data suggest that this part of the conventional view is also overstated.
JActually, Minne,uta's experience was wor,e than our table shows. The State Bank of Minne,ota withdrew all but a ,mall fraction ()f its circulation within a year after opening and did not isslle notes again until il moved t,) SI. Paul in October 11162. lle,ide, that, accofll· ing to the report of Minnesota's statc auditur fl)r IlIbl, "the La em,sc and La Crescent and Chatfield banks maintain no office uf discount, deposit, and circulation in lhi> State. Their cirwlation is entirely confined to Wi,consin" (p. (6). Thus, only 2 of Minnesota's 16 fr<"\! banks remained in operation during the elllire five-year period we consider.
184
Free Banking 11 Tilt; A MERICAN ECONOMIC REVIEW
We estimated the length of time each free hank in our states existed from the reports of condition they filed with the state auditors. First we compiled tahles showing the dates at which each hank appeared in condition reports from the time a state's free hanking act was passed until 1863. 4 Then we assumed that a free bank opened half-way between the date of the first condition report in which it appeared and the date of the previous report and that a free bank closed half-way between the date of the last condition report in which it appeared and the date of the next condition report. We estimated the length of time a free hank was in business as the difference between these dates. These estimates are given in Table 3. Banks which never appeared in a condition report are assumed to have been in existence for zero years. Minnesota is not included in the table because we know most of its banks existed for only a short time, and we doubt whether those banks that existed longer did much banking business during 1860 and 1861. (See fn. 3.) Note that this evidence understates the number of years some free banks were in business, for two reasons. First, since we consider the Free Banking Era as ending in 1863, our estimates ignore the fact that some banks continued to exist after 1863 as either state banks or national banks. Second, since condition reports for New York's free banks were not available until 1843 (five years after its free banking law was passed), we consider all New York free banks as starting business in that year. Despite these downward biases, our estimates indicate that the New York and Wisconsin free bank populations were not marked by large numhers of short-term hanks. New York free banks were in business a mean of 7.9 years (a median of 8 years), and Wisconsin free banks were in business a mean of 4.3 years (a median of 4 years). The free banks in Indiana were in business for far shorter periods. The mean
• These tahles arc availahle from the authors upon requesl.
lJI:CfMRUllfJlJJ
TAn!.F. 3-·-NlJMIIF.R OF FRFF. RANKS ny YFARS IN IllJSINr:SS rOR TJlR[F. S 1/\ II'S Numher of Years in
Rlisinc~s
O.Oh 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
R.O 8.5 9.0 9.5 10.0 10.5 11.0 12.0 13.0 14.0 15.0 16.0 17.0
IR.O 19.0 20.0 Mean Median
Numher of Ranb New York'
Wisc"nsin
Indiana
48
4 0 13 4
JO 18 24
37 19 25 16 14 26 29 23 23 71
13 5 19 5 14 IJ 6 IJ
I 7 3 R 0 2 2 0 2 6
I I
.. , utal ~1 I~ /.~
1.1 R
3 4 3
4~
.1
.1.1
I 0
20
0 2 ()
.1 5 0 0 2 4
35
14 1.1 2<J 7 .15 13 2.1 2 27 4 73 6 35
I
I
14 12 7
14 12 7
I
I
4 2 10 31 Number of Years in 7.9
R.O
4.3 4.0
4 2 10 3I Ru~iness
2.0
1.0
6.3 5.5
Srmrce.v; See Tahle 2. 'The Manufacturer's Bank of Rochester. ",hid, opened in 1856, and the Eagle Rank of Rochester, ",hi('h opened in I 850, merged in 1859 to form the Trader's Rank of Rochester. Consequently. the Manufacturer's Jlank of Rochester was excluded from this calculation. and the merged hank was treated as existing since IR50. hNumber of banks which did not appear on any condi tion report.
length of time a hank existed there wa~ 2 years (the median, only I year). NOlletheless, considering the three states togelher. free banks were in business a mean of 6.3 years (a median of 5.5 years) and only 14 percent were in business less than I year.
185
Free Banking II VoL .. 7J NO.5
ROl.NICK AND Wl:.'lJER: FREE BANKING l:.·R.1
C. 11Ie Safety vf Balik Notes alld NVlelwlder Lossel'
We find Ihl! most exaggeratl!d part of the convl!nlional view of free banks is that concl!rning thl! safdY of the bank notes and the lossl!s to notl!holdl!rs, Many have asserted Ihat free bank notes wl!re genl!rally unsafl!, and some have claimed that in some individual stall!s, thl! losses to noteholders ran into 1111: millions of dollars,s Our data, however, Idl a much diffl!renl story. They indica Ie Ihal free hank noles werl! rdalivdy safe and Ihal Ihe losses 10 nOll!holders wl!re smaller Ihan lIIany have I!stimatl!d, To dl!ll!flnine how safe free bank notes werl! oVl!r liml! for I!ach statl!, we multiplil!d Ihl! circulation of each frl!e bank in each condilion report by its final redemption rate, lotaled the products for all banks for each report date, and divided the sum by the total cir~ulalion of all banks for which we had n:dl!mplion rale informalion. The result, shown in Table 4, is a measure of the expecll!d value of a randomly selecled bank note held until 1863 as of the date of each condilion report. The evidence in Table 4 shows that free bank notes were relatively safe, although the degree of safety varied over states and over timl! within a state. New York bank notes were the safest; the expected value of a randomly selected New York bank note never fell below 99 cenls on the dollar, and for many years this expecled valul! was one dollar. Wisconsin's experience was at first very similar to New York's, but the safety declined over time to a low of 88 cents on the dollar in 1861. (All of Wisconsin's failures occurred in 1860 and 1861; sec our 1982b study,) Indiana's problems wilh free banking occurred within Iwo or three years after its free banking act became law in 1852, This is shown by the expected values of 92 and 95 cl!nts on the dollar in 1853 and 1854, respl!clivdy, 1I0wevl!r, as early as I S56, Indiana's
'Sec, for example, Ihe ,Ialemenl uf Ilugh McCulluch, 1'Il·:-,iJ~1l1 b~Jlk),
{Jf
lhl: UalJk of tile Slate of JIIJialia (nol a frl'l."
ciled in lIammund (I ~57, p. 620).
/vli7
free banking expl!rience was also very similar to New York's, finally, Minnesota ddinilely had the worst bank notl! safety, wilh expected values of less than 50 cents on the dollar through July IS59, However, the safely of Minnesota free bank notes improved substantially after Ihis time; expected values wl!re above 80 cents on Ihe dollar by OClObl!r 1859. Given that the above measure of frel! bank note safely indicates that these notes were not perfeclly safe, Ihe queslion arises: how much did noldHllders losl!? Thl! losses of llotdlOlders in our four free banking slales cannot bl! calculated exactly, Howevl!r, using thl! dala on frel! bank circulation and redl!mption ratl!s, thl! magnitude of such losses can be estimatl!d. Threl! estimates using diffl!rent assumptions are shown in Table 5,6 The first estimate (col. I) is based only on those faill!d banks for which circulation data were available. It is the result of multiplying Ihe last circulation for I!ach bank which faill!d by one minus its redemption rate. Sincl! circulation data were available for only 88 of the \04 failures in our four states, this estimate can be considered a lower bound on total losses, However, it is probably biased upward because some notes included in the circulation numbers we used may have been redeemed at par before the bank failed. The second estimate (col. 3) expands the coverage to include the failed banks for which no circulation data were available. It is the result of multiplying the average loss per bank (col. 2, Lased on col. I) by the total number of failures in each state. We consider this estimate our best approximation of the total losses to the holders of notes of failed free banks, although it is subject to the same upward bias mentioned above, Finally, because in New York and Indiana we could not determine whether or not SOIllI!
"We have divided Ihe New York and Indiana e .• peri· ence in calculaling Ihe average loss per bank because Ihere are subslamial differences in Ihe,e subpcriods for these two statc~. Comhining lhl.! cxpcrkncc leads to !'IligiJlly
and 4.
~lIIalicr \.·~Iilllah,:.'"
or IIIL' lotal
lo!'l~l'.'"
ill cob
I
186
Free Banking 11
101111
Tiff; AMERICAN ECONOMIC REVIEW
m,CHMBliR 198.1
TARI.E 4-FREE HANK Non SAFETY ANI) CIRCUlATION IN FOUR STATES
Slate New York
Wisl'onsin
Indiana
Minnesola
SOllrce.,: See Table 2.
Date of Condition Report 1M3 (Nov.) IK44 (Nov.) 1845 (Nov.) 1846 (Nov.) 1847 (Feb.) 1848 (Mar.) 1849 (Dec.) 1850 (Dec.) 1852 (June) 1853 (Dec.) 1854 (Sept.) 1855 (Sepl.) 1856 (Sepl.) 1857 (Sepl.) 1858 (Dec.) 1859 (Dec.) 1860 (Dec.) 1861 (Dec.) IKli2 (I)ce.) 1853 (.hIIV) 18540"".) 1854 (hIIV) 18550"".) 1856 (JUIl.) 1857 (Jail.) 1858 (Jan.) 1859 (Jan.) 1860 (Jan.) 1861 (Jan.) 1862 (Jan.) 1862 (July) 1853 (Dec.) 1854 (July) 1856 (Jan.) 1856 (July) 1857 (July) 1858 (Jan.) 1858 (July) 1859 (Jan.) 1859 (July) 1860 (Jan.) 18liO (July) 1861 (Jan.) 1862 (Jan.) 1862 (July) 1863 (Jan.) 1859 (Jan.) 1859 (Apr.) 1859 (July) 1859 (OCI.) [860 (Jan.) 1860 (Apr.) 1860 (July) 1860 (Ocl.) 11163 (Jan.) 1863 (OCI.)
Expected Value of a Randomly Selected Dollar flank Note
Note Circulation of All Free flanks
$0.997 0.998 0.999 0.999 0.998 0.993 0.998 0.998 0.999 0.999 0.999 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
$ 3.362.737 5,036.953 5.544.311 6.235,397 5.970.941 8.621.269 10.191.000 13.197.995 14.621.582 21.029.339 21.435.545 23.169.329 26,476.389 22.015.221 23.229.189 24.524.209 23.900.049 25.990.007 35.049.604 $ JOI.748 485.121 786.21(, 740.764 1.060.165 1.702.57!) 2.913.071 4.695.168 4.429.855 4.283.175 1,419,413 1.643.148 $ 3.167.547 5.219.105 1.448.318 1.423.895 1.453.703 1.079.928 1.043.608 [.027.569 1.080.577 1.108.396 1.143.466 [.035.664 971.933 1.109.411 1.223.426 $ 50.000 216.549 298.959 155.258 34.481 38.898 44.38[ 49.145 94.133 100.161
$UXI UKI
0.'1'11 0.991 0.'183 O.9fi4 0.936 0.928 0.896 0.882 1.00 UXI
$0.922 0.949 0.997 0.997 0.992 0.990 0.989 0.989 0.989 0.990 0.990 1.00 1.00 1.00 1.00 $0.456 0.489 0.500 0.810 0.876 0.846 0.820 0.S5!) 1.00 UK)
Number of Free Banks
Average Circulation per Bank
50 65 67 70 70 93 109 130 158 221 232 239 261 266 268 273 279 276 289 8
S 67.254.74 77,491.58 82.750.91 89.077.10 85.299.16 92.701.82 93.495.41 101.523.04 92,541.66 95.155.38 92.394.59 96.942.80 101.442.10 82.763.99 86.676.08 89.832.27 85.663.26 94.166.69 121.278.91 S 37.718.50 48.512.10 41.379.79 n.207.13 35.338.83 37.834.89 42.839.28 47.425.94 41.400.51 39.659.03 22.893.76 27.385.80 $105.584.90 113.458.80 45.259.94 41.879.26 55.911.65 56.838.32 57.978.22 64.223.06 63.563.35 65.199.76 63,525.89 57.536.89 53.996.28 61.633.94 71.966.24 $ 25.000.00 30.935.57 22.996.85 14.114.36 5.746.83 6,483.00 8.87620 9.829.00 18.826.60 20.032.20
10
1'1 2J 30 45 68 9'1 107 108 62 60 30 46 32 34 26 19 18 16 17 17 18 18 18 18 17 2 7 1.1 II 6 6 5 5 5
187
Free Banking II VOl.. 73 NO.5
ROl.NICK AND WEBER: FREI:." BANKING f,'RA
/081J
TAIII.~ 5-EsTlMATEI> LOSSES TO FREE BANK NOTEHOLI>U.S IN FOUR STATES
TOlal Losses of Free Bank Failures with Circulalion' (I) New York Ddore OCl. 11141 In and After OCl. 11141 Wi')l:ull.)in
Indi,ma Ikforc 11156 In and Afler 11156 Minnc.)ota
TOlal
Average loss per Bank
Tolal Losses of All Free Bank Failures
Tolal Losses of All Free Dank and Unidenlilicd Failures
Average loss pa Dollar
(2)
(3)
(4)
(5)
$21,724.29 14,264.96 13,5911.69
$ 325,86435
$ 325,1164.35
271,034.15 503,151.42
3211,093.97 503,151.42
$0.2637 0.2599
17,069.76 5,114.113 18,356.76
3511,46U7 15,344.50 165,210.85 1,639,070.14
768,139.01 3().689.00 165,2\0.85 2,121,1411.60
$ 2112,415.77 (13/15) 213,974.33 (15/19) 503,151.42 (37/37) 204,1137.07 (12/21) 15,344.50 (3/3) 146,854.09 (8/9) 1,366,577.18 (88/104)
0.2444 0.1069 O.15B 0.7050
:)""'.....,.: Sec Table 2. • Fraclions in parentheses indieale number of failed frce banks wilh circulalion availablc/lolalnumber of failures.
banks dos.:d b.:low par, we c.alculated a third (col. 4) as an upper bound on the luss.:s suffered by hold.:rs of free bank notes. In Ihe third estimate, we assumed all of !til: unid.:ntifi.:d banks in these states failed, and we used the average loss per bank (col. 2) to eSlimate their losses. We.:stimate that the total losses to noteholders under free banking in our four states at $1.6 million with a range of between $1.4 million and $2. I million. The breakdown by states is as follows: New York, $597 thousand wilh a range of between $496 thousand and $654 thousand; Wisconsin, $503 thousand; Indiana, $374 thousand with a range uf between $220 thousand and $799 thousand; and Minnesota, $165 thousand with a range of betwee:n $147 thousand and $165 thousand. These estimated losses are slightly higher than those presented by Rockoff (1974, p. ISO). Howe:ver, since he only presents estimated losses through 1860, his totals do not include Wisconsin. Nonetheless, our estimates taken in conjunction with Rockoffs cast doubt on the claims that losses to free bank noteholders in individual states may have run into the millions of dollars. 7 Ine~timah!
7We hoped 10 gel similar dala for Michigan, since il ;. con,idered 10 have had Ihe worsl experience wilh free banking. Michigan's bank commissioner records for Ihe Free Banking Era, however, were deslroyed in a fire,
stead, such claims se:em to implicitly assume that all note:s of failed banks wen: worthless, which was obviously not true:. ti In Table 5 (col. 5), we prese:nt the average loss per dollar fur banks that failed. These losses are derived by dividing the total losses (col. I) by the total circulation of the failed banks. Thus, column 5 is an estimate of the: loss that individual noteholders could expect if they held a note of a bank that failed. In our four states, this expected loss was the smallest in Indiana, ranging between 11 and IS cents on the dollar. The expe:cted loss in New York was approximately 26 cents on the dollar. The expecled losses in Wis~ollsill
and so far copi.:, hav~ nul been localed. Nevcrlhdcss, we did find a bank commissioner reporl daled January III, 11139, thai conlains informalion on over 20 of Michigan's frec banks Ihen being liquidaled. The lhree commissioners wriling Ihe reporl were very confidenl Ihal all bUI 2 of tile free banks had assels more Ihan sufficienl 10 pay llleir nOleholders. (The lwo exceplions were Ihe Bank of Wasillenaw and Ihe Fanners' and Mechanics' Bank of Ponliac.) Thus, lhc losses suslained by holders of Ihe nnles of Michigan's free banls also appear 10 have been overslaled. 'We have considered only lhe losses suslained by nOleholders nOI only for Ihe reason given in fn. 2, bUI also because our dala do nOI permil calculalion of Ihe losses suslained by dcposiwrs or shareholders. FUrllIer. it should be nOled that our loss eSlimales ign"re any delay or inconvenience suffered by individual. in redeeming their nole, al lhe slale audilor.
Free Banking II
188 1090
TilE AMERICAN ECONOMIC REVIEW
and Minnesota were 24 and 71 cents on the dollar, respectively. These results present an interesting contrast between New York and Indiana. Total losses to noteholders in New York came from a slllall percentage of failures involving a substantial loss on each dollar of the failed bank's outstanding circulation. Total losses to noteholders in Indiana, however, came from a larger percentage of failures involving much smaller losses on each dollar of circulation. '". Summary and Conclusion
In this paper we have presented extensive quantitative evidence on the Free Banking Era. We found that each of the four states we examined had a significant number of problem banks and that the experience varied considerably among the states. We also found that many free banks in our four states did not go out of business; and of those that did, many still redeemed their notes at par. Further, we found that most free banks existed for more than five years and that total losses to noteholders have been significantly overstated. Our preliminary conclusion from this evidence is that it is misleading to characterize the overall free banking experience as a failure of laissez-faire banking. However, we also rec()gnize that further work is needed before this period can be properly judged. We would also like to suggest that this future research address the following questions raised by our new evidence. Were the problems with free banks caused by some inherent instability in the banking business, or can they be explained by the laws and regulations that governed free bank activities? Was there enough variation in these laws to explain the different experiences among free bank states, or can the differences be explained by special characteristics of the states themselves? And most importantly, what implications can be drawn from the Free Banking Era about banking deregulation today? REFERENCES nt-ocker, (;ary S., "A Proposal for Free Bank-
ing," unpublished paper, University of
DECEMBER 1983
Chicago, 1957. Cagan, Phillip, "The First Fifty Years of the
National Banking System-An Historical Appraisal," in Deane Carson, ed., Banking alld Monetary Studies, Homewood: Richard D. Irwin, 1963, 15-42. Dillislin, William H., Bank Note Reporters and Counterfeit Detectors, 1826-1866, With a Discourse on Wildcat Banks alld Wildcat Balik Notes, Numismatic Notes and Monographs 114, New York: The American Numismatic Society, 1949. Famll, Eugene F., .. Banking in the Theory of Finance," Journal of MOlletary Ecollomics, January 19RO, 6, 39--57. ""rlednllm, Milton, A Program for Monetary Stability, New York: Fordham University Press, 1960. Hammond, Bray, Banks and Politics in America: From the Revolution to the Civil War, Princeton: Princeton University Press, 1957. ___ , "Banking Before the Civil War," in Deane Carson, ed., Banking OIld Monetary Studies, Homewood: Richard D. Irwin, 1963,1-14. Kareken, John H., and Wallace, Neil, "Deposit Insurance and Bank Regulation: A Partial-Equilibrium Exposition," Journal 0/ Business, July 1978,51,4\3-38. Knox, John J., A History 0/ Banking in the United States, New York: Bradford Rhodes and Co., 1903. Luckett, Dudley G., Money and Banking, 2d ed., New York: McGraw-Hili, 1980. McCulloch, J. lIuslon, .. Misintermediation and Macroeconomic Fluctuations," Journal of MOlletary Economics, July 1981,8, 103-15. Rockoff, Hllgh, "The Free Banking Era: A Reexamination," Journal of Money, Credit and Banking, May 1974.6, 141-67. ___ , Tire Free Banking Era: A Reexamination, Dissertations in American History (rev. doctoral dissertation, University of Chicago, 1972), New York: Arno Press, 1975. Rolnick, ArthurJ., and Weber, Warren Eo, (1982a) "Free Banking, Wildcat Banking, and Shinplasters," Federal Reserve Bank of Minneapolis Quarterly Review, Fall 1982, 6, 10-19. _ _ and _ _ ,(1982b) A New Explanation for Free Bank Failures, Research De-
Free Banking II VOL. 73 NO.5
ROLNICK AND WEBER: FREE BANKING ERA
partment Staff Report 79, Federal Reserve Bank of Minneapolis, 1982. Indiana, State Auditor, Annual Report of the State Auditor, Indianapolis, selected years, 1852-63. Michigan, Bank Commissioner, Bank CommissiOfler's Report, Lansing, 1839. Minnesota, Auditor's Office, Annual Report of the State Auditor to the Legislature of Minnesota, St. Paul, 1860; 1861. ___ , Annual Report of the State Auditor to the Governor of Minnesota, Comprising State Accounts, Banking and Public Lands, St. Paul, 1862. ___ , "Annual Report of the State Audi-
189 1091
tor to the Legislature of Minnesota," Executive Documents of the State of Minnesota for the Year 1863, St. Paul, 1864. _ _ _ , House of Representatives, Journal of the House of Representatives of the Second Legislature of the State of Minnesota, St. Paul, 1860. New York, Bank Department, Report of the Superintendent of the Banking Department, Albany, selected years, 1850-63. _ _ _ , State Comptroller, Report of the Comptrolle.-. Albany, selected years, 1838-50. U.S. Congress, House Executive Documents, Washington, selected years, 1838--63.
[10] Another Look at Free Banking in the United States By
JAMES
A.
to find for muny st:ltes, :lI1d becuusl! IIII! closure rate should give slime indicatilln of the quality and stubility of u stute's b:lIlking system, I present here state-by-stute comparisons of bank "life expectancies." The data come from the same sources as those uscd by Rolnick-Weber (Hol/se Executive DOell/llell/s, 1837 -63). Onc dilliculty in the estimation of lifc expectancies over a tinite sample period is that many of the banks will still be in openttion at the end of the period. Rolnick and Weber assign to such banks their age at the end of the sample period and then calculate the sample means. They recognize the obvious bias in such a procedure, and present their estimates as lower bounds. Their goal was to show that most banks were in business for more than just a few months or a year. Unfortunlltely, one cannot use such calculations to make mellningful state-to-statl! comparisons, because both the length of the sample period and the proportion of flliled banks to the total number of banks varied tremendously lIcross states. I show below that under the assumption that the underlying stochastic process can be represented by a constant hazard rate for bank failures, and that the observations are independent, correction of the truncation bias is a simple matter. By assumption one can ignore the truncation at the sl:lrt of th.: period, and pretend that all b,lIlks in operation at that time opened within the previous year. It then turns out that a consistent estimate of the life expectancy is just the totlll number of bank-years in the sample, divided by the number of failed banks, or, to put it another way, the probability of failure in a year is estimated by the total number of failures divided by the total number of bank-years. Besides simplicity, this procedure has other desirable features as well. One can get reasonable estimates even for relatively short sample periods (as will he :.cen below), lind one can get consistent cstilllatl!s of standard errors as well.
In thl!ir 1910 pllper, Arthur Rolnick and Warrl!n Wl!her document the experience of several states' experiments with fn:e banking in the decades leading up to the Civil War, using data on bank openings and closings, and on redemption rates of failed banks (the rate, per dollar par value, that failed banks were lIhle to plly their note holders). In this note I extend the Rolnick-Weber analysis by comparing the experiences of free and nonfree-han king states with regard to the frequency of bank closings and failures, to give some perspective to their conclusion that free banks were not as short-lived as has been the conventional view. In so doing I on"cr an alternative set of statistics to those provided by Rolnick-Weber on bank life expect;tncies and average redemption rates of failed hanks, statistics that I argue take better account of the truncation problellls inherent in the data. I find that bank closings were much more frequent in free bank states. This was especially so in the first five years after deregulation, but remained true to some extent throughout the period. I. Hank Failure Rates
One prohlem often attributed to the Free Banking Era, and to unregulated banking ilion: generally, is a high rate of hank failures. As correctly pointed out by Rolnick-Weber, what is relevant from a public policy standpoint is not so much failure in the sense of goillg out of businl!ss, but rather IIlOfI! in tl!rms of the inability to redeem notes at their par value. That is, the public policy concern is primarily the provision of a secure currency, not the safl!ty of stockholders' investments. Nonetheless, because data on redemption rates of closed banks are difficult
• (iraJualc .luJelll,
rvla»adIU.CIIS
11l.lilulc of Tech·
""logy, Call1briJj!e, MA 1J213'J. I Ihank I'cler Temin for I,:UI.I 111\1..'11 b,
KAIIN*
ilud A. J. H.oitlit.:k fur ::.upplying ~um~ of the
d ..... /Jill
191
Free Banking II l/ll!
TIlT:' AMliR/CAN F:CONOM/C RFYIEW
Again, the point of this correction is not so much to get beller statistics for the free bank states (after all, if most of a state's hanks did fail, my statistics will not dilfer very much from Rolnick-Weher's). Rather it is t;, get statistics that arc comparahle across states, even when the lengths of the sample periods vary widely. Of course, this is not without its prohlems too. The phenomenon of hank panics raises some doubts ahout the independence of ohservations, and one might expect the ha7.ard rate to decline over time. I take lll'C(lunt, to sOllie extent, of the laller possihility in one of the sets of calculations helow, but mainly it is hoped that this simple correction at least enables one to make stateby-state comparisons; that is, that the biases are not too different from one state [0 the next. This could not he the case with Rolnick-Weber's statistics for the reasons stated above. Suppose that a bank currently in operation has every year a prohability p of going out of husiness, and let Xj denote the total numher of years that hank i operated. Then one may show that
(1)
Pr ( Xj = II) = P (I _ p)" -I,
(2)
Pr(X >II)=(l-p)",
(:~)
j
/:( X,)
=
1/".
Lelling k denote the life expectancy of the hank, it is clear from (3) that an equivalent model may he expressed in terms of k, with p=l/k. Now suppose we ohserve a sample of N hanks over time, hut that we do not necessarily ohserve X,. Instead we ohserve Yj , where )'j = Xj if X, < '1;, and = T; otherwise, where '1; is the numher of years hetween the opening of hank i lllld the last year of the sample. For concreteness, suppose the first M hanks in the sample are observed to go out of husiness, while all the others are still operating at the end of the s,lmple period. From equations (1) and (2), and assuming that the ohservations are independent, the likelihood function for the N ohservations of
Sn'TF:MRI':R 19115
Yj is then M log( p ) + log( 1 - p)
x{
t
;~I
(,\'j - I ) +
f:.
1j} .
j - M' I
This is maximized at p* = {M-IE~_I Yj } -I, so that k * = M-IE~_I r;. Also, it turns out that a consistent estimate of the variance of the estimator k* is M-Ik*(k* -1). Table 1 gives results for the three states used by Rolnick-Weher (New York, Wisconsin, and Indiana), along with results for Minnesota (a free bank state not included in their mean life calculations because the sample period was too short), and for three other states: Maine and Maryland, which had no free banking laws, and New Jersey, which passed a free banking law midway through the sample period. I The results show strikingly that the 'banks in free bank states had much shorter life expectancies (i.e., much higher failure rates). The table also provides estimates of the standard errors, which indicate that the means are estimated fairly precisely_ One might argue that the foregoing analysis overlooks the fact that any major change in the regulation of an industry is hound to cause some upheaval, at least initially, so that it is misleading to characteri7.e the higher failure rates as a permanent elrect of the free bank laws. (The airline industry today is a case in point.) For this reason, Table 2 provides estimates for the free bank states beginning roughly five years after the initiation of the policy (excluding Minnesota, which had free hllnking for only five yellrs in all, and New York, for which the estimates in 'I allempled 10 choc'sc nnn·rree-hank slales Iha. were n~ unexl'C'ptinnal nnd .a~ hrnndly rcprc~('ntati\'c a~ possible. Many nr the Iighlly populated weslern slales had only one state bank ror much or Ihe period. and so would not have provided much inrormatic,". On Ihe other hand, states slich as Massachusetts, with large, wett-estahtished banking systems. would probably have been unrepresentative or state-eharlered banking systems in general (just as. I witt claim later, New Ynrk was not a Iypical rree hank state).
192
Free Banking II /1/13
KAliN: U.S. FREE BANKING
VOl.. 75 NO. -I
TAua 1·- - BANK l.1H' EXI'H:TANOI:S MD
NJ
1837-63
1837-51
21 22 11 61.1 (i8.3)
25 9 3M.7 (12.1)
66 37.M (4.6)
NJ
NY
WI
IN
I M52-63
IM43-62
IM53-62
IM54-63
ME A. Nun-Fr~~-lIank Siaies Perilld I K37-63 Banks al Slarl 55 Openings 59 C1u.ings 46 Mean Life 31.2 (4.5)
10
Tuwl
101
90
TOI~I
I\·IN
Ii. Free Bank Siaies Period Banks al Slar! ()p~nilll\.,
51
('Iosing!-l
25
Mi"illg
II 9.4
M~alll.ik
49 41H) 16U 4X 22.1
(I.X)
(1.1)
1.12 XU 4 7 ..l (U.X)
IM5~
3U H
Xl)
14 II I
Xt.
.lU .U (IU) .-
NolL':
Standard
crr()r~
are shown in
ME
NJ B. Free Bank States Period 11156-63 Bank> at Slart 35 Openings 21 C1".ing. X Mean Life 24.7 (11.5)
1,'11 \(I~
<).j
.1.1
12.9
IIIX)
111.7)
----- . ----
parclllhl!~cs.
TAUI.I: 2--·BANK Lift: ExPECTANCII:S AvnK INITIAl. Yt:AKS
A. Non-Free-Bank Slates Period IM54-63 Bank> at Start 60 Openings 2X Closings 19 Mean Life 37.1 (XA)
63
led to higher failure rates. The results indicate that while there was overall some improvement in the free bank states after the initilll five years, bank failure rates remuined considerably higher than in the non-free-hunk states. 2 In the laller, failure rates were remarkably stable throughout the era.
MD
IM54-63 24
13 5 60.4 (26.X)
II. The Costs of Holding Bank Noles
WI
IN
1857-62
1858-63
45 85 70 7.4 (0.8)
19 2 5 20.7 (9.0)
Nute: Standard errors are ,hown ill parentheses.
Table 1 sutfice). It also presents mean life estimates for Maine and Maryland based on comparahle sample periods, to take account of the possihility that there WliS something dillcrenl uhoUI Ihe IIISO's tlmt might huve
As stated above, the goal of the states' bank regulatory policies was presumahly a sound currency, and the consequent puhlic confidence in the banking system, rather than the safety of bank shareholder's investments. A high rate of bank failures is a symptom of ditficulties, but may not itself have been con\ s~dered. a serious cost from a public policy vleWpOlll1. Unfortunately, data on redemption rutes of failed banks lire not as easy to come hy. The four free bank states documented hy
:2 The inlprovcmcnlS may have b~~n due- 10 Ihl.' W"'l'd,· ing out uf inc..."umpctcIIIS, (0 improved cnfUrL'l..'llll'nl pn,ceduces hy the Slale!'!, llr h.l llIoliilil.:alions uf lilc fn. 'l,'
Imukilltt, laws
it
i~ 1101 ,.'Il~m·
fro III 1II1.'
d')ClIllIl'lIl!\
Free Banking II 884
193
Till:' A MI:'RICAN I:CONOMIC III:'VII:'W
Rolnick- Weher appear to he the only ones that included redemption rates on notes of failed hanks in a systematic way in their reports to the Treasury Department. Therefore in this section no explicit comparison bctween the free and non-free-hank states will be attempted. One should hear in mind, however, that even if redemption rates in non-free-bank states were no higher than those in free bank states (and what evidence is contained in the auditors' reports suggests that they were lower in free bank states), the low failure rates in non-free-hank states imply that the losses to note holders were prohahly negligible. Rolnick and Weber calculate a time-series for each of the four states that they call" the expected value of a randomly selected bank note held until 1863 as of the date of the condition report" (p. 1084). Actually, what they show are the ex post average 1863 values of the notes of banks that are operating in the given year. There arc several problems with this as a measure of the riskiness of note holding. First, as with any ex post measure, it is hased solely on information not availahle at the time that agents are making their decisions, and hence may hear little relation to the actual determinants of the notes' market values. Second, their calculations place wry litlle structure (In the data. Since it is a fairly general property (If f(lreeasts that they arc less volatile than the time-series that they are attempting to predict, it seems likely that Rolnick and Weher's estimates are overly sensitive to the realizations on which they are hased. A more serious problem with their calculations is, (lnce again. Rolnick and Weber's failure to take account of the truncation of their sample, one which lead~ them to make unjustified conclusions ahout the improvement in the performance of free banks over time. The basic problem is that the prohability of a hank's failure hy 1863 will decrease over time, even with a constant hazard rate model as in the previous section. Thus Rolnick and Weher's measure will inevitahly improve over time (and he hiased upward to hegin with). Yet they conclude from their calculations that, for example, "the safety of Minnesota free bank notes improved suh-
S"l'n:MII':R l'IR5
TARI.E 3-AVERAGE REDEMrTlON RAH' (R) ANIl IIOJ.IlING LOSSES ( I.)
R I.
NY
WI
IN
MN
0.95 0.002
0.R7 0.02
0.95 0.D2
OAR 0,12
Note: Calculations arc descrihed in the tex!.
stantially after [July 1859] ..... (p. 1087). Such a conclusion would validly follow from their results only if one were to assume that everyone knew for certain all along that after 1863 all hank notes would be safe forever. This section reports results from an alternative set of calculations. I calculate, for each of the four states, the average redemption rate (denoted R) of failed banks. The expected redemption rate for one year in the future is then just 1 - p(1- R), where p is the failure probability. Put another way, the expected one-year holding loss (denoted I.) is p(1 - R). Table 3 shows the results from this calculation, where p is the inverse of thc life expectancy estimate from Table I, and R is the average redemption rate for the state calculated from the Rolnick-Weber data. Note that this procedure does not actually solve the prohlem of using information unavailahle to agents. The statistics arc hased on the sample as a whole, and therefore make usc of (relative 10 a parlicular ye
Free Banking J1
194
KAliN:
VOL. 75 NO. .J
u.s. FRn: BANKING
they arc not guilty of straining the limits of the data. What emerges is that the losses were not overwhdming (with the ex.:eption of Minnesota), but they were, ex.:ept for the ca~e of New York, nonethdess certainly nonnegligible. III. Conclusions
The preceding st:ctions suggest that fret: banking kgislation often resulted in very high failure rates in those states rdative to failure rates in non-free-bank statt:s. This was particularly so in the initial yt:ars following tht: enactment of the kgislatioll, but remaint:d true to some extent throughout the period. Tht: high failure rates, wmbined with the frequent inability of faikd banks to pay note holders the par value of their nott:s (as dowmen ted by Rolni.:k- Weber) meant that the holding of free bank notes was a moderatdy costly (and risky) a.:tivity. Thus I would caution against Rolni.:k and Weber's rather sanguine view of the free banking experiment. On the other hand, it was certainly possibk for fret: banking tn be su.:.:essful. New York's bank failurt: rate was not much higher than those of non-free-bank states, it expaienced a large in.:n:ase in the number of banks in the state, and the riskiness of its bank notes was negligible. Nonetheless, the experien.:es of other states indicate that sudl SlKcess may have beell the ex.:eption rathn
8115
than the rule. Wis.:onsin fostered growth in its banking industry, but had wntinuing ditliculties with failures and losses 10 note holders. Indiana improved with regard to failures and losses, but ended up with many fewer banks than it had in the tirst year of the poli.:y. Minnesota's problel1\s with noteholder losses were espe.:ially severe. New Jersey probably had the best experience of the states examined in this paper other than New York, with most of its problems in the lirst few years. More importantly, tht: results presented herein show dearly that the rapid growth of a state'> banking infrastructure did not require free banking laws. Maine, alld Maryland to a lesser extent, experien.:ed sigIlilkant growth in the number of .:hartered banks, with less of the .:haos that a,:,:olllpanied su.:h growth in free banking states. The diHkulty here is that su.:h growth was subje.:t to the whims of state legislators, whn need not have been responsive to eC(lllllllli.: factors. REFERENCES Rolnick, Arlhur J. and Warren K Weber, .. New
Eviden.:e A lIleric'lIll
Oil
the
Fr.:e Banking
ha."
Ecollolllic Review, I )cce III ba 19H3, 73, pp. IOH091. U.S. Congress, /lUU.>" I!'x"cIlI;"t! I)U(·UII/('III.\, .. Re-
port of the Secretary of the Trea~lII'Y Oil the Condition of Ihe Siale Banks," various years, I H37 63.
[11] New Evidence on Free Banking in the United States By HUGH ROCKOFF* In their 19RJ paper. Arthur Rolnick and Warren Weher report a number of important calculation~ concerning antebellum free hanking in New York. Indiana. WisC()nsin. and Minnesolll. Their results clarify the dimcnsions of thc free hanking experiments and add further evidence that some portraits of the losses under antehellum free banking are exaggerated. But. while it is important to show that the actual losses were smaller than many have imilgined. the losses and failure rates reported hy Rolnick and Weber will not hy themselves prove entirely persuasive. Skeptics will point to a high overall failure rate of 15 percent. and to failure rates of 25 to more than 50 percent in the western states. as evidence that free hanking was unsuccessful. Part of the problem. I believe. is that Rolnick and Weher present these experiments as repeated trials of the same legislation. Here I show that if one looks carefully at the legal and historical circumstances in which each experiment took place. a case can he made that will satisfy even the skeptics who remain unconvinced by the losses and failure rates presented hy Rolnick and Weher. I illustrate this hy referring first to the experience in an additional state. Michigan. Rolnick and Weher mention Michigan in a footnote (p. 10R9). but their neglect of the particular circumstances attending these experiments leads them to an excessively optimistic view of what happened there. I hclieve the experience in Michigan is important hl'cause it influellced many of the traditional accounts. For example. the famous story ahollt kegs of nails with coins on top heing moved frnm hank to hank one step ahead of the hank examiners. that almost came to summarile antehellum free han king (which is included in the quote from Bray Hammond cited hy Rolnick- Weher. p. IOR3) is un"'Prnfrssor (If Econnmics. R1Itgers University, New Brlln,wkk. NJ OX!)O.I.
douhtedly a reference to Michigan. (See Hammond. 1957, p. 601.) Rolnick and Weher cite a Michigan hank commissioners' report (dated January I XVI) to the effect that most of the free hanks had suflicient assets to payoff at par. and go (In to suggest that traditional pictures of the losses in Michigan are overdrawn. Previolls generations of banking historians have heen familiar with the commissioners' report hut have interpreted it differently. Later material is hard to come by. One reason is that shortly after the commissioners' report was issued, Michigan abolished the bank commissioners' department, and transferred its duties to the attorney general, a fact that in itself tells us something ahout what was happening. There is. however, an old paper hy Alpheus Felch (1880) that gives a very different view from that suggested by Rolnick and Weher. Felch was writing long after the event. hut there is good reason to respect his views. He was in the state legislature when the free hanking law was passcd. and was one of f(lm to vote against it. He was. more importantly, one of the bank commissioners who authored the report cited hy Rolnick and Weher. lIe was also a member of the Supreme Court of Michigan in 1844 when the free hanking law was declared to be in connict with the state constitution. According to Felch. the total amount of currency left outstanding hy the free hanks was not "ascertainahle; hut it could not in all prohahility hl' less than (lIlC million of dollars; and this fell as a dead Inss Oil the cOlllmunity" (p. 123 ).1 Thc document that reich seems to relv (Ill thc most is the Allome,l' (ielle/'a/'s Rel'ort dated December 1839 (Michigan. 1840). This
'In my earlier ~tl1dv (1975, p. (7). I '!tcmpled 1<> create a safe upper bo;md eslimate nf Ihe Inial c·irc"lalion or the free hanks, hut the result \\'a,c; pr<'hahly f()o grc,t an IIpper hound In he hctpflli. In mv olhcr w
196 VOl.. 75 NO.4
Free Banking II ROCKOFF: U.S. FREE BANKING
docume:nt give:s us a picture of Michigan's banking syste:m de:ve:n months afte:r the bank cOllJmissione:rs' reporl. The Report lists 42 banks against which proceedings had begun for disolution of their charters (31 of them fre:e: banks), and bits of information on many of the:m provided by their re:ceivers. The allorney general had some information on 34 of the:se: banks on the basis of which he: estimate:d the:ir outstanding liabilities (mostly note:s) at $1,663,033. In only 14 case:s, however, was the receive:r willing to gue:ss the proportion of asse:ts that we:re: "good." For these: 14 case:s, good assets per dollar of liabilitie:s was 23.6 ce:nts; for the 12 free banks in this group it was 39.2 ct:nts. But both figures may have been overly optimistic in the: light of the on-going depression, and the e:vt:ntual court decision that made furtht:r legal rt:course impossible:. If we blow up the allornc:y general's estimate for 34 banks to represent his total list of 42, and use the figure of 23.6 ct:nts to represent what was eventually paid, then we are looking at losses on the order of $1.6 million. For the free banks alone, this reasoning would yield a lower figure, If we multiply the average liabilities for Ihe free banks in Ihe allorney general's list by 40, Ihe number of free ballks Felch thought got into operation, and assume they paid 39.2 cents on the dollar, we gel lotal losses of almost precisely one million dollars, a number similar to Felch. These figures are still only approximate. But il is clear that they strongly support Felch, suggesting that until further evidence comes to light, Felch's ve:nerable conclusion must stand as Ihe most likely presumption about the fate of the Michigan free: bank notes. The failure of the banks to redeem at par, however, does not necessarily imply that the transfers from the public to the bankers approache:d this magnitude. Many of the note:s may have entered circulation at a heavy discount-in Cincinnati, the notes of Michigan county banks were at a 50 percent discount by August 1838 and, by April 1839, Michigan notes could not be sold at all (Thomas Berry, 1943, p. 459). Later, some of the notes were bought by speculators at heavy discounts, who then brought suits against the
887
bankers for the face value of the: notes, so it is possible that in some: cases the:re: were net transfers from the: bankers to the public. Felch's use of the term "dead los~" may be simply an assertion that the note:s were not redeemed, rather than a reference: 10 u ne:t wealth transfer or loss in output, which is the construction one might be tempte:d to place on the phrase. In fact, it se:ems a re:asonuble conjecture that the: willingne:ss of the: Michigan Supreme Court to declare the law unconstitutional, a decision that broughl the process of suit and counte:rsuit to a halt, was inlluenced by a re:alization thai there: would be lillh: gain in fuirm:ss by further allempts to make the banks or the:ir borrowe:rs pay up. The Michigan experience:, moreOVl!r, was unique in se:veral ways. The fre:e: bunks we:re opened during a period of legal suspension of specie payments so that spe:cie was nel:essary only because: the le:gislation re4uire:d a certain amount to be on hand. This was the circumstance: thut allowe:d the promoters to shift specie from bank to bank to keep one step ahead of the: commissioners. III addition, the law permilled the banks to bal:k up their notes with mortgages on the still llirgely undeveloped land of MichigHll, an Hsscl dillicllli 10 appraise, and of dlluhtflll liquidity. Finally, il must not he f'lrgollen that Michigan ill 1840 was a frontie:r stale:, with a population slightly over 20(),OOO; it had be:en admilled to Ihe Union ollly a few years befor~ the fre:e banking e:xperimenl. The level of financial sophistication both in the general public and in the busine:ss community must have been fairly low. Later, in 1857, Michigan passed a second free banking law, and although the Civil War prevc:nted a real test, there: does not appear 10 have heen any repetition of the boom-and-bust cycle:. But circumstances had changed. For one thing, the second free banking law re4uired the notes to be backed by government bonds -federal or certain selected states, including, of course, Michigan-and Ihe law wus wrillen to force hanks in most circumsllll1ccs to back up each dollar in notes issued with more than a dollar's worth of securitie:s. The point is that to understand antehellum free banking it is necessary to e:xamine
Free Banking II 888
Till, A MliRICAN ECONOMIC REVlIiW
how the general idea was put into practice in each case. The central problem was how to insure confidence in the banking system while preserving the benefits of free entry and competition, the classic dilemma under any fractional reserve system. The mechanisms available for this purpose in the antebellum period were limited by institutional constraints such as the gold standard and by the vision of the legislators. The approach generally taken to reduce the risk of a panic, or restore confidence after one, was to tighten the secondary reserve requirement for notes. This. as we hnve seen. is what hnppened in Michignn. It nlso hnppclll'd in Indiana, where Ihe secondary reserve requirement was tightened in IRS'S by requiring (to simplify a sOl11cwhnt complex law) 110 percent in sec11Illiary reserves, when only 100 percenl was required under the initial law, and by taking into account the lower yield on Indiana sccurities, (see my 1975 study. Table 12, p. R4). In New York the law was amended to exclude nny honds except those of Ncw York. This grently incrensed the security of the note issue since New York state bonds enjoyed n strong mnrket independent of the hanks; although the eligihility of mortgnges remnined n prohlem for n time. Today, of course, we hnve alternntive systems f(;r preventing a loss of confidencedeposit insurance,
197 SI:PTT:MBER 19115
in detail, represented a sophisticated attcmpt to solve the problem of providing a competitive supply of banking services while maintaining the underlying stahility of the system. A second point illustrated by the Michigan example also applies more generally. It is one that I largely overlooked in my earlier work on antebellum free banking, and yet in retrospect seems obvious: the lack of financial sophistication in the communities struck by the worst experiences. Free banking experiments were less successful in Michigan, Indiana, and Minnesota than in Ohio, New York, and I.ouisiana fill' a variety of reasons. Part of thc story l1lust he that the fmmer states were all frontier statcs wherc much of the business community had little cxperience in evalualing the polential dcmand for banking services, and whcre the public had little experience in evaluating the safety of alternative investments. The market tended to discourage the formation of unsound institutions, and eliminate them after they were formed, but it did not do it as smoothly as in established communities. Clearly. if we want to draw a lesson about the wisdom of free entry in modern circumstances, the experi. ence of the developed slates rather than the frontier states is what is relevant. Finally, it is incumbent on the defender of antehellum free banking to show th<1t free entry produced some of the benefils nor· mally anticipated from it. Failure rales are only one dimension of the problem. I tried In do this in my earlier work, citing some corn· parative rates of return in banking, and re· laled bits of evidence. suggesting that free entry improved the allocation of bank capital (197S, pp. SO-65). The best I could do. however, was to outline the case. Much remains to be done. III short, I believe the results presented hy Rolnick and Weber are important. hut read· ers of their paper should not be left with the impression that the case for
198
Free Banking 11
VOL. 75 NO.4
889
ROCKOFF: U.S. FREE BANKING
antebellum free banking even to someone who is distressed by the breakdowns that occurred in Michigan, Indiana, Minnesota, and certain other states.
Hammond, Bray, Banks and Politics in America: From the Revoilltion to the CifJiI lVar,
Princeton: 1957.
Princeton
University
Press,
Rockolf, Hugh, "The Free Banking Era: A
REFERENCES Berry, Thomas Senior, Western Prices Before
1861: A Stl/dy of the Cincinnati Market, Cambridge: Harvard University Press, 1943 . •'elch, Alpheus, .. Early Banks and Banking in Michigan," Michigan Pioneer Collections 2, 1880, 111-29. Friedman, Milton, A Program for Monetary Stability, New York: Fordham University Press, 1959.
Reexamination," JOllfllal of Money, Credit and Banking, May 1974, 6, 141-67. ___ , "The Free Banking Era: A Reexamination," in [)i~·serwti{)ns ill AlIlerimll Economic History (rev. doctoral dissertation, University of Chicago, 1972), New York: Arno Press, 1975. Rolnick, Arthur J. and Webt!r, Warren K, .. New Evidence on the Free Banking Era," American Ecollomic Review, December 1983, 73, 1080-91. Michigan, AI/oTtley General's Report, Senate Document 6, Annual Ses~ion, 1840.
Part III Other Experiences
[12] RICHARD H. TIMBERLAKE
Private Production of Scrip-Money in the Isolated Community "In the company town, or mining camp, . . . United States coin and currency were not in good supply. . . . During the heyday of the old company town, scrip circulated more freely than U.S. currency and was indeed the coin of the realm . . . . Eleanor Roosevelt . . . in the mid-thirties, during [one of] her humanitarian cruslides, attacked the use of scrip by coal mining companies as a very evil thing . . . . Although many mourn the days of a bustling and active coal economy, lillie can be sliid to support the . . . issuance of scrip" (Truman L. Sayre, "Southern West Virginia Coal Company Scrip," in Trade Token Topics, reprinted in Scrip, Brown, 197!l, pp. 343-344).
I. TilE POSSIBILITY OF FREE MARKET MONEY EVER SINCE TilE AIIOLITION OF TIlE OPERATIONAL GOLD
in the early 1930s, the federal government through its agent, the Federal Reserve System, has been almost the sole creator of the monetary base, and has also been the licensing agent for the banks that create most of the demand deposits used in the United States. No money of any significant amount can be created today without some sanction or act of the Federal Reserve System. This condition has encouraged the notion that government is a necessary, or at least desirable, regulator of any monetary system-that without government inSTANDARD
Th~ author is indebted for support and suggestions to the sponsors and participants of the Manhallan Institute Monetary Conference of 1986, especially to David Glasner and Anna SchwartL. My culkague, Price Fishback, and Milton Friedman also made valuable suggestions, as did /fuston McCullllCh and two referees for this journal.
RI(,HARD
H.
TIMIIERI.AKE
is professor o/economics, Ulli,'erl'ily .of Geurgia.
jUl/fllal (~l MUlle)" ere/iii, and Ballking, Vol. 19, No.4 (November Copyright ~ 19~7 by the Ohio State University Press
19~7)
Free Banking II
202 438
:
MONEY. CREDIT. AND BANKING
volvelllent any monetary system quickly degenerates into "chaos. "If this supposition were valid. the evolution of money could hardly have occurred. The barter system that preceded early monetary systems. in which government had no part. would not have heen superseded if the resulting monetary systems were destined to be chaotic. This logic suggests the possibility and perhnps the feasihility of a llongovefllment1l1(llley. lIowever. the IlI'm;tienl efficacy of such 1I systellll'lIllllot be deduced from n theory that merely suggests its possibility. but must be sought from historicnl evidence of monetary arrangements that havc developcd spontaneously in the private sector. This paper examines one such incidence of private money creation--the issue and use of scrip. which occurred primarily in the isolated economic environments of mining and lumbering company towns during the first half of the twentieth century. Fortunately, numismatic collections and records reflect the operational character of the scrip systems in these communities so that some evaluation of their monetary properties is possible. Much of the recent research on the creation of private money has focused on that issued by private banks in the presence of a dominant legal money such as gold. (White 19!14. Sylla 1976. Rolnick and Weber 1982). The issue of scrip. however. had nothing to do with banks. It was issued by private mining nnd lumbering enterprises. While it. too. was redeemable in a dominant money. its issue and acceptance were not critically dependent on any dominant money. For this reason. the phenomenon of scrip issue is e!>pecially revealing.
2. LEGAL RESTRAINTS AGAINST THE ISSUE OF PRIVATE MONEY
Proscriptions against the arbitrary or casual issue of money appeared at the very beginning of this country's political formation. First. the Constitution stated: "No state shall . . . emit bills of credit, [or] make nnything but gold nnd silver coin a tender in payment of debt" (U.S. Constitution. Art. I. Sect. I0). No money except gold and silver was to be the legal tender issue of any governmental unit. Money to be money, however, does not have to be /ega/tender. It cnn be whnt one might call common tender, i.e .• commonly accepted in payment of deht without coercion through legal means. Indeed. privately issued money to exist nt all would have had to be common tender. and would have hnd to earn its ncceptability in a market environment. Even though the states and Congress were constrained to monetiling only gold and silver. the general laws of contract and commercial instruments sanctioned the nppearance of moneys i!>sued by privately owned commercial banks (liurst 1973). In addition. "Nothing in the Con!>titution barred private manufacture of coin. and through the first half of the nineteenth century Congress did not nct against private coinage. . . . General contract law allowed any contractor to issue his notes and coins and circulate them so far as the market would take them" (Hurst 1973).
Free Banking II RICHARD H. TIMBERLAKE
203 : 439
Free enterprise in the issue of common tender money was accidentally encouraged in practice by the federal government's ineptness in establishing a useful denominational spectrum of fractional currency during much of the nineteenth century (Carothers 1967). Private transportation companies-canals, turnpike companies, and railroads-issued significant amounts of such currency between IX20 and IX75. Municipal and state governments did likewise. Redemption of transportation currency when called for was in services rendered, while state and local government currency was redeemed as tax payments (Timberlake 19X I). The paucity of government-issued fractional currency was catastrophically aggravated by the first issues of greenbacks during the Civil War. The metallic v,liues of subsidiary coins rose rapidly above their monetary values in the summer of I X62. and the coins disappeared from circulation. These circumstances provoked not only the ill-conceived issue of postage stamp currency, but also extensive private issues of minor coin (Carothers 1967, Faulkner 190 I). The act that authorized postage stamps as currency in 1862 also outlawed the private issue of notes, memoranda, tokens, or other obligations "for a less sum than one dollar intended to circulate as money or to be received or used in lieu of lawful money of the United States" (Act of Congress, 12 Statutes at Large, 592. July 17, 1862). Then in 1864. even the private issue of gold and silver coin was forbidden, again. "when the coins were intended for use as current money" (Hurst 1973).
3. TilE APPEARANCE OF SCRIP AS AN ECONOMIZING MEDIUM
The lack of adequate denominations in government-produced money was not the only factor that stimulated the private production of money. Shortly after fractional coinage was stabilized around 1885, coal mining and lumbering became major industries. Both coal mining and lumbering enterprises had to be organized in the vicinity of the contributory resources, so were often located in isolated areas with low popUlation densities significantly distant from commercial centers. Coal-producing regions were hilly or mountainous areas where agriculture had been marginal and other commercial development had lagged. "The 'Main Street,' "noted one observer in describing a coal mining community "was often railroad tracks" (Brown 1978). Coal mining entrepreneurs, therefore, had unique problems to contend with in organizing their enterprises. Their common problem was what is known today as a lack of infr:astructure-· no strel!ts, no churches, no schools, no residl!nces, no utilities, and no banks or financial intermediaries. The specialized industries that might otherwise have provided these services were dissuaded from doing so by the high start-up costs "nd the enduring uncertainties of dealing with low-income communities that might be there today and gone tomorrow. Alternatively, the coal mining comp"nies could deal with such conditions because they were in a better strategic position to change uncalculable uncertainties into calculable risks. (Fishback I'JX6, Johnson 1952). Mining companies, therefore, built residences, churches, schools, and water works, and opened company stores or commissaries. In so
204
Free Banking II 440
:
MONEY, CREDIT, AND BANKING
doing, they became both buyers of labor from, and sellers of commodities to, the coal miners and their households. This kind of organi7.ation invited an economy in the community's payments system-the use of scrip in lieu of ordinary money. "Scrip" has become a generic term for the issue of a localized medium of exchange that is redeemable for goods or services sold by the issuer. Originally printed cards or "scraps" of paper. scrip evolved into metallic tokens with many of the physical attributes of official coins. Indeed. scrip in the very beginning was more in the nature of a trade credit. or demand deposit. at the single local general store. Ledger credit scrip. however, gave way to scrip coupon hooks. which "eliminated the tedious bookkeeping chores that were incident to over:"thccountcr credit (day book or journal entries followed by ledger entries)" (Brown 1978). The use of scrip not only implied an issuer-the mining company-and a demander-the miner, it also required a supplying industry. The institutions that supplied coupon scrip were companies already in business printing tickets. tokens, and metal tags for various other kinds of enterprise. They advertised extensively in mining catalogues during the first half of the twentieth century touting the advantages of their own scrip systems. The Allison Company of Indianapolis, for example. noted that when one of its coupon books was issued to an employee. "He signs for it on the form provided on the first leaf of the book, which the storekeeper tears out and retains for the [company] time-keeper, who deducts the amount from the man's next time-check." Then, when the employee buys goods from the company store, "he pays in coupons, just as he would pay in cash. and the coupons are kept and counted the same as cash . . . . The coupon hook is a medium of exchange between the company employees and the company store" (from 1916 Mining Catalog, Brown 1978). Other scrip-producing ticket companies emphasi7.ed the safety of the scrip coupon system in coal mining communities "where little or no police protection is afforded" (adv. of the International Ticket Co. in the Keystone Catalog of 1925, Brown 1978). The Arcus Ticket Company of Chicago advertised a list of advantagcs of scrip to both the employer and employee, one of which for the employer was the fostcring of employee good-will by avoiding misunderstandings on chargc accounts (sic). The advantages to the employee included keeping the" 'head of the house' better informed as to the purchases made by his family from day to day . . . . This frequently puts a check to extravagance and debt" (Keystone Catalog. 1925 in Brown 1978). Local scrip of this type was very similar to modern day travelers checks. The costs of travelers checks were also the costs of coupon scrip: each unit could be used only once. It had to be signed out when it was issued and signed when it was spcnt (nrown 1978).1 IThi~ comparison must he qualified. Many travelers checks. as well as other II.S. currency. are currently lIsed as hand-to-hand media in foreign markets. Sometimes travelers checks rctllrn from ahrnad with more than a dOlen endorscmcnts on thcm. They are called "checks". hut likc food "stamps". they are a quasi currcncy.
Free Banking II RICHARD H. TIMBERLAKE
205 : 441
The transactions costs of coupon scrip eventually encouraged the increased use of metal scrip. This medium became cheaper overall than coupon scrip, in spite of metal's higher initial cost, largely due to the invention and development of the cash register after 1880. Pantographic machines also were instrumental in reducing the unit costs of metal tokens (Brown 1978). Instead of rccdving cash, the scrip-issuing "cash registers" paid out metaltokens, made a record of the pay-out and to whom it had gone, and kept a grand total of the amount issued. The scrip registers would eject a specified "dollar" amount of scrip when a lever like that on a slot machine was pulled. In a 1'.l27 advertisement, the Osborne Register Company (ORCa) of Cincinnati pictured a IO-year-old child who, in a demonstration, issued $600 worth of metal scrip in various amounts to 200 hypothetical employees in 55 minutes, implying an average emission of $3 per employee every 16.5 seconds (Brown 1978).
4. THE POSITIVE-SUM BENEFITS OF SCRIP
The economics of scrip issue, as with all exchange between economic agents, required that both the issuer (the coal mining company) and the acceptor (the employee) benefit from the transaction. The company necessarily had contact with the outside world. It bought machinery and other resources and sold coal in a national market. All these activities required the use of standard money. Scrip was used essentially as a working balance of money with which the coal operator could make advances to his impecunious employees between paydays. It was issued at the request of the miner to the extent of the wages he had already earned, and it was redeemable in standard money on the next payday. The amounts were usually small-five or ten dollars, or even less. To the worker it amounted to an interest-free, small-sum loan that he could get with almost no effort. It enabled him to buy ordinary household goods at the company store. To those workers who had "gone out and got drunk" on the previous weekend, or who had suffered some kind of household emergency, scrip was a blessing only measurable by the cost of its common alternative (Clark 1980, Johnson 1952). Its alternative in a conventional urban setting without scrip was the pawn shop, loan shark, or installment peddler (Johnson 1952). An industrial worker in the same unfortunate position in, say, Detroit, Pittsburgh, or Chicago, had access to money between paydays only by borrowing against his household capital at a pawn shop where he paid exorbitant interest rates ifhe reclaimed his pawned goods. The scrip system could be abused in such a way that a discount would also appear in some scrip transactions. Since the company store did not sell liquorfor the obvious reason that its sale would encourage absenteeism and worker inefficiency, workers would at times obtain scrip from the company clerk and sell it for conventional currency in order to buy liquor. The bootlegger (during Prohibition) or other liquor vendor, whose shop was not likely in the neighborhood
Free Banking II
206 442
:
MONEY. CREDIT. AND BANKING
of the company store, faced significant costs in redeeming the scrip for conventional money. thus giving rise to a discount (Brown 1978. Caldwell 19(9).2 In spite of the obvious advantages of the scrip system to both worker and mine owner. scrip, the company store, and the company town have been universally bemeaned (Brown 1978). The accounts of their operations include contradictions that appear sometimes in the same paragraph. (For example, sec quote of Sayre used as an epigraph. p. I. Brown 1978.) All accounts. while critical of the scrip system. acknowledge. first. that it was issued at the behest of the miner; second, that its issue cost the miner nothing; and, third, that it was r~deemable in standard money on payday. The dogma of scrip's critics was that the company store, in which the scrip had to be spent, raised prices to monopolistic levels and there~ by exploited the defenseless miner (Dodrill 1971). Fishback's and Johnson's studies of prices in company stores versus those in independent stores refute this popular prejudice. Prices were four to seven percent higher, but so were costs. (Fishback 1986; Johnson 1952). The advantage of scrip issue to the mine operator was that it was one worker perquisite he could offer to attract labor into a somewhat unattractive environment. He already offered housing and mercantile services; by issuing scrip against future wages he also provided commercial credit with virtually no interest charges to the borrowers (Johnson 1952). The practice, indeed, was so widespread that it can only be viewed as a traditional perquisite of the trade. A company that did not offer the scrip privilege would have been at a competitive disadvantage. The mine operator thus became a quasi banker. His cost for metal scrip during the 1920s varied from slightly less than I cent to 5 cents a unit for scrip tokem of simple design made in aluminum. In brass or nickel silver and with scalloped edges and more intricate designs, costs could run as high as II cents a piece. (A II these values are unit costs in thousand-unit lots. and are from advertisements of several different scrip manufacturers between 1925 and 1940, in Brown 1978). Scrip sales information from the Ingle Company sales journal of 1928 reveals that the average denomination issued was about $.25 (Brown 1978). Since the average cost per token was only about J cents and could have been even less. an investment by the coal company bank in, say, 5,000 pieces cost it about $150 for the scrip coin, and perhaps $100 more for a scrip-issuing machine. To carry out this same banking function with regular U.S. currency would have required an investment in cash alone of $1,250, as well as substantially greater security costs to protect the money. One observer noted. "The mining company could pay almost its entire payroll in company scrip, disturbing only a few dolla rs of actual working capital" (Sayre, in Brown 1978). Of course, paying out scrip gave workers some additional claims on the working capital oflhe company stores. So
2Scrip wa~ frequently advertised a~ redeemahle only to the worker to whol11 it was originally i~~ued. This condition applied in ~ome mines. "owever. for metallic ~erip. it could hardly have heen enforced. and would have detracted from the utility of any scrip if it were enforced.
207
Free Banking II RICHARD H. TIMBERLAKE
: 443
the monetary economy of using scrip was in part offset by higher costs of merchandising goods.) The difference between the payment system costs of scrip and of real money was a form of seigniorage revenue the coal mine operator realized and shared with his employees. They received interest-free loans; he was able to offer a fringe benefit that tended to reduce what would have been a higher working capital re4uirement. While scrip was usually specialized to one company in a particular community, many coal mining companies had mines in different regions. Their scrip was good in all the different locations where their mines operated. As the scrip-using communities gradually came to experience more extensive commercial relations with each other, their localized scrips became interchangeable. Even some independent stores accepted coal company scrip (Brown 1978). Given the proscriptions against the private printing or coining of money by the Acts of 1862 and 1864, one may wonder how scrip could have been issued and used legally. The key is the word "intended" in the proscriptive laws. The courts ruled that scrip was not intended to circulate as money: first, because it was redeemable only in merchandise until payday; and, second, because it resembled money only superficially and was clearly distinguishable from standard money. (The coin under the court's scrutiny was a 50-cent token, but weighed only onefifth as much as a standard 50-cent piece.) Any token that was redeemable in lawful money on demone/was construed to be illegal, and whether the token in question was coin or pasteboard did not matter (Brown 1978). 5. TIlE ENVIRONMENTS IN WHICH SCRIP APPEARED
The extent of scrip use has many dimensions-temporal, geographical, and industrial. Its most notable occurrence in the twentieth century was in the coal mining regions of West Virginia, in part because the state government passed a "wide open" scrip law some time before 1925. However, it was extensively used in other states as well. The Tennessee Coal Iron and Railway Company, for example, ordered 547,500 pieces between 1933 and 1937 from the Ingle-Schierloh Company of Dayton, Ohio (Brown 1978). Another source lists 20,000 coal company stores in the United States, Canada, and Mexico all of which used scrip between 1903 and 1958 (Dodrill 1971). Numismatic records indicate that scrip was also used extensively in several other industries-fishing canneries, agriculture (to pay crop-pickers), fruit canneries, logging and lumbering companies, and paper companies (Brown 1978, Trantow 197H. Trantow's index lists over 1,100 companies that issued scrip currency in 40 states). One scrip numismatist cites a Chicago newspaper of 1845 that regularly 4uoted the discounted prices of coal scrip, city scrip, canal scrip, railII am indebted to Huston McCulloch for this observation.
Free Banking II
208 444
:
MONEY. CREDIT. AND BANKING
road scrip. Michigan scrip. Indiana State scrip. and Indiana land scrip. as well as the notes of private and chartered banks. Private businesses issuing such scrip numbered in the thousands (Harper 1948). Furthermore. as Brown observed. "The use of paper scrip was much wider than the use of [coin] scrip . . . [but] only a comparatively small amount [of the paper] has survived." Therefore. the extent of scrip use must have been much greater than the vestiges in metallic collections would indicate (see also Caldwell 1969). Just as Brown in his work seemed unaware of scrip that had preceded the issues bycoal companies. Harper in his study of Scrip and Other Forms of Local Money thought that intensive use of scrip only appeared in the United States during the dcpression years. 1932-1935. \lis research uncovered several sources of "depression" scrip: (I) issucs by local governments due to decreases in tax revenues; (2) issues by chambers of commerce after local bank failures as a means of "corralling as large a proportion of the depression diminished volume of business as possible for their membership"; (3) issues by "home-owned stores as a weapon against . . . chain-store competition"; (4) issues by "barter groups as a means by which the unemployed could more conveniently exchange services"; and (5) issues by charitable organizations to needy persons as "commodity orders" for foodstuffs. "Local money in some form." he concluded. "is likely to recur in response to a public demand under substantially similar circumstances." Most of this "depression" scrip had appeared in earlier times-for example. municipal scrip that was redeemable as tax payments. The depression scrip. however. was usually linked to a dated stamp scheme that required the holder to fix low denomination (2- or 3-cent) stamps to the scrip at specified times. The stamps were to provide the revenue to redeem the scrip and to encourage spending. but they added an undesirable burden that greatly reduced the efficacy of the scrip's use. They also detracted from the scrip's effectiveness as an addition to the existing stock of ordinary money (Harper 1948).
6. IMPLICATIONS OF THE SCRIP EPISODE
The phenomenology of scrip issue has significant implications. First. no one had any incentive to leave scrip behind for monetary researchers to count or to analyze. Demanders of such currency would not regard it as a store of value for any time longer than the period between paydays. Suppliers. to whom the scrip was an outstanding demand obligation. would redeem it first if they liquidated. merged. or closed down their enterprises. In addition. everyone who used it and benefited from it was aware of its questionable legality. Archival records of its outstanding quantities. therefore. are almost nonexistent (Timberlake 1981). Scrip's unrecorded existence is emphasized as well by the research that has uncovered its former use. Each scholar who has unearthed one of the diverse scrip appearances has treated the phenomenon as unique. and with good reason. Each one was widely separated in time. place. and circumstance from the others. Yet. each one had characteristics similar to the others. All episodes combined
209
Free Banking II RICHARD H. TIMBERLAKE
: 445
emphasize the feasibility of the spontaneous production of money in the private sector. The coal mining scrip episode adds significantly to the total scrip experience for a number of reasons. First, it lasted for over 50 years, so it was not just a temporary happenstance. Second, it appeared in a wide range of independent communities. In West Virginia alone, almost 900 coal mining companies employing about 120,000 miners issued scrip in one form or another. I n other areas of Appalachia-southern Virginia, eastern Kentucky, eastern Tennessee and southwestern Pennsylvania-the experience was similar. Third, scrip's tenure was not dependent on the previous existence of standard legal tender money. True, the coal company was bound to redeem the scrip on payday, but this guarantee was only a flourish that enabled scrip issuers to avoid violating the proscriptive laws against the issue of private moneys. As it was, many children living in coal mining communities did not see a dollar of "real" money until they grew up and left the area (Caldwell 1969). The self-sustaining nature of the scrip system without recourse to standard money, stemmed from the fact that both the demander and supplier of scrip were active participants in both the labor market and the household goods market at the company store. This intimacy in two markets by both participants enabled them to evaluate wages paid and received in real terms, that is, by the quantity of household goods that the scrip wages could purchase. A decline in the purchasing power of scrip at the company store would simply have indicated to the miner that the real value of his services to the company had declined. He thereupon would have moved to another location or occupation. lfthedecline in real wages was due to an industrial depression or the competitive decline of the coal industry, as occurred simultaneously in the 1930s, both mine workers and mine operators would realize reduced real returns in the mode of any resource owners under similar circumstances. A fourth important result of the scrip system was its reflective emphasis on the returns to the capital structure of the payments system. I n the scrip system the money was supplied endogenously: the coal company banks, the borrowing miners, and the scrip suppliers were all parts of an economy of private ownership. Scrip money was not dependent on any outside money, but was produced under the same conditions and incentives as any common commodity. The mining companies rather than the workers produced the scrip because in working without wages until payday, the workers were implicitly extending credit to the company. Scrip issue was a means of clearing this debt before the regular payday. In addition, the coal mining company had the collateral value of the mined coal to secure the "loan. "4 110th the companies and the workers realized the seigniorage returns from its existence. While the scrip system was small-scale and had a low profile, the government could ignore it because it posed no threat to the government's monopoly over the production of money. However, if scrip issue had shown any tendency to 'I am indcbl<:d to Huston McCulloch for suggesting these details.
210
Free Banking II 446
:
MONEY. CREDIT. AND
BANKING
hecome a national practice. the proscriptive laws against private coinage would surely have been interpreted and enforced much more rigorously.~ An observer of the scrip system might conjecture that the experience of the isolated communities could have ramified into an intercommunity system using some kind of scrip clearinghouses (Le .• scrip banks) if the laws restraining the private issue of money had not existed. Over time, technological and organizational developments could have led to economics of scale and enterprise. Prohahly as few as thrce or four or as many as two d07.en issuers of scrip money might have appeared. Some of the minters of scrip -Inglc-Schierloh. Oshorne. Insuranl'e Credit. Adams, Donnan, and others· would have expanded their enterprises to include management of intercommunity scrip systems and ultimately their probable evolution into credit card systems. Such an extension offunetion would have been analogous to automobile dealers expanding into the car leasing business-a sort of horizontal integration to reap certain economies of scale. Had the scrip system become intercommunal and given rise to scrip-ondeposit in scrip banks necessitating bank reserves and clearing operations, some high-powered scrip into which local scrips could be converted would probahly have appeared. The experience of the ages seems to confirm this evolution (Friedman and Schwartz 1986). Less clear is why the high-powered money has to he issued or regulated by the state. The question of whether or not the market system could, alternatively, produce a private monetary base that wOllld prove to be both stable and serviceable has not heen attempted or allowed, and will remain unimaginable until a general belief in market efficacy becomcs pervasivc. That time as yet seems nowhere near. h I.ITERATURE CITED
Brown, Stuart E., Jr. Scrip. Berryville, Va.: Virginia Book Company. 197R. Caldwell, Walter. Coal Compall)' Scrip. Montgomery, W. Va.: War Printing Co .. 19ft
~In a thought-provoking paper. David Glasner argues convincingly that governmcnlnl assumption of a monopoly role over money enahled governments toenhanee their liscal powcrs. partit'ulmly during war emergencies (Glasner. "Economic Evolution and Monetary Reform."cspecially thc section: "A Rationale for Government Monopoly over Money. ") In short. not only is sci!!niorage an important revenue to the state. but capital expropriation through debasement ofmoney's function as a unit of account may be even more lucrative. ""owcver. the commercial bank clearinghouse system in the United States durin!! the secllml half of the ninetecnth century is an example of a private lender of last resort that produced hase money efficiently at critical times (Timberlake 1984).
211
Free Banking II RICHARD H. TIMBERLAKE
: 447
Fishback, Price. "Did Miners Owe Their Souls to the Company Store'! Theory and Evidence from the Early 1900's." JOllrnal of Economic History 46 (December 1986), 1011-29. Fric:dman, Milton, and Anna J. Schwarz. "lias Government Any Role in Money'!" JOllr11(/1 (!l Monetary t."cunumi('~' (January 1986), 37-62. Glasller, David. "Economic Evolution and MOlletary Reform," (unpublished). Harper, Joel W. Scrip anel Other Forms of Lucal MOlle.v. Ph.D. dissertation, University of Chicago, 1948. Ilurst, James Willard. A Legal HislOr.v of MOlley in the Ulliteel Stutes, 1774- 1970. Lincoln: University of Nebraska Press, 1973. Johnson, Ole S. The In elliS trial Store, Its History, Operations anel Economic Significallce. Atlanta: Foote and Davies, 1952. Rolnick, Arthur J., and Weber, Warren E. "Free Banking, Wildcat Banking, and Shinplasters." Federal Reserve Bank of Minneapolis Qllarter(I' Review 6 (Fall 1982), 10-19. Sylla, Richard. "Forgotten Men of Money: Private Bankers in Early U.S. History.... Journa/ uf Emnomic History 36 (March 1976), 173-88. Timberlake, Richard H. "The Significance of Unaccounted Currencies." Journal of Econumic Jlistory 41 (December 1981), 853-66. ___ . "The Central Banking Role of Clearinghouse Associations." Journal of Money, Credit, al/l/ Banking 16 (February 1984), 1-15. Trantow, Terry N. Catalogue of Lumber Company Store Tokens. Ellensburg, Wash.: Trantow, 1978. White, Lawrence H. Free Banking in Britain. Cambridge: Cambridge University Press, 19M.
[13] TYLER COWEN RANDALL KROSZNER
Scottish Banking before 1845: A Model for Laissez-Faire? EcONOMISTS /lAVE BEGUN TO DEVOTE increasing attention to what the banking and financial system would look like if trends toward laissez-faire, through both deliberate deregulation and financial innovation, continue. One group of authors, who may be labeled the modern free banking school (e.g., White 1984a and b, O'Driscoll 1986, Selgin and White 1987), focus upon the experience of eighteenth and early nineteenth century banking in Scotland as the primary illustration of how an unregulated banking system would operate. For these writers, the Scottish system provides the paradigmatic case of the unhampered evolution of a financial system. In this view, laissez-faire would create a system of banking based upon short-term nominally fixed debt liabilities redeemable in some form of outside (usually commodity) money. There would be a sharply defined money supply. Banks would not be simply pure intermediaries; they would issue non-interestbearing notes on a fractional reserve basis and provide other transactional services. The "free banking" perspective contrasts with the analysis of the "legal restrictions theory of money" and the "new monetary economics," (LRT -NME) (Black 1970, Fama 1980, Hall 1982, Greenfield and Yeager 1983, Wallace 1983, and Cowen and Kroszner 1987a and b, 1988). In this view, our monetary institutions, e.g., the fIXed nominal value, debt-based nature of banking claims, have been shaped by legal
The authors thank Maxwell Fry, Daniel Klein, Lars Jonung, J. Huston McCulloch, Lawrence White, and an anonymous referee for helpful comments. TYLER CoWEN
is assistant professor of economics. University of California. irvine. is a Ph.D. candidate ill the department of economics, Harvard
RANDALL KROSZNER
University. JOllrnal of Money, Credit, and Bunking, Vol. 21, No.2 (May 1989) Copyright e 1989 by the Ohio State University Press
Free Banking II 222
:
MONEY, CREDIT, AND BANKING
restrictions affecting private financial intermediation. Under laissez-faire, the money supply would no longer be well defined, and exchange media could assume such forms as equity shares, similar to today's money market mutual funds, and smaIldenomination interest-bearing assets. Banks would be little more than pure financial intermediaries offering diverse portfolios against which "depositors" would hold claims, much like transferable mutual fund shares (Fama 1980). These assetclaims need not be convertible into an "outside" money such as gold but instead may pay interest or fluctuate in capital value. In this way, it may be possible that the transactions media would not bear a fixed relationship with the unit of account, in contrast with the current system in which Federal Reserve Notes are denominated in dollars (see Cowen and Kros7.ner 1987b and 1988). Without legal restrictions, "it is hard to sec why anyone would hold nOIl-interestbearing currency instead of interest-bearing securities" of similar default risk and negotiability (Wallace 1983, p. I). The demand for portfolio-dominated assetstoday, Federal Reserve Notes-arises because legal barriers restrict the use of other assets for transactions and settlement purposes. If entry into the financial intermediation industry were free, then private firms could "break up"large-denomination Treasury securities into small-denomination bearer bonds to suit transactors' wishes. Competition among intermediaries would ensure that the interest rate paid on the small-denomination bonds equals the rate on large-denomination bonds minus the cost of intermediation. Wallace (1983, p. 4) concludes that under laissezfaire either nominal interest rates would be driven down to the costs of denomination arbitrage, thereby eliminating the wedge between the returns on bonds and non-interest-bearing money, or existing government currency would become worthless since interest-bearing media would portfolio-dominate it. This approach has been criticized for its apparent conflict with the historical record of monetary and financial institutions. White (1984b, p. 707) questions whether a payments system could evolve in which there is no convertibility to an outside money, and equity instruments are used for transactions purposes by citing "the historical fact ... that deposit banking did not naturally grow up upon an equity basis." In addition, White (1987, pp. 450-51) uses the eighteenth and nineteenth century Scottish banking experience to test the predictions of the legal restrictions theory, arguing that the coexistence of non-interest-yielding paper currency with interest-yielding bearer assets during this ''free banking" era cannot be explained by legal restrictions. TIuoughout this period of Scottish history, however, there were a number of legal restrictions affecting the development of the financial system. The Act of 1765 raised impediments to interest-bearing notes and proscribed small denomination notes. The prevailing system of liability and incorporation raised barriers to entry into banking and hindered the development of the equity market. Furthermore, the Bank of England acted as a shadow central bank for Scotland, limiting the autonomy of the Scottish bankers and undermining forces leading away from debt-based fractional reserve banking. These deviations from laisse7.-faire bring into question the appropriateness of Scottish banking as either a model for unregulated payments
213
214
Free Banking II TYLER COWEN AND RANDALL KROSZNER
223
systems or a counterexample to the predictions of the LRT-NME approach.
THE ACT OF 1765
A. Imerest Payment on Currency and the Structure of Bank liabilities: V,e Option Clause The displacement of non-interest-bearing "warehouse receipts" for specie by interest-bearing instruments was thwarted by the prohibition of the option clause in 1765. The option clause gave bank directors the right temporarily to refuse note redemption in specie (see Cameron 1967, pp. 68-69 and Checkland 1975, pp. 67-68). If the bank chose to delay redemption, the bank obligated itself to pay a specified rate of interest on the note until redemption resumed. Typically, the clause would permit a delay of six months and pay a 5 percent (annual) iillerest premium. i The issuers had developed this clause to insulate themselves from short-term disturbances and note redemption "attacks" by rival issuers. Following the Act, all banknotes were required to be redeemable for specie upon demand. 2 Before 1765, the noteholders of banks that invoked the option clause earned explicit interest on their notes. Notes containing thl~ option clause, even if not paying explicit interest, may also be considered an interest-bearing medium. The bank could decide to suspend convertibility and pay interest at its discretion, an event individual noteholders would consider exogenous. The holder of an option clause note thus held a lottery ticket that paid its face value in gold in one state of the world and interest with delayed gold redeemability in the other. Addressing objections to LRT-NME, McCulloch (1986) has suggested that money could bear probabilistic interest, in order to avoid costly calculation of interest on each note. Note issuers could hold lotteries, rewarding those with the chosen banknote serial numbers with lump-sum cash prizes.] In contrast, the form of interest contained in the option clause makes the onset of pecuniary interest payments stochastic, rather than offering a steady stream of ex Ollte expected returns. Once the option clause is invoked, interest during the option mode is nonstochastic. If the competitive forces postulated by LRT -N M E had not been in place, banks could have suspended convertibility without offering any offsetting payments. As noted by Gorton (1985), uncompensated convertibility suspensions were the standard practice in other (less competitive) banking environments (see also Cameron 1967). While the evidence is scant, bank notes containing the option clause do not appear to have circulated at a discount with respect to other Scottish "conversion on 'The Usury Law capped interest rates at5 percent. Powell (1915, pp. 1311-39,246, 329)examines the bindingness of this ceiling. 2 Interest-bearing currency also arose during New York's nineteenth century "'free banking" era, but was ended by legislation in 1830 (King 1983, p. 155). Goodhart (1985, pp. 44, 53-55) provides other instances of interest-bearing currency. I An anonymous referee has pointed out that social and legal sanctions against gambling may have prevented the implementation of a McCulloch-like solution in Scotland during this period.
Free Banking II 224
:
MONEY. (,REDIT. AND IIANKING
demand" notes (Cameron 1967, p. 69 and Hamilton 1956, p. 407).4 Gradually, prior to 1765, the majority of notes came to include this provision. The added insurance against runs and insolvency apparently compensated for the temporary inconveniences, if any, of uncertainty over convertibility into outside money. S The option clause mitigated the instability caused by first-come, first-serve rules for payouts to depositors demanding specie. Diamond and Dybvig (1983) have shown that random noise can cause bank run "bubbles" under "me-first" structures. By providing for orderly suspensions with an interest "bonus" for not being first in line, instability due to runs was reduced. Similarly, the option clause also decreased the likelihood of sudden large loan calls and the attendant costs imposed on bank borrowers (see Bernanke 1983). Instead of allowing banks to use the option clause to protect credit relationships, the Act of 1765 constrained banks to transmit portfolio losses to shareholders and noteholders through the disruptive means of bank runs and institutional liquidation. This constraint on the Scottish system may have helped to create a role for a "lender of last resort" central banker that the Bank of England fulfilled (Rockoff 1986, p. 623). Specie convertibility upon demand effectively forced the circulating media to be fixed-value debt-type instruments tied to gold. Since the notes had to be redeemable in specie, the requirement prevented the possibility of an evolution away from debt-type ties to an "outside money," in particular, gold. Without the option clause and with the requirement of immediate convertibility, note issuers now had no control over the maturity structure oftheir note liabilities. The legal restrictions theory, however, assumes that there are no obstacles to maturity-matched intermediation. In the absence of maturity matching, risk and liquidity considerations may prevent competetive forces from equalizing the rates of return on interest-bearing debt and transactions media. Potential maturity mismatch costs weakened the arbitrage-motivated incentives for paying interest on notes. Such consequences of the restrictions embodied in the Act of 1765 provide an account for coexistence of interest-yielding bearer instruments and bank notes without interest. 6
B. Small- Denomination Notes The Act of 1765 also outlawed the issuance of notes under £ I (= 20 shillings). Wallace (1984, p. 14) has estimated that this £1 minimum would be equivalent to banning notes of under approximately $200 today. Prior to the Act, five-shilling notes alone constituted approximately 17 percent oCthe chartered banks' circulation 'While Scottish notes appear to have circulated at par within Scotland, Adam Smith (1937. pp. 309-10) reports that the notes sometimes were discounted when purchasing exchange on London; that is, Scottish notes were beginning to noat against the English pound as the Scottish innovations evolved. 'During the "free banking" era in nineteenth century Sweden-
215
216
Free Banking 1/ TYLER COWEN AND RANDALL KROSZNER
: 225
and a much larger share of that for the smaller nonchartered banks (Munn 1981, p. 21). The denominational constraint appears to have been binding since the vast . majority of the circulation following the Act consisted of one-pound notes (Cameron ·1967, p. 65). During the convertibility suspension of the Napoleonic Wars, Parliament temporarily lifted the ban on small notes, and they again attained widespread circulation. 7 The elimination of the small-denomination notes inhibited the freedom of entry into financial intermediation and disadvantaged many of the smaller nonbank providers of these services. Contrary to the usual textbook account, the origins of banking in Britain are not found simply with the fractional reserve "warehouse" functions provided by goldsmiths. Instead, banking activities were often undertaken by commercial concerns, particularly in the woolen trade in England and the linen and corn trades in Scotland, as well as by scriveners (see Holdsworth 1925, p. 185). Manufacturers and merchants would issue notes and bills to facilitate their trade and the payment of their employees, while also providing credit and banking services for their customers (much as Sears and J. C. Penney do today). As a consequence of the Act, commercial assets played a smaller role in backing bank liabilities, further inhibiting Scotland's move away from a gold-backed system. 8 The Act of 1765 dealt a severe blow to the numerous smaller financial service competitors. Small-denomination notes and alternative redemption schemes were particularly important to carrying on note-issue on a small scale. 9 The £ I minimum combined with immediate redemption created a much higher variance in the demand on their assets since they had no way to smooth out the now "lumpy" redemption of notes. This increased the minimum efficient scale of banking operations. The two years following the Act saw a sharp contraction in total note issue and the elimination of most of the smaller private bankers as well as an economywide contraction (Mulln 1981, p. 21 and Checkland 1975, p. 123). The scale effect addc:d to the value of the capital market advantages associated with charters of incorporation to which we now turn.
CHARTERS OF INCORPORATION AND LIABILITY RULES
Following the Bubble Act of 1720, free share transferability and, thus, access to the equity market were reserved for chartered corporations. In addition, only 'When the right to issue small-denomination notes was taken away from the private banks and reserved for the state-owned Riksbu"k in the 18705 during Sweden's "free banking" era, the share of private notes in the total value of notes outstanding fell sharply (Jonung 1987, p. 14 and Chart I). I Private banknotes were never legal tender in Scotland, nor in Sweden during its "free banking" era, but nonetheless circulated freely throughout their countries at par (White I984a, p. 39; Jonung 1987). The special legal status of specie, however, may have provided an obstacle to an evolution away from outside money. 'Smaller partnerships and nonbank commercial enterprises issuing small-denomination notes typically inserted the option clause into their notes (Kerr 1918, pp. 75-76; Munn 1981, p. 21). Some also relied upon a variation on the option clause to permit redemption in alternatives to gold: One such note promises to pay the "bearer on demand, in books, coffee, or ready-money, according to the option of the Director, One shilling sterling, value received" (Kerr 1918, p. 75).
Free Banking II 226
: MONEY. CREDIT. AND RANKINO
chartered corporations could obtain limitation ofliability since the common law did not recognize the right to limit liability by private contract. Charters of incorporation required a specific act of Parliament or direct grant from the Crown. This legal setting had two major effects on the development of Scottish financial institutions. First, it raised restrictive barriers to entry in the banking business, thereby reducing competitive pressures (hence relaxing the zero-profit condition) postulated by LRT -NME. Second, it retarded the growth of equity markets, thereby affecting the potential use of equity-based instruments for transactions. Only three of the banking firms in Scotland, referred to as the public banks, obtained charters of incorporation: Bank of Scotland (1695), Royal Bank of Scotland (1727), and British Linen Company (1746).10 The public banks were protective of their special status and effectively lobbied against new charters (Checkland 1968, pp. 149-50). Throughout the "free banking" era, the three public banks were able to block charter applications so that they were the sole banking organi7ations in Scotland with limited liability. II Without limited liability, the non public banks were inhibited from reaching optimal risk sharing arrallgements and had a more difficult time than their counterparts in raising capital (see Carr, Glied, and Mathewson 1986). As a consequence, unlimited liability banks in Scotland were smaller, averaging three branches rather than fifteen for the public banks, and the branches of the public banks were on average ten times larger. In addition, the corporate form appears to have been conducive to survival-none of the public banks failed, whereas about thirty unlimited liability banks in Scotland failed prior to 1845 (Carr and Mathewson 1988, pp. 776-77). Thus, White's (1984a and 1987) characterization of this era as one of free entry is not appropriate. 12 Beyond this special limited liability status, the public banks enjoyed three other advantages. First, important public finance functions were the sole preserve of the public banks. Until the I 830s, only they were authorized to "hold and remit" government revenues, and the Board of Customs and Excise was instructed to accept only their notes (Munn 1981, p. 12; Checkland 1975, pp. 150, 186). Second, the Royal charter conferred the appearance of official sanction on these banks, especially since they were the only banks entrusted with government funds and had uniquely close ties with London and the Bank of England (discussed further in the next section). Third, the uncertain legal identity of unincorporated entities caused difficulties for the nonchartered banks in court actions. Until their standing was 10 During the first quarter of the nineteenth century, these banks controlled about half of the total banking assets in Scotland (Checkland 1975, pp. 240, 424). "Since resources were expended to fight for and against corporate charters, we can infer that the bankers of the time believed that the privileges associated with charters were valuable. The importance of limited liability in banking was emphasi7.ed by a number of contemporary writers such as the prominent banker Alexander Blair (1841). On the cost and difficulty of petitioning for a charter. see Holdsworth (1925. p. 220) and Saville (1956, p. 423). "White (1984a. pp. 41. 142-43) argues that the liability constraint was not binding because the nonchartered banks waited twenty years after the Companies Act of 1862, which permitted mosl types of companies to incorporate with limited liability, to adopt the limited form. For banks. however, limited liability was not effectively available until an Act of 1879 (see Cowen and Kroslner 1987c and earr et al. 1986).
217
218
Free Banking 11 TYLER COWEN AND RANDALL KROSZNER
:
227
clarified in a statute of 1826, for example, the right of nonchartcred banks to bring legal action for fraud and default was in question. The incorporation and liability statutes also impaired the general development of the capital markets. Chartered firms in all industries, not only banking, wished to maintain the advantages of a charter for themselves and successfully blocked many charter applications (see Forbes 1986). Since obtaining a charter typically involved some form of payment to the public coffers, Parliament limited the number of charter grants in order to maintain their value. There were thus fewer firms with transferable shares than if there had been no such constraints on incorporation. Consequently, an equity market was slow to develop. I) The first organized stock market did not appear in Scotland until 1844, one year before Peel's Act was extended to Scotland to prohibit all new entry into note issuance, thereby bringing to a close the "free banking" era. This distortion of the capital market may have reduced the role of equity relative to debt in the evolution of the Scottish financial structure (see Powell 1915). The lack of development of equity instruments and the dominance of debt-based claims thus should not be seen as the "natural" response of an unregulated financial system (see Cowen and Kroszner 1987b and 1988).
THE DEPENDENCE OF SCOTLAND UPON LONDON AND THE BANK OF ENGLAND
The development and structure of bank obligations also was affected by the existence of a lender of last resort in the Bank of England (BOE). As noted above, the prohibition of the option clause coupled with the immediate specie redemption requirement enhanced the role for a lender of last resort by increasing the instabilities of a fractional reserve system which the pre-1765 innovations had been attempting to circumvent (see Rockoff 1986, p. 623). The readiness of the BOE to act as a lender of last resort for the Scottish system provided a source of insurance for a fractional reserve system based on short-term debt tied to gold. The support of the BOE effectively socialized the costs of instability and redemption problems. This undermined the incentives of market participants to mitigate the debt-based instabilities by moving to equity-type instruments and alternative redemption schemes. Although there was no explicit obligation for the HOE to act as a central bank for the Scottish system, the bankers beyond the Tweed relied upon the BOE for credits and gold, particularly in tight times. London acted as a reserve center for the Scottish banks. Scotland effectively had a three-tiered system. The smaller Scottish banks relied on the public banks as lenders of last resort in problem times and, in addition to their London correspondents, sources of specie in normal times. The public banks depended upon the HOE directly in crisis times and upon their London correspondents, who had accounts directly with the HOE, for specie and discount. Il Mirowski (1981) documents that after reaching a high levd of sophistication and efficiency in the first two ,kcades. of. the eighteenth century. the equity market languished for the next hundred years. Patterson and Rletlen (1988) demonstrate that the Bubble Act was a direct cause of the decline of the ~quity market. Powell (1915) also argues that the Bubble Act held back the devdoplllcnt of e4uity IIlstruments. See also a related discussion in Woodward (1985).
Free Banking II 228
:
MONEY. CREDIT. AND BANKING
ing in normal times. 14 In the crisis of 1792-93, for example, both the Bank of Scotland and Royal Bank of Scotland sought the aid of London and received special large credits to ease their distress (Andreades 1924, p. 188; Checkland 1975, pp. 219-20). When another financial crisis arose in 1830, the chartered banks again turned to the BOE for large credits but this time on a long-term basis (Checkland 1975, p. 410).15 The discounting of bills drawn on London constituted a large part of the Scottish bank's business and thus London correspondent banks were crucial to the Scottish banks' operation. Through these correspondent accounts, the Scottish banks had ready access to BOE notes and hence gold, since the London bankers held large reserves of BOE notes that they could turn in for gold upon demand. 16 The few bank panics and failures that did occur in Scotland were sparked by the collapse of the Scottish banks' London correspondents. 17 An episode illustrative of Scotland's dependence upon London involves the suspension of convertibility declared by the Bank of England during the Napoleonic Wars. When the news reached Edinburgh on March I, 1797, the most prominent banking houses met and "spontaneously followed suit" even though they had no legal authority to do so (Cameron 1982, p. 105). Scottish bankers did not permit the Scots pound to float against Bank of England notes for two reasons (White 1984a, p. 46). First, Scotland was now without its primary source of specie, which the Act of 1765 had made so important to the Scottish system. The large holdings of Bank of England notes through their London correspondents and their other London secondary reserves were "immobilized." Thus, the Edinburgh bankers did not believe that they could operate as usual without the assistance of their "shadow" central bank in London. Second, Scottish bank customers "preferred a note convertible into what had become London's basic cash due to the importance of trade with London" (White 1984a, p. 46). The development of Britain as an optimal currency area (Mundell 1961) may have been the best response to the given constraints. The theory of international finance tells us that under fixed exchange rates in a two-<:ountry model, the "small" country loses control over its money supply policy to the "big" country. By maintaining a fixed exchange rate between the English and Scots pound, the Scottish bankers permitted London to have a leading role in their monetary activities. 18 BOE policies effectively controlled the Scottish money supply "Scullish hnnh were prohihited from opernting their nwn Londnn ofrice., hence their relinnce upnn correspondent hnllh in Lnlldon. I~ By 1810, "the principle and ultimate source of liquidity (throughout Britnin] lay in Londnll. nnd. in particular, the Bank of England. This was especially so for the Scottish banks ..." (Checkland 1975, p. 432). See also Clapham (1970, vol. I, pp. 168-69). I. For the large Scottish banks, London also provided an important source of "secondary" reserves of government securities and BOE stock. As of 1806, for example, the Bank of Scotland was the third largest shareholder in the BOE (Clapham 1970, vol. I, p. 352). "See, for example, White (19848, p. 32), Saville (1956), and Clapham (1970, pp. 245-46) on the Ayr Bank collapse and Munn (1981, p. 123) on the Western Bank of Scotland. See also Clapham (1970. vol. 2, p. 309) on the end of the City of Glasgow Bank which precipitated the change in the availability of limited liability for banks found in the Act of 1879, discussed above (see footnote 12). "Scottish banker Alexander Blair (1841, p. 18) reported that London controlled the "state of prices" in Scotland, attesting to the exogeneity of the price level for the Scottish bankers.
219
Free Banking II
220
TYLER COWEN AND RANDALL KROSZNER
:
229
and banking system much as it did in England. Thus, the Scottish system must be considered as a part of the overall British banking system under the aegis of the BOE with the dependence stemming at least in part from the legal restrictions tying Scotland to a redemption on demand structure.
CONCl.UDING REMARKS
Scottish "free banking" banking experience plays a central role in the recent work exploring the consequences of laissez-faire in the banking and financial system. Our examination has questioned the use of the Scottish system as a model of laissez-faire and has found the experience consistent with the LRT -NM E approach. During this period in Scotland, there were impediments to the payment of interest on notes, prohibitions on small-denomination notes, barriers to entry in banking, biases against the development of equity markets, and incentives to eschew innovations and come under the wing of the Bank of England. Prior to the Act of 1765, Scottish bank notes were showing evidence of evolving into an interest-bearing medium, alternative redemption schemes were beginning to arise, and the linkage to an outside money was being attenuated. The challenge now facing the LRT -NME approach is to provide a complete theoretical account of how an unregulated financial system sharing many of the characteristics of Scotland's before 1765 could "naturally" evolve into an LRT -NME payments system (see Cowen and Kroszner 1988). LITERATURE CITED
Andreades, A. History of the Bunk of England. 2nd ed. London: P. S. King, 1924. Bcrnanke, Ben S. "Nonmonetary Effects of the Financial Crisis in the Propagation of the Great Depression." American /:./.'onolllic Revew 73 (June 1983), 257-76. Black, Fischer. "Banking and Interest Rates in a World without Money: The Effects of Uncontrolled Banking." Journal of Banking Research I (Autumn 1970),9-20. Blair, Alexander. SOll/e Ohservations upon the Preselll Slate of Banking. Edinburgh: Johnstone and Fairly, 1841. Cameron, Rondo. Banking in the Early Stages of Industrialization. New York: Oxford University Press, 1967. _ _ . "Banking and Industrialisation in Britain in the Nineteenth Century." In Business. Bunking. a",1 Urban Hislory: Essays in Honour of S. G. Chl!chland. edited by Anthony Slaven and Derch Aldcroft, pp.102 -II. Edinburgh: John McDonald, 1982. Carr, Jack, Sherry Glied, and Frank Mathewson. "Unlimited Liability and Free Banking in Scotland." Unpublished ms., University of Toronto, 1986. Carr, Jack, and Frank Mathewson. "Unlimited Liability as a Barrier to Entry." Journal qf Polilical Economy 96 (August 1988),766-84. Checkland, Sydney G. "Banking History and Economic Development: Seven Systems." Scol/ish Journal of Political EconolllY 15 (1968), 144-66.
- - . Seol/ish Bunkillg: A History. /695-1971. Glasgow: Collins, 1975. Clapham, John. The Balik of England. A l/istory. 2 vols. Cambridge: Cambridge University Press, 1970.
Free Banking II 230
:
MONEY, CREDIT. AND BANKING
Cowen, Tyler, and Randall Krosmer. "The Development of the New Monetary Economics." Journal of Political Economy 95 (June 1987),567-90 (a). _ _ _ . "Recent Developments in the New Monetary Economics." ZRirsclrrift fur Wirtschajtspolitik 36 (1987),207-20 (b). _ _ _ . "Does the Scottish Banking Experience Provide a Model for Laissez-Faire?" Unpublished ms., Harvard University, 1987 (c). _ _ _ . "The Evolution of an Unregulated Payments System." Unpublished ms., Harvard University, 1988. Diamond, Douglas, and Philip Dybvig. "Bank Runs, Deposit Insurance, and Liquidity." Journal of Political Economy 91 (June 1983),401-19. Fama, Eugene F. "Banking in the Theory of Finance." Journal of Monetary Economics 6 (January 1980), 39-57. Forbes, Kevin. "Limited Liability and the Development of the Business Corporation." Journal of lAw, Economics. and Organization 2 (Spring 1986), 163-77. Goodhart, Charles. nle Evolution of Central Banks: A Natural Developmelll? London: London School of Economics and Political Science, 1985. Gorton, Gary. "Bank Suspensions of Convertibility." Journal of Monetary Economics 15 (March 1985), 177-93. Greenfield, Robert L., and Leland B. Yeager. "A Laissez-Faire Approach to Monetary Stability." Journal of Money, Credit, and Banking 15 (August 1983),302-15. Hall, Robert E. "Monetary Trends in the United States and the United Kingdom: A Review from the Perspective of New Developments in Monetary Economics." JOllrnal of Economic literatllre 20 (December 1982), 1552-56. Hamilton, Henry. "The Failure of the Ayr Bank, 1772." Economic History Review 8 (April 1956),405-17. Holdsworth, William S. A llistory of English lAws. vol. 13. London: Methuen Co., 1925. Jonllng, Lars. "The Economics of Private Money: The Experience of Private Notes in Sweden, 1831-1902." Unpublished ms, University of Lund, 1987. Kerr, Andrew W. lIi.ftory of Banking in Scotland. 3rd ed. London: A & C Black, 1918. King, Robert G. "On the Economics of Private Money." Journal of Monetary Economics 12 (1983),127-58. McCulloch, J. Huston. "Beyond the Historical Gold Standard." In Alternative Monetary Regimes. edited by Colin Campbell and William Dogan, pp. 73-81. Baltimore: Johns Hopkins University Press, 1986. Mirowski, Philip. "The Rise (and Retreat) of a Market: English Joint Stock Shares in the Eighteenth Century." Journal of Economic llistory 41 (September 1981), 559-11. Mundell, Robert. "A Theory of Optimum Currency Areas." American Economic Review 51 (September 1961),657 .. 155. Munn, Charles W. nle Scottish Provincial Banking Companies. 1747-1864. Edinburgh: John Donald Publishers, 1981. O'Driscoll, Gerald P., Jr. "Deregulation and Monetary Reform." Federal Reserve Bank of Dallas Economic Review (July 1986), 19-31. Patterson, Margaret, and David Reiffen. "The Rise and Retreat of the Market for Joint Stock Shares Revisited: The Effect of the Bubble Act in 18th Century England." Unpublished ms., Federal Trade Commission, 1988. Powell. Ellis T. The Evolution of the Monev Market. /385-/9/5. London: Financial News. 1915. . .
221
222
Free Banking II TYLER COWEN AND RANDALL KROSZNER : 231 Rockoff, Hugh. "Institutional Requirements for Stable Free Banking." Cato JOllmal6 (Fall 1986), 617-34. Saville, John. "Sleeping Partnership and Limited Liability, 1850-1856." Economic History Reyiew 8 (1956), 418-33. Selgin, George, and Lawrence H. White. "The Evolution of a Free Banking System." &onomic Inquiry 25 (July 1987),439-58. Smith, Adam. An Inquiry into the Nature and Causes of the Wealth of Nations. Edwin Cannan, ed. New York: Random House, 1937. Wallace, Neil. "A Legal Restrictions Theory of the Demand for 'Money' and the Role of Monetary Policy." Federal Reserve Bank of Minneapolis Quarterly Reyiew (Winter 1983), 1-7. _ _ . "Do We Need a Special Theory for 'Money'·r' Unpublished ms., presented at Money: A Search/or Common Ground, 1984. White, Lawrence H. Free Banking in Britain: Theory, Experience and Debate, /800-/845. Cambridge: Cambridge University Press, 1984 (a). _ _ . "Competitive Payments Systems and the Unit of Account." American Economic Reyiew 74 (September 1984),699-712 (b). _ _ . "Accounting for Non-interest-bearing Currency: A Critique of the Legal Restrictions Theory of Money." JOllmal of Money, Credit. and Banking 19 (November 1987), 448-56. Woodward, Susan. "Limited Liability in the Theory of the Firm." Journal of Institutional and Theoreti,·al Economics 141 (December 1985),601-11.
[14] Nor-rES, COMMENTS,
REPLII~S
Scottish Banking and the Legal Restrictions Theory: A Closer Look A Comment by Lawrence H. White The use of Scottish banking experience as counterevidence to the legal restrictions theory of money (White 1987) rests on the view that the Scottish system, before the Bank Charter Act of 1844 closed entry into note-issue, I warranted its traditional characterization 33 "free banking." That is, the regulatory regime approximated laissez-faire closely enough that its institutions, practices, and performance were basically those of an unregulated free-market monetary system. This interpretation of Scottish banking has been challenged by Tyler Cowen and Randall Kroszner (1989). Wishing to defend the legal restrictions theory. whose vision of a laissez-faire payments system does not coincide with the Scottish experience, they argue that the Scottish regime did not closely enough approximate laissez-faire to provide evidence damaging to the legal restrictions theory. They cite three principal "deviations from laissez-faire" (p. 222): (I) the Act of 1765 "raised impediments to interest-bearing notes and proscribed small denomination notes": (2) "the prevailing system of liability and incorporation raised barriers to entry into hanking"; and (3) "the Bank of England acted as a shadow central bank for Scotland." Cowen and Kroszner Ihereafter CK) thus characterize the Scottish system as legally restricted from paying interest on circulating media. as less than competitive, and as dependent upon the Bank of England. I See Gregnry (1964) fnr Ihe lext of the I R44 Act (7 & R Viel., c. 32), Sect inn X c1nse~ enlry into h:lI1k note issue throughout the United Kingdom. , am not sure why Cowen and Kros7.ncr (19R9) rhons(' t R,15 as the end dale of the Scollish free-banking era. Perhaps they believe that Scotland W/IS nol affected until passage of the Scottish Bank Act of 1845,
The author thanks Kurt Schuler. Charles Munn, Randall Kros7.ner. and 'tYler Cowcn for useful commcnts. LAWRENCE
H. WHITE is associate professor of ecollomics lit tire Ullh'ersity
of Mo"ey. Credit, a"d Banki"g, Vol. 22, No.4 (November 1990) Copyright ~ 1990 by the Ohio State University Press
jO/mlt/1
of (1eO/'Ria.
224
Free Banking II NOTES, COMMENTS, REPLIES
527
A closer reading of Scottish banking history, as related in the very sources they rely upon, Checkland (1975) and Munn (1981), undermines these characterizations. The Scottish case docs provide relevant evidence against the prediction of the legal restrictions theory that in the absence of relevant legal impediments non-interest11l:ilring hank notes cannot coexist with interest-bearing assets.2
The Act of J765 The Act of 1765, which banned the option clause and notes smaller than one pound, obviously infringed on laissez-faire, as my 1987 article recognized (White, I()~7, p. 451). CK are correct in noting that the ban on the option clause eliminated a useful contractual means of mitigating potential illiquidity problems and associated instability (sec Dowd 1988), and that the ban on small notes raised a barrier to entry against very-small-scale issuers (see White 1984, p. 30). They are incorrect, however, in stating that the Act of 1765 "thwarted" the "displacement of noninterest-bearing IbanknotesJ by interest-bearing instruments." CK (p. 224) argue that the Act of 1765 accounts for the failure of Scottish banking to deliver the interest-bearing transactions media that the legal restrictions theory predicts for an economy with positive interest rates: "Without the option clause and with the requirement of immediate convertibility, note issuers had no control over the maturity structure of their note liabilities. The legal restrictions theory, however, assumes that there arc no obstacles to maturity-matched intermediation. In the absence of maturity matching, risk and liquidity considerations nwy prevent competetive [sic I forces from equalizing the rates of return on interestbearing debt and transactions media." Even granting its description of the facts, this argulllent is insufficient to account for non-interest-bearing bank notes. While liquidity costs due to maturity mismatch imply a (larger) spread between rates of return, they do not imply a zero nominal rate of return on banknotes. More imp0l1antly, the premise that "note issuers had no control over the maturity structure of their note liabilities" is factually incorrect. CK apparently assume that postdated bearer instruments were outlawed by the Act of 1765. My article (White 1987, p. 451) speculated that they "may have" been so outlawed. An overlooked passage in Checkland (1975, p. 187) indicates that postdated bearer instruments ilclLwlly were 1I0t illegal and were in fact issued. Because of its importance llluotc this passage at length: By 1810, no less serious for a bank than a run on its notes could be a sudden demand for the repayment of deposits . . . . [TJhe traditional form of bank-borrowing from the public, since 170!!, had been the Treasurers' or Cashiers' bond. Under this device, the Bank could take in quickly quite large sums, provided by a relatively few large lenders, friends of the bank, at a stated term and at an agreed rate of interest. But with the ending orthe optional clause in 1765, it became necessary to become more flexible 2Thc Swttish case also casts doubt on the predictions of CK (pp. 221-22) Ihat under laissez-faire debl·based banking claims would vanish in favor of equity-based payment accounts, and that "transactiolls media wuuld not bear a fixed relationship with the unil of accounl." For discussion see White t'!X'!).
Free Banking II 528
MONEY, CREDIT, AND BANKtNG
in deposit gathering. This need stimulated the promissory note, also signed by the Trl'HSIII'l'J", bllt lIIore flcxihle than the homl, ellsier to isslie and readily I'a.l".\",·d{rom htllld /0 hal/d, IlIld so, in H sense, allllost lin extension of the note issue. SlIdl notes were for three or six lIlonths, bearing 3% and 4% respectively. It is prohahle Ihllt cerlHin of the provincial banking compunies anticipated the Bank of Scot lund in this. I El1Iphas is lidded. I
The coexistence of non-interest-bearing bank notes with these transferable interestbearing bank notes is inconsistent with the legal restrictions theory. As CK indicate, the LRT predicts equalization of rates of return between interest-bearing debt and transactions media in the absence of legal barriers to providing equivalent default risk, equivalent transferability, and maturity matching. We can now see that all three of those conditions were met in the Scottish case. A promissory note and an ordinary bank note issued by the same Scottish bank carried identical default risk. Both liabilities "readily passed from hand to hand." A bank issuing three-month and six-month promissory notes could perfectly match the maturities of its assets and liabilities. Yet rates of return on ordinary bank notes and promissory notes differed. Transferable promissory notes did not become the common medium of exchange or otherwise drive non-interest-bearing notes out of circulation. My explanation (White 1987) for the survival of non-interest-bearing noles as transactions media, even in the face of interest-hearing alternatives, is that the timeand-inconvenience cost of computing the present value of an asset like a promissory note in each transaction exceeds the benefit (at least for denominations helow some threshold). The CK explanation, that interest-bearing alternatives were hlocked hy legal restrictions, is rendered untenable by the existence of transferable hank-issued promissory notes. A possible strategy for immunizing the LRT prediction from this counterevidence would he to argue that Scottish banking was not sufficiently competitive to compel note-issuers to pay interest on bank notes. This response would beg the question of why issuers paid competitive interest rates on promissory notes and deposits. 811t CK do emphasize barriers to competition. Accordingly I tum to the question of the competitiveness of Scottish banking.3
SlIIall Notes, Charters (?f II/corporatiol/ and Liability Rilles CK (pr. 224-25) point to the 1765 Act's ban on notes below £1 as a harrier to entry. The small note han did increase the minimum efficient scale of banks (White 1984, p. 30). But it did not rule out effective competition among the numerous larger banks. It remains to be investigated whether competitive pressures in the market for bank liabilities were noticeably reduced. Following Carr, Glied, and Mathewson (1986) and Carr and Mathewson (1988), CK (pp. 225-27) also point to the restriction of limited-liability charters to three particular banks, which made unlimited liability compulsory for other banks, as a )Thc following two seclions draw heavily from White (1990), which replies to several critics of the free-banking interpretation of the Scottish experience.
225
226
Free Banking II NOTES, COMMENTS, REPLIES
:
529
barrier to entry. Clearly such restrictions on liability size and on the contractual risksharing arrangements firms may adopt are potentially barriers to entry (White 1984, pp. 142-43). Thus CK are correct that entry was not completely free. But how much did these barriers inhibit competition in practice? Following Checkland (1975, p. 480) I had previously suggested (White 1984, p. 41) that the restriction against limited liability banks in Scotland was not binding because "the unchartered banks of Scotland chose to retain unlimited liability in the 1860s and 1870s even after limited liability became available to them" through the Companies Act of 1862. CK (p. 226, n. 12) cite this suggestion, and argue to the contrary, citing Carr, Glied, and Mathewson (1986), that for banks "limited liability was not etfectively available until an Act of 1879." Carr et al. point out that while the 1862 Act allowed limited liability for other bank obligations, it retained unlimited liability for note-issues. But this means that the Act of 1862 in fact offered an etfective limitation of liability on all but bank notes, which comprised only 7.9 percent of total liabilities for the Scottish banks in 1865 (Checkland 1975, table 44, p. 743), or 6.6 percent of liabilities plus capital (a figure Carr, Glied, and Mathewson themselves cite Ip. 10]). If unlimited liability blocked the nonchartered banks from "reaching optimal risk sharing arrangements" as shareholders perceived them, as CK follow Carr and Mathewson (1988) in suggesting, wouldn't those shareholders have found a substantial limitation of liability better than no limitation, and therd'ore have taken advantage of the 1862 Act? Carr, Glied, and Mathewson (1986, p. 10) believe that the Scottish joint-stock banks finally adopted limited liability in 1882 because an 1879 Act offered "substantially lessened" liability for notes. On my reading, however, the 1879 Act retained greater shareholder liability for notes than the 1862 Act. Historians of the iudustry have traditionally attributed the adoption of limited liability not to the change in the law, but to the 1878 failure of the City of Glasgow Bank, which evidently changed the perceptions of shareholders concerning the risks they faced. We can also assess the bindingness of compulsory unlimited liability before 1844 by contrasting the three limited-liability "public" banks (the Bank of Scotland, Royal Bank of Scotland, and British Linen Company) with the unlimited-liability banks before 1844. CK note that the public banks were larger on average, and (again following Carr and Mathewson) that none of them failed. They regard these facts as evidence that limited liability made it easier for banks to raise capital and that the corporate form was "conducive to survivaL" These hypotheses are plausible. As contrary evidence, however, several facts about the nonchartered banks as of 1845 (White 1984, table 2.2, p. 37) merit attention: (I) the Commercial Bank of Scotland had more branches than any of the public banks, a greater note circulation than two of the three, and more shareholders than one; (2) the National Bank of Scotland had more branches than two of the three public banks, a note circulation only one percent smaller than that of the second largest public bank, and more shareholders than any of the three; (3) the North of Scotland Banking Company and the Edinburgh and Glasgow Bank each had more shareholders than any of the three public banks; (4) the-Union Bank of Scotland had a greater note circulation than two
Free Banking II ~.1()
:
MONEY. CREDIT. AND IlANKING
of the three public banb. All told, the public banks ranked first, fourth, and seventh in size of circulation. Banks with unlimited liability apparently did not have great difficulty raising capital on a large scale: five had capitals as large as the Bank of Scotland's £ I million, and eight had capitals as large or larger than the British Linen Company's £500,000. Nor was the survival of unlimited-liability banks obviously impaired once they attained adequate size: none of the five above-named large jointstock banks failed while they had unlimited liability. (Two of the eight with the largest capitals eventually did fail some time after 1844.) Still, the banks that made efforts to get legislative charters, and to deny them to other banks, must have wanted them for some reason. CK (p. 226, n. I I) correctly point out that "since resources were expended to fight for and against corporate charters, we can infer that the bankers of the time believed that the privileges associated with charters were valuable." CK cite several possible advantages beyond limited liability, one of which clearly had some value: government customs officers were instructed to accept only the notes of the chartered banks in payment of duties (Checkland 1975, p. 186). CK also claim that "only they [the public banks) were authorized to 'hold and remit' government revenues." The basis for this claim is unclear, since it is not found in the pages of Munn and Checkland that they cite. 4 Checkland (1975, p. 166) in fact relates that the private bank of Sir William Forbes and Company in the late eighteenth century "had the remittance of the excise duties from Edinburgh to London." He adds: "The Royal Bank had long had the remittance of the customs; the British Linen Company that of the revenue of the Post Office[;) the Bank of Scotland had nothing." Thus a charter was neither necessary nor sufficient to benefit from government patronage. CK cite two other advantages of charters: "the appearance of official sanction," which is also noted by Checkland (1968, p. 149), and escape from "the uncertain legal identity of unincorporated entities" before 1826. It is not clear how important these rcally were. But in at least one respect, receivability for duties, the public banks enjoyed politically created rents. The existence of such rents is contrary to strict laissez-faire. Accounts of the Scottish system as free banking have perhaps underemphasized (though they have not ignored) the "preferred position" of the public banks (Checkland 1975, p. 235) in this respect. Contrary to CK (p. 226), however, the charters did not have the effect of "reducing competitive pressures (hence relaxing the zero-profit condition)" throughout the free banking era. While the public banks may have enjoyed a more-or-less advantaged position before the Commercial Bank of Scotland, the first large joint stock bank, entered in 1810, they clearly faced effective competition between 18\0 and 1844. Competition between the public banks and the large joint-stock banks was vigorous in note-issuing, deposit.taking, lending and discounting, inland exchange, 4Cowen and Kroszner cite inter alia Checkland (1975, p. 150), a page not relevant to the question. Perhaps they meant to cite Checkland (1968, p. 149), which contains the statement that "A charter was a useful prelude to gaining a share in the holding and remitting of government revenues, a source of trading funds of great value."
227
228
Free Banking II NOTES, COMMENTS, REPLIES
531
and other aspects of the trade. s Profit margins for all banks were squeezed as competitive bidding for loans and deposits kept the interest spread between these down to I percent (Checkland 1975, pp. 384-88), exactly Neil Wallace's (1983) estimate of the competitive spread in the absence of legal restrictions. Contrary to CK's hypothesis of a "relaxed" zero-excess-profit condition, Munn (1982, p. liS) comments that "the competitive nature of the business" had the result that "protits were reduced and therefore all earning assets had to be mannged with fine attention to detail. There was no room for error or even slackness if dividends were to be maintained." Attempts were made in the 1830s and I 840s to limit deposit interest rates and enhance activity charges through cartel agreements, but such agreements proved unsustainable in the face of strong competitive pressures (Checkland 1975, pp. 449-50; Munn 1982, p. 122). The elimination of excess profits in Scottish banking is evidence of the effectiveness of competition among the banks. CK fail to inquire after this evidence. It lIppears that they simply assume, incorrectly, that the infringements of laissez-faire to which they point precluded effective competition. Supposed Depelldellce IipOIl LOlldoll alld the Balik of Ellglalld
Cowen and Kroszner (1989, p. 227) claim that "the development and structure of bank obligations [in Scotland] ... was atfected by the existence of a lender of last resort in the Bank of England (BOE)." They go on to state that "the readiness of the BOE to act as a lender of last resort for the Scottish system provided a source of insurance." In their view the Bank, acting as a lender of last resort, extended support lit subsidy rates that distorted the banking system: "The support of the BOE clfectivcly socialized the costs of stability and redemption problems." 'Ih act as a "lender of last resort" in the standard sense of that phrase, as described by Humphrey and Keleher (1984, pp. 277-78), is to act as a "backstop or guarantor ... of a fractional-reserve banking system" or to take on "the responsibility of guaranteeing the liquidity of the enlire economy." If this detinition is accepted, Ihe Bank of England clearly was 1101 a lender of last resort for Scotland before IS44. In a few cases the Bank provided loans to Scottish banks. But in other cases (most importantly the crises of 1825-26 and 1836-37) it refused to lend. In the case of the Ayr Bank in 1772, the Bank of England set such stiff terms that its "support" was declined (Checkland 1975, p. 131). CK give as an example of lastresort knding a long-term credit the Royal Bank negotiated with the Bank of England in 1830. But Checkland (1975, p. 444) adds that "in October IIU6 ... the Bank [of England), as part of a general credit contraction, required Ihe Royal Bank of Scotland to pay oft· its advance." This withdrawal of credit in a lime of stringency was not only much to the consternation of the Royal Bank's general manager, but contrary to the behavior of a lender Qf last resort. CK (pp. 227-2S) cite a second supposed episode of last-resort lending by the ;Generalizalions aboullhe Scollish syslem Ihus ne.:d 10 be carefully daled. Munn (19112) discusses Ih.: ,lm,lural I:hangcs in Sl:ollbh banking bel ween II! 10 and 1844.
Free Banking II 532
:
MONEY. CREDIT. ANI) RANKING
Bank of England to the Scottish banks: "'n the crisis of 1792-·93. for example. hoth the Bank of Scotland and Royal Bank of Scotland sought the aid of London and received special large credits to ease their distress (Andreades 1924, p. 188; Checkland 1975, pp. 219-20)." Closer reading of the sources that they cite here shows, however, that the loans in question were granted nof by the Bank of England. hut hy the government, and were granted not to the banks. but to Scottish business firms. Officials of the Scottish banks had gone to London to plea for general relief measures for Scotland, not for BOE loans to the banks. Thus CK seriously misrepresent this episode when they cite it as an example of how "the (Scottish] public banks depended upon the BOE directly in crisis times." Fetter (1978, p. 267) makes a relevant distinction between "supporting the market ex 1'0.1'1. as compared with giving assurance ex ante that credit would be available at some price." The second is an essential part of the lender-of-Iast-resort role of guaranteeing liquidity. Before 1844 (and apparently until the Baring Crisis of 1890) the Bank of England played only the first role, and did not do even that consistently. There is no reason to believe that in making occasional interbank loans the Bank was acting in any way other than to maximize its profits. Contrary to CK, the Bank of England did not extend interbank credit at subsidy rates which "socialized" costs. Because the Bank of England did not assure its willingness to lend, or even act consistently with such an assurance, it did not act as lender of last resort (in the standard sense) to the Scottish system before 1844. This is not just a terminological point. The substantive point is that the Scottish banks could not have relied or depended ex allfe on advances from the Bank of England as a backstop source of liquidity. CK (p. 227) recognize that "there was no explicit obligation for the BOE to act as a central bank for the Scottish system." Neither was there any implicit policy of acting in such a manner. Thus CK (p. 228) are incorrect in calling the Bank of England a "shadow central bank" for the Edinburgh banks, and in saying that the Scottish banks faced "incentives to ... come under the wing of the Bank of England." In fact the Bank of England explicitly rejected the idea that it had lender-of-Iastresort obligations. Fetter (1978, pp. 118-20) describes the Bank's unwillingness to extend credit during the crisis of 1826. In correspondence accompanying the Bank of England's withdrawal in 1836 of its advance to the Royal Bank of Scotland (quoted by Checkland 1975, p. 447), Horsely Palmer of the Bank of England wrote to the Royal Bank that "he deemed it expedient to reduce the Bank's advances to other Banks of issue, therehy making them dependent in such times upon their own respective resources . . . . Every Bank of issue should be prepared to support its own circulation by its own reserve ... without requiring any issues from the Bank of England." Cowen and Kroszner (p. 228) are correct in indicating that the Scottish banks relied on the London financial market to meet occasional liquidity needs. The Scottish banks held "secondary reserves" in the form of "good securities, of a kind that could be realised in London without serious loss" (Checkland 1975, p. 1974).
229
230
Free Banking II NOTES, COMMENTS, REPLIES
533
The ability to sell these assets quickly and at low cost for claims on London banks, which the Scottish banks could in many cases use as a redemption medium, or for gold, permitted the Scottish banks to meet their liquidity needs with less specie in their vaults. Cowen and Kroszner exaggerate the degree of "dependence on London" (CK, p. 22H) this arrangement involved. In any specie-based free banking system we should expect to lind banks striving to economize on their specie holdings, inter alia by holding liquid securities as secondary reserves. In the Scottish case the market in which these securities could be sold happened to be outside of their banking orbit hecuuse London wus Britain's finuncial center, and because the Scottish banks were legitlly excluded fmlll opening branch oll"ices in London. The Scottish ballks did fitcC a hillding legit I restriction here. But this is presumably not what is at issue when "dependence" on the London financial market is cited. Had they opcned London brunch oll1ces, the Scottish banks would presumably have m:\de similar use of the London market. What is at issue is the significance of the Scottish banks' use of the London financial market as a source of liquidity. The Scottish banks relied on the London center in the short-term sense that at any given moment they held securities that they counted on being able to sell there. A surprise "shutting" of the London market would seriously affect them, as it did in the 1797 episode CK cite. But Scottish banks did not rely on London in the long-term sense that access to the London market was essential to the way they economized on specie holdings. Alexander llIair (1826, p. 51), then general manager of the British Linen Company, testified that the Scottish banks might draw gold from Hamburg if it were not available in London. Checkland (1975, p. 194) indicates that they could also draw on Holland to mcet occasional liquidity needs. CK (p. 228, n. 15) quote Checkland's (1975, p. 432) statement that for the Scottish banks "the principal and ultimate source of liquidity lay in London, and, in particular, in the Bank of England." It is true that the Bank of England was the principal source of specie in London. To draw gold from London, the Scottish banks acquired and redeemed Bank of England liabilities. Concentration of London specie reserves in the Bank of England may have distorted the market for specie somewhat. But it probably did not make the constraints on acquiring specie appreciably looser than those the Scottish banks would have faced had there been a plurality of issuing banks serving as specie sources in London. It is not true, as we have seen, that the Bank guaranteed to make reserves available to the market in times of stringency. Interbank loans, such as the Bank of England occasionally made, might have been at least as readily available on similar terms from competitive London issuing hunks. An interbank loan market surely would have existed even without a privileged bank in the financial center, and might well have been more active. III the crisis of 1839 the Bank of England turned to the Bank of France for an emergency extension of credit, to meet severe liquidity needs that had brought the HOE close to suspending payments. The Bank also arranged for sizable credits in lIalliburg (Fetter 1978, p. 175). Yet no one would say that the London banking
Free Banking II 5.14
MONEY. CREDIT. AND BANKING
system customarily "depended" on the Bank of France, or the Hamburg market, as a lender of last resort. Still less would anyone say that the London banking system was "under the wing" of the Bank of France. The ability of Scottish banks to borrow reserve money from foreign financial centers is consistent with, and indeed implicit in, the free banking interpretation of Scottish banking according to a model of a small open economy whose base money (specie) is money throughout the world economy. 6 CK (p. 228) argue that "by maintaining a fixed exchange rate between the English and Scots pound. the Scottish bankers permitted London to have a leading role in their monetary activities. BOE policies effectively controlled the Scottish money supply and banking system much as it did in England." But in fact, unlike the English country banks, the Scottish banks did not regularly payout Bank of England notes as a redemption medium, or regularly hold Bank of England liabilities as an important component of reserves. The Scottish banks did hold working balances with London correspondent banks (which did hold Bank of England liabilities), but not as a major component of their reserves (White \990). The Scottish money stock was not specifically geared to the quantity of Bank of England liabilities. In the long run, Scotland's money stock was determined by the quantity of money demanded at the given purchasing power of the monetary unit. The purchasing power of the monetary unit was determined by global sl)pply and demand for gold. CK suggest that Scotland played the role of a "small" country to England's "big" country in "a two-country model" of international finance. But surely Britain as whole must be regarded as an open economy during the period of an international specie standard. In the short run, the Bank of England may well have had the power to disturb the quantity of lIloney in England significantly enough to cause cyclical effects on the price level, interest rates, and real activity, with spill-over effects on the Scottish economy. (The actual transmission mechanism warrants further research.) Far from being inconsistent with the free-banking view of Scotland, this view of the Bank of England's powers was first developed by the Free Banking School in the I H20-44 period. The School argued in defense of Scottish free banking that cyclical disturbances in Scotland had been imported and had not originated with the Scottish banks. The existence of spill-over effects from London to Scotland would imply that the English and Scottish economies were integrated, as the traditional view of Scottish banking has always recognized. It would not imply. contrary to CK (pp. 228-29) that "BOE policies effectively controlled the Scottish ... banking system." or that "the Scottish system must be considered as a part of the overall British banking system under the aegis of the BOE." A finding of spill-over effects of BOE actions on the United States economy, about which American writers at the time complained, would not led one to deny the distinctness of the English and U.S. banking systems.
~While
(1984, p. II) explicitly argues for the applicability of stich a model to Scotland.
231
232
Free Banking II NOTES, COMMENTS, REPLIES
535
Cunclusio/l Cowen and Kroszner's arguments for dismissing Scottish banking before 1844 as counterevidence to the legal restrictions theory of money do not hold up under scrutiny. Ttll: Act of 1765 did not outlaw interest-bearing notes. Such notes existed, and did not drive non-interest-bearing notes out of circulation. The Scottish system in its heyday was quite competitive. The privileges of the chartered Scottish banks lIlay have been a source of some small rents worth protecting, but they did not impede competition in intermediation or in the provision of inside money. The Bank of England was not a lender of last resort in the Scottish system before 1844. Nor was the BOE a central bank in the sense of providing a reserve base of highpowered money held by the Scottish banks, except perhaps during the Restriction period. The Scottish banks did use the London tinancial market to meet occasional liquidity needs, but this did not imply reliance on the good graces of the Bank of England. At least during the period from 1810 to 1844, then, the free-banking model continues to provide a useful and valid framework for understanding the Scottish banking system. Scottish experience accordingly provides important counterevidence to the prediction of the legal restrictions theory that non-interest-bearing banknotes cannot coexist with interest-bearing assets under laissez-faire.
LITERATURE CITED Andreades, A. A flistory of the Balik of Englallil, 2nd ed. London: P. S. King, 1924. Blair, Alexander. lestimony in the Report from the Select Committee Appoillted to IlIquire illlo the Slate of the Circulatioll of Promissory Notes under the Value of£5 ill Scotlllnd 1I11d Irelallil ... British Sessional Papers, 1826, House of Commons, vol. 3. New York: ReaJex Microprint, n.d. Carr, Jack, Sherry Glied, and Frank Mathewson. "Unlimited Liability and Free Banking in Scotland." MS, University of'loronto, 1986. Carr, Jack L., and G. Frank Mathewson. "Unlimited Liability as a Barrier to Entry." JOllrllal of Political Ecollomy 96 (August 1988), 766-84. Checkland, S. G. "Banking History and Economic Development: Seven Systems." Scottish Jourllal of Politiclll Ecollomy 15 (1968), 144-66. ___ . Scottish Banking: A History, /695-/973. Glasgow: Collins, 1975. Cowen, 'Iyler, and Randall Kroszner. "Scottish Banking before 1845: A Model for LaissezFaireT' Jourllal of Money, Credit, and Blinking 21 (May 1989), 221-31. Dowd, Kevin. "Option Clauses and the Stability of a Laissez Faire Monetary System." Journal of Financilll Services Research I (1988), 319-33. f'Ctter, Frank Whitson. Development of British Monetary Orthodoxy, 1797-1875. Reprint of 1965 cd. Fairfield, N.J.: Augustus M. Kelly, 1978. Gregory, T. E., ed. Select Statlltes Documents lind Reports Relating to British Banking /832lIJ2H, vol. I. London: Frank Cass, 1964. Humphrey, Thomas B., and Robert E. Keleher. "The Lender of Last Resort: A Historical Perspective." Cato JOllmal4 (Spring/Summer 1984), 275-318.
Free Banking II 536
MONEY. CREDIT. AND IJANKING
Munn. Charles W. The Scottish Provincial Banking Companies, 1747-1864. Edinhurgh: John Donald, 1981. _ _ _ . "The Development of Joint-Stock Banking in Scotland. 1810-1845." In Business. Blinking and Urban/fistory. edited by Anthony Slaven and Derek H. Aldcroft. Edinburgh: John Donald, 1982. Wallace, Neil. "A Legal Reslrictions theory of the Demand for 'Money' and the Role of Monetary Policy." Federal Reserve Bank of Minneapolis Quarterly Review (Winter 19S3),
1-7. White, Lawrence H. Free Banking ill Britain. Cambridge: Cambridge University Press.
1984. _ _ _ . "Accounting for Non-interest-bearing Currency: A Critique of the Legal Restrictions Theory of Money." Journal of Money. Credit, and Banking 19 (November
Il)K7), 44H-56, _ _ _ . "What Kinds of Monetary Inslitutions Would II Free Market Deliver'!" ('"t".Imtr//(/1 9 (Fall 1989). 367-91. _ _ _ . "Banking without a Central Bank: Scotland he fore 1844 as a 'Free Banking' Systelll." Forthcollling in Unregulated nanking: Chaos or Order? edited by I~'rrest Capie and Ucon'rey Wood. London: Macmillan, 1990.
233
[15] EXPLORATIONS IN ECONOMIC IIlSTORY
27, 251-276 (1990)
Free Banking during the French Revolution* EUGENE
N.
WHITE
III the early years of the French Kevolutiun, the economic reguliltions of the IIlIciell regime were dismantled. Pursuing laisseL-faire economic policies, the
government permilled the formation of hundreds of banks of issue. While many contemporaries and historians have vilified them, these banks do not appear to have affected price stability or created other major problems. This episode oilers additional evidence on the viability of free banking, but it also shows how such a system can rapidly disintegrate during a political crisis. © 1990 Academic Prm, Inc.
As deregulation of the financial system has progressed, there has been a revival of interest in free banking. However, free banking episodes are quite unusual as governments rarely refrain from intervention. There are, thus, only two well-studied cases, the antebellum United States and Scotland before 1844. Unfortunately, these two experiments do not answer all questions about the performance of an unregulated banking system. I This paper analyzes a forgotten episode in the early years of the French Revolution and offers .. ew evidence on how quickly a stable, free banking system can arise, even in the midst of a political upheaval. This episode is also important for the history of the French Revolution because the suppression of free banking was a key component in the * I have benefited from comments by Florin Aftalion, Michael Bordo, Philip Cagan, Forrest Capie, Michael Edelstein, G. J. Ellis, Stanlt:y Engerman, Hugh KockolT, Anna Schwartz, Lawn:nce H. White, and two anonymous referees, and from seminars at the City University Business School and the Washington Area Economic History Group. I thank Marcia J. Anszperger for technical assistance. Financial support from the National Science Foundation Grant SES-86-06557 is gratefully acknowledged. I The term free bllllkillg has been used to describe 19th century systems where there was relatively easy entry, no central balik, and multiple banks of issue. Although the Icv.:! of regulation was low, they were not wholly unregulated. Governments intervened, in varying degrees, to regulate entry, require the purchase of government bonds, and set the denominations of notes. 251 0014·4983190 $3.0() Copyright
All
ri~hh
~
IWO by Academic
Pn..:~~,
of n:prudul.:lion in any furm
luI.:.
rc~c .. "cd.
Free Banking II
252
EUGENE N. WHITE
transition from the early revolutionaries' laissez-faire policies to the command economy of the Terror. The French experiment with free banking has long been dismissed by historians as a failure. Their conclusions have relied heavily on contemporary complaints of counterfeiting, fraud, and inflation. These criticisms coincide exactly with two common concerns about unregulated banking that Friedman and Schwartz (1986) have identified. The first ohjection to the free creation of bank notes and deposits is that adequate market mechanisms will not arise to enforce contracts to pay, allowing extensive fraud and counterfeiting. The second is that there is the fear that unregulated issue of inside money will endanger price stability. Whereas most recent research (Rockoff 1974, 1975; Rolnick and Weber, 1983, 1984) has focused on the first question, the French experience with free banking permits a close examination of both the fraud and inflation issues. During the French Revolution, there was no lender of last resort and outside money was a fiat currency issued by the government. Each bank was free to issue its own currency and there were no limitations on investment or lending. Newly collected data on the number of banks, their note issue and regional prices show that inflation during the revolution was not a consequence of free banking. A high level of illiteracy made counterfeiting a problem, but it was not so severe as to upset banks' ability to redeem their bank notes. What devastated the banking system were the sharp shifts in government policy that agitated the pUblic. Once confidence in private currency was lost, it was easy to justify closing the banks. The elimination of the banks may seem like a senseless act; nevertheless, it increased the government's chief source of revenue, the inflation tax, by raising the demand for the government's currency. I. THE ORIGINS OF FREE BANKING
Laissez-faire banking that appeared during the French Revolution was a consequence of the revolutionaries' failure to balance the central government's budget. Unable to borrow, the government was forced to cover the deficit by issuing the assignals. In the initial decree that created them on April 17, 1790, assignats were designed to be interest bearing notes backed by the future sale of the church's confiscated lands. However, increased financial urgency forced the government to declare the assignats legal tender, converting the issue of September 29, 1790, into fiat money and abolishing interest on the first issue. A severe shortage of small denomination media of exchange quickly appeared. Creeping inflation and the uncertainties of the revolution had caused most coins to be hoarded, and the assignats were only printed in large denominations, convenient for government expenditures. Until January 1792, the lowest denomination was 5 Iivrn, and these entered circulation very
235
236
Free Banking II FHEE BANKING
253
slowly.2 The government ordered the minting of additional coins using church bells to alleviate the shortage, but the mints were unable to supply an adequate amount (Colson, 1852, p. 263; Bouchary, 1941, p. 10). The shortage of small denomination media of exchange created significant economic dislocations throughout France as the cost of everyday transactions rose. Employers attempted to pay groups of workers with 50-livre assignats. Confronted by bakers and other merchants who refused anything but exact change, workers went on strike (Bruneau, 1902, p. 313; Carraz, 1975, p. 41). in many parts of the country, premia were offered on small denomination assignats (Hubrecht, 1939, 294-298; Archives Nationales AD VIII 4; de Cardenal, 1912, p. 45). While the central government was preoccupied with other matters, many towns, merchants, and manufacturers solved the problem by establishing caisses ptllriotilJlles (patriotic banks) where assignats could be exchanged for smaller denomination notes issued by the caisses. 3 This currency known as billets de conjiallce (bills of confidence) grew very quickly. The caisses patriotiques flourished with relatively little government interference because the revolution had eliminated many of the old regime's business regulations and generally embraced laissezfaire. The billets de COl/fiance were given legal status when, on November 8, 1790, the Finance Committee of the Assembly refused to make them legal tender, declaring that they had the same status as bills of exchange. Although there were some attempts to place all caisse.!>' patriotiques under the supervision of local authorities, they failed. When the National Assembly finally passed legislation governing the caisses in May of 1791, it decreed that all privately issued notes of 25 livres or less ami exchangeable on sight against assignats or copper coin were to be exempt from the stamp tax. The assembly reaffirmed that there was no government guarantee for the solvency of the caisses and that the hillt'ls were not legal tender (Bouchary, 1941, pp. 11-14). This gave entrepreneurs and municipalities a green light to organize (·ai.l'.\'e.\· patrioliqlle.\'. II. THE CHARACTER OF THE CAISSES PATRIOTIQUES.
The pent-up demand for small notes led to a rapid growth in the number of caisse.\' palrio/iqlles. Unfortunately, any description of their activities 2 A 5·livrc note was far too large fur most of the popUlation's (hlily transactions. According to Rude (1959, p. 21), the daily wage for an unskilled worker was 20 to 30 sous and for a skilled worker 50 sous. Hence, a 5-livre note represented 2 to 5 days wages for the vast nuuority of the workers. One livre was equal to 20 sous and each sou was worth 12 deniers. J Th'!se local initiatives had a parallel in the lucal political initiatives known as the revoilltion mlllliciptlie. Early in the revolution, new local guvernments and militias were formed to fill the power vacuum created by the cullapse uf the tlllciell regime. Like the cai.l'St's patriotiC/lies, these new entities were recognized asjilitl· accomplis by the National Assembly (Lefebvre, 1962, Vol. I, pp. 125-127).
Free Banking II 254
EUGENE N. WHITE
has been hampered by a lack of data. The government's withdrawal from regulation meant that it did not collect information on the caisses. Only later when it began to have second thoughts about free banking did official data collection begin. The most complete survey (Bloch, 1910) was ordered by the Minister of the Interior in June of 1792. Although the minister had asked the directors of the departements for detailed reports on the assets and note issue of each caisse. he received very little information as banks refused to cooperate with local officials and local officials ignored instructions. In spite of the absence of official records, the number of caisses in each departement can be fairly accurately determined thanks to the painstaking research of French numismatists interested in tracking down all known billets de cOl/fiance. The first and foremost of these works is Achille Colson's article in the Revue Numismatique (1852). This survey has been extended and corrected by later numismatists. Using their figures and information from various regional history journals, the number of caisses in each departement and some data on note issue has been collected (E. N. White, 1987a) and is presented in Table I. The assembled figures are for the summer of 1792. Their distribution reveals how quickly banks appeared in all parts of France. 4 While the exact number of billets issued will probably never be known, the peak total issue is estimated at approximately 140 million livres, significantly higher than previous guesses of less than 100 million. 5 Many of the first caisses were established by local governments. The caisse patriotique of Tours is representative of these municipally sponsored banks. The operation of the caisse in Tours was supervised by eight commissioners appointed by departmental, district, and town authorities (Bloch, 1910, pp. 158-161). The articles of organization specified • One interesting feature of Table I is the large number of misses in relatively isolatcd departemellts-the Ardeche, Aveyron, Dordogne, Dr6me, Gard, Lot, and Lot ct Garonne. These areas were not known for their financial sophistication. The establishment of a large number of small caisses. including many in very small towns. reveals thc extent of thc severe shortage of small denomination media of exchange. Even towns without mllch commerce with the olltside world were forced to establish a caisse as a consequence of the sharp rise in transactions costs. , The most widely cited earlier estimate is found in Harris (1930, p. 26). The estimate of 140 million was ohtaincd in the following fashion: In most cieparte",ellts it was common for there to be onc large, dominant caisse and many slTIall competitors. In the 36 dt'pmtemellls where there was information on the large cai.~se and many of the small cais.I'('s. the issue of any caisse without data was estimated by the average size of all known small ('aines. For the remaining 46 deplirtemellts where therc was no information on the large (,lIi,fses. an estimatc of their issue was ohtained applying a rcgression of thc note issllc of the largest caisses on demographic variahles for the 36 liepmtemellts. The avcrage isslle of all known small cai.fses was also used to estimate the issues of the small unknown misses in these departemellls (sce E. N. White, 1987a).
237
238
Free Banking II FREE BANKING
255
the details of the design and printing of the billel.\· de cOlljiallce and required them to be signed by four commissioners. The billels and assignats of the caisse were stored in a strongbox in the possession of the receveur (tax collector) of the Tours district. The three keys necessary to open the lock were held by secretaries of the depariemelll, the district, and the town. The cashier was paid 1200 Iivres per year, and he was required to post a surety bond of 15,000 livres. The caisse was open three days a week for business. Its expenses were met by a sliding scale of fees for exchanging assignats for billels, from 4% on notes of 500 livres to Y2% on notes of 5 livres. Any profits from the operation were to be distributed to the poor. This carefully designed organization, aimed at protecting the integrity of the billels de cOlljiatlCe, was typical of many misses pairiotic/lles. However, if all caisses had been like the cah'se of Tours, they would have provided an essential service but they would not have given any reason for the growing chorus of complaints about fraud and speculation. As 100% reserve banks, the caisses palrioliqlles would merely have offered transactions services for a fee to the general pUblic. They appear to have been much like the 100% reserve banks proposed by Friedman (1959, pp. 65-75). Like his banks, creating deposits as they receive reserves, these caisses simply exchanged billels for assignats that were held in strongboxes under the watchful eyes of local officials. If anything, the wisses would have contributed to a slight contraction of the money supply by running small surpluses from accumulated fees. Nevertheless, although most wisses palrioliqlles began as 100% reserve banks, many quickly became fractional reserve banks.!> The freedom to open a bank proved to be irresistible to many citizens. In addition to the caisses organized by the towns, manufacturers, theaters, bakers, printers, grocers, and many others opened their own caisses. Even many municipal caisses were transformed into fractional reserve banks. In Caen, this happened when the government of the cieparlemelll requested a loan (Beranger, 1910, p. 389). The caisse in Rouen (Beranger, 1908, p. 3(4) was begun as a joint stock bank with an initial capital of 270,0()() livres held by 216 stockholders. Although a private institution, its charter included safety precautions to protect its note issue similar to those of the caisse in Tours. The balance sheets of this bank Crable 2) and the Caisse Patriotique de Paris, one of the two largest banks in France (Table 3) show that the caisses operated much like banks of the period in Britain and America. Most of their investments were in bills of exchange and securities. These liquid assets were what most banking theorists believed were appropriate • Hayek (1978, p. 45) has suggested that laissez-faire banking could be introduced today by beginning with 100% reserve banks and then letting the system develop of its own accord.
239
Free Banking II EUGENE N. WHITE
256
TABLE I The Number of Cainn Palrioliqlln and Their Known Note Issue -_._---"---
Deparlemenl
-----~-----------.--------.--
Total caisses
Caisses with known issue
..--- .. _--, .. _......_----_.
Total known note issue (livres) • 0 - _ _ _ _ _ _ _ _ _• _ _ _ _ _
Ain Aisne Allier Alpes Basses Alpes Hautes Ardeche Ardennes Ariege Aube Aude Aveyron Bouches du Rhone Calvados Can tal Charente Charente Inrerieure Cher Correze Cote d'Or Cotes du Nord Creuse Dordogne Doubs Drome Eure Eure et Loir Finistere Gard Garonne Haute Gers Gironde Herault lIIe et Vilaine Indre Indre et Loire Isere Jura Landes Loir ct Cher Loire lIaute Loire Inrericure Loiret Lot Lot et Garonne Lozere Maine et Loire
22 32 14 10
13 74 II 7 2 7 58
41 26 33 4 29 9 9 5 7 12 68 5 82 18 IJ 13
46 43 41 18 37
12 17 7 24 6 4 12 12 2 9 53 51 28 6
18 0 0 I 0 74 I 0 2 4 0 I
22R,O(){)
1,2()()
194,51>5 444,512 722,()()() 6(',O()() 38,340 1,326,9RO
0 0 ()
I
60,000
()
4 3 0 5 0 0 0 0 0 0 0 II 0 0 I
7 I
1,057,426 36,(J()(J 305,250 95,O()()
9,285,692
IO,O()() 830,(J()(J 31O,O(J()
0 0 0 ()
0 3
I,052,70()
()
0 ()
0 Continued overleaf
Free Banking II
240
257
FREE BANKING TABI.E I-Continued
IJepariemetit Manche Marne Marne Haule Mayelllle Mcurlhc Meu,e Morbihan Muselle Nievre Noru Ois..: Orne Paris Pas ue Calais Puy uu Dume Pyrenees Basses Pyr..:nees Haules Pyr..:nees Ori..:nlales Rhin Bas Rhin lIaul Rhon el Loir..: Saon..: ..:1 Loire Saon..: Haule SarI he Seine el Marne Seine elOise Seine Infer-ieure Sevres Deux Somme Tarn Var Venuee Vienne Vienna Haule Vosges Yonnc TOTALS
Source. E. N. While (l9117a).
Tolal caisses
Caisses wilh known issue
III 6 II 3 12 9 12 3 14 34 26 94 64 25 27 6 5 I 2
0 4
2 13 19
()
0
30 15 7 21 9 III 33 13 14 13 4 15
()
Tolal known nole issue (Iivres)
1,1165,141
()
2 5
1,355,000 264,509 79,000
() ()
IIO,O()O 1,431,161 1,629,412
()
3
31,159,1112
() ()
0 ()
0 1411,000
I
o 224,310
()
4 3 3 2 II
72,529 40,000 !!5,411 11,666,626 117,000 2,735,510
0 94,000 106,000
J()
0
16 13
699,665
II
3 6 2
1,666
227
67 ,(H)7 ,251
70,(H)O 50,5(H)
241
Free Banking II
25R
EUGENE N. WHITE TABLE 2
Balance Sheet of the Cai.fu Pa/rio/ique of Rouen June 6, 1792 (in Iivres) Assets Assignats held as reserves Bills of exchange Unissued and returned notes Other
Liahilil ies 2,203,576 6,126.629 681,706 247.961
Stockholders' capital Surplus Chief officer's bond Notes printed
9,259.872
527,850 218,582 20,()()() 8,493,440 9,259.872
Source. Adapted from Beranger (1908), p. 311.
investments. Their reserve ratios of assignats to billets, 26 and 29%, respectively, suggest conservative management that could handle any demand for note redemption, short of a panic. While balance sheets are quite scarce, the remaining ones provide a picture of traditional banking, not one of speculative or predatory financial activities. The fractional reserve banks operated very freely. They competed keenly with the 100% reserve banks, frequently providing superior services. In one town, the municipally sponsored caisse was open three days a week and charged 1% on all exchanges. Its competition was a merchant's caisse that offered to exchange notes for free and was open every day (Beranger, 1910, pp. 342-343). It is thus not surprising that the fractional reserve banks grew much faster and soon dominated the 100% reserve banks. The liberty enjoyed by the banks was virtually complete. Perhaps the most extraordinary development was the appearance of private issuers of coin. Just as the shortage of low denomination paper money led to private issue of notes, so did the shortage of coin open up opportunities for private enterprise. There were nine private issuers of coin in Paris and two in Lyon (Colson, 1852, pp. 447-448). The most well-known of these were the Monneron brothers (Bouchary, 1941, pp. 194-205). Their caisse exchanged 2 and 5 sou copper coins for assignats. These high TABLE 3 Balancc Shect of thc Ct/i.ul' /'a/rio/ique ele /'aris April 3, 1792 (in livrcs) 5,062.5IK) Assignats hcld as rcscrvcs 8,390,150 Bills of exchange Securities on deposit with the city 4,670,614 4,926,715 Securities 1,445,800 Other 24,495,779
Capitlll lind surpills Bank notes in circulation
6,104,1«.7 17,660.812
24,495,679
Source. Adapted from Bouchary (1941). p. 70. The discrepancy betwecn assets lind liabilities is not IIccounted for by Bouchary.
Free Banking II
242
FREE BANKING
259
quality, non-full-bodied coins circulated widely. Other caisses struck a wide variety of copper, billon, and silver coin. During the first year of their appearance, all types of caisses patl"iotiqlles were quite successful. Although they were not legal tender, the billets de confiance often moved beyond the towns and departemellIs of origin. Yet, in spite of the considerable success of the ClIisses patriotiques in solving the shortage of low denomination media of exchange and in gaining the confidence of the public, they gradually came under a cloud of suspicion. III. THE ATTACK ON FREE BANKING
Popular discontent focused on both the issues of fraud and price stability and eventually turned the public against free banking. When the cai........ es patriotiqlles were closed many people concluded, as did the Societe des Nomophiles, that "in a well-run state, the issue of notes must not be shared by the sovereign government and private individuals.,,7 Historians have accepted without dissent the arguments of these critics. Achille Colson (1852) blamed the great variety and overissue of notes for causing capital flight, counterfeiting, and depreciation of the exchange rate. Marcel Marion, author of the standard French financial history (1914), concluded that the caisses patriotiqlles were engaged in speculation and defrauding the pUblic. In 1'he A ........ ignats (1930), the classic study of the revolution's finances, Harris seconded these criticisms. Although Bouchary (1941), Horsin-Deon (1958), and Carraz (1975) are not as critical, they still blame excessive note issue for the rise in prices. Most recently Aftalion (1987) has described the experiment as a national plague fraught with deceit and scandal. While a consensus exists among historians about the evils of free banking during the revolution, there is scant evidence to support their conviction. It is important to note that the National Assembly and Legislative Assembly and their finance committees heard all of the complaints cited by historians but found little merit in them for a long time. Only after political upheavals and rising inflation gave the upper hand to the radicals did the Legislative Assembly and the Convention move to end the experiment in free banking. An assessment of the merits of free banking thus requires a close examination of the twin problems of enforcing contracts to pay and price stability.
A. Counterfeiting, Fraud, and Wildcatting The breakdown of law and order made countelfeiting during the French Revolution a persistent problem. The central government waged a long struggle against counterfeit assignats, but in 1791 and 1792 the problems 1
Quuted in Bouchary (1941, pp. 16-17). Translation of the author.
Free Banking 1/
260
EUGENE N. WHITE
confronting the caisses patriotique.~ were graver. Although impossible to gauge its extent, the tales of bold and brazen counterfeiting suggest that it was widespread. The billet.~ de cOlljiallce were easier to counterfeit because they were often less well-designed and printed than the assignats. The large variety of notes in circulation apparently made detection difficult. Counterfeiting was not limited to producing reasonable facsimiles of real notes. Sometimes the numbers would be altered so that 2 would become 20 or 3 would become 30 (Colson, 1852, p. 373). The low level of literacy, where only 47% of married men and 27% of women (Furet and Ozouf, 1982, p. 26) were able to sign their names, may have permitted crude methods of deception. In one instance (Bouchary, 1946, p. 82), low-denomination notes printed on white paper were dyed red, the color of a higher denomination. Counterfeiters also printed notes of nonexistent caisse.v and circulated them in distant departements. These notes were passed to unwary illiterate peasants, and officials testified (Bloch, 1910, p. 180181) that it was often dangerous to refuse receipt of a note. The caisses patriotique.y did not, however, stand by idly. The Caisse Patriotique de Paris published a table of all its billets and advice on how to avoid accepting counterfeits in the Parisian newspapers (Bouchary, 1946, p. 82). The caisse patriotique in Rouen printed tables of its notes, explaining the types of paper and ink used and giving a list of all the signatures (Beranger, 1908, pp. 288-300). Many other cai.~ses took similar measures to assist and protect the public. These methods seem to have deterred some counterfeiting. More extensive anticounterfeiting measures, like the antebellum American bank note reporters (Rockoff, 1974), did not appear, perhaps because of the brevity of the experiment. In contrast to the French episode, Scotland's experience with free banking seems to have been more successful in avoiding counterfeiting. This may be partly explained by a higher level of literacy and by the exclusion of the less literate classes from using bank notes because of the I-pound minimum denomination. 8 The Scottish prohibition oflow-denomination notes was a consequence of the "small note mania" of the early 1760s when numerous little banks appeared issuing very low denomination notes. There were many complaints about the excessive and varied note issue. Among the opponents of small bank notes was Adam Smith who believed that no bank notes should be issued for less than 5 pounds. Describing problems in Scotland that sound remarkably like those in revolutionary France, Smith argued that:
• Adult male literacy in Scotland is thought to have been about 90% at the end of the 18th century (see GralT, 1987, p. 246).
243
244
Free Banking II FREE BANKING
261
Where the issuing of bank notes for such very small sums is allowed and commonly practised. many mean people are both enabled and encouraged to become bankers. A person whose promissory note for five pounds. or even for twenty shillings. would be rejected by every body. will get it to be received without scruple wllt!n it is issued for so small a sum as a sixpence. But the frequent bankruptcies to which such beggarly banks must be liable. may occasion a very considerable inconvenience. and sometimes even a very great calamity. to many poorest people who had received their notes in payment. (Smith. 1776. p. 3(7)
In 1765, parliament listened to this and other complaints and forbade Ihe issue of notes below I pound (Checkland, 1975, pp. 194-105; L. H. White, 19M, pp. 29-30). Limiting the issue of bank notl!s to the educated and financially sophisticated is one, albeit extreme, way to solve this information problem. It is certainly difficult to justify this solution as thl! extl!nt of fraud committed by "beggarly bankers" in Scotland remains unknown. Only in the United States where fraudulent banks were known as "wildcats" is there some reliable evidence. The practice by banks of issuing mon: notes than they could redeem was thought to be relatively common in antebellum America. However, recently Rolnick and Wl!ber (l9H4) havl! shown that wildcatting was a relatively rare phenomenon. Rockoff (1974) and Rolnick and Weber (l9H3) have measured the losses to noteholders of failed banks and found that they were smal1. 9 The beggarly bankers were very numerous in France, but, surprisingly, there were only three alleged instances of wildcatting banks (Colson, IH52, p. 273; Harris, 1930, p. 26). It is not possible, as in the American case, to measure the losses to noteholders. Yet, the negligible' number of failurl!s up to mid-1792, out of the over 160() cain'es patrioliqlles established, suggests that actual losses to the public were quite small. When the clIisses were forced to close in 1793, only a handful of the non-Parisian banks could not fully repay their noteholders. This was true even when liquidators received more billets-presumably counterfeitsthan had been originally issued. By this measure, the problems of fraud and counterfeiting were relatively minor. The same cannot bl! said about Paris where in 1792 there was one spectacular bank failure and other banks found themselves in trouble. Noteholders of two banks would have lost substantial sums if the government had not intervened to cover the banks' liabililies. However, the lroubles of these Parisian banks, as argued later, were principally the result of government actions that panicked the public rather than any malfeasance by bankers. • According to Rockoff (1974. p. 150). losses to note holders from failed banks for the whole free banking period amounted to $1.9 million. the equivalent of a 2% inflation in 11!60. Rolnick and Weber (19113. p. 10119) calculate total losses at $1.6 million.
Free Banking II
262
EUGENE N. WHITE
B. Price Stability The other great fear was that an excessive issue of billets de cOlljiallce was driving up prices. Many believed that the caisses pnlrioliqlles were directly buying up and hoarding essential commodities.1O Although the assembly generally found no substance in these complaints, some in the government eventually began to listen. In June of 1792, the Minister of the Interior blamed the caisses patriotiques for the rising price level: ' Historians have accepted this charge against the caisses. but there is little supporting evidence. One bank that was singled out for attack was the Caisse Patriotique de Paris, one of the two largest caisus. Initially its operations had been widely praised (Bouchary, 1941, pp. 22-25), but by late 1791 it came under attack in the popular press for speculating in commodities. Its managers took the charge quite seriously. They placed a notice in newspapers denouncing these charges and affirming that the bank's principal business was discounting bills of exchange. The caisse also offered a reward of 100,000 livres to anyone who could prove these allegations. No one came forward to claim the reward, and the bank's reputation remained intact. Likewise, solid evidence to substantiate the charges of speculation by other caisses has never been presented. As this notable example suggests, the rising price level may not have been the fault of the caisses patriotiques. Proponents of free banking Hayek (1978) and L. H. White (1984) have argued that laissez-faire banking should actually help to stabilize the price level. In White's model of free banking under a gold standard, a bank may in the short run issue an excessive quantity of notes, but it will be unable to keep them in circulation in the long run. When an individual finds himself with more notes of a particular bank than he desires he will reduce his holdings by (I) redeeming notes directly with the issuing bank, (2) depositing notes with his preferred bank, or (3) passing them on to other individuals. The overexpansive bank will be forced sooner or later to contract its note issue. This will happen directly if the notes are redeemed either by the public or by the banks at which they have been deposited. The overissue may increase spending, as the third case implies, but this will cause the region where its notes circulate to lose specie as imports from outside the region increase. The overexpansive bank will bear the brunt of the drain in reserves. It is possible for all the banks in a region to overissue. but this would eventually be checked by a loss in reserves. 10 In addition to blaming the caisses for speculation, one pamphleteer accused them of buying up foreign bills with billets, thereby driving up the exchange rate (see Botlchary. 1941. pp. 18-19). 11 Between January 1791 and June 1792, prices increased approximately 50% in Paris (E. N. White, 1989. Appendix I).
245
246
Free Banking II FREE BANKING
263
France was on a fiat standard, but White's analysis can be applied as the assignats served as outside money instead of gold. An inflationary impulse from the overissue of billels is ruled out in the long run because the assignats would be redeemed, but it is entirely possible in the short run. That is precisely what the enemies of the clIisses plIlriolhllll'S claimed was happening. They were afraid that in the short run the clIisse.\· were defrauding the public, overissuing notes to buy up scarce goods, and driving up the price level. Whether the caisses would be checked in the long run was immaterial to them and irrelevant to the political debate. The question is, thus, whether the clIisse.\' palriuliqlles quickly expanded their billets to gain some quick profits in a politically unstable climate or controlled their note issue in the short run in order to maintain their reputation (like the Caisse Patriotique de Paris), ensuring their success in the longer run. If the former is true, then the radical revolutiollaries were correct and the episode provides strong evidence against the desirability of frt:e banking. If the latter is true, the French experiment shows that even in the most adverse conditions market incentives not only produce price stability in the long run but also the short run. Measuring the intlationary effects of the billels de cO/lfill/lce presents many problems. Inflation in revolutionary France varied considerably over time and from region to region. Ideally, one needs cross-section time-series data on the issue and circulation of the billels, assignats, bank reserve ratios, prices, and other economic variables. Only some of these data were collected and preserved. Fortunately, the central government ordered the local authorities to create price indices, allowing intlation to be tracked in each depllrlemelll. 12 Information on the issue and circulation of currency is much more limited and presents a major problem for an analysis of the inflationary role of the ('(/i.\'ses /JlIlrioli£/lies. It is no simple task to determine the quantity of money in circulation in each (/epllriemelli. While the billels de cUllfillllce tended to circulate in their areas of origin, the assignats were a national currency, moving freely across economic regions. The quantity of assignats in circulation had been rapidly growing. In January of 1791 there were 524 million livres of assignats in circulation. By the end of the year, this had grown to 1,369 million livres. 13 The total issue of billcis dl' cO/lfilll/cC was much more modest, reaching a maximum of approximately 140 million livres. Given that this is small relative to the total stock of money, it might be tempting to dismiss the inflationary effects of the billels immediately. 14 .2 These regional price indices are presented and analyzed in E. N. White (19119). " For estimates for the stock of assignats in circulation see E. N. White (19117b) . • 4 The bank notes issued during the free banking era in the United States formed a larger fradion of the lIloncy stock, In IK45, total specie in the United Statcs was $'17 million,
Free Banking II 264
EUGENE N. WHITE
But. they still might have had a significant effect on prices if there was a difference in the velocity between the bilfets and the assignats. This should not be ruled out as the hillets de c01~fiance were the dominant currency the public used in its daily transactions. There are no data for the quantity of billet.~ de conjiallce in circulation in each economic region or departement. What is available is the estimated total note issue in some departements derived from the actual note issue in Table I. The data presented in Table 4, along with the total number of caisses, represent the estimated peak issue of billets, attained during the summer of 1792. 15 The figures are only for the 38 departemellts where there is sufficient information to put together a good estimate of note issue. Using the quantity of billets issued in a departement as a proxy for the quantity in circulation in that departement might be considered a heroic leap of faith. Yet. contemporary sources suggest that most billets were used only locally and did not travel very far. This is not surprising as small caisses would be unlikely to have reputations that would encourage a wide geographic acceptance of their notes. The obvious exceptions were the notes of the Caisse Patriotiqlle de Paris and the Maisoll de Secours, the two biggest Parisian caisses, that had national reputations. However, even the billets of these caisses tended to remain in their departement of origin. When the Maison de Secours was liquidated. only 7.3% of the notes redeemed after January 9, 1793. came from departements other than Paris (Bouchary, 1941, p. 13\). This was relatively late in the redemption process and it is likely that a greater proportion of these notes came from afar compared to those that were redeemed earlier. Thus, this is probably an upward bound on the percentage of notes moving beyond a caisse's home departemellt. The quantity of billets issued within a departemelll may thus closely approximate the quantity in circulation. What is unknown is the quantity of assignats circulating in each deplIrtement at any time. Nevertheless, this may not be vital to determining the effects the billets had on the price level. The assignats and the billets while bank notes amounted to $105 million and deposits $39 million (Friedman and Schwartz, I97(), Table 13). Banking in France was less developed than in the United States. The nneil'n rcll!illll' had only chartered one bank, and free banking had only 2 years of growth. Although the hil/l'ls were only a small component of the money stock, they were controversial as are many financial innovations. " The data in Table 4 is derived from Table I. Only the deparlemenls where there was complete or nearly complete data on issue were used. When a few small C(li.Ul'S in a depnrlemen' had no reported issue, 5()()() livres per missing caisse were added. Thus. the Ain with 18 of 22 misses reporting has its total issue raised from 228,O()() to 24R.()()() livres. Alternative adjustments did not alter the results, apparently because of the small quantity of missing data (see E. N. White, 1987a).
247
Free Banking II
248
265
FREE BANKING TABLE 4 Caisses Pulrioliques and Their Estimated Note Issue by Deparlement
Deparlement Ain Ardeche Ardennes Aube Aude Calvados Cher Cote d'Or Cotes du Nord nordogne Doubs Gironde Indre Indre et Loire Isere Loire! Marne Mayenne Meurthe Meuse Nievre Nord Oise Paris Rhin Bas Rhin HaUl Saone el Loire Sarlhe Seine et Marne Seine elOise Seine Inferieure Sevres Deux Somme Var Vendee Vienna Haute Vosges Yonne Totals
Source. E. N. White (l987a).
Total caisses 22 74 II 2 7 26 9 5 7 68 5 18 17 7 24 9 6 3 12 9 14 34 26 64 2 0 15 21 9 18 33
Estimate I
II
248,000 194,565 582,742 722,000 107,469 1,672,555 170,584 1,(162,426 91,292 1,176,099 150,292 9,320,692 231,168 830,000 627,929 1,082,700 1,892,787 1,360,000 299,509 189,584 245,876 1,721,444 1,664,412 32,003,015 153,000 0 417,832 307,520 150,584 292,756 9,081,316 239,053 2,750,510 109,000 299,522 879,364 105,000 174,907
663
72,607,504
13 14 4 15 16 13
Free Banking II
266
249
EUGENE N. WHITE
de cOIzjiance were not perfect substitutes because of their denominatiorwl differences. The absence of low-denomination assignats was, of course, what had called the billets into existence. Omitting the assignats, whose departmental circulation cannot be measured, the demand for billets can be examined as a component of the demand for the media of exchange. It; The per capita demand for real billets balances in the ith departe111ellt (M1) was specified as
M/ = flTAXES
i ,
UP i , LlTi ) +
ei'
(I)
For the variables explaining the differences in demand across ciepartements, one expects that the higher the level of income as proxied by the total real direct taxes per capita in 1791 (TAXES;), and the higher the percentage of urban population (UP;), the greater the demand for billets. The literacy rate for married men (LIT;) was included to test whether the illiterates had a greater willingness to accept and hold billets de cOIzjiance, as critics claimed. 17 It would have been preferable to have had cross-section time-series data for all (!epartements rather than just partial cross-section data on note issue. This would have permitted the use of regression techniques to control for the spillover of notes from one departemelll to another. Thus, the estimation of the demand for billets and their inflationary effects are simple approximations. Three. versions of the demand for real billets balances model were fitted. The first and second excluded Paris, while the third included it. IR In the ordinary least-squares results reported in Table 5, the demand for money was fairly well identified. Both the proxy for income (TAXES) and the urbanization variable had coefficients with an appropriate sign and magnitude. Although it was not possible to reject the hypothesis that most coefficients on TAXES and UP were different from zero at the 10% level, the variable for literacy appears to be far weaker. The very small positive coefficient offers no support to the idea that illiterates accepted billets de cOIzjia1lCe more readily than literates as many historians have presumed. As the nation's financial and demographic center, Paris is an outlier for all variables. The inclusion of Paris in the third equation cau~;es a marked rise in the R2 but does not alter the estimates for the demalld for real balances.
I. The model used is similar to the one employed by Rockoff (1975). 17 Reinhard provided data on population (1941, pp. 26--28, Table IV) and the percentage of urban population and persons per square league (pp. 48-49). Total taxes was obtained from the Moniteur Universel (May 29, 1791). The figures on literacy come from the Ministere de I'lnstruction Publique (1880, pp. CLXVIII-CLXXI). " There were no data for literacy in five departements. including Paris (see Ministere de I'Instruction publique, 1880).
Free Banking II
250
267
FREE BANKING TABLE 5 The Demand for Billels de COl/fiunce
Constant TAXES
UP
Eq. one
Eq. two
Eq. three
-0.014 (-1.25) 0.127 ( 1.22) 0.213 (3.29)
- (1.(114 (- 0.97) 0.10'.1 (0.8'.1) 0.225 0.13) 0.0001 (0.01)
- 0.022 ( -2.30) 0.133 (1.26) O.2H4 (9.47)
1.1'1' Number of observations F
33 0.287 3.11'.1
37 0.286 6.112
H' Stati~tic
38 0.767 57.55
Nol.:. The numbers in parentlll:ses are I-statistics.
With a stable demand function for billets de confiance, an excessive issue of notes would have led to upward pressure on prices. In Table 6, some simple regressions of price level on note issue are reported. The department totals for note issue represent whatever data were available sometime in the summer months, and thus it is necessary to use more than one month's price level to determine whether there is any relalionship. In the first two regressions where Paris is omitted, the quantity of notes issued has no effect on inflation. The second pair of regressions, which include Paris, suggest that greater note issues may have raised price levels. This may be attributed to the fact that Paris had the caisse.l· issuing the most billets. However, Paris was also the center for the issue of outside money, the assignats. TABLE 6 Prices and Billets de Confiance
Price level Paris excluded Dependellt Varial1le Constant Note issue (thousands of livres)
R' F-statistic Numl1er of observations
Paris included
June 1792
Sept 17'.12
Jlllle 17'.12
Sept 1792
131.H (4.04) O'(lO2 ( 1.06) 0.03 1.13 37
135.t (3.59) 0.002 (1.15)
131.9 (3.69) 0.002 (2.74) 0.17 7.53 3H
135.6 (3.23) 0.002 (2.06) 0.1/ 4.25 311
O.()4
1.33 37
NOI.:. The numbers in parentheses are I-statistics.
251
Free Banking II
268
EUGENE N. WHITE TABLE 7 The Effect of the Billets de Confiance on the Price Level (Dependent variable = Price Level June 1792)
- - - - - - - - - - - - - - - - - - - - - - - - - - - _ . _ - - - .... -. Constant M -
Md
Number of observations RZ F-Statistic
Eq. one
Eq. two
Eq. three
133.8 (36.9) -55.31 ( -0.34)
133.2 (36.8) -35.55 ( -0.23)
135.3 (35.4) -7R.77 ( -0.46)
37 0.003 0.12
33 0.001 0.05
3R 0.006 0.22
Note. The numbers in parentheses are I-statistics.
A more sophisticated means of examining the inflationary effects of the billets de conjiallce is to take into account the factors affecting the demand for money. If a departemellt's note issue was excessive relative to its demand for money, then there should have been upward pressure on its price level. Table 7 reports regression results of the price level in the ith departemellt (PRICE j ) on the difference between the actual notes supplied (M j ) and the predicted demand for notes (M'J) obtained from the equations in Table 5. PRICE j = g(Mj
-
M'J)
+ ej.
(2)
There is no evidence from this exercise to support the argument that the billets de conjiallce were inflationary. Not only do the coefficients for the independent variable have the wrong sign, but it is impossible to reject the hypothesis that they are different from zero at the 10% level '9 • While there are generally no time series for the issue of billets, the record for the preeminent caisse in the departemellt of Calvados has been preserved. The monthly issue of this caisse in Caen and the price level in Calvados are presented in Table 8. A simple regression of the log of the price level (PRICE) on the log of the quantity of hillets (BI LLETS), reported as Eq. (I) in Table 9, reveals why the cllisse.f were held responsible for inflation. There was a close correlation between note issue and the rising price level. However, the growth of hillets could easily have been the result of the increase in assignats that served as reserves. The reserve ratios and other important information are unknown, but there are monthly data on the total assignats issued in France. As some of the total issue flowed into the departemellt of Calvados, the 19 These regressions used the June 1792 prices. The results lIsing the September 1792 prices are even less favorable to the idea that the billets were responsible for innation.
252
Free Banking II
269
FREE BANKING TABLE 8 Money and Prices in Caen
Date June 17'.11 July 1791 August 1791 September 1791 October 1791 November 1791 December 1791 January 1792 February 1792 March 1792 April 1792 May 1792 June 1792 July 1792 August 1792 September 1792 October 1792 November 1792 SOl/reI"'.
Assignats in circulation nationally (million livres)
Caen billets de contiance issued (livres)
Calvados price indl:x (l)ec. 1790 = 100)
983 1062 1114 1164 1265 1332 1369 1433 1502 1549 1598 1667 1706 1748 1821 1972 2082 2192
10,246 I!I ,200 26,220 65,520 155,342 192,130 322,578 487,130 772,530 1,038,419 1,154,080 1,189,543 1,226,943 1,278,643 1,295,643 I,Z95,643 1,31Z,043 1,325,958
109.0 1111.2 113.3
113.3 114.9 114.6 116.3 119.0 IZ3.1 128.2 123.8 128.6 133.8 130.3 134.2 12!1.2 12l!.Z 134.7
E. N. White (19l!7b), Beranger (1910), E. N. White (19119).
local cai.!..se~' would have expanded their note issue like any other fractional reserve banks. To separate the inflationary effects of the assignats and the billets de COl/fiance, the log of the billets of Caen were regressed on the log of the stock of assignats (ASSIGNATS) in circulation nationally.20 Eq. (2) shows that a substantial fraction of the growth of bank notes can be explained by the increasing supply of assignats. The price level was then regressed on the residuals of this equation, the unexplained issue of billets (UN EXPLAINED). These results of Eq. (3) indicate the hypothesis, that the unpredicted component of billets had no effect on the price level, cannot be rejected at the 1% level. There is, however, a distllrbingly high level of ulltowrrclation in Eqs. (2) and (3). This is partly a consequence of a dramatic slowdown in the issue of bank notes after March 1792. As discussed below, this change in behavior was the result of a crisis that caused banks like the caisse in Caen and the Caisse PlItriotiqlle de Paris to pursue more conservative 20 There is no measure of the assignats in circulation in Calvados. It is assumed that any national increase in assignats leads to a proportional increase in the assignats in circulatioll in the departmellt.
Free Banking II
270
EUGENE N. WHITE TABLE 9 Billets de Conliance and Innation in Caen June 1791 to November 1792 (I)
In(PRICE) = 4.31 + 0.039In(BILLETS) (74.1) (8.6) R Z = 0.823 N = 18
F-statistic = 74.8 Durbin-Watson = 0.90
(2) In(BILLETS) = -35.19 + 6.56In(ASSIGNATS) ( - 6.99) (9.53) RZ = 0.850 N = 18
F-statistic = 90.85 Durbin-Watson = 0.19
(3) In(PRICE) = 4.81 + 0.015UNEXPLAINED (279.6) (0.53) R Z = 0.02 N = 18
F-statistic = 0.3 Durbin-Watson = 0.17
(4) In(BILLETS) = -61.71 + 10.28In(ASSIGNATS) ( - 26.5) (31.6) R Z = 0.99 N = II
F-statistic = 998.4 Durbin-Watson = 1.95
(5) In(BlLLETS) = 11.67 (17.8) R Z = 0.73 N = 7
+ 0.32In(ASSIGNATS) (3.7)
Durbin-Watson = 0.95 F-statistic = 13.4
(6) In(PRICE) = 4.76 - O.OIUNEXPLAINED (294.8) ( - 0.10) R Z = 0.01 N=II
Durbin-Watson = 0.23 F-statistic = 0.01
(7) In(PRICE) = 4.87 + 0.28UNEXPLAINED (535.1) (0.59) R Z = 0.06 N = 7 Noll'.
Durbin-Watson = 2.53 F-statistic = 0.34
The numbers in parentheses are I-statistics.
policies. Breaking the sample at this point and estimating Eqs. (4) and (5) eliminates some of the autocorrelation. The change in the coefficients from Eq. (4) to (5) shows how the increase in billets in response to an increase in assignats was substantially reduced after the crisis. Yet. as seen in Eqs. (6) and (7), the unexplained issue of billet.f still has no recognizable effect on inflation in either subperiod. In spite of the paucity of observations, it seems clear that inflation was being driven not by unscrupulous local bankers but the excessive issue of assignats. The caiS,fes palrioliques were not wildcat operations,
253
254
Free Banking II FREE BANKING
271
their cautious note issuing practices suggest that they were primarily concerned with soundness and their future reputation.
IV. WHY THE EXPERIMENT FAILED The story of the demise of the caisses patriotiques has always been told as a consequence of their disruptive activities; however, there is scant evidence for fraudulent or inflationary banking practices. It thus remains to be explained why the government and the public were convinced that the experiment in free banking should be abruptly terminated. A close inspection of the events reveals that sudden action by government undermined public confidence in the billets de confiance, paving the way for their abolition. Traditional exposes of the evils of the caissej' patriotiques have focused their criticism on the Parisian caisses, especially the Caisse Patriotique de Paris and the Maison de Secours.21 Yet, although these two caisses were assailed by critics, the public continued to use their notes with confidence for a long time. Serious problems did begin in early February 1792 when the notes of these caisses lost acceptability in the departemellt of the Eure-et-Loir. This crisis abated when the cais:!;es quickly dispatched agents to freely exchange assignats for their billets (Bouchary, 1941, pp. 32-33; Horsin-Deon, 1958, pp. 267-268). While excessive note issue was blamed for this crisis, the immediate reason was that the government's receveurs de district (district tax collectors) suddenly refused to accept the billets for taxes. The origins of this decision have never been adequately explained, but fear of accepting counterfeits may have precipitated this action. In the previous month, a large cache of counterfeit billets of the Maison de Secours were discovered by the police in Paris (Bouchary, 1946, pp. 83-84). News that many counterfeits had already been put into circulation probably prompted the receveurs to take this perfectly legal step. Even though confidence was restored by the caisses' prompt action, the opponents of the caisses took this opportunity to press their attack. In Paris, the Legislative Assembly's Comite des finances agreed to reexamine the question of regulating the caisses patriotiques and produced a report on February 25, 1792. While initially conceding that the billets lie confiwu:e might have contributed some to inflation and speculation, the committee reported that most fears were ill-founded. The report (Archives Nationales, AD IX 495) then proceeded to argue that any government regulation would be ineffective and an infringement on the rights of property and the principles uf liberty.
" Harris (1930, pp. 24-26), whose treatment of the cail.se~· ptllri(}liC/lle~' is brief and dismissive, mistakenly believed that these two caisses were one.
Free Banking II
272
EUGENE N. WHITE
Although the report reaffirmed the legal standing of the hillets de cOIllimlce, the debate grew more bitter and partisan. As war seemed likely ami economic dislocations mounted, more memhcrs of the assclllbly listened sympathetically to the critics of the caisse.v. On March 2M, a proposal to require the caisse.v to submit to government regulation and deposit their reserves, printing plates, paper, and ink with the local authorities was presented to the assembly (Horsin-Deon, 1958, p. 270). This extreme act was not adopted by the assembly, but a panic broke out the day after it was seriously debated. The hardest hit bank was the Mai.voll de Secours. When it closed it doors, noteholders rioted and had to be driven off by the national guard. The cashier ned the bank with some funds. leaving the owners with only 60,000 livres in reserves. The noteholders advised the Mayor of Paris of the bank's imminent ruin. He appealed to the assembly. On March 30, the assemhly grudgingly voted the city a loan of 3 million Iivres to assist any troubled caisses. The assembly's quick response averted an incipient nationwide panic. but the assembly was now convinced that regulation was necessary. On the same day as it granted the loan, it ordered the municipalities to examine all caisses and forbade any further note issue except by government-supervised caisses. The disintegration of the central government's authority meant that this decree went largely ignored. In fact. there was a boom in the number of new nominally government-supervised caisses. It is important to note that many caisses weathered these troubled times quite well. The Caisse Patriotique de Paris trimmed its sails. reduced its note issue from 27 to 17 million livres. and published its balance sheet. The government might have wished to terminate the cai.vses patriotique.v, but it ,:ould not do so during the summer of 1792 as few lowdenomination assignats had been printed and coins were still hoarded. What the assembly failed to learn from the events of March was the effect of any of their actions on public confidence. On August 22. the assembly decreed that all bearer notes were subject to the registration tax (droit d' enregistrement) (Horsin-Deon, 1958, pp. 281-282). Although it was uncertain whether this applied to the billets de cOI!fiallce, it prodllced a panic in Paris until the assemhly made it clear that bilkts were not included. Popular unrest in Paris erupted in the jOllrnee.v of August and September. pushing the assembly and. after September 20. the Convention in a more radical direction. On September 3, the private issue of coin was forbidden. 22 On September 14. the assembly accepted a plan to retire the Parisian billets de conjiance. The public was put to the great inconvenience of long lines; but apart from the Maison de Secollrs, only one 22
Some privately issued coins continued to circulate afterward (Caron. 1912. p. 215).
255
256
Free Banking II FREE BANKING
273
other caisse required government assistance to reimburse the public. The Caisse Patriotique de Paris finished its exemplary record by quickly retiring its entire issue, leaving a large surplus that was distributed among the shareholders (Bouchary, 1941, pp. 72-73). The convention took the final step on November 8 and voted to liquidate all caisses. Examination of the ca;sses and retirement of the notes was to take place by January 1, 1793. While the gradual note redemption had worked moderately well in Paris where a supply of low-denomination assignats was ensured, the provinces bitterly complained about the new redemption plan. There was a dearth of assignats of lesser value and the public had considerable difficulty in returning the variety of notes they held to the issuing caisses. This forced the convention to continually grant delays before all billets de culljiallce were finally declared worthless. Note redemption dragged out for months. In Caen, for example, only 71% of the notes that had been in circulation on November 8 had been retired by mid-August 1793 (Beranger, 1910, pp. 336). Most billets were redeemed for assignats by the end of the year, but some notes continued to be exchanged well into 1794. This huge clearing operation had many problems, some counterfeits were accepted and notes coming from other departemellts clogged the mail. But, in this undertaking, it is remarkable that very few of the over 1600 caisses required government assistance to liquidate. This achievement offers further evidence for the essential soundness of the (:a;sses patr;otiques. Most historians have assumed that when the government closed down the ca;sses patriotiques it acted in what it believed to be the best interests of the public-protecting it from bank notes of dubious quality. Yet the government actions may not have been so high-minded. By late 1792, well over half the government's expenditures were being covered by the issue of assignats. The success of such an inflation tax depended on the willingness of the public to hold money as prices rose. If the government could find some way to increase the willingness of the public to hold assignats, the inflation tax could be increased. 23 Forcing the retirement of the billets de confiance caused people to increase their holdings of assignats, thereby promoting the government's financial strategy. While this was not the stated goal, the benefits to government coffers were quite real. A simple measure of the tax is the increase in the stuck of assignats divided by the average price level: (M, - M, 1)/O.5(PI + PI_I). This index, using Paris prices, is shown in Fig. I for July 1791 to
2J One principle of "infiationary finance" is that "the infiatillll that results from a given government deficit will be lower than otherwise if the availability of close substitutes for government money is curtailed" (Nichols, 1974, p. 423).
257
Free Banking II
274
EUGENE N. WHITE 120,--------------------------------------------
90
•.e
~
'0
80 70
•c
80
E
50
~
40 30 20 December
July
1791
1791
FIG. I.
JUly
1792
December
1792
Revenue from inflation: July 1791-December 1792.
December 1792. 24 The big jump in government revenue occurs in late August and early September. This is when the plans for closing the caisses in Paris are first revealed. The forced shift from hillet.f de COIIfiallce to assignats, thus, produced a significant increase in the innation tax for the government. V. CONCLUSION
The experiment with free banking during the French Revolution, although brief, provides additional evidence on the success offree banking, France welcomed both large sophisticated banks in Paris and the "beggarly bankers" in the countryside. Counterfeiting may have been a problem, but however numerous counterfeits were, they did not frustrate the final reimbursement of noteholders. The French experience also yields some evidence on the eITects of free banking on price stability. noth cross-sectional and time-series data provide little support for the traditional view that the clIis.H'.v pat,.iotiC/lIl·.\' overissued bank notes. The vast majority of cais.ves were not wildcat banks attempting to make a quick fortune. They were sound banks interested in establishing a reputation for prudence. Although the government was loath to admit it. the principal generator of inflation was the rapid issue of assignats. The French experiment does not oITer a definitive 24 The price index is found in E. N. White (1989). Using other regional price indice~ reveals similar increases in the inflalion tax. The slock of a~signats in circulalion is found in Table 8.
258
Free Banking II FREE BANKING
275
answer to the desirability of unregulated banking. It does show that free banking was reasonably successful even in the most trying political times. Unfortunately, while free banking was viable economically, it could not survive the radicalization of revolutionary politics that brought an end to laissez-faire economic policies.
REFERENCES Aftalion, F. (19117), I. economie de la Revolution franraise. Paris: Pluriel. An:hives Nationales, AD VIII 4, Letter from Valenciennes, and AD IX 495. Beranger, J. (19011), "Billets de Confiance de la Societe Patriotique de Rouen." GC4Z/!IIe Nllmimwtiqlle Fran~'aise 12,2113-317. Bcrangcr, J. (1910), "Bons des communes de la Periode Revohllionnaire: la Societe I'atrioti'lue de Caen." Gazelle Numisllwtique Fralll;aise 14, 335-412. Bloch, C. (1910), "La Verification des Caisses Palriotiques." Billie/in d'Histoire EcO/wmiqlle de la IUvolution, 134--197. Bruneau, M. (1902/1977), Les Debllts de la Revollllion dans les DepartemenU" dll CII", et cie /"Incire. Paris: Slatkine-Megariolis. Bouchary, J. (1941), Les compagnies jinancierel' a Paris cllI XVIII siede. Paris: Marcel Riviere, Vol. 2, Bouchary, J, (1946), Les FClllx-McJllnayellrs de la Revolution Frcmraise, Paris: Marcel Riviere, Caron, p, (1912), I.e mOll/wie et Ie papier-IIuJllnaie. Paris: Imprimerie Nationale. Carraz, 1<. (1975), "A propos de I'emission de billets de confiance 11 Chalon-sur-Sa6ne." ReVile J'Hil'toire Economic/lle et Socia Ie 53, 37-60. Colson, A. (11152), "Tableau des Billets de Confiance." Rel'ue Nllmismatique, 257-2117, 344-468. Checkland, S. G. (1975), Scollil'lI Banking, A History, 1695-IY73. Glasgow: Collins. De Cardenal (1912), "Les Billets de Confiance du Departement de la Dordogne (17911796)." I.e Revolution Fran~'aise 63, 311-{i1, 131-147,232-251. Friedman, M. (1959), A Program for MCJllelt/ry Slt/bility. New York: Fordam Univ. Press. Friedman, M., and Schwartz, A. J. (19116), "Has Government Any Role in Money." JOIll"lwl of M,,"ewI"Y I:;COIWlllics 17, 37-60. Friedman, M., and Schwartz, A. J. (1970), Monetary Swtistics of the Vlliteel Sit/tel'. New York: National Bureau of Economic Research. Furet, F., and Ozuuf, J. (19112), Reading and Writing, LiteTCIcy in Fl"Clnce from Call'ill to Jules Ferry. Cambridge: Cambridge Univ. Press. Grall, H. J. (1987), The Legaciel' of Lileracy. Bloomington: Indiana Univ. Press. Harris, S. (1930), 11,e AssignaU". Cambridge, MA: Harvard Univ. Press. Hayek, F. (19711), DenatiollCliisaticlII of MOlley-11,e Argllment Refilled. London: Institute of Economic AIl"airs. 2nd cd. Horsin-Deon, E. (1958), "En marge des assignats. Les billets de confiance." Rel'lIe d'lIi.I'lOire EccJllolllique et Socia Ie 36, 2511-293. 11uhrc.:ht, <.. (19),), "Les Assignllts illIonlcilux ilU dchut lie Iii Revolution (17111)···17<)21." AIIIWlt'l' 1/;.\'IOl";cllI<'.\' cl" III R,II'"llItioll 1·"'C11I~·Cli.l·e 94, 2111)-.10 I. Lllfaurie, J. (11)49), "Les Assignats et Ics l'apiers-Monnaic em is pilI' Ics Governclllcnts Revolutionnaircs (l7~17911)." Blllletill de la Societe pOllr Ie Eli/tic' clc' PLlpit'I"-JUIJIIllClie 4, 1-7. Lefebvre, G. (1962), The French Revolutioll. New York: Columbia Univ. Press. Vol. 1. Marion, M. (1914/1921), Histoire Fi/lClnciere de Ie France depllis 1715. New York: Burt Franklin. Ministere de I'lnstruction publique (18110), StCltistiqlle retrospective. Ewt reCClpill/lCltij et
259
Free Banking II
276
EUGENE N. WHITE
("omparalif i/l(liqllflnl, par deparlemenl Ie '/timbre de.v cm!;oilll.v qlli de Il'lIr maria!?1' tIIlX XVtr, XVIII' el XIX'" .vih-ll's. Mmlilellr Ullil'l'rsel, (May 27, 1791).
01/1
.viplIll {"a("le
Nichols, D. A. (1974), "Some Principles of Innationary Finance."' ./mmlll/ (!r /'on,;,.,,/ Economy 82, 423-430. Reinhard, M. (1941), Eludl' de la Populalion pendant la Revolllliml /'1 {"F:mpi,.,.. (JlIP: Imprimerie Louis-Jean. Rockoff, H. (1974), "The Free Banking Era: A Reexamination." ./ournal of MOlley Credil and Ballkill!? 6, 141-165. Rockoff, H. (1975), "Varieties of Banking and Regional Economic Oevelopment in the United States, 1840-1860." Journal of Economic /li.Vlory 35, 160-181. Rockoff, H. (1975), 11,e Frel' Bankin!? Era: A Re/'xamitlllliml. New York: Arno Press. Rockoff, H. (1986), "Institutional Requirements for Stable Free Blinking." Tire Calo ./ollrlIal6,
617~34.
Rolnick, A., and Weber, W. (1983), "New Evidence on the Free Banking Era." Americall Economic Review 73, 1080-1091. Rolnick, A., and Weber, W. (1984), "The Causes of Free Bank Failures: A Detlliled Examination." Journal of MOllelary E"conomi(",v 14, 267-291. Rud~, G. (1959), The Crowd in lire French Rel'ollllion. Oxford: Oxford Univ. Press. Smith, A. (1976/1937), The Weallh of Nalions. New York: Modern Library. White, E. N. (l987a), "Estimates of the Note Issue of the Caisses Patriotiques." Mimeo. White, E. N. (l987b), "The Stock of Paper Money." Mimeo. White, E. N. (1989), "Measuring the French Revolution's Innation: The Tableaux de D~pr~ciation." Mimeo. White, L. H. (1984), Free hanking ill Britain: Theory, E"xperiellce alld /)ebale, 1800-1845. Cambridge: Cambridge Univ. Press.
Part IV Private Clearinghouses
[16] THE ORIGINS OF THE SCOTTISH NOTE EXCHANGE BY CHARLES W. MUNN
Department of Economic History, University of Glasgow
between 1750 and 1772 witnessed a conT tinued expansion of the Scottish economy in which the HE PERIOD
tobacco trade and the linen industry made the most spectacular progress. Entrepreneurs from these groups were responsible for the formation of many of the early provincial banking companies which were set up, essentially as a result of customer demand, to provide short-term finance for these businesses. All of these new banks issued their own notes. In an article in the June 1974 issue of The Three Banks Review Mr. R. N. Forbes described the initial response of the Royal Bank and the Bank of Scotland to these new developments. l The chaotic situation which existed by 1765 was largely regularised by an Act of that year. This legislation prohibited the use of the 'optional clause', whereby banks had deferred payments in specie for their notes for six months. It also banned the issue of notes under £1 in value. Furthermore, summary diligence could be taken against any bank which refused to pay its notes on demand. 2 The issue of bank notes is the central theme in the early history of Scottish banking. The banks circulated their notes by discounting bills and by making advances on cash accounts for customers for which interest was charged at a maximum rate of 5 per cent, as stipulated by the usury laws. R. N. Forbes, 'Early Banking Excursions', The Three Banks Review, June 1974, No. 102. I S George III c. 49.
1
45
264
Free Banking II
A small commission was also charged on bills of exchange to meet the expenses of postage. If notes were returned for payment several book entries were possible. There was either a corresponding decrease in cash balances or advances. It was also possible that the notes could be accepted as a deposit on which interest would have to be paid. In the latter two cases the result was a diminution in the earnings of the bank. The note issue then was the major source of banking profits. The early records of the banks reveal that no effort was spared to put notes into circulation and to keep them there. The note issues of banks provided a convenient medium of exchange for customers and indeed for anyone who came into possession of them. They were not subject to the disadvantages of a gold coinage. The Act of 1765 ensured that note holders could secure, on demand, 'lawful money of Great Britain'. In practice, however, most of the notes returned to a bank were applied in the reduction of debtor balances rather than in the withdrawaJ of specie from the reserves. Banks always tried to dissuade customers from withdrawing specie or from purchasing bills on Edinburgh and London with their own notes because these practices reduced the circulation and the profits. By 1765 many of the initial difficulties which had developed with the early growth of the system had been overcome, but one major problem remained. A bank was only obliged to pay specie for its notes at its head office. There was no way in which a provincial bank could be made to pay specie or bills for notes in another provincial city or in Edinburgh. A bank, based in the Capital, if it had taken provincial notes in the course of business had to go to the expense and risk of sending them to their place of issue for payment. A provincial bank was similarly burdened if it accepted the notes of another provincial bank or of one of the 46
Free Banking II
public banks. The consequence of this was that banks did not generally take the notes of others in the course of business. Although most provincial bank notes had a purely local circulation in and around their place of issue, nonacceptance by other banks was a serious inconvenience for merchants who sold goods over a wide area and who were offered the notes of several banks in payments. This problem became acute as the pace of business accelerated in the years under discussion. The exception to nonacceptance of other banks' notes was the arrangement which had existed between the Bank of Scotland and the Royal Bank since 1751;1 these banks accepted each other's notes and exchanged them on a regular basis. The Dundee Banking Co. had made a trial of taking Edinburgh and Glasgow notes in 1764 but this practice was ended during the retrenchment caused by the withdrawal of small notes in 1765. 2 This article sets out to describe the events and to examine the motives which led, in 1771, to the general acceptance of each other's notes by the Scottish banks. The note exchange, formed as a result of this event, was based in Edinburgh. It was attended by the public banks and by the agents of the provincial companies and was the 'clearing house' for bank notes. As such it was a unique historical phenomenon, being the first of its kind in the world. Some banks saw it principally as a limitation on the amount of business which they could do. Others saw its potential as a controlling influence over the dangers of excessive note issues, and there is no doubt that it had a moderating effect on the volume of notes which a bank could keep in circulation. Nevertheless in the long term there was a net expansionary effect on business as a whole .
._ - - -------------.-----------Royal Bank of Scotland, Minutes 27/7/1751. I Dundee Banking Co., Sederunt Book, 1763-1776, 30/7/1764 and 24/8/1765. 1
47
265
266
Free Banking II
Table 1 sets out the names, dates of opening and note circulations of provincial banks in existence in 1765 together with similar details for those founded between 1765 and 1772. TABLE I SCOTTISH PROVINCIAL BANKING COMPANIES 1765-1772
I. Ship Bank, Glasgow 2. Arms Bank, Glasgow
Opened
Circulation
1750 1750
£ 65,000 200,000
Date of Figure 1761 contemporary estimate 1763 1768 estimated 1768
64,000 1761 Thistle Bank, Glasgow 32,000 G. Dempster & Co., Dundee 1763 15,000 1763 J. Macadam & Co., Ayr 44,000 Perth United Banking Co. 1766 Johnston, Lawson & Co., 15,000 estimated 1766 Dumfries 1768 43,000 8. Aberdeen Banking Co. 1767 10,000 estimated 1767 9. General Bank of Perth 1772 200,000 10. Douglas, Heron & Co., Ayr 1769 estimated 25,000 II. Merchant Bank of Glasgow 1769 Notes: I. The businesses of Johnston, Lawson and Co., and John Macadam and Co. were taken over by Douglas, Heron and Co. in 1771. 2. The Perth United Banking Co. was an amalgamation offive very small banks, all of which issued small notes prior to the Act of 1765. 3. Estimates are based on the size of other balance sheet figures or on notes retired by neighbouring banks. Sources: Minute books, general ledgers, balance books and letters of banks. 3. 4. 5. 6. 7.
Between 1765 and 1770 several attempts were made to overcome the difficulties created by the non-acceptance of bank notes by other banks. Early in 1768 the Aberdeen and Perth United banking companies agreed to a mutual acceptance and exchange of notes. 1 This had the effect of encouraging the use of notes rather than specie by those merchants who traded between Perth and Aberdeen. The note exchange was affected in Edinburgh by the agents of 1
Aberdeen Banking Co., Letter Book, 1767-1769, 30/1/1768.
48
Free Banking II
these companies. The ledgers of the Perth United show that the notes of the Dundee Banking Co. were accepted and exchanged at irregular intervals, seemingly by direct remittances between the two towns. l Nevertheless a start had been made. Unfortunately, nothing is known about the situation in Glasgow where, by 1769, four banking companies had been founded. It seems likely, however, that some accommodation had been reached, at least amongst the three older companies which had developed a close relationship during the 'optional clause' period in the early 1760s. 2 By 1768 some provincial banking companies had made a positive start towards following the example set by the public banks in accepting the notes of neighbouring banks. There remained, however, a good deal of mistrust between companies which were, geographically, further apart. In one particular case this mistrust was well founded. The 'aristocratic' Thistle Bank of Glasgow, in order to increase its own circulation, employed agents, on a commission basis, to 'pick up' the notes of others in circulation and replace them with Thistle notes. This was achieved by approaching shopkeepers and merchants who held notes and effecting an exchange, possibly with some slight monetary advantage as an incentive. The circulation was further augmented by risk banking, i.e. by discounting longdated bills which would normally be declined by a bank. Undoubtedly, some of the problems arose as a result of over-zealous agents who, without the knowledge of their head office, and because they were being paid by commission, were anxious to do as much business as possible even if it meant taking risks. Equally there can be no doubt of the later complicity of head offices who were keen to force their own circulation and embarrass their competitors.
-------------------------------------------
Perth United Banking Co., General Ledgers, 1768, 1769. Wm. Mure of Caldwell, Selections from the family papers preserved at Caldwell, Part 2, Vol. 2, 1885, p. 3.
1 I
49
267
268
Free Banking II
The British Linen Co., which had begun to assume a banking function, also engaged in 'note picking' of this nature. A compromise was reached between this company and the Aberdeen Banking Co. in 1768 whereby the former agreed to withdraw all its agents from the counties of Aberdeen, Banff and Kincardine. Each agreed not to molest the other's circulation. The best documented record of such early bank wars is that involving the Aberdeen Banking Co. and the Thistle Bank. In the years 1767-1768 the latter appointed agents in the Northern counties which the Aberdeen bank, although a new arrival on the banking scene, was concerned to establish as its own, exclusive territory. The Thistle Bank collected Aberdeen notes and demanded specie in return for them. These demands caused the Aberdeen directors to retaliate in like manner. In addition to regular agents in the Northern counties they also appointed note pickers in Edinburgh, Glasgow, Brechin, Montrose, Stirling, Falkirk, Ayr, Kilmarnock, Dumfries, Hawick and Kelso. l They decided that they would 'persevere in the present measures against the Thistle Bank till they withdraw their agents here'. Always at pains to establish their purity of motive they claimed that they had 'no wanton inclination' to distress the Thistle Bank and that they were 'desirous upon reasonable terms to live in harmony with all their neighbours'. They were determined to adhere to this resolution even 'if they should throwaway all their profits'.2 In December 1769, however, the Aberdeen and Thistle Banks agreed to submit their dispute to arbitration. The arbiters were Lord Gardenstone for the Aberdeen Banking Co. and Baron Mure of Caldwell for the Thistle Bank. Both were partners in their respective concerns. The discussion dragged on until R. S. Rait, History of the Union Bank of Scotland, Glasgow, 1930, pp.lSS-lS7. I Aberdeen Banking Co., Agents' Letter Book, 1767-1771, 12/7/1768.
1
SO
Free Banking II
August 1770 by which time several proposals of the arbiters had been rejected. By then both sides had been exhausted by the hostilities. The Aberdeen directors stated to the Thistle Bank cashier that their conduct in future would be 'entirely regulated' by his.! The Aberdeen Banking Co. minutes made little reference to the Thistle Bank thereafter. Some indication of the scale of the dispute was given in the Glasgow Chronicle of 1769. It was reported that: the Bank of Aberdeen made a demand on the Thistle Bank of Glasgow for £41,000 stg; which was immediately paid in gold and silver specie. a
The necessity imposed upon a bank of making provision for paying specie on this scale placed an enormous strain on its resources. At its annual balance in February 1768 the Aberdeen Banking Co.'s ratio of specie to demand liabilities was 61·2 per cent; in 1769 it was 46·3 per cent. In the latter year profit and loss account carried a debtor balance of £962.3 The public banks in Edinburgh tried to remain aloof from this bank war. They still regarded the provincial banking companies as a group of upstart merchants who had no right to engage in banking. Nevertheless, they did become involved on the periphery of the fight. The Thistle Bank, in order to pay demands for specie, collected the notes of the Royal Bank and Bank of Scotland and presented them for payment in gold and silver. This hostile action did not provoke any retaliatory measures from the public banks who seemed content to pay their notes and let the provincials continue to fight amongst themselves. Although the war between the Aberdeen Banking Co. and the Thistle was Aberdeen Banking Co., Minute Book, 1767-1776, 17/8/1770. Scots Magazine, 1770, p. 333. • Aberdeen Banking Co., Balance Books, 1768, 1769.
1 I
51
269
Free Banking II
270
probably the most vicious, all the provincial companies were involved at some time either on the offensive or defensive. Credit for the first real attempt to extinguish the conflagration must go to the most maligned of the early Scottish banks - Messrs. Douglas, Heron and Co. (the Ayr Bank). When this concern opened for business in November 1769 it wrote to all the Scottish banks assuring them that it was their intention 'to hurt as little as possible any bank or banking company in Scotland'. They intended 'receiving payment of notes offered to them which are issued by any company of whose credit they have a good opinion'. In particular, they requested that other banks would accept their notes in payment and exchange at stated periods. The request met with a positive, if cautious, response from some of the provincial companies. The Perth United, Aberdeen, Dundee, Glasgow Ship and Glasgow Arms banks all agreed to accept Ayr notes in payments and to exchange them and those of each other regularly in Edinburgh over a trial period. l The British Linen Co. later joined this arrangement. The response from the public banks, on the other hand, was entirely different. The Royal Bank replied that: from a principle of living at peace with everybody in (the banking) trade and to prevent all suspicions of taking up notes for unfriendly purposes, they do not receive in payment any notes but those of the Bank of Scotland. 2
They promised to retire any of their notes which were taken by Douglas, Heron and Co., and hoped for a 'good understanding' between the two concerns. The Bank of Scotland, in keeping with its agreement with the Royal, replied in a similar vein. The choice of Edinburgh as the location of the exchange was significant. It reaffirmed the role of Edinburgh as the Aberdeen Banking Co., Minute Book, 5/1/1770. • Royal Bank of Scotland, Minutes, 16/11/1769.
1
52
Free Banking /I
commercial capital of Scotland even from the point of view of the provincial banks who might have been expected to see Glasgow in this capacity. Furthermore, it provided a firm foundation of experience on which to build when the exchange became general in 1771. The exchange was to be carried out weekly by the Edinburgh agents of the banks involved. This was undoubtedly a far cheaper and more efficient method than by each bank remitting parcels of notes to several different banks and receiving the same in return. Unfortunately, no details of the volume of business done or the manner of settlement have survived. It seems likely, however, that debts were settled by payment in bills drawn on agents in Edinburgh. For a few months the exchange worked well, but by May of 1770 Douglas, Heron & Co. were suspected of note picking. The main accuser was the ever-sensitive Aberdeen Banking Co., which declined to exchange its notes from Douglas, Heron and Co. at Edinburgh from June of that year.l A note war of varying degrees of intensity ensued between Aberdeen and Ayr but the note exchange continued to exist although, for the moment, not to grow. In May 1771 the public banks decided to break with tradition and accept the notes of the provincial banking companies in payments. This of course involved an extension of the note exchange. The fact that a number of the provincial banks had already been exchanging their notes in Edinburgh encouraged the directors of the Royal Bank and Bank of Scotland to think that they would continue to do so in an exchange which embraced all the Scottish banks and banking companies. Several reasons may be adduced for this radical change of policy on the part of the Edinburgh banks. Customer demand was probably the most important. In a joint letter to the Aberdeen Banking Co., they claimed that the measure 1
Aberdeen Banking Co., Minute Book, 11/6/1770. 53
271
Free Banking II
272
'was adopted at the general desire of the Merchants and Traders about Edinburgh'.l This 'general desire' was not just that the traders' inconvenience of non-acceptance would be overcome, but more particularly that the note issues of the provincial banks would be controlled if they were required to retire their notes regularly. An anonymous writer in the Scots Magazine of 1770 advocated just this line of action. He believed that paper money had: animated the spirit of industry more and more; and the rapid progress of industry and manufactures increased, in proportion, the demand for money. Nevertheless, he was aware of the dangers of an unfettered multiple note issue. He believed that the public banks had been altogether too alooffrom the real needs of the Scottish economy and that: by confining their circulation within too narrow bounds, Banks have sprung up in every corner, so as even to become a public nuisance. As a lover of his country he could not help deploring 'the pestiferous itch that has infected men to deal in banking', and his remedy for stable growth was the acceptance by the public banks of the notes of the provincial companies. This, he claimed, would have the effect of keeping Edinburgh notes in circulation and confining the issues of the provincial banks within proper limits.2 In addition to a desire to satisfy their customers the public banks were also motivated by a desire to bolster their falling profits. The Bank of Scotland complained that: the business of banking is now very different from what it was formerly and that it is very difficult to keep the company's notes in circulation with profits owing to the late erection of many new banking companies.3 1
Aberdeen Banking Co., Minute Book, 13/9/1770.
2
Scots Magazine, 1770, pp. 353-354.
8
Bank of Scotland, Minutes, 9/12/1771. 54
Free Banking II
The Royal Bank were aware that their 'first and most obvious improvement' should be 'the extension of their circulation' .1 The factor which enabled the public banks to change their policy so dramatically was the infusion of new men into their courts in the years 1770 and 1771. This was particularly true of the Bank of Scotland which admitted a number of Edinburgh private bankers to the directorate. 2 These men stood to gain from the admission of provincial banking companies to the exchange. As private bankers they were often the agents of provincial banks and as such would represent these concerns at the exchange for which they would be paid a commission. The public banks took pains to befriend the provincials and to assure them of their purity of motive. Gradually, over the months from May to August 1771 all provincial banks were informed that their notes would be taken by the public banks and that they would be expected to change them once a week in Edinburgh. The Bank of Scotland and the Royal Bank agreed to a form of letter to be sent to the provincial companies which stated: as nothing is meant by this measure hostile against them it is hoped the consequent transactions will be carried on in the most amicable manner.3 Declarations of good intent were backed up with more material attractions. In a letter to the Dundee and Glasgow banks the Bank of Scotland informed them that if at any time they: shall have occasion for the assistance of the Bank of Scotland in the way of discounts the Directors will accommodate them Royal Bank of Scotland, Minutes, 4/12/1771. • N. Munro, History of the Royal Bank of Scotland, Edinburgh, 1928, pp. 398-415 and C. Malcolm, The Bank of Scotland, Edinburgh, n.d. pp. 68-69 and 293-302. a Bank of Scotland, Miscellaneous Documents, Bundle 539. 1
55
273
Free Banking II
274
as far as they can with any degree of convenience, and in general do everything in their power to preserve a good understanding between the two societies. 1 As subsequent events in 1772 proved, both of the public banks were prepared to help the provincial companies out of their difficulties. Two decades of hostility or at least mistrust had made some of the provincial banks wary of the embrace of the public banks. The ever-cautious Aberdeen Banking Co. resisted all overtures, maintaining that it was merely a 'neighbourhood' bank and could not agree to an exchange of notes at Edinburgh 'without acting in contradiction to the very spirit of their institution'. The fact that they had done so in 1770 was quietly forgotten. They agreed to pay their notes at Aberdeen as the law required them to do but hoped that the public banks would 'continue to decline receiving the notes of this Company as formerly'. The other reason for staying out of the exchange was that the practice of note picking by the agents of other banks continued unabated: a practice which . . . will defeat every measure that can be proposed either for establishing banking on a proper footing or for promoting that harmony that ought to subsist among all banking societies. 1 A further reason for non-participation in the exchange and for discouraging other banks from accepting their notes was that the Aberdeen directors believed that such a line of action would give a free rein to their note circulation. If other banks refused to accept Aberdeen notes and this had no adverse effect on their customers' desire to take them, then these notes would tend to stay in circulation longer, thereby increasing the profitability of the note issue and maintaining the amount of business that could be done. A
._---_._-_.__ ..._ - - - -
1 I
Ilank of Scotland, Minutes, 10/7/1771. Aberdeen Banking Co., Minutes, 14/8/1771.
S6
Free Banking II
bank which took part in the note exchange on the other hand, was required to tailor its business to the amount of notes which it could keep in the circle. In 1774 the cashier of the recently founded Hunters and Co., bankers in Ayr, regretted: the unlucky resolution of the Edinburgh Banks to take in our notes, by which they do us too much honour .... This has for two months past kept us both bare and busy, by exchanging every Monday at Edinburgh with the three banks there, and prevents us from discounting so freely as we have done.!
The note exchange was an effective check on excessive issues of notes by participating banks. The Dundee and Perth United companies were of a similar mind to their friends in Aberdeen. They resisted the embrace of the public banks who resorted to a period of 'rough wooing'. The tellers of the Royal Bank and Bank of Scotland were sent north to demand payment for notes of the provincial banks. On the 18th September 1771 they demanded payment for £5,400 of Dundee notes. The following day they demanded £15,300 from Perth.2 These figures represented approximately 18 per cent and 30 per cent respectively of these banks' total note issues. 3 The cashiers of the banks wrote to their friends in Aberdeen to inform them of what had happened, whereupon the Aberdeen directors, fearful that they might be presented with a similar demand, decided to join the exchange 'so far as to make a tryall'.4 The Perth United and Dundee companies were similarly impressed by the advantages of exchanging their notes regularly in Edinburgh and joined the exchange. The reason for the willingness of some provincial companies to exchange R. S. Rait, up. cit. p. 176. Aberdeen Banking Co., Minutes 20/9/ J77 J and 24/9/ J 771. I C. W. Boase, Centllry of Banking ill Dllrulee, Edinburgh, 1867, and Perth United Banking Co., General Ledger 1771. 'Aberdeen Banking Co., Minutes, 24/9/1771. 1 I
57
275
276
Free Banking II
with one another in Edinburgh but to reject a similar measure with the public banks lies in their expectation of the scale of the exchange. They feared that there was a greater tendency for their notes to find their way to Edinburgh and to the public banks than to other provincial towns - hence their initial resistance to the general exchange. In June 1772 the Bank of Scotland decided not to accept any more Aberdeen notes in payment. The reason for this is not clear but it may be that the Aberdeen Banking Co. was suspected of infirmity and the Bank of Scotland was merely playing safe in view of the impending crisis.1 The Royal Bank followed suit. Neither concern accepted Aberdeen notes again until February 1774. On that occasion the Aberdeen directors rehearsed all the old arguments and declined to join the exchange. 2 They continued to retire their notes as they were required by law to do, but this was achieved by the occasional remittance of parcels of notes to Aberdeen and the return of sight bills on Edinburgh in exchange. It was not until the Royal Bank became Edinburgh agent for the Aberdeen Banking Co. in 1782 that the northern bank rejoined the exchange. 3 When a branch of the Bank of Scotland opened in Aberdeen in 1780 a local exchange of notes was agreed upon. The Aberdeen Banking Co. was an exception. In general the provincial banking companies settled down to live in peaceful coexistence with the public banks. When new banking companies were formed in the next half-century often one of their first actions was to apply for admission to the note exchange. The banks accepted that, although the exchange might be a short-run limitation on their business, the long-run benefits were greater. As the public enjoyed greater confidence in the issues of banks and the practice of Aberdeen Banking Co., Minutes 5/6/1772. Ibid. 25/2/1774. a Royal Bank of Scotland, Minutes 17/6/1782.
1
2
58
Free Banking II
note picking ceased, bank notes came to replace specie in circulation to such an extent that banks were able to operate on fractional specie reserves which were often only 2 or 3 per cent of demand liabilities. Therefore the note exchange was a permissive factor in business expansion. In 1782 the Royal Bank became Edinburgh agent for several provincial banks and in the following year a branch was opened in Glasgow. The Bank of Scotland and the British Linen Co. on the other hand opened several branches in the provinces and competed directly with provincial banks in several towns. From 1789 the provision which had been made of the local exchange for the Aberdeen Banking Co. was extended to other companies. 1 It seems that the concession was intended primarily for the British Linen Co. although it was also grudgingly offered to the provincial banks. When the Montrose Banking Co. was founded in 1814 it applied for an exchange with the local agents of the British Linen Co. and the Bank of Scotland. The former agreed but the latter, at first, refused. It was not until George Kinnear, the bank's Edinburgh agent and a director of the Bank of Scotland, interceded on their behalf that a local exchange was set up.z Similarly, when the East Lothian Banking Co. set up a branch in Haddington in 1810 the terms of the local exchange were dictated by the directors of the Bank of Scotland without reference to the wishes of the Dunbar men. 3 The effect of the note exchange as a control on the overissue of notes was attested by the witnesses of the 1826 Parliamentary Committee on Promissory Notes in Scotland and Ireland. Alexander Blair, Secretary to the British Linen Co., thought that it was 'impossible that there should be an British Linen Co., Minutes, 14/5/1789. Court of Session Papers, CS311/2142, Box 207 - Letters of Geo. Kinnear to Montrose Bank, 11/10/1814 and 31/1/1815. I East Lothian Banking Co., Sederunt Book, Scottish Record Office BI8/18/14, 30/7/1810 and 6/8/1810. 59
1 I
277
278
Free Banking II
over issue of our notes under the system of note exchanges which exists at present'.l Other witnesses were of the same opinion. The note exchange played an important part in developing the stability which was such an important feature of the Scottish banking system. It was not, however, an infallible guarantee of safe practice. Messrs. Douglas, Heron and Co. in 1771 found that their notes were being returned upon them quickly from the exchange. Rather than tailor their business to the amount of notes which they could effectively keep in circulation they resorted to the 'practice of drawing bills of exchange upon London' to pay for their returned notes. The number and value of these bills increased until, when the bank failed in June 1772, they exceeded £600,000. 2 The note exchange acted as an indicator of a ban k's business capacity. Those who ignored it did so at their peril. Parliamentary Papers, Evidence to the Select Commil/ee on Promissory Notes in Scotland and Ireland, 1826, evidence of Alex. Blair, p. 48. I Adam Smith, Wealth of Nations (Harmondsworth, 1970), p. 413. 1
60
[17] RICHARD H. TIMBERLAKE, JR. *
The Central Banking Role of Clearinghouse Associations "MOST OF TillS [clearinghouse I currency was illegal, but no one thought of prosecuting or interfering with its issuers . . . . As practically all of it bore the words 'payable only through the clearing house,' its holders could not demand payment for it in cash. In plain language, it was an inconvertible pap\!r money issued without the sanction of law, ... yet necessitated by conditions for which our banking laws did not provide . . . . When banks were being run upon and legal money had disappeared in hoards, in default of any legal means of relief it worked effectively and doubtlessly prevented multitudes of bankruptcies which otherwise would have occurred" (Andrew 1908b, p. 516).
I.
FREE MARKETS IN THE PRODUCTtON OF GOODS AND MONEY
Most professional economists believe that the free market system is the optimal means for allocating resources and for distributing products and incomes. However. economists have traditionally made an exception of money. Even among the classicists, money was viewed as a special case requiring some amount of government regulation over its supply. The difference in outlook is most manifest when the ljuestion is raised to university students (or to any other group of people); "What would happen if everyone were permitted to issue money without constraint'!" The 't am indebled 10 William Heranek, Rubel'l Dince, G~'()rgc Sclgin, I.awrcnce While. Illy wife, lIilllegard, alllilhe edilur un.1 referees uf Ihe jUlll'ml1 fur nl~lIy 1I~"flll '"/!/!L"liu"" Lilllil:uiuII~ of ~I'a.'e prevenled formallrcalmclII of allihe sugge.liulls Ihal were lIIade. I, uf cum.e, acklluwledge resl'",,,ihilIly lilr Ihe paper's remaining imperfeclions. RI(,IIAIW
II.
TIMIlHltl.AKH, JI(' i.~
/"'/!/i',uol' of pm III"", Vl/iI','nil.\' ,!J' (;,'ol'gi" ,
JlI/trllIIl of Money, Credit, lind BlIllkillg, Vol. 16, No.1 (February 19K4)
Copyright © 1984 by the Ohio Slate University Press
Free Banking II
280 2 :
MONEY. (,lmI>IT. ANI> IIANKIN
typical answer is "chaos." Yet, if this question substitutes "ballpoint pens" for "money." the answer becomes a conventional market analysis with no chaotic implications. Somehow. people who envision chaos when they see money in a free market environment neglect the demand side: everyone imagines universal creation of cost less legal tender currency that would become spent immediately in a frenzy of price explosions. Legal tender, however, is a coercive device of government and would have only a very constrained role in a free market monetary system. Rather. acceptance of privately created money would be voluntary; and those who supplied it would have to offer market inducements--redemption in some form or other-to earn market acceptance (Hayek 1978, pp. 19-20; Klein 1974, p. 423). If free market money is permitted only in Economists' Wonderland, private enterprise central banking is regarded as even more fanciful. Indeed, most economists would deny its feasibility, let alone its desirability. t But necessity in the practical world of the nineteenth century was the mother of innovation. Banks in that era faced the traditional problem of holding enough reserves to protect themselves in an emergency while operating conventionally with minimal reserves to maximize returns as cost-recovering enterprises (Mints 1945. p. 247). In the absence of a central banking institution. the commercial banks could not effectively cope with emergency conditions in financial centers. However. the circumstances of their existence prompted their defense; and the defense they developed in the United States was an extension of the functions of the clearinghouse associations. This paper traces the development of the clearinghouse association from its inception as an economizer of payments to its policy function as a quasi-centml banking institution. This evolution occurred over a fifty-year span from 1857 to 1907 and came to an end in the post-I907 era with the organization of the Fedeml Reserve System. Of particular importance is this question: Given that the clearinghouse system had developed some well-defined central banking functions. why did innuential decision makers in the banking fraternity and the government r"shion " formal central bank-the Federal Reserve System-under governmental auspices. when the clearinghouse system, with some simple modifications. could have been refined and continued as an effective lender of last resort?
2.
INSTITUTIONAL ORIGIN OF THE CLEARINGHOUSE
A number of economisls. most of whom were professionally active in the early twentieth century. have treated thoroughly the origin and evolution of clearinghouse associations (Cannon 1910; Hepburn 1924; Myers 1931; Redlich 1951; Sprague 1910; White 1935). Their accounts are much in agreement-as they should be since clearinghouses are no more mysterious nor complicated than grocery stores. Instead I For e'Ulmple. consider this statement "For most economists. the debate regarding the desirability or Igovernmentall central banking is over. Central banking doctrines have become the conventional wisdom" (Frangul 1980. p. 212).
Free Banking II RICHARD H. TIMBERLAKE. JR.
281 : 3
of each bank eSlablishing a transactional relationship with all other banks, every bank sends a representative to one place-the clearinghouse-where its debit items arc deared against its credit items. Then the balance is struck, and payment is due from debtor banks to creditor banks. Originally, one bank in the association was assigned the "central" administrative role for dearing the other member banks' accounts. Each bank kept part of its specie (and later, greenback) reserve as a deposit with this bank, which in turn issued clearinghouse certificates of an equivalent amount to be used in the seltlement of daily balances. These certificates were in large denominations-$5,OOO and $IO,OOO-to expedite the dearing process (Myers 1931, p. 96; Cannon 1910, p. 44). At this stage, the issue of clearinghouse certificates was strictly in lieu of conventional bank reserves. The certificates were only a technical expedient for economizing the payments system.
3.
DEVELOPMENT OF CI.EARIN(iIlOLJSE LOAN CElnll'ICATJ:S
The p'lIlic of I X57 stimuluted the extension of de'lringhollse isslles. Reserves in New York banks had dedined during August of thut year, and, when a prominent bank failed, commercial-paper rates approached panic levels. At first, the banks wanted to curtail loans-the usual means of meeting an internal currency drain. However, the clearinghouse banks agreed to "increuse their louns so that the clearing-house balances of all of them would be increased proportionately and would cancel each other without reducing the slender stock of specie" (Myers 1931, p. 97); that is, the banks agreed in concert to "cry down" their reserve ratios. Concomitantly, the country banks hud druwn down the bulances with which their notes were customarily redeemed in New York City. So the city banks refused to honor the notes, that is, accept them as means of payment. A policy cOl11miltee of the New York Clearing House Association (NYCHA) then issued a circular suggesting the propriety of including in the clearinghouse settlements the currently irredeemable notes of the country banks. The committee subsequently allowed the issue of clearinghouse IUlIll certificates 2 against these notes, which the creditor city banks had deposited in the bank acting as the "central" bunk (the Metropolitan Bank). The country banks, who could not redeem their notes iml11ediately, ugreed to pay 6 percent interest on them as "loans" from their city correspondents, and the city banks then used them as collateral for the new dearinghouse loan certificates. Thus, the notes as a basis for issues of certificates becume equivalent 10 specie in the seltlement of clearinghouse bulances (Myers 1931, p. 9H; Redlich 1951, pp. 158-59). The state bank notes, although not redeemuble in specie, were bused on securities deposited with New York State uuthorities. Therefore, no extra security was needed when they were made the basis of clearinghouse loan certific'ltes. The city bunks in 2Ctearinghouse cenilicales must I>c distinguished from clearinghouse /0<11/ ccniJicales. The formcr wcre Ihe conventional issues made strictly in lieu of specie. kgallendcr notes. or "Iher legal reservcs fur sellkment of clearinghouse halances. The laller were i,,"cd unly in clllcrgcnl'ies un Ihe h"sis uf lu"ns IIWU~ lu memher hill,k!\ hy dC'll'iJlghuu~c polh.'y ~·ulllluilh:CS.
Free Banking II
282 4
:
MONEY. CREDIT. AND BANKING
turn were able to furnish more of their own notes to pay depositors and to extcnd credit to borrowers (Redlich 1951, pp. 159-60). The issue of clearinghouse loan certificates temporarily made the clearinghouse itself a fractional reserve institution: certificate issues of the clearinghouse-its demand obligations-substantially exceeded its specie reserve. As the panic subsided, the city banks "began to redeem country-bank notes. sending them home for redemption at the rate of one-fifth of the total each month. The certificates were retired as the collateral Istatel bank notes were rcdeemed . . . . They bore interest at 6 percent and were a safe investment. It finally took a resolution of the Clearing House Association to persuade the holding banks to part with Ithe last of! them" (Myers 1931. p. 98).
4.
CItARACTERISTICS OF CLEARINGItOUSE POLICY
The precedent established in 1857 was made the basis for all the subsequent issues of loan certificates (hrough 1907. In every instance, though. some ncw wrinkles were added. In Novemher 1860, when a stringency developed in anticipation of the Civil War. the NYCHA took in New York State Bonds. U.S. Trcasury notes. and bills receivable as collateral for the issue of loan certificates. The tcrms for this and future issues were that the maximum value of the loan certificates was limited to 75 percent of the collateral securities face value, and the rate a borrowcr bank paid was at an annual rate of6 percent (Myers 1931, p. 1(0). The committee also had the power, according to the agreement. "to equalize thc specie Ireserve of the member banks) 'by assessments or otherwise,' treating it as a common fund to be used for mutual aid and protection." In addition. any loss caused by the nonpayment of collateral was to be borne among the clearinghouse banks in proportion to their capitals. Myers's further comment was that "a higher degree of centralization would have been hard to obtain even with the aid of a strong central bank" (p. 1(0). The "pooling" of reserves provision was not popular. The more conservatively managed banks that had a stronger reserve position than their fellows objected to it as inequitable. They felt that pooling denied them the rewards for their caution.' It also could be evaded by altering what were held as "reserves." National bank notes were not subject to pooling; so many banks accumulated these notes while paying out greenbacks which were subject to pooling (Sprague 1910, pp. 120-21). Pooling was also unnecessary. Clearinghouse loan certificates cost the borrowing banks an annual rate of 6 percent (plus a small administrative charge), and paid 6 percent to the banks that acquired them through the clearing process. By this mcans, the banks with stronger reserve positions loaned to those who felt deficient in reserves-a procedure that anticipated the contemporary federal funds markct. It provided individual banks an incentive to use reserves they might otherwise have ·'If pooling had become the norm. the "economics of the commons" would have prevailed. No individual bank would have felt any constraint in making demands on the "pool."
283
Free Banking II RICHARD H. TIMBERLAKE, JR.
;
5
sequestered; and, as well, it augmented the total reserves in the system at just those times when the demand for bank credit, as expressed by market interest rates, was greatest. Ideally, the interest rate charged by the clearinghouse association would r.llion the amount of accommodation sought and provided.-I Thus, a market clement rationed clearinghouse accommodation in a way that pooling of reserves would seriously have compromised if it had become a part of conventional policy. Another innovation during the panic of 1873 was the issue of irredeemable "certified" checks to stretch the reserve base. These checks did not have cash on deposit as the basis for their issue. They were simply a quasi-currency. To prevent anyone from cashing them and thereby reducing bank reserves, the clearinghouse policy committee adopted this resolution: "All checks when certified by any bank shall be first stamped or written 'Payable through the Clearing House' .. (Sprague 1910, p. 54). This resolution put certified checks on a par with clearinghouse loan certificates. The hanks ,Iccepted them as settlelllent IIledia hy COllllllon consent through their clearinghouse association, but did not have to redeelll thelll with legal tender. During this era, the New York City hanks held a good fraction of the reserves of the interior ("country") hanks, and IIHlIly of them competed for deposits hy paying a nominal interest rate to get them. This practice was seen both then and later as reprehensible because it allegedly induced the interest-paying banks to make risky loans and extend their loan portfolios unduly as a means of recovering the costs of their interest payments. An unusual demand for the conversion of deposits into greenbacks or other legal tender supposedly put such a stmin on the more radical hanks that cash payments were restricted or suspended. When the clearinghouse subsequently issued loan certificates, the clearinghouse banks, ohserved Sprague, "were converted, to all intents and purposes, into a central bank, which, although without power to issue notes, was in other respects more powerful than a European central bank, because it included virtually all the banking power of the city" (pr. 50-63; 90-97; see also, Redlich 1951, p. 158).
5.
INTRODUCTION OF OTHER CLEARINGHOUSE MEDIA
In 1873 the policy of the NYCHA spread to many other "reserve" cities that were secondary money centers (Chicago, St. Louis, New Orleans, Baltimore, Washington, etc.), as well as to other cities or towns that were strictly "country" (Dubuque, Knoxville, Concord, Harrishurg, etc.) (Sprague 1910, p. 63). Minor panics occurred again in 1884 and 1890, and a major one in 1893. Certificates were issued in all of them. Indeed, by 1893, the NYCHA anticipated the need for certificates and began issuing them in the sUlllmer before any suspensions or restrictions occurred. ·New YurK slale usury laws limiled Ihe maximum rale uf inieresilimllhc dearinghuuse c'u"ld <'Ilarge lu (> percen!. Inleresl rales being whal Ihey were in Ihal era uf 'Iable prkes, 6 perl'clIl was all effeclivc ,·u"strain!. Fur general policy purpuses, uf cuurse, markel illieresl ralcs shu"ld lIul have heen lilllil,'" hy a pu/ilically inspired ceiling.
Free Banking II
284
MONEY. CI(EDIT. AND BANKING
Other clearinghouse media made their appearance at this time. Hepburn noted. from statistics he compiled as Comptroller of the Currency, that $100 million of clearinghouse loan certificates, clearinghouse certificates, certified checks, certificates of deposit, and cashier's checks in all denominations were issued. "all designed to take the place of currency in the hands of the public" (1924. p. 351). Sprague estimated that $300 million in money and "substitutes" for money were added to the supply outside the banks. In addition to Hepburn's list. Sprague reported that "in factory towns pay checks became an acceptable part of the circulating medium" (1910, p. 197). The episode in 1893 thus saw a significant extension of the large denomination loan certificates to other bank-issued items of small denomination. Once this practice started it became ubiquitous and the denominational depreciation had to he extended. For example, issues by the Atlanta Clearing House Association Hnd others were in denominations as low as $10. Since these notes could be used hy anyone, the banks paid them out over the counter to depositors, who in turn bought ordinary goods and services. Since the change people received in the form of legitimate currency tended to deplete bank reserves, the clearinghouse associations had to supply certificates in smaller and smaller denominations. finally as low as $0.25. In addition. "a considerable amount of clearing-house certificates were taken out lissuedl in the Southeast in cities where no clearing-houses existed"I!1 (Cannon 1910, pp. 109-12).
6.
TIm LEGALITY OF CLEARINGHOUSE ISSUES
Naturally, the issue of lower denominational currency of different types raised some questions concerning their legality. A law passed in the early 1860s had prohibited the privale issue of token and subsidiary coin, and the national bank act had a like effect on state chartered banks' notes by taxing them 10 percent a year. In Hepburn's opinion, clearinghouse certificates (and clearinghouse loan certificates as well) were not subject to the 10 percent tax, but all of the other currencies isslled during the stringency of 1893 were. He then made a revealing comment. "This temporary currency," he wrote, "performed so valuable a service ... in moving the crops and keeping business machinery in motion, that the government. after due deliberation, wisely forebore to prosecute . . . . It is worthy of note." he concluded, "that no loss resulted from the use of this makeshift currency" (1924. p. 352; italics added). When a government "wisely" decides not to enforce a law because violation of the law is universal and everyone is better off in the breach, the law itself riltlst be ill-advised. Yet, Hepburn did not draw this conclusion. His usc of the adjective "makeshift," furthermore, suggests a spurious character to the issues of currency that the actual train of events did not support. If the panic of 1893 initiated the extension of makeshift currency, the panic of 1907 saw its full flowering. During the latter year, the NYCHA issued $IOO-odd
285
Free Banking /I RICHARD H. TIMBERLAKE. JR.
: 7
million of clt:aringhouse loan certificates (see Table I), while the Chicago Association issued over $39 million worth of currency items, including $7.5 million clearinghouse checks in denominations of $1, $2, $5, and $10. In Philadelphia, the clearinghouse issued certified checks for wage payments to depositors who opened special payroll accounts. The checks were printed up by the American Bank Note Company I! I. In many other towns, such checks were issued in small denominations and made payable to "bearer." Sometimes, clearinghouses issued blank checks so the banks Clluld put on them the most suitable denominations (Andrew 190Mb, pp. 497-502; Cannon 1910, pp. 123-31). Andrew, in his famous article, cited this episode as "the most extensive and prolonged breakdown of the country's credit mechanism which has occurred since the establishment of the national banking system." He referred to the "ingenious invention of multifarious other substitutes for legal currency during the weeks of hoarding and suspension," thus ~:onveying by editorial implication his disapproval of the developments he chronickd ( IIJOXh, p. 497). Nonetheless, Andrew's account is scholarly and thorough. lie reported that many state auditors and comptrollers had wrillen to the hanks under their jurisdiction encouraging them to restrict payments of legitimate cash and to payout the various array of "substitutes" in restricted amounts. This indulgence was only to be permitted to "solvent" banks. But, wrote the Indiana Slate Auditor in a "P.S." 10 his leiter circulated among Indiana banks, "the question of your solvency is to be determined by yourselves upon an examination of your present condition" (p. 499). One can only imagine the soul-searching and painstaking self-study the banks engaged in prior to issuing their quasi-currencies!
TABLE I l.IMN CE~TIHCATIi tSSUES AND RESE~VES 01' TilE NEW Y()~K CLiiAIUNlllIOlJSE BANKS:
Slil.liCTEU DAniS Al!!'.regalc J!>su~
y, ...
Dale uf Finol I~!)uc
IM60 IM61 IM63 1864 1873
November 23 Seplember 19 N()vember 6 March 7 Seplember 22
7.3M 22.6 11.5 17.7 26.6
6.86 22.0 9.61 16.4 22.4
1884
May t5
24.9
21.9
1890
November 12
16.6
15.2
1893
June 21
41.5
38.3
1907
OClober 26
101.0
88.4
(Smillion)
SUUkCL:..S: SprOiGuc (IYIU, pp. -432-33' alKl Uniteu SIOIlto::!o CUiliptrulkr of
Maximum OUblanlfmg (Smilliunl
Ih~
Currclll.:Y ,ISIJ), 1'XJ1).
Kt~rV\."s
ur
NYC N.llinn:al Ibllks
46.9 (Seplclllber 12) 70.7 (Juno: 20) 92.5 (Ocluber 2) 99.0 (July 12) II! loll (Dcl·cmber 3)
286
Free Banking II 8
MONEY. CREDIT. AND RANKING
Andrew's admittedly incomplete data revealed that in cities over 25,000 population, clearinghouse issues of all kinds totaled $330 million. The issues hegan in late October 1907. and the last of them was retired by late March 1908. Clearinghouses in at least 71 of 145 cities made such issues. For cities under 25,000 population. Andrew's data are even more fragmentary. He recorded $4.5 million clearinghouse issues, including some in places such as Willacoochee, Georgia, hut he acknowledged that his accounting covered' 'only a small fraction of what actually existed in the smaller localities" (p. 506). In the largest cities the amounts issued in 1907 were 3 Y2 times the amounts issued in 1893. In all cases, the note issues were secured by collateral which had I Ij, to 2 times the dollar value of the notes (p. 5(9).5
7.
SOME CONTEMPORARY VIEWS
or CLEARINGHOUSE CENTRAL RANKING
To a banker, such as James Cannon, who was intimately involved with clearinghouse operations. the issues were not only defensible but highly recommended. lie denied that the loan certificates were a currency. Rather, he defined them as "temporary loans," and cited one Comptroller of the Currency who had called them "due-bills." In addition, Cannon argued, the courts in Pennsylvania had "decided that Ithe issues I shou Id not be regarded as money. A tax on them Ithe court decision readl would have been 'a serious blow to one of the most effective and ingenious contrivances devised, [andl a direct violation of the spirit of the law'" (1910. p. 116). Cannon concluded that loan certificates were "one of the finest examples the country has ever seen of the ability of the people when left to themselves to devise impromptu measures for their own relief" (p. 96). One can endorse Cannon's conclusion without accepting his denial that the certificates were currency. Clearly, the small denomination items were media of exchange. They were used for transactions of all kinds by everyone, and they were reissuable. The clearinghouse loan certificates may not have seemed to be conventional currency, since they were not transacted by anyone except banks. They were on that account, however, the equivalent of bank reserves, that is, high-powered money, and therefore even more potent than the small-denomination items. Horace White agreed with Cannon on the utility of the clearinghouse operations. Nonetheless, White, in contrast to Cannon, recognized that all the issues of 1907 were illegal: "Some were engraved to resemble bank notes or government notes." he stated. "and these were doubly illegal; some were as small as twenty-five cents, and these were trebly illegal" (1935, p. 383)5' Fritz Redlich remarked that where the initial clearinghouse loan certificates were "extra-legal," the various issues in 1907 were illegal. and presented "new dangerous problems [unspecifiedl. " Legalization of clearinghouse loan certificates, he claimed, would have been' 'predicated on the incorporation of clearing houses under a federal law" (1951, pp. 167-681. ~Generally. the lower the denomination of the currency issues. the greater the collateral villuc of the securities supporting them. hWhite implied here that issues of currency were in violation of any or all of three laws: the COllstitution. the national bank act. and the prohibition against private issues of fractional currency.
287
Free Banking II RICHARD H. TIMBERLAKE, JR.
9
1I. THE CHOICE BETWEEN THE CLEARINGHOUSE SYSTEM AND THE FEDERAL RESERVE SYSTEM
The widespread issues of clearinghouse currency associated with the panic of 1893 stimulated the only proposals for organizing the clearinghouse system on a legitimate and formal basis in the manner suggested by Redlich. Several l>lans appeared in the middle and late 1890s, the most comprehensive of which was one by Theodore Gilman (1898) of New York. Gilman noted that a reserve maintained outside the commercial banking system might not ever be needed "ITJherefore, it should not be provided by capitallreserves) withdrawn from productive use. It will cost nothing and will be just as serviceable if it is provided by law as a power which may be used in case of need." To provide this potential for the banking system, Gilman proposed what he termed a "grade of banks higher than our ordinary commercial banks," by which he meant a banking institution that would be one stage above commercial banks and able to furnish accommodation when needed (pp. 44-45). His choice for this role was the clearinghouse association incorporated under federal law. A clearinghouse in each state would have the power to issue notes in much the same fashion as what was already being done in practice. The notes would have a maximum value of 75 percent of the security collateral on which they were based. They would be guaranteed, first by the bank issuing them, second, by the associated clearinghouse banks in their home state, and, third, by all the clearinghouses who would receive the notes at par. "This protecting structure," Gilman observed, "would rise noiselessly over the heads of the people without the sound of axe or hammer" (pp. 46-47). 7 He emphasized an additional safeguard in the clearinghouse system beyond the note and collateral guarantees. The clearinghouse loan committees would be "conservative" in granting loans because of their "pecuniary interest as stockholders in banks which they would endeavor to protect from loss on their contingent liability as guarantors" (pp. 71, 124). In short, the private clearinghouse "central bank" would operate prudently because it would face a bottom line. "The kind of currency Ithe clearinghouses) make for themselves," he observed, "[should be] ... good lenough) for the public" (p. 119). Gilman also argued that the clearinghouse notes he envisioned would always be redeemed on demand at any clearinghouse and would therefore circulate at par (p. 126). He may have oversold his scheme with this pledge. It implied that no restrictions or suspensions would occur in the future. Or he may have meant only that if a note-holder were willing to go to the time and trouble of redeeming his note at the clearinghouse, he could do so. Gilman saw the new currency as a temporizing device that would promote elasticity in the monetary system. It would allow time for goods to be sold, so that the 1'fhe clearinghouse agreements already in force called for sharing any re>ulling losses frum the issues in proportion to the participating banks' capital and surplus or by the pruportiunal amuunt uf their clearings during the previous year.
288
Free Banking II 10
:
MONEY. CREDIT. AND BANKING
c1earinghousc l:UlTCIH:Y would return to the hank~ liS paYllIent for liquidating the loans that gave rise to the original issues of notes. Gilman also relt that the interest charge of 6 percent to hanks who took out clearinghouse currency "would act as a check upon their issue. and they would not be taken so much for profit as for protection and necessity" (p. 157). Gilman's plan was introduced as a hill in the House of Representatives by Ben L. Fairchild (Rep. - N. Y.) in lanuary I R96. It was then referred to the House Committee on Banking and Currency, hut it never appeared again, and no popular movement ever developed in support of his scheme. One Treasury official who pointedly objected to the clearinghouse strategy was Leslie M. Shaw. Secretary of the Treasury from 1902 to 1907 (Timherlake 1l}7R. pp. 171-85). The clearinghouse loan certificates. he stated in an address delivered in 1905. are a "plea of guilty to an indictment charging bad management locally or bad legislation nationally" (Shaw 1908, p. 279). Yet, Shaw's prescription for appropriate policy during a panic logically supported the clearinghouse method. First, he wrote, the supplemental currency should be identical with the legitimate currency that already existed. Second, he continued. the relief should he capahle of immediate and local application. Finally, he concluded, the additional currency should be retired promptly when "the demand therefore ceases" (p. 293). In point of fact, the clearinghouse issues fit Shaw's criteria reasonably well. Their use as currency was ubiquitous; they were issued in a Illultitude of localities on the perceptions and judgments of thousands of bankers, and the notes never stayed out more than a few months. Nevertheless, Shaw regarded thelll as undesirable because in his opinion they raised an alarm that something was wrong and thereby caused "hoarding" (p. 294).8 The Comptroller of the Currency. lames H. Eckels. in his Rcport for IR93 was more sympathetic. He lauded the clearinghouse associations for their actions during the previous summer. "The service rendered by them." he wrote "was invaluable •... and to their timely issuance ... is due the fact that the year's record of suspensions and failures is not greatly augmented" (U .S. Comptroller of the Currency 1893. p. 15). Eckels denied that the issues were currency: hut he had tongue in cheek. "If they had been used as currency." he argued, "the banks issuing thelll would have heen fined" (p. 16). Since the banks were not fined hy the governlllent. the law must not have been violated. This syllogism is more reasonahly viewed as pragmatic accommodation to events than logical defense of the law. The comptroller in 1907. William Ridgely. proved to be more of a legal constructionist. lie noted that the law. under Section 5192 of the Revised Statutes, declared that clearinghouse certificates ;11 tire /JOJJeJS;OIl of allY ballk belollg;IIg to the clcal';lIglroliJe associat;oll were lawful money (U.S. Comptroller of the Currency IY07. "Furthcr on in his articlc. Shaw contradictcd his own implication that thc governmcnt ~1](JUld ovcrsee a supplcmcntal. hcavily taxed, national bank note currency. "Governmcnl officials." he stressed. "arc eXlremcly cautious. The rejection of a proposition never causes trouble. Atlirnmtive acts only arc investigated and censured. Technical objections are as good as valid ones with the average oureau official" (p. 295). Given this fact of bureaucratism. a governmcnt agency could hardly oc expeclcd 10 furnish relicf for a typical crisis following thc norms that Shaw had specificd.
Free Banking II RICHARD H. TIMBERLAKE, JR.
289 :
II
p. 64). He begged Ihe question of those issues that were used universally as media of exchange outside of banks. Later in his Report he discussed the possibility of legalizing the clearinghouse issues as an "emergency circulation." He found "merit" in this suggestion, but rejected it as a "half-way measure." The full development of the clearinghouse idea, he argued, "should carry us further ... to the inevitable and logical conclusion ... , which is that we should have a national central bank of issue and reserve lacting under governmental authority)." This institution would be "more systematic and efficient" and would have none of the "disadvantages" of the clearinghouse system. Ridgely did not specify how or why the government central bank would be more systematic and efficient; he simply offered unsubstantiated assertions. For example, .. if we had had such a Igovernment centmll hank in opemtion in 1907, no such hallk panic as we have h,ld would have hecn possihle" (p. 75). Olad Ridgely been ahle to witness the monetary and banking events of 1929-33, he might have experienced a fair amount of intellectual indigestion.) Most amazingly, his next paragraph is a description of how well the clearinghouse system generally operated.
9.
RESTRICTION AND SUSPENSION OF CASH PAYMENTS
The most characteristic feature of a panic-the attempt to convert demand deposits into legal tender-stimulated the restriction of cash payments by the banks. Restriction took many forms: for example, limitation of cash payments to a nominal amount ($25 to $200) per transaction; or this amount per day or per customer. Often payments were "discretionary" with the banks and therefore negotiable with the banks' managers. Its purpose was to induce as much friction as possible into any transaction that would provoke an "internal drain" of bank reserves. Sprague argued that the clearinghouse banks in New York assumed the indulgence of restriction too readily (1910, pp. 273-77). They faced reserve requirements of 25 percent and were prohibited from making new loans when reserve ratios reached this minimum. However, the law did not restrict them from paying out reserves even though they were in a reserve-deficit position. In his article, William Dewald (1972) has reargued Sprague's analysis and found it valid. The New York banks, Dewald notes, "had made a fetish out of their twenty-five percent reserve ratio and held closely to it. ... Reserve deficits were small during crises, and New York banks in fact held enormous reserves on every occasion when they suspended payments. It was this policy that Sprague attacked and it was this policy that the legal authorization to issue clearing house 'currency' promised to change" (p. 940). Sprague, Dewald notes, recommended that reserve banks be required to maintain payments as long as they held any reserves. This principle had been repeatedly urged as well by the Comptroller of the Currency (p. 939). The conclusion of the Sprague-Dewald thesis on restriction-of-payments policy is that, far from being complementary to the issue of c1e'lringhouse currency, it increased uncertuinty and generally upset the payments systcm. It witnessed the
290
. Free Banking II 12
:
MONEY. CREDIT. AND BANKING
substitution of less efficient mcdia for conventional money and also cxtcnded the pcriod of disequilihriulll (pp. 914. 941 -42). 'I A contrary view on restrict ion comes from Friedman and Sd1\vart/. ( I WI]). and also from Margaret Myers in her earlier study (1931. p. 143). Given a bank panic. Friedman and Schwartz argue. "the fairly prompt restriction of payments was a therapeutic measure that almost surely kept the contraction from being even more severe and much more protracted than it was . . . . (Tlhe failure of one (unsound) bank did not set in train a chain reaction. Restriction of payments thus protected the banking system and gave time for the immediate panic to wear off. as well as for additional currency to be made available" (1963. pp. 163. 1(7). The alternative was for the fractional reserve banking system to experience a "black hole" type of collapse in the face of the increased demand for money with the m:companying declines of prices. wages, income. and business activity (pp. 167--6XI. Friedman and Schwartz correctly focus on the inherent instability of the fractiomll reserve banking system as the primary danger to the monetary system and thc economy. Nonetheless. given that clearinghouse issues were already in progress. the I'l'Oml" restriction of payments may have heen ill-advised. Why not require the clearinghouse banks to nm down their reserves to a very small minimlllll with graduated rate penalties on their reserve deficiencies'! A law that provided explicitly for such an action would have countered the timidity of the hanks and. in conjunction with the clearinghouse issues. encouraged them to maintain the level of their current portfolios as well as their cash payments. Ultimately, they might have hecn forced to suspend: but before they did so, their reserves would have heen largely utilized in meeting internal and external demands for liquidity.
10.
MONETARY REFORM LEGISLATION
The immediate legislative reaction to the events of 1907 was the Aldrich-Vrceland Act of 1908, which called for the grouping of ten or more national hanks into a National Currency Association. The associations anticipated from this act wcre not to operitte itS c1caringhouses. hut were to issue "emergency currency" just the way the clearinghouses had done in the past. The only difference was tlmt the new institutions' currency issues were to be under the administration of the Secretary of the Treasury, who had some discretion in allocating the notes to different sections of the country. The currency outstanding was to be taxed at an annual rate of 5 percent during the first month. and at an additional I percent each month thereafter until the annual rate reached 10 percent (Friedman and Schwartz 1963. p. 170). The Aldrich-Vreeland Act marked a critical turning point in United States policy toward the monetary system. As Redlich observed, the act "turned its back on the clearing-house loan certificate instead of legalizing it. . . . [The I suh-committee "Spntgue also favored the "pooling" of reserves in order tfl reduce the tOfl-ready tendcnl'y of Ihe banks to restrict or suspend. Dewatd apparently agrees (p. 941). However. this provision would have vitiated an important market constraint (see above. p. 6). Sprague's principal remedy requiring banks to use their reserves. in conjunction with clearinghouse issues. would have been sufficient to ,top 'my p,mic.
Free Banking II RICHARD H. TIMBERLAKE, JR.
291 :
13
[Report) of the Commillee of Banking and Currency declared the power to issue loan certificates dangerous as long as clearing-house associations were not under government control" (1951, p. 168). The allegation of "dangerous" flies in the face of the clearinghouse system's actual performance. The most extraordinary fact associated with the several clearinghouse episodes between 1857 and 1907 is that the losses from all the various note issues, spurious and otherwise, were negligible! The only loss reported in any of the accounts here considered was $170,000 in Philadelphia in 1890 out of an issue of $9.7 million-1.8 percent (Sprague 1910, p. 392). Few of the economists who analyzed clearinghouse operations even noted in passing this astonishing record, and none used it as an argument for continuing the system. When the Federal Reserve Act was debated in Congress in 1913, the sentiment again reflected the notion that the government should have some control over the currency. In a typical statement, Congressman William A. Cullop of Indiana COIltrasted the new Federal Reserve Board with the clearinghouse executive commillee: "Every fair-minded man," he declared, "would prefer that the control over the currency be vested in seven men [on the Federal Reserve Board) selected from different parts of the country than to have it remain in the hands of the five managers of the New Yark clearing house . . . . To this self-constituted and unauthorized organization [the Federal Reserve Act) . . . deals a deadly blow in order to secure for us industrial and commercial freedom" (U.S. Congress 1913a, p. 332). Cullop's hyperbole demonstrated the decided influence populism still exerted in monetary affairs. Hepburn's comment on the passage of the Federal Reserve Act was that this "measure ... , while open to criticism, is a vast improvement upon the old system and as good as could have been fairly expected under all the circumstances" (1924, p. 397). Margaret Myers was even more positive: "The passage of the Federal Reserve Act," she wrote, "ended the long struggle with monetary problems which had harassed the New York money market and the rest of the country throughout history" (1931, p. 421).
tt. RECAPITULATION A fractional reserve banking system by its very nature contains an element of instability. An increased demand for currency can so deplete bank reserves that the supply of bank-created money (deposits) declines. This possible reaction means that the supply of money is at times unstable: an increased demand for one kind of money may result in a decline in the total quantity of money supplied. The banking industry recognized this problem at an early stage. Since both the money-supplying industry-the banks-and the users of money-firms and households-were harmed by the results of bank-aggravated crises, everyone had an interest in reducing or eliminating this structural inslilhility. The solution, arrived at through force of circumstances and some financiul innovation, was clearinghouse creation of temporary currency. The banking industry
Free Banking II
292 14
:
MONEY, CREDIT, AND BANKING
simply reinstituted itself as an ad hoc central bank. and through its clearinghouse associations issued more currency. Since the laws proscribed private issues of money. clearinghouse currency had to be issued discreetly. It thcn worked so wcll that no one had any incentive to prosecute its issuers. It was the private-moncyproducing industry's answer to a pronounced need. It was, as well. constrained by market factors-interest rate charges on its issue, and the real stake the clearinghouse banks' directors had in seeing to it that the clearinghouse association did not make costly mistakes. In addition, the gold standard system regulated and constrained the monetary base. Bankers, who knew the rules of the gold standard game. realized that clearinghouse issues were only a temporizing device. They knew they had to restrain bank credit appropriately when their reserves were threatened. Nonetheless. the issue of clearinghouse currency put the brakes on the development of an unstable bank credit contraction. It did not prevent the dcmise of inefficient banks; it only stopped the fractional reserve collapse that might otherwise have occurred. One may then wonder why this system was rejected. Why was a government central bank superimposed on the banking industry when the clearinghouse method had proven so effective? The accounts of the times suggest several answers. First. the clearinghouse issues by 1907 had become recognizably illegal. This awareness could have meant simply that the laws were bad. Instead, the realization endowed the clearinghouse currency with a hucksterish quality. Because clearinghouse currency was illegal. it was makeshift funny money-never mind the logic of the laws prohibiting it. Second, since clearinghouse currency was market-regulated and because it was created and accepted voluntarily, many contemporary observers did not recognize its utility and integrity. They wanted an "official" currency similar to national bank notes or greenbacks. Third, the clearinghouse currency seemed to arise out of nothing. Everyone could understand the principle of a quasi-public central bank standing by with a reserve obtained from contributions by participating banks and ready to be applied when a crisis threatened. But few could understand how a clearinghouse system could create emergency currency without sceming to havc an emergency reserve on which to base it (Sprague 1910. p. 234). It was altogether too mystifying an operation for common understanding and acceptance. It smacked of Wall Street legerdemain. Last. the clearinghouse issues were associated with the restriction or suspension of cash payments. Sprague clearly del11onstrated that the two actions were not at all necessarily related (1910. pp. 260-77). Nonetheless, the popular mind was only too willing to apply here the fallacy that correlation implies causation. Many advocates of a central bank saw the Federal Reserve System as an evolutionary development of the clearinghouse associations. "This bill. for the most part," said Robert Owen, the sponsor of the Federal Reserve bill in the Senate," is merely putting into legal shape that which hitherto has been illegally done" (U .S. Congress 1913b. p. 904; see also Willis 1922, p: 251). The Federal Reserve alternative, however, was critically different from the clearinghouse system. It introduced a discretionary political element into monetary deci-
293
Free Banking 1/
RICHARD H. TIMBERLAKE, JR.
15
sion making and thereby divorced the authority for determining the system's behavior from those who had a self-interest in maintaining its integrity. LITERATURE CITED
Andrew, Abram Pialt. "Hoarding in the Panic of 1907." Qllarter/yJoIITlla/ o/EclIIwmics 22 (February 1908, 290-99 (a). ___ . "Substitutes for Cash in the Panic of 1907." Qlltlrterl.v )mmlll/ 0/ EnJl/omi/'.v 22 (August 19(8),497-516 (b). Cannon, Jallles Graham. C/elITillg HUII.ve.I·. Sellate OOl'lImelll Nllmb,'r -191. 61 st Congress, 2nd Session, National Monetary Commission, 19 Hl. Dewald, William G. "The National Monetary Commission: A Look Back." )mmlll/ uf MUlley, Credit, alld Btlllkil/g 4 (November 1972),930-56. Frangul, Ramzi. Book Review of Origi/ls ofCelllm/ Btlllki/lg i/l the Ullited SWtes: JOIITlla/ 0/ Fi/ltlllce 35 (March 1980), 212-14. Friedman, Milton, and Anna J. Schwartz. A MOl/etary Hil'lOry 0/ The Ullited SUItes, UJ67-1960. Princeton: Princeton University Press for the National Bureau of Economic Research, 1963. Gilman. Theodore. A Grmled BlIl/killg System. Boston and New York: Houghton Mimin, 1898. Hayek. Freidrich A. The De-NlIIilllllllizlIIioll uf MO/ley. 2nd ed. London: Institute of Economic Affairs, 1978. Hepburn. Alonzo Barton. History 0/ Coi/lage 'Illd Cllrrem'Y il/ tile UI/ited SllItes 'Illd the Perell/lial CUlllest for SOlllld MOIII!)'. rev. ed. New York: Macmillan, 1924, Klein. Benjamin. "The Competitive Supply of Money." )ollTllal of MOlley. Credit. tlml lJallkillg 6 (November 1974),423-54. Mints, Lloyd W. A History of Ballkillg Theory i/l Gre'" 8,.ilai/l I/I/el the Ullited Slates. Chicago: University of Chicago Press, 1945. Myers, Margaret. The New York MUI/ey Market. Origil/s 'Illd Del'elopmellt. Vul. I. New York: Columbia University Press, 1931. Redlich, Fritz. The Moldil/g ufAmericclII BI/I/kil/g, Mel/ 'Illelle/etls, Part II, 18-10-19/0. New York: Hafner, 195 J. Shaw, Leslie M. Cllrrellllsslles. New York: Appleton, 1908. Spragul!, Oliver M. W. History .uf Crises UI/der tile NlIIiollal 81/l/kil/g System. Semlle Doclllllelll Number 538. 61st Congress, 2nd Session, National Monetary COlllmission, 1910. Timberlake, Richard H., Jr. Origil/s of Cel/tml B'lIIkillg ill the Ullited SlIltes. Cambridge: Harvard University Press, 1978. United States Comptroller of the Currency. AIII/utli Report. Washington, D.C., 1893. _ _ . AI/III"" Report. Washington, D.C., 1907. United States Congress. CIJIIHrel'siCIIUlI Rt·l'lml. 63rd Congress, 1st Sessiun, Appendix, II) 13 (a).
___ . COl/gressiolltl/ Record. 63rd Congress, 2nd Sessiun, 1913 (b).
White, Horace. MOl/ey (/lui 8al/kil/g, Revised and enlarged by Clmrles S. Tippens "nd Lewis A. Froman. New York: Ginn, 1935. Willis, lIenry Parker. "Feder,,1 Reserve Ulll1ks." In TI,,' 1'/"'0')' (l1It111i.~w(l· ,!f U(llIkillg. edited by C::harleli F. Dunbar. New York and Lundon: G. I). I)utnam's Sons, 11)22.
[18] Competitive Monies and the Suffolk Bank System: A Contractual Perspective* DONALD J. MULLINEAUX UI/iversity (!f" l\el/llldy
Lexington, Ke1llucky
J. Introduction In recent years, a number of monetary economists have begun to focus research on an old topic: the economics of private money or "free banking." Much of this literature is motivated by a sense that competitive markets may well prove superior to government regulation in coping with the problems inherent in a "laissez-faire" monetary system. Noting this, Robert King, in his recent survey paper, suggests that: "The central focus of future research in this area must be the types of monetary institutions that the private market will deliver" [10,123]. By monetary institutions, King presumably means the contractual mechanisms utili7,ed in production and exchange relationships involving private money. This paper examines one mechanism -the so-called Suffolk Bank System--which "the market" did deliver in New England from around 1820 to just before the Civil War. While most historians have looked favorably on the Suffolk mechanism I, almost all have mischaracterized the underlying rationale for the system. I n particular, most analysts have interpreted the establishment of the Suffolk Bank's note redemption program as a response to Gresham's Law ("Bad money drives out good money"). However, the conditions of 19th century banking in New England (and elsewhere as well) prior to the establishment of the Sulfolk system rule out any application of Gresham's Law. Prior studies have also neglected the importance of evolutionary changes in the Suffolk Bank system. Over the /irst five years of its existence, what we will label Suffolk I was largely unsuccessful both in terms of its profitability and its capacity to control the behavior of the banks participating in the system. Several contract adaptations introduced in 1824 accounted for a sharp reversal in the fortunes of the Suffolk Bank system; it is Suffolk II which has gained the accolades of bank historians. I ronically, Suffolk II iI/creased the prospects for a Gresham's Law kind of outcome (of the sort emphasized by Akerlof [I]). However, the contract structure of Suffolk II involved properties sufficient to allow that Suffolk "The author wishes to thank James Moser and Arthur Rolnick for helpful comments and suggestions, I. lIorace White's view is typical: ", , , it brought banking in New England to a state of responsibilities, order, and solvency unknown before" [20,302]. A more recent analysis of the SulTolk system by Van Fenstermaker and Filer [18] finds that the SulTolk mechanism had no signilicant elfect on the growth rate of money, hut that it did illcrease the bank nole to deposit ratio. This finding is consistent with our interpretation of the Suffolk contract as a mechanism ror enhancing the quality of bank notes.
884
Free Banking II COMPETITIVE MONIES AND TilE SUFFOLK IlANK SYSTEM
management 10 monitor and control the behavior of the participating banks. Consequently, bad banks did not drive good banks out of t:xistt:nce. Analysis of tht: Sullolk arrangement from a contractual perspt:ctive focuses on the issue most nitical to the discussion of "free banking": the ability of "the markt:t" to control the behavior of bank managers. Hayek [9] and White [21] have argued that market forces arc capab\t: of controlling banks, provided that bank liabilities are convertible into some more "basic" money originating outside the banking system. Milton Friedman [5] exemplifies the counter view: ,\ liduci"ry currellcy ostellsibly cOllvertible illto the IllOlictary COllllllodity is thercforc likely to he overissued from time to time and convertibility is tikely to become impossible .... A liduciary currency would thus probably tend through increased issue to degellerate into a wlllmodity currency ·into a literal paper standard - there being riO stable e4uilibriulll price level short of thul at which the money vulue of currency is no gl. caler Ihan tlial of Ihe paper it contains. And in view of Ihe negligible cost of adding lCroS, il is nol clear Ihal Ihere is uny Jinite price level for which this is the case.
Most bank historians agree with Friedman's viewpoint. Bray Hammond [ll] argues that "... the Jacksonians glorified in what was a triumph of laisser faire in a field where lainer faire had no place. Sovereign and unified control of the monetary system is needed in any economy, whatever freedoms may be proper otherwise." We shall argue that analysis of the Suffolk Bank system suggests that the truth lies "somewhere in the middle." That is, the Sulfolk system shows that markets--in the sense of exchange relationships characterized by the complete ahsence {~lhierarchy- were unable to control and discipline the behavior of 19th century bank managers. Nevertheless, in New England the private market did deliver an alternative "governance structure"~ (the Suffolk Bank system) which was largely effective in achieving what "the market" could not. This mechanism involved many of the features of a more modern contractual device-the franchise system. Since a franchise has been recognized as a hybrid of "the market" and "the lirm," we argue that the Suffolk experience shows that banking involves transaction characteristics which require some form of hierarchical "controL" Consequently, Lawrence White's [21] claim that a simple note-exchange device provides sullicient control of free bank managers is questionable. Nevertheless, the Suffolk Bank systcm was a systcm of private rather than government control, and it was reasonably successful. However, the Sulfolk system was: (I) relatively unique to New England; and (2) brought down by a blend of market and extra-market "competition." The Sulfolk's failure to survive in New England may provide some hints concerning the necessity of government intervention in banking markets. The organization of this paper is as follows. Section II provides an overview of the market for bank notes during the early 19th century as background to the analysis of the Sulfolk Bank system. Section III demonstrates that Gresham's Law, ill the traditional sense of the term, cannot provide a rationale for the Sulfolk system. Section I V describes till: evolution of the Sulfolk system, emphasizing the contractual elements of the Sullolk II arrangement. Section V argues that the post-I 1124 Suffolk system can be interpreted as a fram:hisc contract through which the Sulfolk Bank provided a "trademark" or brand name 2. In Oliva Williamsun's work 12J] 011 economic institutions, this lerm is used 10 describe medlanisms which arc: innul!Il~C and l:ontroilfansacliulls costs. A contract is a well·ddincd type of gO\lc:rnance structure, but other IIH;l·h.U1i~lll~ (implicit i.lIId lIunslamJard CUUlral:ls, for examplt:) playa critical rok in aJkclilig IranS.1Clions costs.
employed (0
295
Free Banking II
296 886
DOl/aid J. Mullineaux
backing to notes issued by participating banks in New England. The final section analyzes the demise of the Suffolk Bank system and draws some tentative conclusions.
II. The 19th Century Hank-Note Market The typical bank note contract in 19th century U.S. banking involved a promise by the issuing bank to pay the bearer on demand a specified sum of "lawful money." Lawful money consisted of specie (gold or silvcr). The engraved note instrument contained the name of the issuing hank, the amount of the claim, and the signature of an ofliccr of the bank. Notes were issued by both the First and Second Banks of the United States and by state-chartered banks in various denominations and were the primary media of exchange in small exchange transactions. Notes were placed into circulation when banks made loans or when they were paid over the counter for specie deposits. Rank notes served as a medium of exchange because they were convertible into specie. Notes did not necessarily exchange at par (face value), however, for several reasons. First, notes had to be transported to the issuing bank in order to collect specie. This cost drove a wedge between the value of a note in exchange outside the bank and its specie value at the bank. Thus, while notes normally exchanged at par at or near the location of the issuing banks, the notes of geographically distant banks (labelled "foreign notes" in the 19th century) typically traded at a discount which increased with distance [20]. Second, the promise of a bank to redeem notes at par was subject to some risk of default. Such defaults were not infrequent, and, on several occasions, involved practically all commercial hanks (the socalled "suspension" periods). Consequently, the notes of riskier banks traded at larger discounts than the notes of banks judged liquid and secure enough to redeem at par. A third factor accounting for discounts on certain bank notes was the cost of authenticating the note contract. Counterfeit notes were prevalent during the 19th century prior to the formation of the national banking system, in part because of the large number of different bank note issues. The ease of "passing" counterfeit bills no doubt increased with the geographic distance to the falsely claimed bank of issue, since the signatures of bank oflicers and engraving structures on bills were less likely to be familiar over substantial distances. The presence of discounts on bank notes provided incentives for the establishment of firms which traded bank notes in a secondary market at prices suflicient to cover the costs (including a normal profit) of holding risky assets and of transporting them to the issuing banks to collect specie. Dillistin [3] describes how these note brokers (or "shavers") often published "bank-note reporters" which detailed bid quotes on various hank notes and also contained information concerning outstanding counterfeit bills. Newspapers also puhlished quotations of hank-note discounts and counterfeit-bill information. Redlich [ 14J notes that by the 1820s, a numher of commercial hanks had entered the note-hrokerage business and consequently also quoted discounts on the notes of other hanks. The hank note market ofthc 19th century essentially functioned as a floating exchangerate system, with each note convertihle into a dominant money (specie) at a rate (discount) determined by transportation costs, the perceived riskiness of the operating bank, and the costs of authenticating the note's validity. If the market was efficient, the sizc of the discount on a particular bank's notes provided some information concerning the "quality" of the note in terms of its purchasing power. Thus, a bank which pursued a systematic policy of depre-
297
Free Banking II 887
COMPETITIVE MONIES AND TilE SUFFOLK IlANK SYSTEM
ciating the value of its notes (via overissue) would find that the market "revealed" the bank's strategy via a larger discount on its notes. White [20] notes, for example, that the notes of banks judged riskier than their competition traded at larger discounts than the notes of safe banks. This is not surprising, since note brokers possessed strong incentives to monitor the riskiness of the banks whose notes they bought and sold. But if the market signaled changes in the 'Iuality of certain bank notes via the price systelll, then (,reslwlII's I.aw could not havc applkd to the note market.
III. The Suffolk Hank System and Gresham's Law The Suffolk Bank system has been interpreted by notable historians, such as White [20], Hammond [8], Redlich [14], and Dewey [2], as a private-market response to Gresham's Law. In particular, each of these authors claims that the notes of Boston banks were driven out of circulation by the "inferior" notes of the country banks. Hammond asserts, for example, that "Everywhere, of course, out-of-town notes had to be accepted by the city merchants, though they were of varying and uncertain value, dirty, and hard to be redeemed. Being inferior to city bank notes, they drove the latter out of existence" [8,549]. Dewey notes that: "The consequence was that the bills of country banks obtained the entire circulation even in Boston" [2,79]. The source of this view appears to be Whitney's The Suffulk Balik which notes that: At this time (I !!24) the city was flooded with country money. The circulation consisted almost entirely of the notes of banks outside Boston. With more than one halhhe banking capital of New England, the Boston banks supplied only one twenty-fifth of the bills in use [22,10].
The figures quoted by Whitney were in fact an estimate of the total amount of country bank notes in circulation (based on an assumption of note/capital ratios at country banks) and an estimale of the so-called "permanent" circulation of Boston banks. No details were provided by Whitney concerning the method used to identify "permanent" notes. In fact, the ratio of Boston to country-bank notes was considerably larger than Whitney and many bank historians have suggested. From 1812 to 1844 the ratio of Boston to country notes averaged 77 percent. In 1824, when the Suffolk joined the other Boston banks in a joint venture arrangement for redeeming "foreign" notes, the ratio was 76.3 percent, or roughly equal to the average for the entire period. The notes of Boston banks, indeed, never approached extinction. Nor would Gresham's Law have applied in any case. A necess,lry condition for Gresham's Law is that tht: competing monies t:xchange at a fixed price relative to specit: or some "base" money.' Country bank notes exchangt:d at variabk rates, so that price adjustments could preclude "bad money from driving out good." In n:ality, the institution of the Suffolk Bank system improved the prospects for a versioll of Gresham's Law to come into play since the effect of the (Suffolk II) system was that wuntry bank notes traded at par. That is, the Suffolk system created conditions such that bank notes wen: priced in exchange as if they were perfect substitutes for each other. As 3. Friedman and Sl.:hwartz{4], among others, hay!! cmphasilt:d the: necessity of a lixcu cXl.:h:.mgt! ment fur the upcr~tion of (jn:,slldm's Law.
nilt!
cllviron-
Free Banking 11
298 888
Donald J. Mullineaux
Klein [12) has emphasized, if note holders were unable to differentiate the quality of one bank note from another by some mechanism other than price, each bank would possess incentives to depreciate the quality of its notes. The costs of such a strategy would be borne principally by other banks. Compctition would, via a form of Gresham's Law emphasizcd by Akerlof [I) in his classic "lemons" paper, drive those banks which did not depreciate the quality of their currency out of business. Since all banks would possess the same incentives, the market would, like Akerlors used car market, break down. Ranks would complctely depreciate the quality of their notes, producing the infinite price level predicted by Friedman in a free banking regime. The deterioration of product quality is the predicted outcome in any industry where consumers, because of high information costs, cannot distinguish among products in an industry. Money is a product the quality (in terms of purchasing power) of which cannot be determined by direct inspection, e.g., the si7,c or color of a currency docs not inl1uellce its purchasing power. One mechanism which does allow differentiation among similar commodities is the establishment of a "trademark" or "brand name." We shall argue that the Suffolk system evolved into a mechanism through which the Suffolk Rank "franchised" a trademark to the participating banks. As the next section shows, this was hardly the conception of the Suffolk maintained either by its management or by the participating banks, at least in the early stages of the system.
IV. The Suffolk Bank SY!item: Version I versus Version" The Suffolk Rank was founded in Roston in 1818; it was the seventh hank to he chartered in that city. Roughly one year after opening, the management decided to enter the "foreignmoney" (country-bank note) business. The only other bank purchasing country hank notes at the time was the New England Rank. In Suffolk I, the bank purchased country-hank notes from merchants. individuals and other banks at a discount. It allowed the issuing bank to purchase these notes from the Suffolk at this same rate of discount, provided the issuing bank maintained a permanent deposit with the Suffolk of $S,O()() and agreed to make additional deposits to cover those notes redeemed at the Suffolk. If a hank chose not to participate in the arrangement, the Suffolk purchased its notes at a discount and presented them promptly to the issuing bank for redemption at par. The source of the profitability of the Suffolk I scheme apparently stemmed exclusively from its ability to invest the permanent deposits supplied by the patticipating banks (on which the Suffolk paid no interest). The discount was passed on to the participating banks, although it is unclear how the cost of returning redeemed notes from the Suffolk to these banks were apportioned. Apparently, the Suffolk absorbed these costs·. Whitney notes that competition with the New England was "lively" and that the typical discount declined from one percent to less than one-half percent soon after the implementation of the program. At the end of 1820, the Suffolk halted all purchases of the notes of nonparticipating banks since this aspect of the business had become unprofitable. lIe also reports that by 1822, "the profits of the (foreignmoney) business had hecome so small that the directors felt obliged to reduce the expenses of the bank" [22,8]. 4, Otherwise. there was no incentive to participate --the Sull'olk would recapture the discount through deliverycost rees.
299
Free Banking II COMrETITIVE MONIES AND TilE SUFFOLK IlANK SYSTEM T~blt
I. Circulalion uf NUlr.!S of Ma""achu,Clh Hanks (Thousands)
Dale
(I) Ih)Slon Banks
IH2.\ IX24 I Xl:; IH2h
$1,.\54 1,7')7 1,91H 2,206 2,103 2,067 2,07H 2,171 3,464 .,,060 2,K24
I~n
IH2H 1~2Y
IH30 IX.!I IH.l2 IH.lJ
(2) Non-Bo,wn Ballks
$1,775 2,046 2,17.\ 2,344 2,K33 2,KIK 2,670 2,'153 3.675 4.06.1 5,065
Ralio of NOII-Boslllll 10 BmwlI Banks
1..11 1.14 1.1.\ 1.01>
US 1..16 1.2H
1.36 1.06 1..1.1 1.7'J
Source: I.ake [iJ,IMX).
The management of the Suffolk Bank was unsatisfied not only by the lack of profits in the "foreign money" business, but also by a sense that country bank notes claimed an inordinate share of the circulation in the Boston market. Two of the directors, in a letter to the other Boston banks, noted that "it might be expedient for them in common (the Boston banks) to adopt, with a view of checking the enormous issues (~f' coull/ry, and especially Eastern (Maine), paper and of securing to the bills of Boston banks ajllst proportion of the circulation ..." ([8, II], emphasis added). j The Suffolk management attributed "the prodigious credit enjoyed by the country banks" not to "any superior confidence in the stability of these institutions", but rather to the discount on their notes which was "founded on the very dilliculty and uncertainty of means of enforcing this payment." The Suffolk's proposal to its sister banks was to cut the size of the discount and to generate a fund, "to be assessed in proportion to their respective capitals, ... for the purpose of sending home the bills of the banks ..." such that " ... the profit or loss shall be in common ...." The proposal (Suffolk II) was approved, a fund of $300,000 was established (payable in the notes of the so-called Associated Banks), the Suffolk was appointed as the agent bank for the association; and the discount was reduced to 1/4 of one percent. Within a year, the discount was eliminated and the notes of country banks "in good standing" were accepted at par at the Sufli.)lk. As noted in the previous section, the management's rationale for establishing. such a scheme was flawed. The note circulation of country banks was not "inordinately" large, but rather rellected the equilibrium decisions of households and businesses given prevailing payment habits and exchange rates." And a reduetion in the size of the discount on country notes should have been expected to increase their circulation, since Boston prices in terms of country notes accordingly would be reduced. 7 I ndeed, when they traded continuously at 5. Thus, the alfinity uf many bankin~ historians for a Gresham's l.aw inll!rpn:latiun may havt: been kindled by Ihe lIIallagcnll!lIt'!i view that the Un:sham prim.:iple was indeed at work in Hoston in HS25. O. The Suiti.>lk's management apparently failed to appreciate the releVath':C: of the silabk dt'po}'il balances of the Hustoll banks, buth absolutely and relative tu the coulilry bauks. Over the period 1M 12··IM44, fllr cXiJlllple, depu~ib of Uu~tOIl b"nks aVl.!faged $5.2 Inilliun, while Massal.:huSodts country bank deposits avclaged $1.7 milliull. A!; a r~llil) hI Ci1pit~J, Iotai ballk liabiltth':S (nutes plus lkpo~its) ill liOslOn avt:ragcd 62.6 pt:n:clll OWl this period, whik Ihl.: Si.lIlle filHo fur cuuntry banb averaged 64.3 pacellt. rhus, cuuntry banks wac han.lly much kss slahl!.: lhall thcir (ilY cuunJcrpans, by this lIu.:asuclo: at least. 7. iJuslOll bank mallagers were not alonc in misjUdging 1/11: likely elli.:cls uf this sysh:lll. Cuuntry bankers vigorously up posed the system, often in colorful language, siw.:e they, lUo, believed their circulatioll would be curlaikJ.
300
Free Banking II 890
Donald.!. Mullineaux
par. country bank notes were perfect substitutes for Boston bank notes. As Table I slHlws. a relative increase in country bank circulation was apparently what resulted. at I('a~t after IR2o. While the ratio of Boston to country bilnk notes did increase in the initial ~tilge of Suffolk II, Lake [IJI argues this probilbly reflects the fact that the Boston banks paid interest on some bank notes during the years I R25'-27 in yet another attempt to hoost their local circulation. Whatever the rationille for the ilrrangement, the new system WilS considemhly more successful in other respects than its predecessor. The volume of country hank notes received at the Suffolk rose from $300,000 a month in June, I R24 to over $2 million a month hy the end of 1825. Steady growth in note redemption continued over the life of the Sufl'olk svstem; by 1858, its monthly redemptions exceeded $30 million per month. Whitney [221 reports that the husiness was "very remunerative"; from 1825 until the end of the SuflC)lk systelll (1858), the Suffolk Bank was invariahly the highest priced hank stock in Boston. Whitney reports average annual dividends of a little over 8 percent from 1847-IR46 and 10 percent from IR34-1858, versus 5.7 percent from IR19-182J, the period ofSufl'olk I. The mechanics of the Sufl'olk II system estahlished in I R24 were similar in part to Suffolk I." A permanent, interest free. deposit was required of participating hanks (although the size of the deposit was made to vary with the capital of the bank. with $2,O()O the minimum size deposit) and additional balances again were to be deposited to cover note redemptions. The Suffolk redeemed the notes of participating hanks at par (after IH25) and dehited their deposit balances accordingly. There were two significant innovations to the Suffolk 11 scheme, however. First, the Suffolk permitted a bank to borrow temporarily (via an overdraft) against its permanent halance, reserving the right to send home that bank's hills for specic redemption should its borrowings exceed the permanent deposit. In addition, the Suffolk accepted as deposits (or loan repayments) the notes of any participating hank judged in good slanding, i.e., banks were not required to deposit specie to their clearing accounts to replenish temporary balances. The latter provision of the new contract was especially important si lice it allowed thc Suffolk to offset thc notes of one participating bank against those of anothcr on its halancc sheet, i.e., to act as a clearinghouse for bank notes. In effect, the Sufl'olk 11 arrangcment involved "collective net settlement" of bank notes, whereas Suffolk I involved "bi-Iateral nct settlement"! In a collective settlement arrangement, a central agent (the Suffolk) maintains bookkeeping information on the net credits and debits of all the participants and effects the necessary transfers of funds as debits and credits on its balance sheet. In a bi-Iateral scttlement, each pair of banks illdividually calculates its net credit or debit position and the necessary transfer is eflected either in specie or as a credit or debit on their respective books. The Sufl'olk I arrangement forced each New England bank to form a correspondcnt relationship with the Suffolk or settle net debit balances in specie. But this system did not allow participants to present the notes of other country banks as deposits and consequcntly did 8. The joint venture aspect of the Suffolk II system wa' ,hort-lived, cnding in 1826. Apparently. the A"ociate Banks believed their deposit requirements were too large for the services rendered and profits obtained. (Director's Records of the Suffolk Bank. March 14, 1826). The Suffolk reduced the permanent depo,it requirements of the Ilmton hanks. all of which continued to participate in the system except the Massachusetts Hank, in a series of steps from $30,000 or more in 1825 to S5.000 by 18,15. 9. (j;uhade and Silocr [6J discuss the critical differences hctween these two kinds of clearing arrangemcnts.
301
Free Banking II COMPETITIVE MONIES AND TilE SUFFOLK BANK SYSTEM
891
not result in .IIlY netting out of claims on the books of the Sulrolk. This kind of bi-lateral scttil:mcnt system economizes som.:what on the shipment of speci.:, but not as Illuch as a collective system. lndced, the enhanced prolitability of the post··1824 SulliJlk mechanism proh'lhly rellccts the reduced costs of sp.:cic shipm.:nts und.:r this syst':llI. Walk.:r cOlllmcnts on th.: "popular fallacy" that the Sulfolk system increastd the amount of th.: Sulrolk's sp.:ci.:, noting that the notion was "a fiction."J() A well-run collective net settlement system economizes on specie, which n(l doubt accounts for the grudging acceptance of the system by the New England banks and the rapid increase in the volume of note clearings following the implementation of the second Suffolk system. The efrectiveness of the Sulfolk's note clearing and redemption mechanism also depended on the ability of the Suffolk to influence and control the risk-taking activities of the participating banks. The elimination of the discount on bank notes meant that "the market" was no longer capable of signaling a change in the riskiness of a particular bank through a change in the specie price of its notes. This increased the incentives for each bank to overissue, i.e., depreciate the value of its notes. A systematic policy of overissue by Bank XYZ would be promptly revealed to the Suffolk, however, in the form of adverse clearings against XYZ. The Sulfolk could respond by presenting XYZ's notes for specie once its balance dropped below its required permanent balance (reserves), forcing a contraction in note issue. In the extreme, the Suffolk could suspend or expel a bank from the system which it judged as excessively risk. Dewey notes that a bank "ceased to be in good standing when thrown out of the Suffolk" [2,91]. The Suffolk also engaged in "jawboning" or "moral suasion" with its member banks. Hammond [8,553] reports, for example, that the Sulfolk "scolded its country correspondents like bad boys and admonished them as if their souls and not merely its earnings were at stake." It often commented to the members concerning "pressures on our money market", sometimes backing its advice with policy changes. Whitney makes note of a circular distributed to participating banks "informing them that .1Il account of tht: scarcity of money and in order to have some control over its own funds, overdrafts must be limitt:d to $10,000" [22,24-25]. Tht: Suffolk was even sympathetic on occasion, telling one bank: "We regret the necessity of these measures (elimination of overdrafts), but the deranged state of mon.:y matters makes them unavoidable". In essence, the Sullolk employed many of the risk control techniques (reserve requirements, discretionary lending, and moral suasion) utilized by other regulatory organizations such as clearinghouses in the latter 19th century [7] and by modern central banks. II The basic similarities of these mechanisms for monitoring and IU. Walk« noleS Ihal: .. rhe whole circulalion of New England banks oul of Boslon in OClob«. IH56, was SJ4.J,5(J2,J~b.
This, then. is the amount the Sulfolk is to cl!deem at sigill, as fast as il makc;:s its appli!arance in the city of what is the propurtion of its specie: to this sum of $39 millions'! Why it's less than O1i~ ('('''' ull/he dol/ur!" (ilalie> ongillal) [I ~.161. J l. ·1 hl!~e dc\'il.:c~ afC aililed at "\.fuality assurancli!" in the sense of pccdictablt: purchasing power. The Sulrulk I1lcl:hanislll alsu addressed the problem of countt:rfcit notes in an interesting and apparently successful fashion, Till! Sutlolk CllllLCacted out this aspect of 4uality assurance to the uUicer in charge of the foreign money dc:partment, ptJying him a lixcd sum ($4,25U in 1~26) to conduct tJlI aspects orlhe nott-exchange business (including hiring his own statt). lie was rcyuin.:d tu inspect I!a~h note's authenticity and to persunally assume <.ill risk uf loss in the forll1 of cuulltcrfeib, mutilall..:d ..:urrl!lH.:y, t:tc. Whitlley reports that such losses averaged less than $150 a year through nUb, "such grl!at cafe ,lIId wi..ItchJulne:ss had he exen.:iscd" L22,27]. 1he fact that the Sulfulk managt:d tht is:,ue: of monituring, and controlling the '-Juantil), of inauthl.:lltil.: Ill)h:s at such low I.:osts suggests that Klein':, obse:rvation th ... , "During lhl! free banking I.:ra. t..:IJulIlcrkll Hules were: a more: signiJicant prublem than the: fraud associate:d Wilh ovcri:,sue , . ."[11,450 I prob ... bly did f1o1 hold ill the New England Cd:,e. lJO)10Ii.
Hut
Free Banking II
302 892
Donald J. Mullineaux
controlling individual bank behavior suggest that a certain kind of contract appears to be delivered in both private and public settings when the production of money is at issue. In the next section, we argue that the SulTolk system represents a specific example of a generic kind of contract arrangement --the franchise.
V. The Suffolk System as a "Franchise"
A franchise contract typically involves an agreement in which the franchisee pays a certain sum of money for the right to market a product or service developed by the franchisor. Franchise contracts usually involve a number of elements, including the specification of the degree of management control by the franchisor, an indication of the assistance provided by the franchisor to the franchisee (e.g., training, on-going advice, site selection, etc.), an agreement concerning a royalty payment (usually as a percentage of sales), and a termination clause (which typically indicates the franchisor can terminate at will). A franchise contract falls somewhere between interfirm exchange (markets) and intrafirm exchange (hicrarchy). Most economic arwlysis of franchising have appealed to capital markets in explaining the existence of the franchise system. That is, the franchisor raises cllpital from franchisees in order to expand his operations. As Rubin ll5 J has noted, however, capital market theory is inconsistent with such an hypothesis. Both franchisor and franchisees would be better off if the franchisor formed a portfolio of shares of all the outlets and sold such shares to the franchisees. An alternative explanation for franchising emphasizes agency costs (especially the costs of monitoring and controlling agents). According to this view, the franchisor sells a "trademark" and the essence of the franchise contract involves attempts to estahlish incentives to maintain the quality of the trademark and control the extent of deviations from the established quality level. The franchisee requires incentives to be eflicient (to avoid shirking) and the franchisor needs incentives to maintain quality (to avoid depreciation of the trademark). The franchisee's incentive is a share in the profits of the enterprise. The franchisor's incentive to monitor and police the quality of the "trademark" illso is provided by share in the profits or revenues of the enterprise. The frilnchisor's role essentially is to control an externality problem. An individual franchisee has a strong incentive to cheat on quality maintenance, since the cost of such cheating is borne largely by other franchisees. While a franchise contract typically grants the franchisor the required mechanisms to "police" a trademark (such as screening of potential franchisees, inspection rights, and a termination clause), without a continuing share of total sales or profits a franchisor would sulfer diminished incentives to police the trademark. A well-designed franchise contract thus grants propertyrights to the parties in the areas each can elliciently control. We now consider how the Suffolk Bank system ciln be interpreted as a franchise mcchanism. Clearly, the Sufl'olk Bank did not develop some unique prodllct which other agents wished to market. The Sulfolk, indeed, was an "ordinary" commercial hilnk. purchasing assets and issuing liabilities much like other banks. We contend, however, that the Suffolk's scheme did involve an attcmpt to estahlish and maintain a "trademark" on hank notes. The trademark consisted of redemption at par, which essentially signaled that such a bank note
303
Free Banking II COMPETITIVE MONIES AND TIlE SUFFOLK BANK SYSTEM
!\93
was "as good as gold." The Suffolk provided direct maintenance of the trademark by redeeming at par the notes of any participating bank "judged in good standing."'! As with modern franchise contracts, the Suffolk possessed the capacity to screen potential participants and could terminate a bank from the system, apparently at will.i.l As noted ubovo:, the Suffolk provided a great deal of management advice to the member banks. One letter to the President of Woodstock Bank claimed that " ... too large a portion of your loan is in accomodation paper (long-term loans), which cannot be relied upon at maturity to meet your liabilities ... Since you are now placed at the head of the institution we hope you will take measures to change the character of your loans" [22,35). Another bank was admonished that " ... if all the banks in New England were moving at the same rate as you are, it would require more than all the capital of the banks in this city to supply their wants" [22,3!\]. According to Whitney, "Constant warnings like the above run through all the correspondence of the bank." The very nature of the system also involved the Suffolk in near continuous "inspection" of the quality of bank notes. And the Suffolk could respond to an observed depreciation in quality--signaled by abnormally large clearing-balance debits-by limiting overdrafts and by presenting the notes of the offending bank for specie. Whitney notes that for the member banks, "Their overdrafts on the Suflolk enabled them to do so (extend their circulation); but at the same time it put them completely in its power." Finally, like modern franchise arrangements, the Suffolk system involved a franchise fee (the permanent deposit) and an arrangement for sharing in the benefits of the trademark. The rationale for a revenue or profit sharing arrangement in a franchise is the necessity for the franchisees to motivate the franchisor to monitor and police the quality of the trademark. In the Suilolk system, this was produced through the overdrafting mechanism and the "price" charged for the trademark provision services. The Suff·olk clearly possessed strong incentives to monitor and control overdrafts.'~ Much of the correspondence referenced in Whitney concerns the Suffolk's management of overdrafts, which was highly discretionary. The Sutl"olk's "return" to any increased circulation via overdrafts was the interest it earned on ~uch loans. Yet if the Suffolk itself overextended such loans, it ran the risk not just of depreciating the trademark, but of bankruptcy. If the Suffolk Bank was largely responsible for restraining excessive note issues by New England banks, we might ask "what prevented the Suffolk from depreciating the quality of ilS own noles?" The threat of bankruptcy and the Suffolk's substantial investment in "n:pulalion capital" were no doubt relevant factors. Another, and less obvious one, was the fact that lhe Suil"olk "uiLi I/O interest on the permanent balances of the member banks. As a result, 111\: Sufrolk earned what amounts to seignorage on whatever increased circulation 12. I lir.! Sutlolk II iurallgcment did not require the: maintenance: of an account at the Sutiulk. A counlry bank was only rCl.juil eJ to cl!u..:cm thc:ir notes at par at SUIII~ Uo),on bank. The: Sutl'ulk wuuld in turn prcsl.!nl the note) of SUL:h banb to il) Uu)lon
Free Banking l/
304
Donald J. Mullineaux
894
that the Suffolk system provided. The potential loss of these additional monopoly profits further limited the Suffolk's incentive to depreciate quality itself. The Suffolk system thus can be viewed as a hierarchical mechanism. involving private control of individual bank managers, with a contractual mechanism structured much like a franchise contract. Such a contract economi7.es on information and other agencyrelated costs. Gorton and Mullineaux [7] have analY7.ed the commercial-hank clearinghouse (CHCH) as a related contractual mechanism for coping with control problems associated with a different, but related, bank product: demand deposits. Thus. there appear to he certain information-related problems associated with the production of money which call forth hierarchical mechanisms to economi7.e on these costs. In his discussion of free hanking. White [2 I] contends that free entry and exit into the money production industry and convertibility to base money are sufncient conditions to produce a stable monetary system. !lis discussion of note exchange facilities and clearinghouses fails to note the substantial elements of hierarchy, e.g., regulation, which obtain in real-world system such as the Suffolk or the 19th century CHCf/s, however. History seems to suggest that free markets. in the sense of a complete absence of hierarchy, "don't work" in the long run where money is concerned. This does not of course directly imply a necessity for government-provided regulation. It is a fact, however, that neither the Suffolk system nor the CHcH mechanism possessed longterm survival properties. The CHCH was supplanted by government regulation (the estahlishment of the Federal Reserve System); the Suffolk, however, was a victim of competition. In the final section, we examine the demise of the Suffolk system in an attempt to glean some hints concerning the feasibility of private hierarchical control of the money industry.
VI. The Demise of the Suffolk Bank System In November, 1858 the management of the Suffolk Bank announced it was withdrawing the Suffolk Bank from the note exchange business. While the proximate causes of the demise of the Suffolk system are varied [13], if! terms of our contractual perspective, the Suffolk became, rather ironically, a victim of Gresham's Law. In particular, a "bad" regulatory institution (The Bank of Mutual Redemption) drove out a "good" one (The Suffolk Bank). As almost all historical commentators on the Suffolk have noted, from its onset the system was unpopular with a large number of the participating banks. The activities of the Suffolk were seen as "a war against the country banks" involving a "Holy Alliance" of Boston banks. The country banks responded with more than words, seeking relief in the courts, petitioning state governments for laws permitting delayed redemption in specie, appealing to merchants to move their accounts from Boston to New York. and. from time to time, "retaliating" against the Suffolk by collecting a large volume of Boston bank notes and presenting them for specie. II In 1835. a petition was presented to the Massachusetts legislature for the chartering of a new bank, owned by the country banks, to perform services similar to those rendered by the Suffolk. The bill was postponed by a fairly narrow vote and did not come up for renewed discussion in the following year. IS. Thus the hypothesis that firms expend resources to avoid regulation applies to private, as well as gcwcrnll1cllt regulation.
Free Banking II COMPETITIVE MONIES AND TilE SUFFOLK HANK SYSTEM
305 895
In 1855, following several years of effort, a number of country banks and a single Boston bank were successful in obtaining a charter to form the Bank of Mutual Redemption (BMR), the lirst "banker's bank" in the United States. Its stated purpose was "to redeem our currency at par in Boston." Its stock was to be subscribed and held exclusively by New England banks. It took almost three years to raise the necessary capital, and the B M R opened its doors on August 23, 1858. Shortly afterwards, several of the banks which held stock in the new bank withdrew their deposits from the Suffolk and announced their intention to redeem at the BM R. The expectation was that each institution would clear and redeem notes for the members of its system, and the Suffolk and the BMR would exchange with each other through the Boston clearinghouse, which was then in operation. lo The Suffolk refused, however, to present the notes of nonmembers to the BM R since the latter maintained no permanent deposit with the Suffolk. The Su/lolk announced its intention to send such notes home for redemption in specie "at such times and in such manner as to us may seem best." L.ake notes that: "This refusal to recognize the Bank of Mutual Redemption raised a storm of hostile criticism which surpassed in virulence that of any previous period" [13, I97]. The Suffolk's action was criticized in one Boston newspaper as "arbitrary, tyrannical, impertinent and unjustiliable.~'J7 Many country banks undertook systematic policies of substantially dc:laying the payment of specie claims by the Suffolk, with substantial support from the popular press. Mure pressure was brought to bear on the Su/lolk when the Boston Clearinghouse reversed a prior decision and voted on October II, 1858 to admit the BMR. On the same day the Massachusetts Bank Commissioners advised the Suffolk that it should either continue to receive the notes of all the banks which had withdrawn and present them at the BMR, or it should simply decline to accept such notes. The Suffolk at lirst agreed to exchange with the BMR, but then abruptly announced a week later that: "The business of assorting country money will not be continued by this bank after 30th November, 1858." Historians (Lake and Redlich, for example) suggest this may have been a strategic move to regain support from the Boston banks, which might in turn collectively influence the country banks to rejoin the Su/l"olk system. By then, more than half the banks had defected to tile 8M R. This tactic failed, and the Suffolk again reached an exchange agreement with the 8M R. It lasted roughly a year, when the Suffolk again withdrew from the arrangement over a dispute concerning the notes of certain Rhode Island banks. For all practical purposes, the Suffolk had given up the foreign money business. The Suffolk's demise has been interpreted by several economic historians largely in psychological terms. Redlich contends that: "Its defeat was caused mainly by the fact that when once it had gained a monopoly it became dictatorial and highhanded. Unwilling to adapt itself to changing conditions, it became unable to find its way back when the business became competitive" [14,77]. Similarly, Lake states the "fundamental explanation for the final collapse of the system lay in the autocratic attitude of the Suffolk Bank towards the country banks, an attitude which was a natural but unfortunate result of the outstanding success of the system" [13,206]. The management of the Suffolk, of course, saw things differently, contending that: 16. Correspondence between the managements of the Suffolk and HM R shows that the Suffolk initially agr.ed to such an arrangemenl. 17. BoS/un Allus alld Daily Bee, Sept. 24, I SSS.
Free Banking II
306 896
Donald.l. Mlillineaux ... we (cannot) be expected to stand out against public opinion, prejudiced and excited, after it (the note exchange business) becomes unremunerative, and ha7.ardous to the stockholders ... and ... the bank does not wish to stand in the way of a trial of the attempted experiment of /I foreign money system, to be conducted on less stringent principles [22,59).
While a more "politic" stance by the Suffolk might have precluded the BMR from obtaining a charter, our analysis suggests that the after-the-fact instincts of the Suffolk management were appropriate. From a contractual standpoint, the BMR arrangement was considerably less advantageous than the Suffolk system. Stockholding banks were not required to maintain any permanent deposit, for instance, and while other member banks had to keep the same deposits a:; those required by the Suffolk, the B M R offered to pay interest of 3 percent on such deposits. The B M R also paid the same interest rate on temporary balances in the clearing account. In addition, the BMR directors voted in May, 1859 that overdrafts would not be permitted "except by special agreement." The B M R contract therefore lacked the strong incentive properties contained in the Suffolk system to monitor and police the trademark of par redemption. Members could avoid franchise fees by purchasing stock in the OMR; such a policy reduced the incentive of BMR to carefully screen applicants. Without overdrafts, the B M R had reduced incentives to monitor behavior,l. and its policy of paying interest on deposits reduced its seignorage or monopoly profits. In short, the Suffolk was quite correct in judging the BMR contract "less stringent." Lake reports that "the policies of the bank (OM R) were fadrom conservative." Adverse balances at the clearinghouse were met by pledging its own notes for loans from other banks and specie reserves were brought to the required level by borrowing on the same security. The Bank Commissioners also instituted actions against the BMR in 1862. The BMR did manage to maintain the par redemption system, however, although the efficacy of the system was never thoroughly tested because of a general suspension of specie payments in 1861 and the passage of the National Banking System Act in 1863 which resulted eventually in the elimination of state bank notes. In our jUdgment, then, the Suffolk's exit from the note exchange business was a result of "competitive" pressures in which a lower-quality system for maintaining a trademark on hank notes "drove out" II higher-qulllity one. The historical evidence of course speaks less thlln conclusively on this point since the OMR system was never thoroughly tested. The latter arrangement lacked the distinctive franchise-like properties inherent in the Suffolk contract, however, and its long-term efficacy seems open to doubt. The evolution of the Sufl()lk mechanism seems to satisfy Stiglit7.'s suggestion that in the presence of asymmetric information and other agency-related costs there is a need to find "institutional arrangements that reduce the free-rider problem, that extend control to those for whom information is relatively cost less, and that provide incentive structures such that those who are in a position to exercise control take into account the consequences for all the enected groups" [16,148]. The Suffolk system did all of these things, and did them rather well. Yet it was a victim of a factor (competition) which economists normally consider positive rather than negrtive. " substantial literature now exists which argues that in the presence of imperfect information, there may not he a competitive equilibrium. (Stiglitz [17J is an early example.)
IR. 'J h",. in nllr view. Redlich', .i"d~l1enllhallhe overdraft policy of R M R wa, "more .""tio,,," '14. 771 than the Sulfolk'!; is an inaJlrroJlri~lte conclu~ion.
307
Free Banking II COMPETITIVE MONIES AND TilE SUFFOLK BANK SYSTEM
897
In efrect, monopoly or elements of imperfect competition are required in such settings to produce excess profits, which provide the incentive for a firm not to depreciate quality. I" Does all of this constitute an argument for government regulation of the money-production business'l A thorough examination of this important issue is beyond our scope here, but the historical experience of the Sufl"olk Bank system is at least suggestive of the need for government regulation. It is now widely recognized that in the presence of informationrelated problems, the simple exchange process will not generate etlicient results and more complex arrangements may be called for. If facilitating such informationally complex exchanges requires the inclusive membership of an entire group, governmental institutions may be necessary. At a minimum, it requires some form of collective, hierarchical mechanism for resolving agency-related problems and economizing on such costs. Neither theory nor historical experience offers much succor to those who argue that simple (nonhierurchical) exchange is an optimal mechanism for the production of money. I~.
Klein ami 1..111« 112] demOnSlral. formally Ihal increased price is a means of assuring conlracl performance.
A 1l111lpcrfocIIling lirm I()sc~ a discounted ~trc:am of n:nls which must exceed the wealth gained fWl1ll:thirking. IlIlhe ~;.l!!lC 01' lhe Sultolk Bank, the rC4uircmcni for interest free dcpu~its created scignoCilge·likc pro Ii Is which served the Siamc fUIIL:liun.
References I. Akelluf. Gwrge. "Th. Markel for Lemons: QualilY IJncerlahlly and Ihe Markel Mechanism." Q"""('r/y JowlluloJ i:i.'OllolUics, August 1'J70, 4H~·-500. 2. Dewey. Davis K. St",. /Junking Ikfure the Civil Wur. Washinglon. O.C.: Governm.nl Prinling Ollice. I~IO (N",lluual Munetary Commission). 3. Dillislin. Willidm. /Jank Nute Nepo"eTl'and ('mll/t.-feit Detectors. 1826-1866. New York: American NumisIn",i!..' Sodety. 1,:)49. 4. "dedman, Millon and Anna Schwarlz. A ,\fonetary /liS/ur)' of the United Stute•. 1867-196U. Chicago: lJnivcc:)ilY ul Chi~ago Press, 1':)71. 5. Friedmdll, Miltun .... Progrum.!(}, l\Jo",,'ar.~' SwbiUlY. New Yock: Fon.lharn lJnivl!csity Prt:ss, 1959. C, (iacbaJc, Kenneth and William Silber, "The Payment SySh!1ll and I>ollll!slic t:,,-chaugc Rates: Tcchnulugical VchU~ In:tlitutiunal Challgc." Jou,,,al of AIOllel"'>, I:.i.'olwm;('j', January 1979, I·M22. 7. (junon. (jary amlD'lIIilld J. Mullineaux. "Th. JoinlProducliun of Cunlidence: Corumorcial Bank C/c,,,inghuuses .md Ihe n,cory of Hierarchy." ManuscriPI, 1~~5. ~. 1I,."nJll~md. Heay. BUIlk.s UII" P(JIiIil's i" A",,'fit'a. Princetun: I)riuccton IJUiVl!fSily Press, 1951. 'J. lfiL)'ck. FricJril.:K A. 111t: Ot'lUuiullalizUlioll of 4\1(11"'Y. London: Institul": of Econumic Affairs, 1976. 10. King. Ruben, "On the Ecunomics of I"rivate;: Moncy," Journal O/,\lUIJt/:lr,l' J:.:cOIwmi,'j', July 19M3, 127-SH. II. Klein, Iknjamin ••• rhe;: Competitive Supply of Moncy." Journal of fI./UfJey. Cred;1 umi BunkilJg. Nl)Vcmbcr 1~74.
423- 53. 12. Klein. Benjamin and Keilh l..mer. "The Kole of Markel Forc.s in Assuring Conlraclual Performance." Jtmflwl v/ Polilh'ull:.'\'OIwmy, August 19HI, 615-41. 13. Lake. Wilfred S .• "Th. End of Ih. Sulrolk Syslem." Jou",ul ull:.'·",wmk lIi,wr.!'. November 1947. IH3··2U7. 14. Itcdlieh. Frill. 171e Molt/i"g olAmer;"an /Junking. New York: lIafner Publishing Co., 1~31. IS, Kubin. I)aul, .• n,,: fhc:ory ur the Firm and the Structure of 'hI.! Franchise ('oillract." Jouflwl of l.ilu' Ulul l:.i·..,I",,,;'·.I. April 197Y. 223,-33. lb. Sliglill. Joseph 10., "Credil Mark.ls and Ihe Conlrol of Capilal." Jou",u/ uf Money (,,,,,/it /Junking. May 19~5. /33 -52. 17. Sliglitl, Joseph E. "Information and Economic Analysis," in Currt!1I1 1:.'c.'olJomic Probieml', cditc:tJ by Mkhad Parlin and A. Kober! Nubay. Cambridge: Cambridg. IJnive"ily Pr •••• 1975,27'-52. I~. VanFenslermaker. J. and John E. Filer. "lmpacl of Ihe Firsl and Second 1I'lnks uf Ihe Uniled Siaies and Ihe Sulfulk Sy.!lll!m UII New England Hank Money: 17'J1-' IM37," Jour"ul 0/ A/(J'I'·.I~ (''''dil, und LJallAiliK. February Il)Ko, 2~ 40. I~. Walker. Amasa. 111t! NUlllre uwl U.es uf Money afltl Mixed Currenq. lIoslon: Craslus. Nichols and Company. 1~57.
",,,I
308
Free Banking II 898
Donald J. Mullineaux 20. While. Borace. MOlle), alld Bollkillg. lIIu.."atrd ".I' Americall lIi,,'ory. Sih ed. Boslon: Ginn Publishing Co .•
1915. 21. While, Lawrence. Free Bollkillg ill Bri,aill: Theory. F.:rperir'Il·(' alld Oehatt'. 1800-45. Cambridge: Cambridg. Univ."ily Pre .., I'IK4. 22. Whilney, D. R. 77,. Sujjolk Balik. Cambridge, Muss.: Riverside Press, 1878. 23. Williamson, Oliver, "Transaclions Cost Economics: The Governance of Contractual Relations." JourII.1 ojl:"w alld Ecollomics, October 1979,233-61.
n..
[19] Competitive Monies and the Suffolk Bank System: Comment
Donald J. Mullineaux's piece 16) in the April 1987 issue of this journal is in many ways a useful contribution to the emerging literature (whose importance is stressed by Gorton II) and by King 14)) on the institutional features of private monetary systems. The historical example of the Suffolk Bank note-exchange system of antebellum New England certainly deserves the attention he pays to it. Because of the importance of the Suffolk experience, Mullineaux's interpretation of it merits careful scrutiny. Mullineaux offers the Suffolk system as evidence against the proposition that private markets alone, with free entry and exit and contractual convertibility of bank notes into a common outside money, can provide monetary stability. From the elements of hierarchy present in the Suffolk system, elements he deems essential to a stable monetary system, and from the evidence he thinks the Suffolk experience provides on the poor survival properties of unregulated banking hierarchies, he infers (albeit tentatively) that government regulation of hanking may be retjuired for monetary stability. We believe that there are several serious problems with Mullineaux's argument. The major problems include: (I) a potentially misleading classification of hierarchies as "non-market" phenomena, which confuses two different meanings of the term "market" and promotes a misunderstanding of the arguments other economists have made; (2) an overestimation of the importance of hierarchical structures in free banking, based on Mullineaux's failure to appreciate the role of legal restrictions in shaping the particular banking arrangements he discusses; and (3) an unjustified inclination toward government intervention based on the unsubstantiated claim that a laissez faire monetary system supports too little institutional hierarchy to deliver stability.
I. Markets versus the Free Market System Mullineaux 16, 885) motivates his discussion by citing contmsting polar views: on one side, that free-market (or laissez-faire) forces are completely adequate to discipline bankers' so as to deliver monetary stability without any government role; on the other, that laissez faire is completely inapplicable to money and that unified government control is a necessity. I-Ie then indicates that analysis of the Suffolk Bank system "suggests that the truth lies 'somewhere in the middle.' " In this context one would presume that he means "suggests that the system most conducive to stability lies somewhere between laissez faire and complete state control." Curiously, this is 1101 what Mullineaux means. His next sentence remls: "That is, the Suffolk system shows that markets-in the sense of exchange relationships characterized by the comp/ele absence of hierarchy-were unable to control and discipline the hch'lvior of 19th ccntury bank managers." A similar statement appears later 16,894): "History seems to suggest that free markets, in the sense of a complete absence of hierarchy, 'don't work' in the long run where money is concerned." These statements promote confusion by confounding two independent claims: (I) that a 215
Free Banking II
310
216
COMMUNICATIONS
laissez faire monetary system will of necessity evolve institutions embodying a substantial degree of hierarchy; and (2) that the necessity of hierarchy is somehow antithetical to laissez faire. We discuss the first claim in the' following section. At present we wish to object that the second claim is simply false. The capability of laissez faire to deliver monetary stability does not require that "markets" in the sense of completely non-hierarchical exchange arrangements alone must do the job. lIierarchical structures are in no way logically excluded from playing an important role in a free banking system. On one level this is obvious: the simple institution of the firm involves hierarchy, after all, and no one associates laissez faire with the absence of firms. Nor are franchise systems, whose hierarchical features Mullineaux emphasizes, at all inimical to laissez faire. Mullineaux's real concern, of course, is with the hierarchical features of note-exchange or clearing institutions.' Our point here is that the presence of hierarchical features in private clearing institutions is fully consistent with laissez faire. This has been recognized in the literature. Recent work of ours [8, 9J discusses the hierarchical features of free banking clearinghouses that should be expected to evolve in a laissez faire setting. Works by Trivoli [II J and Timberlake [10], who earlier drew attention to the Sutfolk and commercial bank clearinghouse (CBCH) systems, are specifically concerned to show how free market forces have promoted monetary stability by evolving special private hierarchical structures.'
II. How Much Hierarchy Does Free Banking Require? Though some degree of hierarchy clearly has a role in free banking institutions, we fear that Mullineaux seriously overstates its importance. The overstatement results from his generalizing unduly from the special structures of the Suffolk and CBCH systems. Mullineaux overlooks the fact that these special arrangements evolved in settings significantly constrained by government regulations. The Suffolk system was conditioned by legal restrictions against entry and particularly against branch banking in New England. Had branch banking been allowed, a noteexchange arrangement less hierarchical than the Suffolk system would almost certainly have arisen. More opportunities would have existed for decentralized note-exchange. More importantly, since branching would have allowed all banks to have offices (for direct redemption) in Boston, numerous banks would probably have participated as co-equal partners in forming an organized note-exchange facility. The accumulation of country banknotes in Boston would never have become a problem, because active note-redemption would have been routinized much sooner. That at least is how events unfolded elsewhere, particularly in Scotland and Canada, where notes were competitively issued without restrictions against branch banking. Mullineaux's hypothesis that an efficient note-exchange system must resemble a franchise operation, with one bank playing a special central role, is not borne out in those countries' experiences. The clearinghouses of the National Banking era took on the special hierarchical functions I. It is true, as Mullineaux charges [6, 8941, that White's [I3J discussion of note exchanges and clearinghouses fails to emphasize the hierarchical or quasi-regulatory aspects of the Suffolk or the later U. S. commercial bank clearinghouse systems. This omission should be understandable, however. in light of the fact that White's historical work concems British monetary institutions and not American. The note-exchange institution of the Scottish free banking system was far less hierarchical. This raises the question of whether the Scottish or the Suffolk system is the more relevant model. i.e., of how much hierarchy is actually required by a free banking system. This question is pursued below. 2. Hayek 12J. perhaps the best known recent discussion of monetary laissez faire. does not discuss clearinghouse institutions because Hayek. contrary to Mullineaux's 16, 885] characterization of his position. does not think it necessary that bank liabilities be convertible into some more basic outside money.
311
Free Banking II COMMUNICATIONS
217
for which they are noted today also in response to legal restrictions. The restrictions in this case were a prohibitive 10% tax on notes issued by state-chartered banks (the same tax which finally ended the New England note-exchange system) and bond-collateral requirements on notes issued by federally chartered National Banks. These restrictions made it ditlicult or impossible for commercial banks acting individually to satisfy their customers' varying demands for currency. The supply of eligible collateral bonds tell dramatically in the last part of the 19th century, making note-issue by national banks increasingly difficult and costly [7; 14). Because the CBCHs were able to circumvent the It:tter of the laws restricting note-issue, they took on the extra function of issuing currency in order to aid their membt:rs during emergencies when the public's demand for currency was especially high. In other nations where note-issue by commercial banks was basically unrestricted (Scotland before 1844, Canada before 1914) clearinghouses did not issue currency or serve as lenders of last resort to their members.) Thus it is doubtful that the "substantial elt:ments of hierarchy" in the Suffolk and CBCII systems highlighted by Mullineaux are inherently required by privatt: mont:lary or free banking systems. On the contrary, a broadt:r historical perspective suggests that a high degree of hierarchy is an indirect consequence of particular sorts of regulation. More work needs to be done to explain satisfactorily the international differences in note-exchange arrangements. (The non-hierarchical system of Sweden's private note-issuers, as mentioned by Jonung [31, also deserves attention in this elfort.) But it is clear that a Suffolk-style hierarchy cannot be presumed efficient as a gent:ral proposition.
III. The "Survival I'roperties" Argument Mullineaux [6, 894) recognizes that the presence of hierarchical structures in a private monetary system does not "directly imply a necessity for government-provided regulation." Nonetheless he concludes [6, 897) that "the historical experience of the Suffolk Bank system is at least suggestive of the need for government regulation." He draws this conclusion ufter observing that the Suffolk system and other historical free-market monetary arrangements huve had poor "long-term survivul properties" [6, 894). The Suffolk system was of course discontinued. But Mullineaux's view that its successor (the system administered by the Bank for Mutual Redemption) was "a lower-quulity system" is not supported by the evidence. As Mullineaux notes, the BMR competed successfully with the Suffolk for the note-redemption accounts of country banks by offering to pay interest on those accounts. It then, in his words, "did manage to maintain the par redemption system" ,6, 8961 until the general suspension of 1861. Statements of the BMR's inferiority put forth by the management of the Suffolk Bank are cited, but of course these are not unbiased observations. Mullineaux thus offers no evidence to support his notion that the BMR system was of lesser quality than the Suffolk.' He admits [6, 896J that "the historical evidence of course speaks less thun conclusively on this point [the supposed inferiority of the BMR at maintaining bank note qualityJ since the BMR system was never thoroughly tested," but we would say that the evidence speaks not at all 3. r'Or an account of how Canadian banks prior to 1914 deal! with the surt of increases in currency demand that forced U. S. clearinghouses to issue currency, see 1121. 4. We also note that no theoretical rationale is given as 10 why an inferiur system fur maintaining pruduct quulily should be expected to drive out a superior one. Klein and Letller 151, which Mullineaux cites in this context, contains no such resull. The banks, after all, did have an interest in maintaining product (banknote) quality.
312
Free Banking II 218
COMMUNICATIONS
on this point. Mullineaux is left mcrely with the unfalsifiable assertion that the BMR's "long-term ellieacy secms opcn to doubt" because it lacked the franchise features he deems necessary. While not providing a ranking of the BMR relative to the Suffolk, the available evidence docs indicate that free-market institutions succeedcd in promoting monetary stability in New England until the outbreak of the Civil War. During the war the Federal authorities, who aimed primarily at raising revenue, imposed the regulations which scuttled the system. Yet for Mullineaux (6, 8851 "the SulTolk's failure to survive in New England may provide some hints concerning the necessity of governmcnt intervention in b,mking markets." We cannot sce how it may do so once the Suffolk's demise is understood. The Suffolk, as we arc told, "was brought down by a blcnd of market and extra-market 'competition'." We have seen that its nmrket competition, the Bank for Mutual Redemption, was not demonstrably inferior at maintaining bank note quality. By "extra-market 'competition' " is probably meant the Federal banking legislation of the Civil War. Mullineaux similarly cites the non-survival of the private CBCHs of the National Banking era, which wcre victims of the Federal Reserve Act. But surely one would not want to say that any free-market arrangement eliminated or replaced by government regulation must have deserved it. Governments may serve interests other than promotion of the public welfare; and even wellintentioned governments may make mistakes. The mere enactment of banking legislation does not demonstrate its goodness.
IV. Conclusion
To summarize: the role of hierarchical structures in a free banking system, including banking firms and arrangements for note and check clearing, surely merits exploration. Mullineaux is to be applauded for turning his efforts in this direction. BlIt recognition of a natural role for hierarchy does not imply that free-market forces cannot provide an enective mechanism for the production of monetary stability. Mullineaux's suggestion of a contrary view, however tentatively expressed, seems to be based on incorrect interpretation of, and invalid inferences from, the case of the Suffolk system. George A. Selgin University (!f" /long Kong Hong Kong Lawrence H. White University (!fGeorgill, Athells, Georgia and Nell' York Ullil'ersity, New York
References I.
Gor1011.
Gary. "Banking Theory <.Inll Free Banking Hislory: A Review Essay." Journal of MOnel",.\, Economics,
September 19K5, 267-76.
.
2. lIt1yck. F. A. /J('Ilatioltali.mrioll (~r MOlley, 2nd cd. London: Institule of Economic AfT;:lirs. 197R. 3. Jonllng, Lars. "The Economics of Private Money: The Experience of Private Notes in Sweden 1831-1902." Manuscript. Lllnd University. 19K7. 4. King, Robert. "On the Economics of Privatc MOJlcy." jOIlrt/a/ ({M(Hlrlar." Economics. July 19X3. 127-SH.
313
Free Banking II COMMUNICATIONS
219
5. Klein, Benjamin, and Keilh Leffler, "The Role of Markel Forces in Assuring Conlraclual Performance." JOllrnal 1981, 615-41. 6. Mullineaux, Donald J., "Compelilive Monies and Ihe Suffolk Bank Syslem: A Conlraclual Perspeclive." Souther. Economic Journal, April 1987, 884-'17.' 7. Noyes, Alexander Dana. History ollh. Natiollul-Bank CII"ellcy. Washinglun: Governmenl Printing Office, 1910. 8. Selgin, George A. The Th.ory 01 Free BUllkillll: MOlley SlIl'l'ly ",,,Ier COllll'''titi.,. Notel,w", lI'luwa, New Jersey: Rowman & Allenheld, forthcoming. 9. - - - , and Lawrence H. While, "The Evolulion of a Free lIanking Syslem." ECOIIOlllic III'IIIiry, July 1987, 439-57. 10. Timberlake, Richard H., "The Cenlr~1 Banking Role of Clearinghouse Associations." JOllrnal qfM(mey. Credit, and Banking, February 1984, 1-15. II. lrivoli, George. Tire SlIffolk Balik: A S/ud.v 01 (/ Fr••-EnteTl'ri". CIo'(/rillll SY.,/tlll. London: Allum Smilh Inslilule, 1979. 12. Wells, Donald R., and L. S. Scruggs, "Canadian Banking Before 1914: Implicalions for Financial Deregulalion." Manuscript, Memphis State University, 1986. 13. White, Lawrence II. free Ballkillll ill Britaill: Tlreory, t:.rl'eriellce, wul /Jel"'t" IH()()-1H45. C",nbridge: Cambridge University Press, 1984. 14. - - - . "The Growing Scarcity of Bonknoles in Ihe Uniled Slates, 1865-/913." Manuscripl, New York University, 1987.
of Political Economy, August
[20] Competitive Monies and the Suffolk Bank System: Reply
I. Introduction George Selgin and Lawrence II. Whilc's ('ollnllcnl 171 on my analysis Ihl or Ihe SIIII.,lk Balik syslem IiH:uses primarily on Ihe Icnlalive conclusions I drew concerning Ihe "olenlial cllicacy "r free hanking in the United States. They argue that there is nothing in the hislorical experience of the Sufrolk or commcn:ial hank clearinghouse (CBCII) systems which constitutes evidcnce against the hypothesis that private markets can provide monetary stability. Selgin and White (S-W) go on to suggest that the hierarchical reatures of these systems were primarily n product of legal restrictions and regulation. As SoW note, my conclusions on the issue of regulated versus /a;sJ(':fa;re hanking were stated weakly. They remain tentative. but no less so in light of S-W's analysis. In their haste to defend free hanking. SoW appear to have missed the main point of my article. And their conclusion that the Suffolk and CBCII systems were occasioned by legal restrictil'ns cannot he supported by analysis or the historical evidence. I welcome this opportunity to c1nrify some issues concerning the relevant perspective for resenrch involving unregulated hanking systems. a litemture to which S- W have made notahle contrihutions. My reply is structured around the three prohlems they identify with my analysis.
II.
Confllsin~
Terminolol!Y
SoW note that there is some pmsihility for confusion in my analysis. since the term "market" is used in two different contexts. In particular. the term "nmrket" cnn he used hroadlv to connotc the ahsence of !!overnmental restrictions (a Inissez-faire system) or the term can he used in a narrower sense to renect exchnnge tmnsnctions mediated hy mnrkets mther than hy authl1rily (hiemrchy). My contractual analysis of the Sun'olk system employs the nan-ower usage-n~ S·W nole. I define mnrket exchanges as those chnmcterized hy the complete ahsence of hierarchy. Given this context. it is hardly "misleading" to classify hierarchies as "nonmarket phenomt'nn." Williamson 1101 is perhaps the strongest proponent of this view. and the samc author 1II1 also has demonstrated how laissez-faire cnn deliver mnny different rorms of hierarchy to solve dillerent kinds of tmnsactions-cost related prohlems. Indeed. the fundamental message of my arlicle is thnt the infornmtion-related transactional characteristics or money require some form of contracillal device or hierarchical mechanism ror any monetary system to function. lIaving demonslralcd thai some form of contract is required in a particular setting. it is certainly consislent to ql1e~li"n whelher a laissc7.-lilire systcm ("Ihe market." in the hroad sense) will deliver the required conlmct. Buchanan III and Goldherg 121 argue convincingly that there can he no ,,"(,.\""'1,/;(11/ Ihat the market will deliver an efficient outcome.' I agree strongly with SoW that hicrarchicalmcchanisms
I. BlH.'hamm SICIICS. for cxumplc: "When the slancJcucJ assumptions cue nut uc~crirti\'c . the simple c'l..·hal1~c prot'c'", will nol gcncr3tc efficient rc!\uhs. and more complex ;'lmmgcmcnts mel), he called for. BUI thc~c ilrrilngcl11l'n" may ... till
220
315
Free Banking II COMMUNICATIONS
221
arc notlu!('t'.mlrily "antithetical" or "inimical" to laissez faire. but I also colllend that. depending on the nature of the factors that prompt hierarchy in the private sector. there may be some lessons to be learned concerning the appropriateness of regulation. In my view. s-w cOlllinue to underestimate the imponance of these factors. as I will detail in the ti.lllowing section.
111. Thc l(cI"li\lc IlIIllorlallcc of lIicrarchy ill I'.·CC
Uallkill~ Sy~lcms
S-W claim Ihat I "seriously o\lerstate" the imponance of hierarchy in free banking systems "by generalizing unduly from the special structures of the Sulfolk and CBCII systems." s-w claim that these institutions are largely a rellection of legal restrictions on entry and on branch banking in New England. S-W do not spell out how entry restrictions are relevant to the issue at hand. but they claim that. "since branching would have allowed all banks to have offices (for direct n:demption) in BoslOn. numerous banks would have panicipated as co-equal panners in Ii.muing an organized note exchange facility" [7.2161.' S-W cite Scotland and Canada as examples of CIlulllries with less hierarchical note exchange systems. S-W also argue that the CBCH systems which developed in the laller half of the 19th century were also largely products of regulation (the 10% tax on state bank notes and bond-collateml requirements on notes issued by National banks). In my judgment. S-W have seriously overstated the role of regulation in expillining the hierarchical 1I11ributes of the Sutfolk and CBCH systems. The basic problem which both the Suftillk lind CBCH systems were addressing involved the I-I exchange rate between bank notes (Sutl'olk) and specie and between bank deposits (CBCII) and specie. These I-I exchange mtes imply that the payments-reillted lillbilities of all blinks arc perfect substitutes for ellch olher. In this selling. each bank faces incentives to depreciate the quality of its notes or deposits. Some mechanism or contract is required to control the tendency to depreciate quality and enforce the viability of the I-I exchangl: rate. It is the cllse that 1I branch bank faces higher costs to a systematic policy of overissue. since there would likely be a more rapid drain on reserves in a branch network than in a unit bank. Nevenheless. a branching system only reduces but does not t'lilllil/(I/{! the incentive to overissue. Nor does a simple (non-hierarchical) note exchange system eliminate the incentive to overissue. It seems much more likely that the primary reason why the Scollish banking system did not suffer systematic problems with overissue rellccts the unique contractual design of that system. In particular. the unlimited liability feature at most banks in Scotland severely discouraged overi~sue.j No U.S. banks operated under such a contmctual constraint. although sOllie states did illlpose a "double-liability" on bank shareholders. S-W's arguments concerning the CBCHs arc especially unconvincing. While clearinghouses be eXiJmiucd in a cuntractual fcank:work . . . . In fadli(alin~ Ihc~c cOlllplex cxchangl!s. which I1Ii1Y rClIuin: (he in\'.'lu~ivl!
1I1I.!lIlbcl"!)illP uf Ihe whole group, CUIlC(;llVt!-govcmmcntal inMilulions may b\! called for" i 1.2271. GulJberg Ul)h:~ Ihal . lilihm: lu apprcl.'ii.IIC Ihe compl..:xily uf 4:Ulllra4.:lual arrangcllu:IlIS in the private )..:",'101' i!lo .. pi to ICOJve th..: analy:-.I unduly !)OJllguiuc it) 10 the cflkacy uf privale marKel ~ululiuus lU problelll) in Ihe n:gulalury )c~Iur" 12, 4171. 2. The aoj..:divc "<.:u-c4ual" is. of CUUf!loC, 4uil\! rdl!vanl 10 Ihe nature uf the hil.!rar~hy. GWllUl 131 ~uggC:-'b the dli.lfh.:n.:J halll!lo OJppcar 1o have been "dominanl players" in the Sculli~h hankiul! !ty!th.:m. whidl implie!lo Whilc Il}1 Hlay hil\O'; ulld~n::-.lilHilh:d liIe- dcgrcc uf hicran:hy in lhill :-.y:-.lcm . .J. Tlli!'. L'tJlllladuul OI:-'PCt.:t uf Ihe St.:uubh ball~illg !tY!'olem i!t Wli.ll elI/Oll"-'/ IIIC S,.:ulli~h 1101\: cXI,,:hallgc !'Iy!'ol\:llI lu he ic)) hicran.:hit.:al. The ri!lok!lo a~!tocialctl wilh uVl!ris!tuc anJ potential raHun..: wen: very largely borne hy bauk )IuL'LhuIJcr~. A:-. (;orhm 131110lc). Ihi:'!. unlimitcd liabililY !!.f.:hclUc; (I) WiQ ibcil a pn.xJU~1 uf ~I .. IC iIlICr'v'ClHioli. anJ l2J 1101 1I\!4.'\! ... ~arily uplilllallhllll'" fI!tk·:-,lialilig pcr)pc. . . li\l\!.
316
Free Banking 11 222
COMMUNICATIONS
did "clear" a small volume of notes, their primary function and activity revolved around demand deposits. as Gorton and Mullineaux [41 have demonstrated. Deposits were not, of course. subject to either the 10% tax on notes or the bond-collateral requirements which SoW cite as the causes of the hierarchical functions of CBC/ls. Gorton and Mullineaux 141 link the specilic hierarchical aspect of the C13CHs to the transactionul llttributes of demand deposits. which dilTered from those of notes. SoW also miss the salient point of currency issue by the CBClls. In pllrticular. since this form of currency was a joint liability of all the numbers, the CBCI-f became even more hierarchically structured during these periods. I can think of no way to attribute this panic-related shift in contractual structure to regulation. I heartily concur, however, with SoW's conclusion that "more work needs to be done to explain satisfactorily the international dilTerences in note· exclmnge arrangements" 17. 2171.
IV. The "Survival I'roperties"
Ar~ument
As I noted in my article. no convincing policy conclusions can be drawn from the collective demise of the Suffolk and the CBCH clearing mechanisms. After all, these represent but two "observations." Reasoning along similar lines. it seems reasonable to point out that the Scottish system represents one observation. and as Gorton 131 emphasizes. many of White's conclusions concerning that historical experience are open to alternative interpretation. S·W again cite regulation as the factor which accounts for the discontinuation of the moni· toring aspects of the Suffolk and CBCH systems. In my article. I agreed that regulation was the proximate cause of the demise of the CBCH system. But regulation is not itself an "uncaused" phenomenon. and it is worth investigating what factors prompted regulation. This is. of course. a large task and one my article did not undertake. Instead. I argued. on theoretical grounds. thnt the supplanting of the Suffolk by the BMR represented an historical example of "a had system driving out a good one:' If. indeed. had systems drive out good ones. then there is a stronger c .. ~c for regulation on the grounds of the government's comparative advantage in enforcement. S- W argue that the evidence "speaks not at all" on the issue of the inferiority of the BMR systcm. But this claim ignores the fact that the Suffolk itself was distinctly unsuccessful in the note-clearing business under an alternative contractual scheme. As McCloskey 151 has suggested. the essence of economic history is to "tell a convincing story" and arguing from a sound theoretical perspective is surely one way of being convincing.' In sum. my article should not be reviewed as "a hrief for bank regulation." Rather. it is an attempt to analyze an existing historical episode involving "self.regulation" from a particular perspective. The prior work of Selgin and White tends to ignore the information-related contractunl features of hank products and of banking instilutions. They continue to do so in their Comment. Consequently they have overlooked. in my jUdgment, some critical issues involving enlilrcement and incentive problems which arc quite central to "the free banking debate." Donald J. Mullineam: U,,;w'r.l';'Y /(('''/111"/.;,''
or
"('.ri"K''''', /("/11111'/.;," 4. Sclp.in 'lOti While Ptl i~ a !!()od cx'lmplc of thcory-hclsctl "story-telling." BUIlhe story at ion of the information-related characteristics of money and monctary institutions.
a~ilin Ic;wc:oO
Oll( (.'on,hll'r-
317
Free Banking II COMMUNICATIONS
223
References I. Buchallan, Jalll<:., "A Conlraclual Paradigm t"r Applying ECOII
}"""",I
[21] GARY GORTON DONALD J. MULLINEAUX
The Joint Production of Confidence: Endogenous Regulation and Nineteenth Century Commercial-Bank Clearinghouses TilE FEASIDILITY OF (,I{(VATE-MARKET ARRANGEMENTS for the production of money has resurfaced as an important research question (see King 1983 for a review essay). In an early and influential contribution to this literature, Benjamin Klein (1974) emphasized the critical role of consumer confidence in laissez-faire monetary arrangements, and he analyzed "brand names" as potential devices for insuring confidence in private monies.' He noted that if monies could not be differentiated, each producer would have incentive to overissue and would do so, unless constrained by some mechanism involving monitoring and control of individual bank behavior. In this regard, Klein notes (p. 441) that "many banks became members of private protective and certifying agencies, which performed some functions similar to present-day central banks." Commercial-bank clearinghouses (CRCHs), for example, utilized regulatorylike tools such as reserve requirements, deposit-rate ceilings, and bank examinations to influence and control the behavior of member institutions. 2
The authors thank the New York Clearinghouse Association for access to their archives, an,l (;crtrude Bcck of Ihe NYCIIA for assistance wilh the an:hives. They also thallk Midlael Hordo and Illembers of the slaffs of the Federal Banks of I'hiladdphia and Cleveland for comments on UII earlicr draft. 'Vauhc1 (1977) claims thlll guarunlees, rather than brand name backing, arc more likely to he provided ill a compctitive money-production environment. '(;orton (19115b) .lIld Timberlake (19114) havc called still more explicit attention tothe strong similaritics bctwcen the activities of ninetcenth l'cntury CBClls and today's Fedcral Reserve System. Neither of thc,c authors explored in depth the reasons why clearinghouses took on rcgulatory-like activities, however. GAKY (JotU'ON
is assiSlalll
pr(~lessur
(ifilllallce. 11,e Whartoll School. Ulliversily (!l
Pelllll~I'/I'lI/lia.
' 'It
I )ONAI.t>.J. II.I.lNEALJX is Ihe I'I!Hil.l' c!ll\ellll/d.:y. JUI/I'I/(/1
0/ MO/ley,
Cupyright
©
DII POIli prcdt's.l'Or o/hankillg allti/illClIIcial serl'ic'es,
Crt'elil, allel Ballking, Vol. 19, No.4 (November 19!17) 19117 by the Ohio State University Press
lIlIi-
Free Banking II 458
:
MONEY, CREDIT, AND RANKING
Based on Klein's analysis, it is somewhat unclear: (I) what motivated commercial hanks to voluntarily participate in such arrangements or, (2) why CBClls were involved in the production of monetary confidence. In this paper, weargue that the evolution of the CBCH renects an endogenous "regulatory" response to the problems associated with the asymmetric distribution of information in the banking industry. The nature of these information prohlems was related to the product mix in the banking sector~in particular, to the proportion of demand deposits relative to hank notes. The capacity of "the market" to monitor and control the hehavior of hank managers was increasingly erodcd as demand deposits came to supplant hank notes during the nincteenth century. The set of actions of the CBell represent the suhstitution of hierarchy ("private regulation ") for a market-hased mechanism of control. That "organizations" may dOIllinate markets as allocation and control devices is hardly a new idea (Coase 1937, Williamson 1975, and Stiglit7. 19R5). In section I, we disclIss the importance of the hanking product mix during the nineteenth century from the viewpoint of information costs. Section 2 descrihes the role of the CBCII as a monitor/supervisor which provides valuahle "screening" services to hoth memher hanks and the pUblic. Section 3 examines the behavior of the CBell during financial panics. In response to the unusual information costs associated with a panic, the CBell increased the amount of private rcgulation. The eBeli then reverted to its simpler organi7.ational form following the conclusion of a panic. Private regulation declined and the role of "the market" as a control mechanism increased. Section 4 concludes.
I. IlANK NOTES. BANK DEPOSITS, AND INFORMATION COSTS
Bank notes involved a contract between the hearer and the hank to redeem the face value of the note in specie at the hank. The specie value of a h;,nk note to a seller accepting it in exchange was simply the expected value of a hank's specie promise less t he costs of co lIecti ng specie at tha t bank. Even if t he ex pected specie value of a note was par, the collection costs drove a wedge hetween the par value of a note and its value in exchange for goods. This wedge created an incentive for note-broker businesses to form offering to exchange hank notes for gold or the notes of other hanks at discounted rates. Brokers could profit hy collecting specie at par at the issuing hank. Such firms indeed did form, and a secondary market in hank notes emerged. The size of the discounts quoted on notes presumahly varied with the geographic distance to the issuing hank, the perceived riskiness of that institution and the quantity of counterfeit notes of that institution believed to he in circulation relative to the total issuc (Gorton 19RA). In "hank-note rcporters," hrokers puhlished information on counterfeits along with current quotes on various notes. Secondary market makers also had strong incentives to monitor the quality of the assets hacking hank notes since they collected specie in hulk as the source of their profitahility. Their price quotations in turn revealed their information to
319
320
Free Banking II GARY GORTON AND DONALD J. MUI.LlNEAUX
: 459
buyers and sellers of bank notes. Indeed, merchants commonly consulted banknote reporters in reaching judgments ahout the exchange values of particular bank notes. Competition among note hrokers and publishers of note reporters presumahly enhanced the information quality of these price signals (Dillistin 11)49, White IX95). To the extent that brokers returned notes to the ballk of issue, they also performed a clearing and collection function. Thus, while hank notes typically exchanged for goods and services at a discount, the overall variahility in these discounts was constrained hy the self-correcting responses of hanks, note brokers, and consumers to the recurring signals provided by the secondary market in hank notes. A demand deposit, unlike a hank note, is hoth a claim on a hank and on an agent's account at that hank. This complicates the information required to price a check claim on that deposit. In an exchange mediated hy check, the seller of goods must consider (I) whether the check writer has sufficient funds for the check to be collected; (2) whether the check writer's bank can exchange for specie; and perhaps (3) whether his own bank can exchange for specie at par. While the identity of a buyer "doesn't matter" with use of a bank note (in the absence of counterfeits), a check-based transaction is agent-specifi.: with respect to risk. The contractual characteristics of demand deposits accordingly increased the transactions costs associated with this product. These costs in turn precluded the development of a secondary market in claims on such deposits. Such a market would require pricing agent-specific claims on a bank. It would prove extremely costly for specialist note brokers to acquire information on the reliability of individuals as well as banks. Yet such information is necessary to price such a claim since the agent issuing a check can overdraw his balances. Banks were hetter able than note brokers to handle the information-related disadvantages of checks. Banks could delay specie payment on checks, for instance, until after checks were collected. This required an accounting system, hut such a system was a necessary adjunct to producing demand deposits. Also, hanks could assume that some proportion of the checks collected would be held as deposits rather than paid in specie. These deposits could fund income-prodUCing assets. Brokers could not offer deposit-type accounts, at least not without the risk of being considered a bank, and therefore having to suhmit to chartering n:quirements and perhaps other regulations. The contractual differences between bank deposits and notes effectively precluded brokers from competing with hanks in the collection of deposits. Accordingly, no "secondary market" in check claims was formed. As a result, the information production of the note brokers concerning the "quality" of individual banks ht:came increasingly less available as the volume of dt:posits increased rclative to notes. Ilolders of hank liahilities thercl'ore could monitor hank heha viol' only ill a direct and costly fashion. Banks in the cities had a larger proportion of their liabilities as deposits than as bank notes as early as the late eightt:ellth century. The Balik of New York rep()rtt:d in 17lJ I that it had 50 percent more deposits than notes outstanding. Data ht:G1I11e n:gularly available in the 1~30s and show a fairly steady decline in the
Free Banking 1J 460
:
MONEY, CREDIT, AND BANKING
notes/ deposits ratio. I n New York state, for example, the notes/ deposit in ratio was 1.2 in 1837,0.74 in 1847, and 0.31 in 1857 (Redlich 1951). Nationally. the trend was less pronounced. The ratio fell from 0.85 in 1835, to 0.79 in Ili45, and to 0.67 in 1860 (see Historical Statistics of the United States, p. 995). Given their informational disadvantages, it may seem curious that deposits came to dominate bank notes rather early in the century, even before the establishment of the first eBCH. Hut demand deposits do possess certain well-known advantages over hank notes. They are less subject to theft, for example. In addition, writing checks avoids the cost of making change and provides proof of pa Ylllcnt. Anot hcr Icss commonl y recogni/.ed fea tu re of usi ng chccks ra t her tha n notes to make payments is that checks exchanged against currency or goods and services in local markcts at a fixed price. While the specie pricc of a particular hank's notes could vary dramatically over time and space, deposits, when acceptable to sellers in transactions, exchanged at par in local transactions. But if deposits were to prove viable in exchange, some mechanism for providing confidence in performance by banks was necessary. This was especially the case since a uniform exchange rate for deposits created incentives for banks to "cheat" hy hack ing deposits with inferior assets. There was no secondary market to "reveal" such hehavior as there was with bank notes. The formation of the CBCH not only reduced the costs of clearing checks. it solved the information prohlem created by the missing market. by il1tl'malizillf!, the secondary market in a unique organizational form. With the CBell. the apparent defects of the demand deposit product could be turned into distinct advantages.
2. "I III' CLEARINGIIOIJSE AS A MONITOR/MANA(,!'R
The CBCH was not initially formed to deal with resource allocation prohlems which markets handle poorly. Its function was to economize on the costs of check clearing. Prior to the New York CHCH formation in 1853, commercial hanks collected checks and othc, instruments by a daily exchange and settlement with each other bank. Once the clearinghouse formed, the exchange was made with only one party the clearinghouse itself. Gihhons (1859) estimates that for Ncw York City hanks the cost of "conducting this vast amount of husincss did not exceed eight thousand dollars a year,"which constituted roughly 0.02 percent of deposits in the New York CBCII at the end of 1854. While the clearinghouse was organized to produce a simple product, checkclearing, it was also capable of produeing a hy-product-information. When demand deposits dominate hank notes, banks have an exploitahle information advantage over their customers concerning the quality of hank liahilities. Banks face incentives to hack deposits with high-yielding, risky assets. Customers want to ohtain information about the true quality of hank deposits, hut face free rider prohlems. The direct statement of the hank lacks credihility since a "had" bank has no incentive to reveal its true condition. Customers would clearly gain if
321
322
Free Banking II GARY GORTON AND DONALD J. MlJl.l.INEAUX
: 461
some form of credible supervisor monitored the quality of bank liabilities and disseminated relevant information. Such a supervisor would need enforcement powers to correct contract deviations. The supervisor, in olher words, would act as a substitute for the price system; hierarchy (authority) would replace the Ill" rket. Such a system would be implemented if it were in the welfare interests of the banks as well as their customers. The gain to an individual bank from industry supervision is identical to that for employees in a firm: colleagues can shirk only at a higher cost. Even though workers see compulsion as costly, they are better off in a number of circumstances by accepting it (Stigiitz 1975). This becomes more true as shirking by colleagues reduces the return to an individual worker or increases his risk. When deposits dominate, banking is characterized by just such a condition, since shirking by one bank can lower the return to another directly. A "bad" bank's failure or suspension, for example, would induce bank customers to monitor the quality of their own bank's liabilities. 3 The cheapest way to monitor was to exercise the deposit contract. But if large numbers of customers chose to monitor at once (a bank run), even a "good" bank ran a substantial risk of failure. This externality problem strengthened the demand for supervision, other things eq ua!. The "best" banks would favor monitoring even aside from externalities since disclosure of their status may allow them to capture "ability rents." There are strong reasons in favor of quality measurement by the banks themselves. Bank measurement need occur only once per measurement period, for example, but customer measurement involves a great deal of duplication. In addition, bankers possess comparative advantages in jUdging the quality of the assets backing deposits. The CBCH was well positioned to provide monitoring and supervision services to the banking industry. The form of the New York clearinghouse, embodied in its U~54 constitution, included a number of aspects similar to institutions commonly identified today as providing screening services, mainly educational institutions. The clearinghouse required, for example, that member institutions satisfy an admissions test (based on certification of adequate capital), pay an admissions fee, and submit to periodic exams (audits) by the clearinghouse. Members who failed to satisfy CBCB regulations were subject to disciplinary actions (fines) and, for extreme violations, could be expelled. Expulsion from the clearinghouse was a clear negative signal concerning the quality of bank's liabilities. It suggested that in the collective judgment of the banking colllmunity, the probability of nonperformance in the exchange process by the expelled bank was uncomfortably high. The ability of the C13Cllto audit a Illember's books (to measure quality) at any moment provided strong incentives for prudellt behavior by each bank and thus strengthened the credibility of the CBell signals. 4 Moreover, without access to the clearinghouse a bank had to 'Suspcllsioll was a lcmporary .kfauh
Oil
the conlracl 10 cxchange bank li .. bililics for spccic.
It j i bbolls "riles: ··Wil h knowledge of lhcsc facls (debils ill excess of spccie b" la nccs for a suslaincd
period). the CUllllllillcc visils lhe bank, ano invcsligales ils affairs. If lhey arc found III be Iwpdc>sly imol\'cd, il i, suspclldcd frolllihe cxdwilge allhe Ckarillg lIousc a las I hili\\, III ilS eredil" (pp. J I ~"20). i>i'llli"al lrulll lhc ckaringhou,c requircd only a lIlajorily Yole.
Free Banking II 4(,2
:
MONEY. CREDIT. AND RANKING
clea r its checks in the more costly manner used prior to the existence of the CBCH. Consequently, expulsion was a potent enforcement threat. The CBCH also increased the value of other information signals. Each hank in New York City was required by law to publish on each Tuesday morning a statement showing the average amount of loans and discounts, specie. deposits. and circulation for the preceding week. Ranks were also required to puhlish quarterly statements of condition. The existence of the CHCH prevented banks from publishing inaccurate statements and from engaging in excessive "window dressing" of balance sheets,5 The advantage of the CBCH organi7.ation were such that within a decade a large numher of new local clearinghouses were formed. Thcsc typically organi/cd a long lines similar to the New York CBCH, but some extelllied their roles beyond that of monitoring to regulating hank behavior. The Buffalo and Sioux City c1earinglwuscs set interest-rate ccilings on deposits which could he paid hy memher banks (Cannon 1910), The New York CBCH did not employ fixed reserve requirements as a supervisor-enforced constraint on members until 1858. when a 20 percent "coin requirement" was estahlished against "net deposits of every kind" (llammond 1957. p. 713). Reserve requirements were also soon thereafter estahlished in Philadelphia, The reserve requirement did 110t apply against circulating notes, The CBCII also monitored the extent to which memhers purchased or horrowed specie from external sources to meet claims. Memher banks were. in effect. under implicit contract to the CRCH to avoid "excessive liability management."h These activities of CRCHs served to enforce the fixed local exchange rate of one-to-one between specie and demand deposits. By credihly supervising member bank activities and hy reducing the costs of clearing checks. eBClls helped demand deposits become the preferred bank product on the liahility side, But one problem remained: how would bank liability holders monitor the monitor?
J, TIIECI.EARINGIIOtJSE DtJRIN(; BANKIN(; PANICS
The behavior of CBCHs was consistent with a hierarchical form of nrganintion focused principally on supervisory kinds of activities, But. while the costs of
~"It was only when thc Clcaring IIousc rt'('ort!.,' were hrought to such perkction as to !!iH' the means of analysis and test beyond dispute. that the positive integrity of thnsc statcmcnts collld he guarantecd to the puhlic"(Gibhons IH59, p, 325), The CBCIl would also investigate rumors ahollt the states of particular mcmher banks, In response to rumors. the CBCB would audit the hank and puhlish the rcsults, Thcre arc many examples of this in thc New York City Cicarin!!hollsc Association. Cil'aringhouse Committee Miml/I',\' (hereafter. Miml/",~), ·"A pllsiti\'e principle. or rule ofrinancial govcrnment. has hccn dct11llnst1'lltcd hy thi.action n!'tlw Clearing 1I0nse Iln the city hanks tlmt is. 'he rcstriction of loans. hy the nCl'essity of rnaintainin!! a certain avcrngc of cllin.fi'tml re,HWrl'('.f lI'i,lri" ,Ir.. ha"k, Borrowing from duy til day \\,il1no longcr do, It cunnot he concealcd,"(ltalics original. Gihhons IX59. p, 321.)
323
324
Free Banking II GAR Y GORTON AND DONALD J. M LJl.l.INEAUX
:
463
member-bank "cheating" were raised by the CACH, it could not eliminate all incentives to cheat. Indeed, by raising the public's percep!ion of the quality of the "average" bank, the CRCH raised the benefit of cheatlllg along with the cost. There remained some incentive, thcrefore, for bank cllstomers to engllge in their own monitoring of bank behavior. A banking panic may be seen as an instance of customer monitoring. Exercising the deposit contmct's option feature en masse represents a cheap way for bank customers to monitor the ability of their bank to perform, and, in effect, to monitor the monitoring of the CBen. Banking panics were large-scale attempts by bank customers to convert deposits into specie or currency. While the precise causes of banking panics remains a point of dispute, it seems clear that, because of the information asymmetry created by demand deposits, depositors had to rely on aggregate or nonbankspecific information to assess the riskiness of deposits. Increases in business failures or the failure of a single large financial firm could cause depositors to "run" on all banks seeking, in a single act, to withdraw deposits and measure the performance of their individual banks and, implicitly, the performance of the CBCH (Gorton 1984). From a bank's point of view, there are potentially large costs to such measurement by its customers. The customers can only be convinced of the value of demand deposits if the banks can transform them into specie or currency. With hank notcs, the secondary market signaled the value of bank portfolios in an efficient manner. But without a secondary note market, bank claim holders had to rely on non market methods of evaluation. In part because of the high cost of obtaining information on the quality of bank loans, this portion of a bank's assets can be deemed illiquid. If the sale of such illiquid assets is required to meet depositors' demands, then a bank may incur substantial losses. In other words, the excessive measurement by customers which occurs during a panic effectively makes illiquidity the same as insolvency. With cost less, full information, the banking system would never face problems during panics because bank assets could easily be transformed into any other desired securities. But in that case there would never be a panic to start with hecause depositors would never need to monitor. With an information asymmetry, banks would value some mcchanism which allowed for their assets to be transformed into some other security in such a way as to signal to depositors their vallie. The CBell provided such a mechanism by inventing a new security, the c1earinghollse loan certificate. Ill(: first issue of clearinghouse loan certificates occurred during the panic of I X57; they were issued in every subsequent panic through 1914. The process was ~t raiglMorwa nl: a policy committee of the CBC /I first aut horil.ed the issuance of loan certificates. Member banks needing specie orcurrency to satisfy customers' demands could then apply to the clearinghouse loan committee for certificates. Borrowing banks were charged interest rates varying from 6 to 7 percent and were required to present "acceptable collateral" to be "discounted" by the CBCIl. The loan certificates had a fixed maturity of, typically, one to three months. The important feature of the certificates was that memher banks could
Free Banking JI 464
:
MONEY, CREDIT, AND BANKING
usc the loan certificates in the clearing process in place of currency, freeing currency for the payment of depositors' claims. 7 The mechanism of the loan certificate produced a more hierarchical organizational form of the CBCH during panics than existed otherwise. Indeed, during panics when the loan certificate process was operating, the CBCH behaved much like an integrated firm allocating resources by hiearchical decision. I n fact, the loan certificates were claims on the clearinghouse, ajoint liability of the member banks. If a member bank with outstanding loan certificates failed, the loss (in excess of the value of pledged collateral) was shared by the remaining memhers of the cncH.R The loan certificate process in effect internali7.ed the missing market within a hil"lllrchiclII form. While depositors fllced lin informlltionllsymllletry.the hllnks themselves were in a position to cope with this problem. The clearing process itself provided information on members, as did clearinghouse audits and member bank reports. Also, banks had the specialized knowledge to value bank assets. Most importantly. individual banks had an incentive to lower the probability of other members' failures because of the information externalities. This meant in practice that no member banks were allowed to fail during a period of panic. Instead. members were expelled from clearinghouse memhership for failure to repay loan certificates after the panic had clearly ended and their failure would result in weaker externality effects. The loan certificate process was available to all members. and consequently. is accurately described as a coinsurance arrangement. But this meant that resources had to be allocated to members, even those which the CBCIf perh
Rel'o" of the U.S. Treasury. 1914. p. 589. In the pre-Civil War. "bills receivahle. stocks. honds. and other securities" were acceptahle. Also see Sprague (1910), PI'. 4.12-.1.1. 'In New York the first explicit record of how loan certificates were to function, /l(i/llIIl'.I'. No· vcmher 21. 18(,0. docs not mention this. It was made clear cluring the Panic of 1907 (Mi/llIlI'S, (ktoher J I. 19(7) which was apparently the only occasion When, after the panic. memhers (two hanks) could not repay loan certificates. 1I0wever. during the first panic the eHClls faced after (,ormation. a particularly lucid statement of this was adopted hy the Boston C" (Octoher 15. IR~7) .. , he agreement is quoted in Redlich (1951), p. 159. "In IInston the original IR57 agreement included the fnllowing: And it is further agreed . . . that the Ckaring \louse Committce may at any moment call upon any hank for satisfactory collateral sccurity, for any halnnce thus paid in hills in· stead of Specie: and each Hank herchy agrees with the Clearing 1I0use Committee. and with nil and each of the other Banks to furnish immediately such security wheo (kmanded. Quoted in Redlich (1951), p. 159.
325
326
Free Banking II GARY GORTON AND DONAl.D J. MULLINEAUX
:
465
option feature of deposit contracts. Suspension amounted to default on the deposit contract, and was a violation of banking law. Nevertheless, suspension occlilTed on eight occasions during the nineteenth century. III In banking panics aftcr 185.1, the ('Bell played the central role in deciding whether and when suspcnsion was appropriate. 11 Suspension signals that the CBell believes further liquidation of bank assets to acquire currency or specie is not in the welfare interests of either the suspending banks or their customers (Gorton 1985a). The transformation of the CBCH into a single firm-like organization during panics was signalled by suspending the weekly publication of individual bank statements, and instead, publishing the weekly statement of the clearinghouse itselL I2 1n this way, the clearinghouse avoided identifying weak banks. But, more importantly, with the loan certificate process at work, the aggregate information was the appropriate information. Also, the CBCH did not publish the identity of banks borrowing through the loan-certificate process. Cannon (1910, p. 90) reports that "attempts on the part of the business community were made in vain to discover what banks had taken out in certificates." For this organizational structure to be successful, the amount of currency released from use in the clearing process through use of loan certificates had to be large enough to signal to depositors that the one-to-one deposit exchange rate was, in fact, correct. But the amount of currency released was limited, and so, during the panics of 1893 and 1907, the clearinghouses directly monetized bank portfolios by issuing loan certificates, in small denominations (as low as 25~), directly to the pUblic. This allowed all the banks' assets to be monetized, if needed.13 Depositors were willing to accept loan certificates in exchange for demand deposits (rather than currency) because the loan certificates, being claims on the CBCH, insured depositors against individual bank failure. In this way, the problem of bank-specific risk arising frol11 the information asymmetry was solved, leaving only the risk that the CBCH would fail. But the circulating loan certificates were neither bank- nor agent-specific, so a secondary market could and did quickly develop, allowing the risk of CHCII failure to be priced. This secondary (footnote 9 continued): In New York the ell Committee had the "power to demand additional ,ecurity either by an exchange or an increased amount at their discretion." (Milllll~S, November 21, I K60). But beyond this was power to direclly allocate resources by making re4uisitions on individual banks (Mifllllt'S, October 21, 19(7). Also, see MillUles, October 18, 1907; October 21-22, 1907; January 9, 2K, 1907; February I, 190!l. '''Suspension of wnvertibility occurred during Augllst 1814, Fall 1819, May I K37, October I!l57, Sqllenlber I K7J, .July I K93, and October 1907. Suspension also OCCUlTed in the I K6()s though this was not related to a Illajor banking panic as in the other cases. Loan certilicales were issued during every pilnic ai'ter the fornwlion oftheCBCII, including 1860and 1884. [)uringthecrisesof 1895and IK% Ihe New York City CBeli authorized the issuance of loan cerlificates, but no member banks applied (l.(IUfI ('ommillet' Atillltles, December 24- 31, 11195; August 24, 11196). II For eXilmple, the Marine National Bank was punished for acting on its own by unilalerally 'lJ,pending in t-.lay, I K84 (Alitlllles, May 6, I K84). The New York CHef! avoided sllspension during tile Panic of 1884. "E.g., [Olill C"mmill~e Mimtles, January 30, IK91; June 6, IK93; November I, 1907; and Atill111<'1, Novelilber I, 1907. "(jolt
Free Banking II 466
:
MONEY, CREDIT, AND BANKIN(;
market served as an index of confidence. Initially, a currency premium existed in exchanges of certificates for cUHency.14 Over the period of suspension, it gradually subsided until reaching zero, whereupon suspension was lifted. In this way, a market signal was sent from depositors to CBClls. During banking panics, the CBCH was operating a miniature capital market, allocating resources by nonmarket means for the benefit of the collective of firms. But once the period of suspension was over, the CBCH reverted to its more limited organi7.ational form. Only by reverting back to the more limited organilatiol1al form could the CBCII restore the rroper incentives for hanks tojointly monitor each other on a continuous basis. Surrose that once the more hierarchical form of organilation had been adopted during a panic, the CBCII did not revert hack to its more limited form. Then individual banks, knowing that the loan certificates were available, would have an incentive to make riskier loans since each would believe that the risk could be spread over the other members through the loan certificate process. Clearly, this would not be viable. During the period of suspension when the risk pooling arrangement was in effect, however, banks have incentives to make more risky loans, free-riding on the CBCH. No mention of such a problem appears in the archives of the New York Clearinghouse Association or other sources. The prohlem apparently didn't exist because memher hanks had no funds to make new loans. During panics hanks attempted to liquidate loans of existing customers to generate cash. If a memhcr did engage in making riskier loans, however, it was exposed to the risk that the maturity of the loans would he longer than the suspension period, making free-riding less likely. Also, the CBell required daily reporting of all balance-sheet changes during a panic period. Only by reverting back to the more limited organizational form did individual hanks have the incentives to monitor each other. The externalities from individual bank cheating provided the incentives, and the resulting monitoring made it possible for the panic-form of the CBCH to be effective since the risk exposure of the members had heen limited during nonpanic times. Consequently, the changing organi7ational form and degree of regulation of the CBell was an integral part of the production of demand deposit services. In the ahsence of rI market to monitor product quality, hank firms were required to jointly produce "confidence" in deposits, hut this required a delicate halance between hierarchy rind maintenance of market incentives.
4. CONCI.lJSION
Analysis of the CBCIl system focuses attention on the issue most critical to the discussion of competitive hanking: the ahility of "the market" to control the hehavior of hank managers. Hayek (197(,) and White (f9R4) have argued thrlt market forces are capahle of controlling hanks, and consequently rreserving "Sec Sprague (1910). pp. 57, IR7, 2RO-RI.
327
328
Free Banking II GARY GORTON AND DONALD J. MULLINEAUX
: 467
confidence in the system, provided that bank liabilities are convertible into some outside money. K kin (1974) has emphasized the role of brand names in establishing and maintaining confidence concerning convertibility. We have argued, however, that, hecause of information asymmetries, the market's capacity to control hank behavior depends on the banking product mix. In particular, the rising ratio of deposits to bank notes during the nineteenth century resulted in (I) increased monitoring costs for bank customers, and (2) more significant externality problems among banks. The CHCH, originally formed as a simple collective to reduce the costs of collecting checks, became involved in monitoring activities and established mechanisms of managerial control. In fact, the CHCH "regulated" bank behavior. Our analysis provides a more complete and consistent explanation for the role of private institutions such as the CBCH in the creation of monetary confidence, which has been noted by Klein (1974), Timberlake (19g4), and Gorton (1985).15 It also suggests that the conclusions of Hayek (1976) and White (1984) concerning the efficacy of markets as control mechanisms in banking may be valid only under certain conditions concerning information costs and monitoring technologies.
I.ITERATURE CITED
Cannon, Jamcs G. Clearing I/ollses. (U.S. National Monetary Commission) Washington: Governmcnt Printing Office, 1910. Coasc, Ronald H. "Thc Naturc of the Firm." EcotlollJica 4 (November 1937),386-405. Dillistin, William. Blink NOlI' Reporters ami COllll1e~leil Delec/O/'s. 1826-1866. New York: American Numismatic Society, 1949. (iibbons, James S. The Ballks of New rork. Their Dealers. Ihe Clearillghollses. alld Ihe I'llllie 0/1857. New York, 1859. liorton, Gary. "Banking Panics and Business Cycles." Mimeographed. The Wharton School, University of Pennsylvania, 1984. _ _ _ . "Bank Suspension of Convertibility." JOllfllal (!l Alollela/,,,, Economics 15 (March 1985), 177-94, (a). -----. "Clearinghouses and the Origins of Central Banking in the U.S." JOllrnal (!l i:'(,(Jlwmic l/is/(J/'Y 42 (Julie 1985),277-84, (b). _____ . "Inside Money and Contracting Technologies: An Empirical Study." Mimeographed. The Wharton School, University of Pennsylvania, 1986. Ilammond, Bray. Bunks and Polilics ill America. Princeton: Princeton University Press, 1957. Hayek, t'rit:drich A. Von 111e J)t!f/UliOllalisalion o/Motley. London: Institute of MonctMY Affairs, 1976. King, Robert. "On the Economics of Private Money." Journal of A/olle/ury El'Otlomics 12 (.Iuly 1983), 127-5S.
"M ullineaux ( IlJH7) analyzes the role of a different priv
Free Banking II 46R
:
MONEY. CREDIT. AND BANKING
Klein. Benjamin. "The Competitive Supply of Money." journal (Jf MOl/ey. Credil. an'/ /Jankillg 6 (Novemher 1974),423-53. Mullineaux, Donald .I. "Competitive Monies and the Suffolk Bank System." SOlilhenl F;collomic journal 5] (April 19117),11114-911. Redlich, Frit7.. '1he Molding o/,AlllericlJlllJankillg. New York: IIafner Puhlishing COI11pany. 1951. Sprague. Oliver M. W. Ifistory (!/'Crises ullder Ihe Natiollal/JankillK Srstelll. (U.S. National Monetary Commission). Washington: Government Printing Office, 1910. Stiglitz, Joseph E. "Incentives, Risk and Information: Notes toward a Theory of Hierarchy." lJell Journal (J/,I:c'ollomics 6 (Autumn 1975), 552-79. _ _ _ . "Credit Markets and the Control of Capital." journal (!/' Money. Crcdil. alld /JallkillK 17 (May 19115). IJJ-52. Timherlake, Richard 11.,.1 ... "TheCcntralllanking RoleorCicaringi1ouse Associatiolls." .Iolmwlo/'/IJ(J//(·Y. Credil. all'/ IJallkillg 16 (Fehruary 19114), 1-15. Vallhcl. Roland. "Frec Currency Competition." IVl'ltl1'irlsc/wflil'hes !lr,,"i\' I IJ (ScrtCl11her 1(77),4.15- 59. White, Lawrence II. Free BallkillK ill Brilaill; '1hf'urr L'<:/Jcricncf' and DeIlIJIf', 18()()-45. Camhridge: Camhridge University Press, 19114. Willi
329
[22] Competitive Currencies, Legal Restrictions, and the Origins of the Fed: Some Evidence from the Panic of 1907* STEVEN HORWITZ St. Lawrence Ulliversity Calltoll, New York
I, Introduction In the last several years, economists have produced numerous studies examining both the theoretical opemtions and historical manifestations of unregulated banking systems. Recent examples of historical investigations are the studies by L. White 1471, who explores the Scottish experience, Selgin 1331 on China, and Timberlake 1431, Gorton 1121, and Mullineaux 1201, all of whom examine the role of clearinghouses prior to the creation of the Federal Reserve System in the United States. Timberlake and Gorton emphasize the role that currency substitutes played during banking panics of that era. The present paper seeks to explore the issuing of currency substitutes during the fall of 1907 in greater detail. In particular, it seeks to explain precisely why currency substitutes were needed and how they became accepted and to examine the implications of the panic for both traditional and present arguments over the historical justifications for the Fed.
II. I.egal Restrictions and the Inelasticity of National Bank Notes The legal restrictions allecting pre-Federal Reserve banking provide the key to understanding banking panics and the use of currency substitutes in that era. The National Currency Act as passed in 1~63, and subsequently amended, allowed individual national banks to issue notes under cellain provisions. Banks were required to purchase certain United States government bonds in the open market and deposit them at the Treasury in exchange for notes in amounts equal to 90 percent of the lower of the par or market value of the dcposited bonds. (After 1900 thc limit was changed to 100 percent.) Ostensibly, bond collateral was lIsed to redeem notes of railed banks, but it also meant an enhanced market Illr li:deral debt, an important concern given the level or debt in the 1~6()s due to the Civil War 146,111. Under this system. the market price or approvcd bonds was the out of pocket cost of note *Thc authur wuuld like IU CXICIUJ mure than Iht.! usuallhank!t tu Geurge Sl.!!lgin fur his cununcnlS and I.!llcuuragclllcnl on severat drah. "I' Ihis paper. Thanks als" lu Don Lavuie, Lawrcll':c Whil~, Rkhard Tilllberiake, Kevin Grier, Ihe Workshop in Auslrian Empirical Sludics al Geurge Mason Univcrsily. culh.Xluiul1l panicipanl!) at (korgc Masun and New York University, and several anonymuus rdcrr.!cs lor their COllulll.!nls un I.!arlicr dralh. The usual (aveal applies in Ihe slrongcM ICrllI) po))iblc.
639
Free Banking II 640
Stel'('/1 Ho/'witz
issue for banks. while the difference in interest between bonds and foregone alternative assets was a measure of opportunity cost. Consequently. the amount of notes in circulation tended to vary with bond prices rather than with the demand for currency. Higher bond prices or higber rates of return on alternative assets meant that note issue was more costly. Since there was no necessary connection between low bond prices and high currency demand. this system often led to currency shortllges where banks were unable to supply the required amount of currency to depositors and would-be borrowers.' Other stipulations also prevented the supply of currency from responding elastically to clllInges in dellland. There was a significant time lapse before new notes could be issued by mllional banks. since banks often had to wait for new bonds to be issued in order to take but more notes. In addition. redemption and contraction of currency did not occur when tbe banks desired it. Finally. a prohibitive ten percent tax on state bank note issue made it impossible for adjustment from state issues to offset shortages of national bank currency. During panics these problems were exacerbated. The two main difliculties faced by national banks were: I) purchasing acceptable bonds at prices that made note issue profitable. and 2) avoiding delays before notes actually went into circulation. Concerning the first problem, a prominent national bank president said, "The real difliculty ... has been to get acceptable securities ... one 1New Yorkl bank had to borrow from banks in California. This shows the effort the banks are making to take out circulation" 124,21. A. Barton Hepburn of Chase National Bank noted that "it is extremely diflicult to secure the loan of government bonds which are necessary ... to secure an increase in the circulation of any national bank" 124,21. When bonds could be purchased, the expense involved was often enough to make further note issue unprofitable. Frank R. Vanderlip of the National City Bank reported that "in order to get United States 2 per cent bonds we have forced the price of them to the highest point on record" 124.21. A. Barton Hepburn noted similar difliculties: "To purchase ... 2 per cent bonds. Ibanksl will have to pay $108 for them. By having to pay this price for these bonds to bring out circulation the financial situation becomes aggravated rather than relieved" [24,2). Hepburn recognized that further note issue was badly needed and that high bond prices prevented banks from buying the necessary bonds. The requirement that notes be backed by specific government bonds funnelled the banks' buying power into a single bond market and raised prices there to prohibitive levels. If banks had instead been allowed to issued notes based, like their deposit-credits, on their general assets, marginal costs of note-issue would not have risen any faster than marginal costs of depositcreation, lind banks would have been indiffcrent at the margin as to which form their liabilities took 1341. The result of the inability of the banks to create enough currency was that they were often forced to give depositors reserve media instead of inside money. This reserve drain caused country banks to replentish their reserves from their deposits at city banks. who did the same to the New York blinks. shrinking bank balance sheets and raising interest rates throughout the system. I. If anYlhing. Ihe bund prices·currency demand problem would be e,acerbaled under Ihe bund collaleral syslem. because higher bond prices lendcd In accumpany increased demands fur cash. Ihus making nole issue more expensive precisely when il was needed Ibe mos1. The treatment of the cost of nole issue is udmiHcdly brief. It is not my inlcnlion 10 reformulate the exact cost of note
issue. f1owever. il seems Ihallhe Friedman and Schwartz/Cagan calculalions are mislaken in concluding Ihallhe shurtage of n()IC~ under Ihe Nutinnal Banking System is inconsistent with profit-muximi7.ing behavior. The additional bunlens crcaled by Ihe bond collaleral requiremcnls IIpp,,,cnlly mellnl somc addil;on"1 cosl 10 Ihe banks ,h., has been overlookcd in previous slud;es. More dela;led Ircalmenls of Ihese is
331
Free Banking II
332
COMPETITIVE CURRENCIES AND TIlE ORIGINS OF TIlE fED
641
Once bond collateral was acquired, actual shipment of currency from the Office of the Comptroller could still involve a lengthy delay. During panic situations the Treasury was especially hard-pressed with numerous panic-associated responsibilities. According to a banker qlloted in the New Yurk Times, currency shipments during the 1907 panic were delayed "owing to the great pressure of work being put upon ... the Treasury staW' 124,21. Therefore currency shipments took longest when they were most desperately needed. According to Hepburn, during panics, banks might have to wait thirty days or more after depositing bonds before actually getting hold of new notes. So even if the banks could find acceptable bonds at a profitable price, the crfect of their efforts to increase circulation might not be felt for a month, mainly because of the need to rely on centralized governmenWI approval. This, in turn, meant excessively long printing and transportation delays. These delays would not have occurred in the absence of legtll restrictions, as the banks's choice of assets would not have to have been reported to, deposited at, or upproved by the Treasury.'
III. The Panic of 1907 and the Use of Currency Substitutes The panic of 1907, like previous crises, involved both general problems of currency supply and special stringencies associuted with the fall harvest. The country bunks needed significantly more currency during harvest season as the transactions demand rose sharply. Due to the inelasticity of note issue, notes (or reserves, including Greenbacks, if sufficient national bank notes were not available) often had to come from city banks, causing shortages in the city 121. These seasonal difficulties worsened in 1907 as a result of the questionable banking practices of some individual members of the New York financial community. The failed attempt of F. A. Heinze to corner the copper market using depositor funds from his Mercantile National Bank, led to the bank's failure to meet its clearinghouse obligations in mid-October. One of Heinze's directors at Mercantile, Charles F. Morse, was involved in other financial dealings, and when his connection with Mercantile became known, his banks were hit by runs. Morse's connection to the Knickerbocker Trust Company led to a run on it on October 24, which, in turn, led to runs on other banks in succeeding days. Over the next two days various banks and other financial institutions, including the Treasury, moved to help the stricken banks. Normally banks cleared by using large denomination certificates representing actual holdings of high-powered money. On Saturday the 26th, the New York Clearinghouse Association issued clearinghouse loan certificates, which allowed the banks to clear among themselves without relying on their usual reserve assets. Instead, the loan certificates represented "loans made to member banks by clearinghouse policy committees" 143,31. The use of certificates freed high-powered money for depositor use and eased the reserve shortilge facing the banks. By November 6th, the failed institutions were in new hands and were settling their obligations to former note holders and depositors. The bank runs ended and stock prices recovered. There was a premium on currency, with prices as high as $104 for $100 face value currency \38,281 I. By early November, this currency shortage had become the only, but significant, reminder of the
2. In an interesting anicle un many of these saUle issul.!s, Miron requirements were a major (;allSC uf problems. ttc
IIKI
docs not seem 10 think the bond collatcml
explains the t:fcalion of the Feu th,,' oC<':UI'CU Juring pallies. Fur a JiM:ussiOIl of tliis view, sec section six below.
as a rcsponsl! to inh..:rcsl nUl! Ihll.:luations
Free Banking II 642
SI£'\'£'II
!forll'ilz
earlier bank runs. Because or the prohibitive cost of note issue, banks were unable to directly prevent such a shortage by increasing the supply of notes. Instead, the banking system and the non-bank public devised ingenious and somctimes amusing methods for evading the law and meeting the demand for currency. These methods included the use of currency substitutes.) The most common currency substitutes, comprising over 80 percent of the total, were the various clearinghouse currencies 13,SISI. Along with clearinghouse settlements between the banks, c1earinghouscs also issued such loan certificates in small denominations for use directly by the public. These were marked with "payable through the clearinghouse" and were backed by assets of member banks deposited at the clearinghouse. The member banks agreed to accept them on deposit, though the clearinghouse itself held only a percentage of cash reserves for the deposited securities.' Clearinghouse certificates, for interbank use, had been issued in previous panics, but had not circulated among the public because they were of large denominations.~ Though these certificates were important, and are the focus for Timberlake 1431, other currency substitutes more closely resembled currency that might have been issued in tbe absence of bond collateral requirements. Also, by implication, these other substitutes were more obviously a violation of the letter of the National Banking Act, as well as laws prohibiting issues of private money. Another kind of currency substitute used as hand-to-hand money by the public was the negotiable cashier's check. These were written in small, convenient denominations of $S, $10 or $20 and had no special reserves backing them. Since they were checks, they had to be payable to some person or entity. According to Andrew 13,SIOI, the banks made the checks payable to "bearer" or "through the clearinghouse" or "in exchange" or the like. A number of banks were concerned about the legality of circulating checks in this manner, so they made the checks payable to "John Smith, or hearer" or "Richard Roe, or bearer." By naming a specific person, banks made these currency substitutes look less like notes. Had the authorities regmded thcse small denomination cashier's checks as bank notes, they would have held them to he in violation of the law, which required bond collateral for notes (in the case of national banks) or a ten percent tax (in the case of state banks). Still another currency substitute, not issued by banks at all, was the negotiable pay check, also known as scrip" In normal times payment by any form of check was unusual. When firms (in their elfort to cooperate with the banks) attempted to pay in ehecks during the panic, employees were disappointed. The firms then attempted, hy various innovations, to overcome this reluctance. First, like cashier's checks, scrip was issued in small denominations. For example, a worker might get paid $SO in four $10 checks and two $S checks. Unlike cashier's checks though, the scrip was not a direct liability of either a specific hank or the clearinghouse. Though it was drawn through the finn's bank, in that it often had the hank's or clearinghouse's name on them, scrip was clearly a /illhilily of Ihe firm Ihal iss lied i/13,S121. Also, scrip was negotiable and passed hand to hand; .1. This scclion is boscd on Andrew's 1.11 SI",ly. which delves more dceply inlo Ihe nalure. Iypcs and amounls of clIn-cney suhstitllics. 4. Some certificales and checks were fOllnd in cilies Ihal previously hod no clearinghouse 13. 5(7). Thol such clI~ency was accepled waS probahly due to Ihe hanks in lown lisling Ihemselves on the acl,oal noiCS. This indicales Ihal the Si7.C ,lnd organi7.ation of ;1 central clearinghouse were not necessary conditions for the successful use of currency suhstitulcs. These "clearinghouse" certificates also point to the huge role that brand names and reputations phlycd in hanking: during this time period. 5. The role played hy these currency c;uh!'ltiltllcS parallcl the Wfty in which private mintings of coin have hecn used to mcel shortages as for hack as Ihe IRlh century 14; 491. The woys in which Ihese coins became accepled arc quile similar 10 the process noted hclow. 6. Timhcrlakc 1441 gives an ex.tended disclIsc;,ion of the n~turc of scrip and some historical manifestations.
333
334
Free Banking II COMPETITIVE CURRENCIES AND HIE ORIGINS OF TIlE FEU
643
local merchants accepted it knowing that it could be transferred or accepted by banks as a deposit credit 123,21.' Eventually, when the need for the currency substitutes diminished, scrip was turned in for deposit credits. Businesses went to other extremes to combat the currency shortage. In Omaha, Nebraska, a streetcar company was forced to pay its workers wilh 600,000 nickels from its fare boxes 125,11. The St. Louis streelcar company had earlier done its Onwha counterpart one better by paying employees with live cent fare tickels lhat could be used in exchange for checks or goods at local Slores 141,11. The fare tickets circulated fairly widely for several weeks, evi(kntly because they had a redemption value as streetcar rides.
IV. Currency Substitutes and the Acceptance of Competitive Hank Notes Most currency substitutes were illegal, and banks, the government and lhe public knew it, yet no serious attempt was made to stop issues of such money. Nor did creditors take advantage of the law to put pressure on debtors to pay in legal currency, which would give them a premium profit. If bond collateral requirements were so much in the public interest, why did the Treasury not step in and prohibit (or tax if appropriate) the illegal asset currency? More fundamentally, why was the general public willing to accept such currency at the same value as national bank currency if bond deposit currency was less risky? Another interesting question relates to lhe banks: why did they judge the cost of issljing currency substitutes, including (in cases) the prohability of prosecution multiplied by the penalty, to be less than the cost of issuing legal currency,! To answer the last question, reference might he made to the previous evidence of the high cost of legal note issue during panics. An explanation of the public's acceptance of currency substitutes must, on the other hand, rest on consideration of the fundamental nature of money and the competitive actions of the banks. The essence of money is its general acceptability. Historically, stones, shells, tobacco, cigarettes and cattle have all served as money. Indeed, as Mises 119,951 put it, something can become money only "through the practice of those who take part in commercial transactions." The banks's principal competitive method for gaining general acceptability of the currency substitutes was to earn the puhlic's trust.' For the banks, this often meant deploying their brand name capital. In both Pittsburgh and St. Louis, where substitute issue was heavy, pay checks and circulating cashier's checks were endorsed by the clearinghouse. This endorsement was advertised both on the notes themselves and in public statements 128,11. In St. Louis the banks advertised their willingness to provide copies of their official signatures to local merchants in order to prevent fraud 140,21. The Pittsburgh clearinghouse took out an ad announcing that it had hired Pinkerton security to guard against fraud in pay checks 128,11. A St. Louis hank director walked a customer down to a railway ticket agent's office to personally endorse a check 139,61. In Philadelphia, banks routinely certified the worthiness of pay checks at employers's requests 125,11. Such procedures were aimed at promoting the acceptability of the various currency substitutes by showing that the banks stood behind them. Note also that the banks had an interest in promoting checks because acceptance protected their legal cash resources. 7. Timberlake f43.IOI notes the "perceptions
1371 and Frankel 1HI.
335
Free Banking II 644
Slel'(,1I lIol'll'il:
A second method for getting the substitutes accepted was education. Pittsburgh banks took out a large newspaper ad explaining how pay checks worked and how they could be deposited in order to open up a checking account for the depositor. The ad also pointed out the conveniences of using checks as well as the safety measures taken 131,61. A bank teller in 51. Louis explained to a recalcitrant customer how he could pay his employees in cashier's checks: "Give the men checks. They're good at the butcher's and the baker's" 139,61. Another educational device was the explicit printing of instructions and conditions on the back of checks 130,51." In addition to stric.:tly informational properties, the process of a good becoming money involves a geometrically expanding circle of acceptance. The more others arc willing to accept good x as money, the more likely I am to accept it. In Menger's 1171 story of the origin of money, this snowballing process is essential to a medium of exchange becoming generally accepted.'" People want to hold the most generllily accepted of assets in order to make purchases more easily. As Ihey sec Ihe gains llIade by Ihose who hold, and then spend, the llIore marketable assels, people illlilate that behavior and attelllpt 10 llIake silllilar gains. This whole process multiplies unlil one object is universally (or nearly so) accepted as money." This process is revealed in Ihe acceplance of the currency substitutes during the fall of 1907. Store managers, not wanting to lose business due to the greater likelihood of people having to lender wage checks, actually competed briskly wilh each other in their attempts to encourage holders of such checks to usc them at Iheir establishments. A 51. Louis jewelry store was one of the first to advertise that checks would be accepted. When a competitor followed suit, the store outdid them by ofTering a ten percent premium on cashier's checks over other forms of paymentactually reversing the usual currency shortage condition 142,181! In Pittsburgh, a reporter for the POSI noted, "a strange desire among all of the storekeepers to be the first to let their customers know ... the fact that wage checks would be as good as cash" 129,11. In general, merchants were all very willing to accept wage checks, for reasons similar to the banks's. Not accepting them would reduce sales, whereas acceptance gave them a medium that could be redeposited like cash. These reports demonslrale Ihe relationship between the willingness of the merchants to accept and Ihe llttitudes of the public and Ihe banks. The grealer proportion of merchants that would accept substitutes, the more confidently and commonly the banks would issue them in times of demand. The merchants would then read this as evidence of bank confidence, leading to them bcing more likely to continue to accept suhstitutes. Analagous arguments hold for the public's attitude toward both the hanks and the mcrchants. Tbis process as a whole is an example of the snowballing process discussed by Menger. Like money itself, the currency substitutes are
9. These infornwlional aspc<.:ls tic into the litemttlre on Ihe origin and natuTe of rnonclary exchange. Bnmncr and
Mellzer 161. Oslrny alld Slarr 1271 allo Alehian III.
Ii"
exa",ple. all slress Ihe illformalional proper1ies of goods Ihal
hecoll1e IIlcdia of exchange. Brunner and Mch7.cr and Alchian expressly argue that Ihe lower the C(lst of recngni7.ing the qtlalily of a J!otld. Ihe more likely il is 10 hecome
it
mc(liuJn of exchange. The allcmpls by the
b.mks and mcrchanls in
I<}07 to provide this Iype of information ahout qtwlity did lower the cost of using the suhstitutes and facilitated their
a,,'ecpt;!ncc.
10.011 Ihe Mellgerian Iheory of mOlley. sec Jones 1151, Nagal"ni 12tl and O'l)riseoli 1261, as well as par1 of Ihe discll~sion
in Richter 1.121.
II. II shoulo be noled Ihal Ihe Mellger slory refers
10
Ihe origin of outside money. or whal Mises 119. 5261 calls
"money in the ",'TroW sense." The currency substitutes were inside money, or "money in the broad sense," and this dincrence limits Ihe applic:lhility of the Menger story. Whereas Menger was trying 10 explain how any medium of exchange would ever arise, here we .are examining why !,or/;t'll/ar media become more or less accepted. The two evolutionary processes sccm 10 be parallel and Mcnger's theory can provide insighls on the viability of particular Iypes of inside money.
336
Free Banking II COMI'ETITIVE CURRENCIES AND TilE ORIGINS OF TIlE fED
645
converged upon when people are left to themselves to resolve the need for an acceptahle medium of exchange.
v. Dank Runs, Urand Names, and Hank-Specific Information To assess the relevance of the panic of 1907 lor the possibility of competitive note issue, two distinct problems that confronted banks during the panic need to be distinguished. The first was th.: attempt by depositors to turn bank promises into non-promises (what we might call an "outside money run"), due to a loss in conlidence. The second difficulty was another sector's demand to turn deposits into notes (which we can call a "currency run"); i.e., simply switching the form in which they wished to hold bank liabilities. The outside money runs in 1907 were centered around the Heinze-Morse hanks. The difficulty facing the system then (and facing unregulated banks) is how to prewnt localized runs from spreading to the whole system. One way banks tried to prevent contagion in 1907 was through advertising and use of their brand name capital. In addition to advertising the acceptability of the currency substitutes, banks advertised the soundness and trustworthiness of their institutions. Advertising trust and confidence was common practice in an era before federal deposit insurance programs. During the panic, banks immediately resorted to stronger and more innovative ways of advertising, especially through the usc of their brand names. After the panic Ilrst began in New York, there was a marked increase in general advertising by banks. There were also changes in the kind of ads they ran. It became more common for banks to list their directors and owners in their advertising. They also offered such standard information as length of time in business and volume of husiness. In normal times a bank might only rarely advertise its balance sheet. During the panic, on the other hand, various issues of the New York Timel' indicate that advertisements frequently included abbreviated balance sheets. In the issue of October 25, 1907 122,151. at the height of the panic, there was a full page of bank ads (compared to the usual quarter or half page). The ads included short versions of balance sheets and long detailed lists of bank personnel, including specific information on other business cOllnections of the board members and management. The banks, like the public, were sensitive to concerns about interlocking directorates, and any connection with anyone questionahle was had fin business. Such advertisements appear to be attempts to provide the kind of bank-specifk inlimnation discussed by Gorton 112]. Bank-specific inlimnation allowed the public to determine whether problems with a given bank were individual or symptomatic of wide-spread difficulties. Without such inlormation, the failure of a single bank, perhaps appropriately due to faulty management, could become a multi-bank run, as the public misperceives a local problem as a systemic one. A few banks also provided such inlilrmation hy refusing to accept deposits from customers who had drawn the funds off of sound banks. Winmill and Fish Bankers, for exampk, ,Idvenised that it would not accept deposits from "reputable banks and trust companies to their unnecessary embarrassment" 122,151. As Gorton 112,2HOI notes, informational externalities give hanks incentives to prevent competitors from failing. Winlllill ,Ind Fish's policy can he viewed ,:IS a rational response to that incentive and a way of internalizing the costs of confidl!nce deterioration. Such activities can also shed light on Gorton and Mullineaux's 1131 argument about the role of the clearinghouses during the panics. They argue that clearinghouses are a way for banks to internalize the information externalities that result from a lack of hank-specific III1fonn.ltion. By switching to a hierarchical mode of organization (rather than a market one) during panics,
Free Banking II 646
St('\,('1/ !ion";tz
contagion could be prevented. While the evidence prevented here corroborates that point to an extent, individual hanks also had ways of handling these externalities. both by advertising their own linancial situation and hy refusing to accept deposits drawn off of sound hanks. Gorton and Mullineaux argue that individual hanks would hide hehind the clearinghouse in panic situations, hut the evidence indicates that this is only part of the story. Banks also used individual tactics to internalize information externalities." The panic also has additional evidence on the role of clearinghouses during crises. At the outset of the panic, the New York Clearinghouse Association made halance sheets of memher banks that had suspended payments availahle for puhlic scrutiny. The clearinghouse also made puhlic a list of all memher hank directors and their various husiness connections. The Wall Street )o/ll"/la/ made it standard practice to puhlish this clearinghouse information as soon as it was released 1451. This was a way in which the clearinghouse, as a central information source, could provide hank-specific information. All of these clearinghouse activities lend support to Timherlake's 1431 description of the clearinghouses as ellicient, non-governmental central banksY The actions taken to issue the loan certilicates and other currency substitutes to meet the rise in the relative demand for currency were hoth necessary and elTective. In fact, total losses from all issues of clearinghouse currencies were virtually nil 143,131. Despite the illegality of the currency substitutes, they circulated widely and easily, just as theorists of competitive note issue indicate they might." The way in which clearinghouses became fractional reserve institutions that issued the needed currency, and the fact that the currency was accepted by the pUblic, together indicate how competitively issued currency along with a well developed clearinghouse system might hehave under similar circumstances.
VI, Legal Restrictions on Note Issue and the Origins of the Fed Inside money runs (shifts in the currency-deposit ratio) are only a problem when legal restrictions bias the relationship between deposit and note liabilities.'~ The bond collateral requirements meant that changes in depositor liability preferences required changes in the composition of bank assets, which made producing the desired liabilities unprofitable. Currency substitutes were one way 12. II is interesting Ihat Gorton and Mullineaux ID. 466-67J and Mullineaux 1201 argue that Ihe hierarchical role of the clearinghouse somehow C3!'OtS doubt upon arguments that the market can effectively discipline free banks. As Selgin and White (36. 2161 point out, this is true only if one views "the market" as simply responses to price signalS. The markel, viewed more hrmuJly as a nexus of voluntary exchanges, involves all kinds of hierarchies, including Ihe idea of a bank (or firm) itself. The movcmcnt from markcllo hierarchy simply indicalcs Ihat price coordination is too costly, not thai mnrket coordination is. 13. Though one need not agree wilh Gorton 112, 277) that Ihe Fed was .• imply a nation.liled clearinghouse system. Aside from traditional clearinghouse activities, the Fed had a monopoly of note issue and could provide the government with an instrument of inlhttiolllUY finance. 14. One other consideration is the role of branch banking. Interstate branching was basically illegal in the U. S. at the lime. though it was legal in Canada (which aho had asset backed notes) and Canadian banks felt no effects from the panic. Nationwide branching would also have made funds transfer easier, so that any possible local run could be handled smoothly without the drain on reserves caused by the country banks. Additionally. the fact that over 80 percent of the value of the currency substitutes were issued by the New York Clearinghouse. and backcd by the diverse portfolios of its membcrs, provides support for the hcnclits of the kind of diversification that large branched banks can provide. It can be argued that the New York substitutes succeeded partially because of the assets that backed them. One would expect competitive currencies 10 be more viable when the issuing banks have more diversified portfolios, like those that branched banks can have. 15. See Selgio's 1341 extended discussion for more on the relationship belwcen competitive note issue 3ml the relative demand for currency.
337
338
Free Banking II COMPETITIVE CURRENCIES ANI> THE ORIGINS OF TilE I'EI>
647
banks evaded these legal restrictions on note issue and profitably provided the desired liabilities. The currency-deposit problem was not a demand for base money in exchange for notes, but a demand for notes in exchange for deposits 143, II J. The extra time and trouble involved in the issue of currency substitutes (instead of conventional bank notes) were a major cost of the era's restrictive banking laws. The risk of legal penalties was also a consideration; shortages and runs had to get fairly severe before banks became convinced that the authorities would look the other way. Had blmks been able to issue notes backed by their general assets, they could have easily and quickly made the switch from deposits to notes simply by printing new notes. Such new notes could also serve to meet the demands from the banks where confidence was lost, thus preventing a ripple effect on outside money due to those runs. Since both notes and deposits would be backed by the same assets, no new assets need be purchased to meet the switch in liability preference, and the currency runs could have been avoided.'· Actual experience with the issue of currency substitutes gives important testimony to these conclusions about the viability of note issue deregulation. To the extent that concern over the elasticity of the money supply was an important factor in the debate over the creation of the Federal Reserve System, this last point comes into play. Friedman and Schwartz [9,190] argue that elastic meant "subject to substantial change in quantity over short periods for reasons other than immediate profit to the issuer." Phrasing the question this way is misleading because there should not be any reason why elasticity and bank profits are contradictory. Given the legal restrictions of the National Banking System, the two were incompatible, but if the national banks had been able to issue currency without the bond collateral requirements, then their own profit maximizing behavior would have led to an elastic currency." Friedman and Schwartz's [9,190] three conditions for elasticity," and implicitly the rationale for the Fed, overlook a fourth possibility-removing the bond collateral requirements.'" Recent literature on the origins of the Fed, for example Miron 118], emphasizes the seasonality of interest rates under the National Banking System. The justification for the Fed was the need to smooth interest rates. However, it can be argued that legal restrictions were one likely cause of such seasonality. Because the bond collateral requirements led to currency shortages, and because the shortages led to reserve drains, bank balance sheets were subject to heavy fluctuation. The harvest season drain on currency and reserves may well have led to the interest rate movements that the current literature is concerned with. In the absence of legal restrictions, shifts in the currency-deposit ratio would have posed no problem, and the resulting drains on reserves and interest rate seasonality would not have occurred. Further deregulation of note issue could have been a possible alternative to the Federal Reserve System as a way to smooth out seasonal interest rates. 20 t6. Sclgin (341 also notes how the currency·deposit ratio can play havoc with the total supply of money under central banking. Since currency (in the U. S.) is also ballk reserves, increases in the public'S currency-depo.it ratio mean a loss in bank reserves and a mulliplicd cOlllraclion in thc 10lal supply uf muncy. Such fluctuations would nol occur with unregulated banks, sinc.:c currency would be inside, and nul outside, mUllcy. 17. Kleill (161 and L. White 1471 oller preliminary sketches of the prulit maximizing comlitions tilr compelilive note issue. Selgio 1351 provides a more complele picture of lhu~e cundiliulls. i8. These conditions are: I) some body or budies to supervise ur cuntrol the crcalion and rctircl1lc.!nt uf fcdc.!ral Reserve money and (0 haudle the mechanical details; 2) some mcans fur crca,ing and retiring Federal Kcscrvl! moncy; 3) some crileria to replace profit in determining the amounllu be ncated or retired. 19. Admittedly, friedman and Schwanz wrote lhose liul!s UVl!r 25 years agu when cenlrali7.ed conlrol over Ihl.!' money supply appeared to be lhe only way 10 solve apparently endemic claslicily problcms. However, rc,-=cnt advancc~ in the theory of cOlnpetitivc banking now Dlake the passage SCCIII sUlllcwhal dated. 20. or course, any cornplele explanalion or Ihe origins of Ihe Fed would have lu ,leal wilh a myriad of ulher historical factors in addition 10 cunsideratiuns of economic theory slrklly dclincd. II
Free Banking /I 648
SI('I'('/1 Horwi/Z
VII. Conclusion Empirical cases of unregulated note issue are difficult to find, since instances of such hehavior have hcen fcw ;111<1 far hetween. The currency suhstitutes issued hy the national hanks during the panic of 1907 provide a fairly good example of many of the phenomena that might be expected with competitive note issue. The acceptance of these suhstitutes also provides some evidence that the public would use competitive currency hacked hy general bank assets. When evidence or puhlic acceptance is combined with the ability of competitive currency producers to easily switch from deposit to currency liahilities, the removal or the legal restrictions on note issue presents itself as a feasible alternative to solving the elasticity and interest rate problems that have traditionally justified the creation of the Fed. Further research in this area should help to more completely determine the costs and benefits of each option.
References I. AJehian, Annen A., "Why Money"" JOllmal of MOlley, Credil and Ballkillii. February 1977, 133-40. 2. Amlrew. A. PiaU, "The Influence of the Crops Upon Business in America." Quarlerly JOllmal oJ Ecollomics, May 1906, 323-53. J. - - - , "Substitutes for Cash in the Pallic of 1907." Qllarlerly JOllrnal of Ecrmomics, August 190X, 497-519. 4. Barnard. 13. W.o "The Usc of Private Tokens for Money in the Unilctl Slates," Quarlerly Journal oJ 1~'('mlOmics. August 1917.600-34. 5. Ilell, Spurgeon. "Prolit on Natiollaillallk Notes." Amerimll Ecollomic ReI·iew. March 1912.38-60. 6. Brunncr. K.. rl and Allan II. Mciller. "The Uscs of MOlley: Money in the Theory of an Exchange Economy." Amer;ca" Rconomic Rl'~';('H'. December 1971,784-805. 7. eag,m. Phillip. /)el"I'",illlllll.< alld I'fji'ct., of C"(I11/Ies ill I"e MOlley Stock, 11i75-19fiO. New York: National Bureau or Economic Rcsc:.rch. )9()5. X. Frankel. S. Herbert. TlI'o P"ilo.wl'"ie., ,,(Money. New York: SI. Martin's Press, 1977. 9. Friedman. Millon and Anna J. Schwar17.. A Monelary llistory oJ I"e Ulliled Siales, 11i67-1960. Princeton: Princeton University Press, 1«)63. 10. - - - a n d - - - . Mon,'lary Slali.'li", o{ II", Uniled Sillies. New York: Columbia University Press. 1970. II. Goodhart. C. A. E .• "Prolit on National Bank Notes 1900-13." JOIll'/laloJ Polilical Ecollomy. October 1965, 516-22. 12. Gorton. Gary. "Clearinghouses and the Origin uf Cent",1 Banking the United States." JOIll'/lal oJ Ecollomic lIi.'lOry. June 1985, 277-X3. 13. - - - and Donald J. Mullineaux, "The Joint Production of Conlidence: Endogenous Regulation and Nineteenth Century Commercial-Bank Clearinghouses." Jourl/aloJMolle)', Credil (/Ild Bankill/i, November 1987.457-68. 14. James. John A., "The Conundrum of the Low Issue of National Bank Notes." JOImwl oJ Polilical Ecollomy, April 1976,359-67. 15. Jones, Robert A., "The Origin anti Development of Metlia of Exchange." Journal of Political Eco/Unny. August 1976, 757-75. 16. Klein, I3enjmllin. "The Competitive Supply of Money." Journal of Money, Credit a"d fJauki,,):. November 1974.423-53. 17. Menger. Carl, "On the Origin of Money." /:'collomic JOllmal, June 1892.239-55. IR. Miron. Jeffrey A .... Panics, the Scasonalily of Interest Rates. anti the Founding of the Fed." America" Econnmic R""iell', March 1986. 125-40. 19. Mi~cs. Lutlwig von. 111t' n,rory (!rMol/('Y (111(/ Crrdit. Indimwpolis: Lihcrty Press, 19KO. 20. MulliIlC'IIIX. Donald J. "COlllpctilive Monies ,Inti the SufTolk B::ll1k System: A Contractual Perspective." Sou,h· er" I~'co""",ic Journal. April 1987, KK4-98. 21. Nagatani, Keiw. MOllel"ry 11leory. New York: North·Holland Publishing Company, 197X. 22. Nell' York TillleI, October 25, 1907. 2.1. - - - , Novemoer 3, 1907. 24. - - - . November 12, 1907. 25. - - - . November 17,1907. 26. O'Driscoll, Gerald 1'.• "Molley: Menger's Evolutionary Theory." HiS/ory oJ Polilical Ecollomy. Winter 1986. 601-lfi. t
339
Free Banking Jl
340
COMPETITIVE CURRENCIES AND TilE ORIGINS OF TilE FED
649
27. OSlruy, Joseph M. and Ross M. Slarr, "Moncy and Ihe Dccenlralizalion of Exchange." l:."cunol1/t'lr;cu, November 1974, 1093-·1113. 2H. P;/Isburgh p()SI, November 2, 1907. 29. - - - , November 3, 1~07. 30. - - - , Nuvember 5, 1907. 31. - - - , November 6, 1~07. 32. Richler, Rudulf, "The New InSliluliunalisl Ecunomics Applied 10 Monelary Ecunumics." JOI(rtlul of illS/;II/I;on,,1 alul Theorel;ml EL"tJllomics, february 19HH, 20H-24. 33. Sdgin, George A. "Free Banking in China: IHOO-I~35." Manuscripl, George Mason Universily, I~H6. 34. - - - , "Accommodaling Changes in Ihe Relalive Demand lur Currency: free Banking vs. Cenlral Banking." CUIO JounUll, Winler 19H8, 621-41. 35. - - - . The 71.eory of Free Hlowa, New Jor.ey: Rowman and Lililelidd, 1~8H. 36. - - - alld Lawrence H. While, "Compelilive Munies and Ihe Sulfulk Bank Syslem: Commenl." SOli/hem Economic lU1m",I, July 1988, 215-19.
37. Sinunel, Georg. The Philosophy of MOl/ey. Boslon: Roulledge and Kegan Paul. 197H. 38. Sprague, Oliver M. W.. lIiswry of Crises Under Ihf Nal;/JI1t11 Hunk;1/1i !(vSlem. Fairlield, New Jersey: AugusIus M. Kelley, 1977. 3<). SI. LOllis Globe-Dell/ocral, OClober 29, 1907. 40. - - , Oetohcr 30,1\/07. 41. - - , OClober 31,1\/07. 42. - - - , NovemberH, 1907. 43. Timberlake, Richard H., "The Cenlral Banking Rule ofClearinghuuse Associaliuns." J",mlal ,{Money, Cred;1 alld Hankil/g, February 19l!4, 1-15.
44. - - - , "Privale Produclion of Scrip-Money inlhe Isulaled Communily." J(l/Irt1al ({ MOl/ey, Cred;1
47. While, Lawrence II .. Frfe Bankil/g in Hri"'il/: Theory, 1:.:'I'(·r;"I/(·(', <11/(1/)('/111/(' IXIXJ-IH45. Camhridge, Cambridge Universily Press, 1~l!4. 4H. - - - . "The Growing Scarcily of Banknules in Ihe Uniled Siales: 18('5-1~13." Manu.cripl, New Yurk Universily, 1987. 49. Wooldridge, William C. UI/de Sam Ihe M(JIIOI"'/y Mall. New Ruchelle, New Yurk: Arlinglon House, 1970.
Name Index Aftalion, F. 242 Akerloff, George 294, 298 Alexander, William 4 Andreades, A. 219', 229 Andrew, Abram Piatt 279,285,286, 333 Atkinson, Edward 81-3 Bastiat, Frederic 81 Baxter, W.T. 31 Becker, Gary 178 Benton, Colonel 106 Beranger, J. 238, 241, 243, 256 Bemanke, Ben S. 215 Berry, Thomas 196 Biddle, Nicholas 106 Bilgram, Hugo 24, 33, 34, 35 Black, Fischer 23,25,43,47,48, 50, 53, 54, 150 Bloch, E.N. 237, 243 Bouchary, J. 236, 241-3, 245,247,254,256 Bronson, Isaac 99, 100, 101 Brough, William 83-5 Brown, Stuart E. 203-8 Bruneau, M. 236 Bryan, William Jennings 82 Bryant, W.C. 67 Buchanan, James 314 Butler, Benjamin Franklin 59-63 Byrdsall, L. 67, 95 Cagan, Phillip 152, 181 Caldwell, Walter 206, 208, 209 Cameron, Rondo 214, 216, 219 Campbell, Duke of Argyll 4 Cannon, James G. 280, 281, 284, 285, 286, 326 Carey, Henry Charles 59, 67-9, 81, 108 Carothers, Neil 203 Carr, Jack 217, 225, 226 Carraz, R. 236, 242 Chaddock, R. 168 Checkland, Sydney G. 214, 216, 217, 219, 224, 226-30, 244 Clark, C.R. 205 Clark, Truman A. 27 Cleveland, Grover 82 Coase, Ronald H. 319
Cole, A. 163, 170 Colson, Achille 236, 241-4 Colwell, Stephen 31 Conant, Charles 87 Cowen, Tyler 43,47,212,213,218,220, 223, 228-30, 232 Cullop, William 291 Curtis, Edward 95-6 Curtis, George 108 Dewald, William 289 Dewey, David R. 297, 301 Diamond, Douglas 215 Dillistin, William 296, 320 Dilworth, James 85-7 Dodrill, Gordon 206 Dorfman, J. 69 Dowd, Kevin 224 Dundas, Henry 11 Dundas, Sir Lawrence 11 Dybvig, Philip 215 Eckels, James H. 288 Einaudi, Luigi 30 Einzig, Paul 30 Eleazar, Lord 101, 102, 103, 107 Elliott, Gilbert 8 Ellis, Edward D. 110 Ellis, Howard S. 29 Engerman, Stanley L. 134 Fairchild, Ben L. 288 Fama, Eugene 23, 25, 26, 47, 50, 54, 148, 150, 153, 174, 212, 213 Faulkner, Roland P. 203 Felch, Alpheus 195, 196 Fetter, Frank W. 229, 230 Fishback, Price 203, 206 Fisher, Irving 27,48,51,53, S5 Fisher, J. Greevz 24, 36 Flagg 107 Forbes, Kevin 218 Forbes, R.N. 263 Forman, Joshua 100, 101, 102, 103 Frankel, S. Herbert 29 Friedman, Milton 149, 156, 157, 178, 179, 210, 235, 238, 290, 295, 338
342
Free Banking II
Fullarton, John 64 Gallatin, Albert 102, 159 Gardenstone, Lord 268 Gibbons, James S. 171,321 Gilman, Theodore 287, 288 Glied, Sherry 217, 225, 226 Goldberg, Victor 314 Gorton, Gary 157,214,304,309,316,319, 324, 326, 328, 330, 336, 337 Gouge, William 58 Gray, John 31 Greene, William B. 32 Greenfield, Robert I. 23, 31, 39, 47, 48, 50, 52, 54, 55, 212 Hall, Robert E. 23, 25, 47, 48, 50, 54, 146, 147, 212 Hamilton, Alexander 31, 58 Hamilton, James A. 101 Hammond, Bray 93, 181, 182, 195,295,297, 301, 323 Harper, Joel W. 208 Harris, S. 244 Hayek, F.A. 66, 146, 147, 175,245,280, 295, 328 Henry, Duke of Buccleuch 8, 10, 11 Hepburn, Alonzo Barton 280, 284, 291, 331 Hildreth, Richard 59, 62, 69-72, 108 Hoard, Charles Brook 107 Holdsworth, William S. 216 Horsin-Deon, E. 254, 255 Hubrecht, G. 236 Hume, David 95 Humphrey, Thomas H. 228 Hurst, James W. 202, 203 Jackson, President Andrew 63, 94, 106 Jefferson, Thomas 58 Jevons, William S. 27, 54 John, Stuart 31 Jonson, Ole S. 203, 205, 206 Kareken, John H. 29, 179 Keleher, Robert E. 228 Kent, James 108 King, James Gore 108 King, Robert 294, 309, 318 Kinnear, George 277 Kitson, Arthur 24, 29, 35-8, 42, 47, 51-2, 54 Klein, Benjamin 155, 280, 298, 318, 328 Knox, John 181 Krooss, H.E. 58, 63 Kroszner, Randall 43, 47, 212, 213, 218, 220,
223, 228-30, 232 Laidler, David 29 Lake, Wilfred S. 300, 305 Law, John 3, 12 Leggett, William 59, 63-7, 69, 71 Locke, John 95 Lothrop, Edwin H. 110 Lucas, R.E., Jr 66 Luckett, Dudley 181 MacVickar, John 97-103, 105, 107 Madison, James 58 Mann, Abijah Jr 106, 107 Marcy, Governor William L. 67, 95, 96, 102 Marion, Marcel 242 Mathewson, Frank 217, 225, 226 McCallum, Bennett T. 28, 29, 53, 54 McCloskey, Donald N. 316 McCulloch, 1. Huston 179, 214 McCulloch, J.R. 62, 70, 100, 104 McElroy, Wendy 27 McVickar, John 58, 59-63, 64, 69 Menger, Carl 28, 81, 335 Meulen, Henry 24, 25, 29, 37-42, 47-50, 52, 54 Miller, Patrick 11 Mints, Lloyd W. 78, 280 Miron, Jeffrey A. 338 Mises, Ludwig von 28, 36, 334 Mitchell, W. 163 Montesquieu, Baron de 30 Moran, Charles 79-81 Morris, Robert 31, 99 Morse, Charles F. 332 Mullineaux, Donald J. 304, 309-12, 316, 330, 336, 337 Mundell, Robert 219 Munn, Charles W. 216, 217, 224, 227, 228 Mure, Baron 268 Mushet, Robert 65 Myers, Margaret 280, 281, 282, 290, 291 Newcomb, Simon 78 Noyes, A.D. 72 O'Driscoll, Gerald P.J. 28, 212 Ogden, David B. 106, 107 Oswald, James 8 Owen, Robert 292 Parnell, Sir Henry 62, 65, 103 Paterson, William 3, 12 Patinkin, Don 35
Free Banking II
Patterson, W.G. 105 Penney, J.C. 216 Pitt, William 97 Powell, Ellis, T. 218 Prime 108 Proudhon, PJ. 32 Raguet, Condy 67-9, 99, 108 Ramsay, William 11 Redlich, Fritz 67, 69, 280-83, 286, 287, 290, 296, 305, 321 Ricardo, David 62, 97, 99, 100, 105, 107 Ridgeway 81 Rockoff, H. 72, 174, 181-3, 187,215,218, 235, 243, 244 Rolnick, Arthur 166, 190, 191, 193-5, 197, 202, 235, 244 Root, L.C. 166 Rousseau, Jean-Jacques 95 Rowe, Nicholas 29 Rubin, Paul 303 Samuelson, P. 153 Sargent, Thomas 23, 24, 29 Schwartz, Anna J. 65, 156, 157,210,235, 290, 295, 338 Scott-Moncrieff, Robert 16 Sears 216 Selgin, George 212, 314, 316, 330 Seymour, Henry 37 Sharp, J.R. 67 Shaw, Leslie M. 288 Sherwood, Sidney 79 Simmel, Georg 29 Smith, Adam 3-22, 64, 66, 81,94,97,243 Smith, Vera 72 Smith, W. 163, 170 Soddy, Frederick 38 Spencer, John Canfield 105 Sprague, Oliver M.W. 280, 282-4, 289, 291, 292 Steuart, Sir James 3-8, 11, 12, 16, 17, 20, 21, 22, 31, 93, 98
343
Stigler, G.J. 153 Stiglitz, Joseph E. 319, 322 Sturtevant, Julian M. 79-81 Sumner, William Graham 78 Sylla, Richard 202 Timberlake, Richard H. 162, 203, 208, 288, 310, 328, 330, 333, 337 Tobin, J. 152 Trantow, Terry N. 207 Trivoli, George 310 Tucker, Benjamin 24, 36 Van Buren, Martin 67, 100, 107 Vanderlip, Frank R. 331 Venit, Abraham H. 99 Verplanck, Guilian 104, 105 Walker, Amasa 301 Walker, Francis 78 Wallace, Neil 23,24,29,47, 179,212,213, 215, 228 Ward 108 Weber, Warren 166, 190, 191, 193-5, 197, 202, 235, 244 Webster, Pelatiah 58 Westrup, Alfred 24, 32 White, E.N. 237 White, Horace 87, 280, 286 White, Lawrence H. 28, 29, 40, 42, 60, 70, 202, 212, 213, 217, 219, 223-6, 230, 244-6, 295,297,304,314,316,320,328,330 Whitney, D.R. 297, 300 Whittick, William 24, 32 Williams, Aneurin 48, 52, 54, 55 Williamson, Oliver 314, 319 Willis, Henry Parker 292 Wilson, George 79-81 Wright, Austin Willard 85-7 Yeager, Leland B. 23, 31, 39, 47, 48, 50, 52, 54, 55, 212 Young Samuel 102, 103, 104, 109
