B R O O K I N G S PA P E R S O N
William C. Brainard and George L. Perry, Editors
2005 Ryan D. Nunn, Statistical Associate Theodore Papageorgiou, Assistant to the Editors Michael Treadway, Editorial Associate Lindsey B. Wilson, Production Associate
BROOKINGS
INSTITUTION
WA S H I N G T O N , D . C .
Copyright © 2005 by THE BROOKINGS INSTITUTION 1775 Massachusetts Avenue, N.W., Washington, D.C. 20036 ISSN 0007-2303 ISBN-10: 0-8157-1350-9 ISBN-13: 978-0-8157-1350-0 Authorization to photocopy items for internal or personal use or the internal or personal use of specific clients is granted by the Brookings Institution for libraries and other users registered with the Copyright Clearance Center Transactional Reporting Service, provided that the basic fee is paid to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For more information, please contact CCC at (508) 750-8400. This authorization does not extend to other kinds of copying, such as copying for general distribution, for creating new collective works, or for sale. Specific written permission for such copying must be obtained from the Permissions Department, Brookings Institution Press, 1775 Massachusetts Avenue, N.W., Washington, DC 20036; FAX (202) 797-6195; E-mail
[email protected].
THE BROOKINGS INSTITUTION
The Brookings Institution is a private nonprofit organization devoted to research, education, and publication on important issues of domestic and foreign policy. Its principal purpose is to bring the highest quality independent research and analysis to bear on current and emerging policy problems. The Institution was founded on December 8, 1927, to merge the activities of the Institute for Government Research, founded in 1916, the Institute of Economics, founded in 1922, and the Robert Brookings Graduate School of Economics and Government, founded in 1924. Interpretations or conclusions in Brookings publications should be understood to be solely those of the authors.
Board of Trustees John L. Thornton Chair Strobe Talbott President Zoë Baird Alan R. Batkin Richard C. Blum Geoffrey T. Boisi James W. Cicconi Arthur B. Culvahouse Jr. Kenneth W. Dam Vishakha N. Desai Thomas E. Donilon Mario Draghi Kenneth M. Duberstein
Lawrence K. Fish Cyrus F. Freidheim Jr. Bart Friedman David Friend Ann M. Fudge Jeffrey W. Greenberg Brian L. Greenspun William A. Haseltine Teresa Heinz Samuel Hellman Glenn Hutchins Joel Z. Hyatt Shirley Ann Jackson Kenneth Jacobs Suzanne Nora Johnson
Michael H. Jordan Harold Hongju Koh William A. Owens Frank H. Pearl John Edward Porter Steven Rattner Haim Saban Leonard D. Schaeffer Lawrence H. Summers David F. Swensen Larry D. Thompson Laura D’Andrea Tyson Antoine W. van Agtmael Beatrice W. Welters Daniel Yergin
Robert A. Helman Roy M. Huffington James A. Johnson Ann Dibble Jordan Vernon E. Jordan Jr. Breene M. Kerr Marie L. Knowles James T. Lynn Jessica Tuchman Mathews David O. Maxwell Donald F. McHenry Robert S. McNamara Mary Patterson McPherson Arjay Miller Mario M. Morino Maconda Brown O’Connor Samuel Pisar
J. Woodward Redmond Charles W. Robinson James D. Robinson III Judith Rodin Warren B. Rudman B. Francis Saul II Ralph S. Saul Henry B. Schacht Michael P. Schulhof Joan E. Spero Vincent J. Trosino John C. Whitehead Stephen M. Wolf James D. Wolfensohn Ezra K. Zilkha
Honorary Trustees Leonard Abramson Elizabeth E. Bailey Rex J. Bates Louis W. Cabot A. W. Clausen William T. Coleman Jr. Alan M. Dachs D. Ronald Daniel Robert A. Day Bruce B. Dayton Charles W. Duncan Jr. Walter Y. Elisha Robert F. Erburu Henry Louis Gates Jr. Robert D. Haas Lee H. Hamilton F. Warren Hellman
B R O O K I N G S PA P E R S O N
2005
William C. Brainard and George L. Perry, Editors
Editors’ Summary
ix
Articles OLIVIER BLANCHARD, FRANCESCO GIAVAZZI,
and FILIPA SA International Investors, the U.S. Current Account, and the Dollar Comments by Ben S. Bernanke and Hélène Rey General Discussion 62
1
50
and KENNETH S. ROGOFF Global Current Account Imbalances and Exchange Rate Adjustments MAURICE OBSTFELD
Comments by Richard N. Cooper and T. N. Srinivasan General Discussion 141
67 124
MICHAEL DOOLEY and PETER GARBER
Is It 1958 or 1968? Three Notes on the Longevity of the Revived Bretton Woods System Comments by Barry Eichengreen and Jeffrey A. Frankel General Discussion 204
147 188
SEBASTIAN EDWARDS
Is the U.S. Current Account Deficit Sustainable? If Not, How Costly Is Adjustment Likely to Be?
211
Comments by Kathryn M. E. Dominguez and Pierre-Olivier Gourinchas 272 General Discussion 282 DEAN BAKER, J. BRADFORD DELONG,
and PAUL R. KRUGMAN Asset Returns and Economic Growth Comments by N. Gregory Mankiw and William D. Nordhaus General Discussion 325
316
289
Purpose
Brookings Papers on Economic Activity contains the articles, reports, and highlights of the discussions from conferences of the Brookings Panel on Economic Activity. The panel was formed to promote professional research and analysis of key developments in U.S. economic activity. Prosperity and price stability are its basic subjects. The expertise of the panel is concentrated on the “live” issues of economic performance that confront the maker of public policy and the executive in the private sector. Particular attention is devoted to recent and current economic developments that are directly relevant to the contemporary scene or especially challenging because they stretch our understanding of economic theory or previous empirical findings. Such issues are typically quantitative, and the research findings are often statistical. Nevertheless, in all the articles and reports, the reasoning and the conclusions are developed in a form intelligible to the interested, informed nonspecialist as well as useful to the expert in macroeconomics. In short, the papers aim at several objectives: meticulous and incisive professional analysis, timeliness and relevance to current issues, and lucid presentation. Articles appear in this publication after presentation and discussion at a conference at Brookings. From the spirited discussions at the conference, the authors obtain new insights and helpful comments; they also receive searching criticism about various aspects of the papers. Some of these comments are reflected in the published summaries of discussion, some in the final versions of the papers themselves. But in all cases the papers are finally the product of the authors’ thinking and do not imply any agreement by those attending the conference. Nor do the papers or any of the other materials in this issue necessarily represent the views of the staff members, officers, or trustees of the Brookings Institution.
Correspondence
Correspondence regarding papers in this issue should be addressed to the authors. Manuscripts are not accepted for review because this journal is devoted exclusively to invited contributions.
Subscriptions
For information on subscriptions, standing orders, and individual copies, contact Brookings Institution Press, P.O. Box 465, Hanover, PA 173310465. Or call 866-698-0010. E-mail
[email protected]. Visit Brookings online at http://bookstore.brookings.edu. Brookings periodicals are available online through both Online Computer Library Center (contact OCLC subscriptions department at 800-848-5878, ext. 6251) and Project Muse (http://muse.jhu.edu).
Panel members and staff:
Dean Baker Center for Economic and Policy Research Olivier Blanchard Massachusetts Institute of Technology William C. Brainard Yale University J. Bradford DeLong University of California, Berkeley Michael Dooley University of California, Santa Cruz Sebastian Edwards University of California, Los Angeles Peter Garber Deutsche Bank Francesco Giavazzi Universitá Commerciale Luigi Bocconi Paul R. Krugman Princeton University Maurice Obstfeld University of California, Berkeley George L. Perry Brookings Institution Kenneth S. Rogoff Harvard University Filipa Sa Massachusetts Institute of Technology ________ Ryan D. Nunn Brookings Institution Theodore Papageorgiou Yale University Michael Treadway Brookings Institution Lindsey B. Wilson Brookings Institution
Panel advisers participating in the seventy-ninth conference:
Martin Neil Baily Institute for International Economics Richard N. Cooper Harvard University Benjamin A. Friedman Harvard University Robert J. Gordon Northwestern University N. Gregory Mankiw Harvard University William D. Nordhaus Yale University Edmund S. Phelps Columbia University
Guests whose writings or comments appear in this issue:
Henry J. Aaron Brookings Institution Ben S. Bernanke Board of Governors of the Federal Reserve System Kathryn M. E. Dominguez University of Michigan Barry Eichengreen University of California, Berkeley Jeffrey A. Frankel Harvard University Pierre-Olivier Gourinchas University of California, Berkeley Gian Maria Milesi-Ferretti International Monetary Fund Richard Portes London Business School Hélène Rey Princeton University T. N. Srinivasan Yale University
Editors’ Summary The brookings panel on Economic Activity held its seventy-ninth conference in Washington, D.C., on March 31 and April 1, 2005. This issue of Brookings Papers on Economic Activity includes the papers and discussions presented at the conference. The first four articles address the position of the United States in the global economy, an increasingly controversial subject in the research, financial, and policy communities. Since the early 1990s, U.S. current account deficits have grown almost without interruption, reaching $666 billion, or about 6 percent of GDP, in 2004. The U.S. international investment position is now one of net indebtedness approaching 30 percent of GDP, and in recent years a substantial portion of the buildup in net debt has come in the form of additions to dollar reserves by foreign central banks. Some observers see the present situation as unsustainable and warn of an abrupt depreciation of the dollar, which could destabilize financial markets and disrupt the global economy. Others are more sanguine, arguing that the present situation reflects the relative strength of the U.S. economy, consumer and business preferences, and rational financial decisions, all of which could evolve so as to make any needed adjustments gradual. Each of the four articles takes a different approach to analyzing the situation, focusing on issues that the authors see as key. The first article models portfolio choices and how they moderate the pace of adjustment in exchange rates and current accounts. The second stresses the relative price changes that will be needed, both in the United States and abroad, to move the U.S. current account toward balance. The third considers the motivations of policymakers in China and elsewhere for accumulating dollar reserves. The fourth assesses the likelihood of an abrupt depreciation of the dollar and the economic instability that might result in the United States and abroad. The volume concludes with an article on the possible impact of slowing labor force growth on stock market returns. ix
x
Brookings Papers on Economic Activity, 1:2005
The u.s. international investment position is affected by developments in both foreign trade and international capital flows—the market for imports and exports of goods and services and the market for foreign and domestic assets. The sustainability of the U.S. current account deficit and the consequences of reducing that deficit depend on features of both those markets. Most economic models that have been used to analyze the current account deficit assume imperfect substitutability between foreign and domestic goods and services but perfect substitutability between foreign and domestic assets. These assumptions carry strong implications for how the economy adjusts to new developments. In the first article in this volume, Olivier Blanchard, Francesco Giavazzi, and Filipa Sa provide a distinctive analysis that allows for imperfect substitutability between domestic and foreign assets and between domestic and foreign goods. With this feature, movements in exchange rates and asset prices have potentially important effects on the portfolios of international investors and strong implications for the speed with which exchange rates adjust to shocks. Compared with popular discussion and with earlier, simpler models, this rich specification provides a better understanding of past developments in the U.S. current account balance and the dollar exchange rate and a more realistic framework for assessing future prospects. In its simplest form the authors’ model has just two regions—the United States and the rest of the world—each of which supplies interest-bearing assets. The wealth of each region is given by the value of domestic assets plus net claims on foreigners. Investors diversify their portfolios, holding both foreign and U.S. assets, but exhibit home bias: given equal expected returns, they place a larger fraction of their wealth in domestic than in foreign assets. As a result, a shift in wealth to foreigners reduces the demand for U.S. assets, causing the dollar to depreciate. Similarly, an increase in private or government demand for dollar assets causes the dollar to appreciate. Because of imperfect substitutability, the relative returns on foreign and U.S. assets can vary with changes in relative supplies or shifts in the distribution of world wealth, and uncovered interest parity does not hold. In the model the effects of a depreciation on the path of the current account balance and changes in U.S. net foreign indebtedness are conventional. The current account balance is the sum of the trade balance and net interest earnings. Dollar depreciation improves both, immediately reducing the dollar value of net interest payments and eventually reducing the U.S. trade deficit. Changes in U.S. net foreign indebtedness reflect the sum of the
William C. Brainard and George L. Perry
xi
current account balance and the revaluations of U.S. and foreign portfolios that arise from exchange rate movements. In the real world, asset values and therefore net debt will also change with changes in domestic interest rates, but the model ignores these so as to focus on exchange rate movements, which are the key for understanding the model’s distinctive implications. Whereas the response of the current account in the model is quite familiar, the effect of depreciation on asset demands is quite different than in conventional models where assets are perfect substitutes. Depreciation of the dollar reduces U.S. net indebtedness directly, increasing the dollar value of foreign assets held in U.S. portfolios while decreasing the value of U.S. assets in foreign portfolios. If assets were perfect substitutes, these changes in portfolio shares would be of no importance, and the expected returns on U.S. and foreign assets would always have to be equal. With fixed domestic interest rates, the expected change in exchange rates would then be zero. In such a world, real exchange rate changes are always unexpected. With imperfect substitutability, in the absence of compensating changes in expected relative rates of return, investors in both regions will want to rebalance their portfolios following an unexpected exchange rate movement. Thus an unexpected depreciation of the dollar in response to a trade shock actually increases the relative demand for U.S. assets, reducing but not reversing the depreciation. Unlike in the case of perfect substitutability, the expected returns on U.S. and foreign assets do not have to be the same after the initial adjustment. Rather than jump all the way to a new equilibrium from which no further change is expected, the dollar undergoes a sharp initial, unexpected depreciation followed by a more gradual, expected depreciation. The expected depreciation merely reduces the desired shares of U.S. assets in investors’ portfolios rather than causing massive flight from dollars. The rate at which the dollar depreciates after its initial response to an adverse shock depends on the elasticity of asset demands with respect to the relative rates of return: the lower the elasticity, the more gradual the depreciation and the improvement in the current account. Since observed outcomes are always the result of past and present shocks, the dynamics of adjustment toward the steady state are of particular interest. The authors analyze two representative cases. In response to a shock that increases the trade deficit, such as an increase in U.S. economic activity or an enlarged preference for imports, there is, as explained above, an initial, unexpected depreciation of the dollar, followed by a gradual
xii
Brookings Papers on Economic Activity, 1:2005
further, anticipated depreciation and an increase in U.S. net debt. How much of the depreciation is immediate and how much takes place on the subsequent path of adjustment depend on the response of trade to the depreciation and on the responsiveness of portfolio demands to the anticipated changes in relative rates of return. The less substitutability between foreign and U.S. assets, the smaller will be the initial depreciation, and the more rapid the subsequent depreciation. However, the eventual depreciation in the new steady state is the same, and large enough to generate a sufficient trade surplus to offset the higher interest payments on the larger debt. The second case involves a response to a shock that increases the demand for U.S. assets, such as an increase in demand by foreign governments. In this case the reduced supply available to private portfolios leads to an initial dollar appreciation. This enlarges the trade deficit, adding to the future flow of dollar assets supplied. The subsequent path is one of a gradual, anticipated depreciation and increase in net debt. Despite the initial favorable portfolio shift, the new steady state requires a weaker dollar, since, as in the previous example, the trade surplus must be larger to offset the interest payments on the now-larger debt. The authors suggest that the U.S. experience of recent years can be understood as responses to shocks like those just described. In their view a shift in private portfolio preferences toward U.S. assets led initially to an appreciation of the dollar. Independently, a shift in the preferences of U.S. consumers toward foreign goods worsened the trade balance by more than can be explained by exchange rate and income effects. As described above, both kinds of shifts predict an eventual sustained dollar depreciation to a level below that prevailing before the shift. Although the accumulation of reserves by foreign governments has supported the dollar against some currencies, the authors argue that the United States has entered the depreciation phase of the adjustment that their model predicts. To assess future prospects, and in particular how large an eventual dollar depreciation should be expected, the authors quantify their model using estimates of present wealth, assets, portfolio shares, and net debt for the United States and the rest of the world, together with estimates of model parameter values based on existing empirical studies and some assumptions about adjustment speeds and policy preferences. For 2003 these estimates include the following: U.S. assets of $36.8 trillion, foreign assets of $33.3 trillion, and U.S. net foreign debt of $2.7 trillion; 77 percent of U.S. wealth
William C. Brainard and George L. Perry
xiii
invested in U.S. assets, and 71 percent of foreign wealth invested in foreign assets. In the model these shares imply that a transfer of one dollar of U.S. wealth to foreigners leads to a decrease of 48 cents in demand for U.S. assets. The estimated trade elasticities imply that a 1-percentage-point reduction in the ratio of the trade deficit to GDP requires a depreciation of 15 percent. Armed with these quantifications of their model, the authors use it to predict where the U.S. international position is headed. First they calculate the exchange rate adjustment that would be needed to maintain the present net debt position as a steady state, under the implicit assumption that the economy has already adjusted to past shocks, and introducing no important asymmetries between foreign and U.S. interest rates or growth rates. In this case the ratio of the current account deficit to GDP that can be sustained indefinitely is given by the economy’s growth rate times the ratio of net debt to GDP. With 3 percent annual growth in U.S. GDP, maintaining a net debt ratio of about 25 percent requires reducing the current account deficit from its present 6 percent to 0.75 percent of GDP. With annual interest rates at 4 percent, this requires a depreciation of the dollar of 56 percent. The authors note some important qualifications to this calculation. To the extent that the present current account deficit reflects J-curve effects in response to the dollar’s recent depreciation (in which a depreciation at first worsens the current account balance before improving it), it overstates the additional depreciation required. Noting that the current account continued to worsen for nearly two years after the depreciation of the mid-1980s began, they estimate that a similar path this time would mean that only a 34 percent further depreciation is needed. They also note that if the U.S. net debt ratio were allowed to stabilize at a higher level than the present, the equilibrium current account deficit could be larger. As an alternative way to assess the dollar’s prospects, the authors undertake dynamic simulations of the response to trade and portfolio shocks in which the equilibrium debt-to-GDP ratio is endogenous. Simulating permanent shocks to the trade deficit, they calculate that a 1-percent-of-GDP shift away from U.S. goods increases the equilibrium net debt ratio by 17 percentage points and causes the dollar to depreciate by 12.5 percent. Simulating shifts in asset preferences, they calculate that, in response to a shift that raises the share of U.S. assets in both U.S. and foreign portfolios by 5 percentage points, the dollar initially appreciates and then eventually
xiv
Brookings Papers on Economic Activity, 1:2005
reaches an equilibrium depreciation of 15 percent with a 35-percentagepoint increase in the net debt ratio. A striking feature of both simulations is how long it takes to reach equilibrium. After fifty years the adjustment is still far from complete, with the dollar still above its pre-shock level after the shift toward dollar assets, and the depreciation only about two-thirds complete after the shift in trade away from U.S. goods. Although they question the realism of these extraordinary adjustment periods, the authors believe they do correctly show that the adjustment process can be very long. Such gradualism contrasts with the predictions of some observers that the dollar is likely to fall abruptly in the near future. To evaluate this possibility, the authors examine under what conditions their model would predict a faster depreciation than in their baseline simulations. As discussed above, the anticipated rate of depreciation is faster, the less substitutability there is between U.S. and foreign assets, with the extreme case of constant shares providing an upper bound. For this case the authors show that the anticipated rate of depreciation depends on the change in the ratio of U.S. net debt to U.S. assets: a faster rise in the debt ratio requires a more rapid depreciation to maintain portfolio balance. In a situation where the net debt ratio is rising by 5 percent a year, and with a ratio of gross assets to GDP of 3—both rough approximations of recent experience—they calculate an anticipated rate of depreciation of 2.7 percent a year. This estimate is based on anticipated portfolio shares remaining constant. In the model, however, the rate of depreciation will also be affected by any anticipated change in the relative demand for U.S. assets—a shock imposed on top of the constant-shares assumption in the previous calculation. If the demand for shares of U.S. assets in foreign or domestic portfolios is expected to decline, the expected depreciation can be much faster. For example, if the share of U.S. assets demanded in foreign portfolios is expected to decline by 2 percentage points over the coming year, the expected depreciation rises to 8.7 percent. The authors note that there is considerable disagreement about the share of U.S. assets that foreigners will want to hold in the future. Some observers argue that foreign central banks will continue their recent policy of adding to dollar holdings. Others see a latent demand for U.S. assets by private Chinese investors who are currently restrained by capital controls. Although the authors consider these outcomes possible, they find it more likely that the relative demand for U.S. assets will decline in the near
William C. Brainard and George L. Perry
xv
future, as foreign central banks stop pegging the dollar or diversify their portfolios away from U.S. assets, or both. The calculations just provided for a shift in shares are then relevant. The authors also observe that the longer the peg continues, the larger both the initial and the eventual depreciation will be. The depreciation of the dollar since its 2002 peak has been very uneven against different currencies: the dollar has fallen 45 percent against the euro, 25 percent against the yen, and not at all against the Chinese renminbi. To investigate how future adjustments would impact each of these important currencies, the authors extend the essentials of their model to include four regions rather than just two. The analysis focuses on the interrelations among the United States, Japan, the euro region, and China, ignoring the rest of the world. The authors assume that half the U.S. current account deficit is with China and a quarter with each of the others, values that approximate recent actual deficits. These deficits transfer wealth, and how that wealth is invested drives exchange rate movements. The model allows for two special features of the Chinese economy: capital controls on private financial capital inflows and outflows, and the pegging of the renminbi to the dollar. Asset preferences in each of the other three regions are allowed to differ, but all are assumed to have the same marginal response to changes in expected returns, and interest rates measured in the domestic currency are assumed to be the same in each. The authors illustrate the main forces at work using a simplified version of the model in which asset demands do not depend on expected returns. For a given U.S. current account deficit, the more dollar assets China holds, the smaller is the appreciation of the euro and yen. Surprisingly, if China holds only dollar assets, a U.S. current account deficit actually causes the dollar to appreciate against both the euro and the yen, since most of the U.S. deficit is with the region with extreme dollar preferences. If only Japan accumulates dollars, both the yen and euro appreciate, with the yen appreciating more. In this case a transfer of wealth to Japan leaves the real effective exchange rate of the euro unchanged, as the euro rises against the dollar and falls against the yen. The authors also use this framework to analyze the effects of prospective changes in China’s policies. If China stops pegging but maintains capital controls, it will have a zero current account surplus, which would require an appreciation of the renminbi against the dollar. Reserve accumulation would then cease, and the U.S. current account deficit would have to be
xvi
Brookings Papers on Economic Activity, 1:2005
financed entirely by investors in Japan and Europe. This shift in wealth accumulation away from the region with extreme dollar preferences would strengthen the euro and the yen against the dollar. A diversification of China’s portfolios away from all dollars would have a similar effect. The same qualitative results are also found in simulations that allow for the endogenous response of portfolio choices to expected relative returns. Thus, in the authors’ analysis, China’s pegging to the dollar has limited the appreciation of the euro and yen against the dollar, in contrast to the opinion of some commentators that it has increased the pressure on the euro to appreciate. The authors briefly address the connections between domestic fiscal and monetary policy and the U.S. international position. As the U.S. current account and budget deficits have risen together in the past five years, they have frequently been paired in discussions of needed policy changes, with some commentators identifying the latter as the cause of the former and calling for reduced fiscal deficits as a possible substitute for depreciation. The authors point out, however, that these are complementary changes rather than substitutes, with interest rates a key link between the two. With the dollar depreciating under the pressure of excessive current account deficits, demand for U.S. output expands, requiring a combination of higher interest rates and fiscal deficit reduction to maintain domestic balance. Because higher interest rates would limit the immediate depreciation while requiring more in the future, smaller budget deficits are the appropriate balancing change. But, if fiscal policy is tightened without dollar depreciation, the economy is likely to weaken. The authors conclude by summarizing the implications of their findings for understanding the recent past and projecting the future. In their view the path of the dollar since the late 1990s has been supported by increases in the demand for U.S. assets, first by private investors for equities and more recently by central bank demands for bonds. A shift in preferences away from U.S. goods has also contributed to growing trade deficits in this period. Imperfect substitution in portfolios helps account for the gradualism of exchange rate adjustments and for the persistent U.S. current account deficits that have been observed. The model predicts that a gradual depreciation of the dollar will be the prevailing trend for an extended period. However, if the expected demand for U.S. assets falls, as it would if central bank policies changed, the decline in the dollar would be more abrupt. Similarly, the gradual depreciation could be interrupted by a temporary
William C. Brainard and George L. Perry
xvii
appreciation if investors’ preferences shifted toward dollar assets, although the resulting larger trade deficits would lead to an even larger depreciation eventually. For the same reason, a rise in U.S. interest rates would strengthen the dollar only temporarily and require a larger depreciation in the longer run. The authors thus reason that a better policy mix would combine a reduction of budget deficits with a reduction of interest rates to maintain growth. Turning to China, the authors argue that eventually the government will find it difficult to continue to sterilize interventions and will abandon its dollar peg. But the longer the peg will have supported the dollar, the larger the eventual dollar depreciation will have to be in order for the United States to service the larger accumulated foreign debt. The authors also observe that a large dollar depreciation would not necessarily be a major problem for the United States. By improving the trade balance, it would permit a reduction of budget deficits without causing a recession. However, dollar depreciation might pose a bigger problem for Japan and Europe, which are already growing slowly and which have limited scope for expansionary stabilization policies. Some lay commentators have suggested that eliminating the federal budget deficit would automatically reduce today’s massive deficit in the U.S. current account. In a 1987 paper, James Tobin identified this as one of eight “myths” about exchange rates and the current account, because it ignores the fact that improvements in the current account balance have to be earned in competition with foreign producers and will, if employment is to be maintained, require changes in exchange rates and terms of trade. In the second article in this issue, Maurice Obstfeld and Kenneth Rogoff pursue this theme. They first provide a wide-ranging discussion of recent economic developments, concluding that the U.S. current account deficit will before long have to be substantially reduced, if not eliminated. They then model the price adjustments that would be required to change import and export patterns in the United States and abroad so as to eliminate or substantially reduce the U.S current account deficit without reducing aggregate economic activity. Although most analysts recognize that improving the trade balance will require a real depreciation of the dollar, less attention has been paid to the likely need for changes in the relative price of traded and nontraded goods both in the United States and among its trading partners. In earlier work
xviii
Brookings Papers on Economic Activity, 1:2005
Obstfeld and Rogoff have argued that these adjustments are, if anything, likely to be larger than the changes in the relative prices of domestic and foreign tradable goods—the terms of trade. It is easy to show why this might be so. Without changes in production anywhere, eliminating the U.S. current account deficit, which today stands at roughly 6 percent of GDP, implies something like a 20 percent reduction in U.S. consumption of traded goods. Assume for simplicity that the traded goods of different countries are perfect substitutes, so that exchange rate changes do not change the relative price of different traded goods, but only the prices of nontraded goods relative to traded goods within countries. Then, with a unitary elasticity of substitution between traded and nontraded goods and hence constant shares, this 20 percent reduction in consumption of traded goods requires a fall in the price of nontraded goods relative to traded goods of the same percentage. In foreign countries, where, under these assumptions, consumption of traded relative to nontraded goods has to rise, the relative price of the latter must also rise. If the traded goods of different countries are not perfect substitutes, the calculations are more complicated, and the required terms of trade and real exchange rates need to be determined simultaneously. But the qualitative nature of the needed adjustment is the same. To capture the salient features of the current international environment, the authors develop their model by assuming three world regions, representing the United States, Asia, and Europe, all linked by trade and by a matrix of international asset and liability positions. This enables the authors to model asymmetries in the trading relationships between regions and to analyze the implications of dividing the improvement in the U.S. trade account between Europe and Asia in different ways. The model is short run and static. Each region produces two goods: a nontraded good consumed only by its residents, and a traded good that is both consumed domestically and exported. Hence there are a total of six goods in the world economy. The regions are endowment economies with flexible prices, implicitly assuming factor immobility between sectors and full employment. The preferences of consumers, and in particular the elasticities of substitution among the different goods, play the central role in determining price adjustments associated with changes in the current account. Four commodities are available to consumers in each region—their own region’s traded and nontraded goods and the traded goods of the other two regions. The authors model goods preferences in each region by means of two
William C. Brainard and George L. Perry
xix
constant-elasticity-of-substitution (CES) consumption indexes: the first is an index of overall traded good consumption derived from a bundle of the three traded goods, and the second aggregates this index with nontraded good consumption to provide a utility measure for total consumption. Although the functional form of these CES functions is the same across regions, the weights on the commodities differ. In particular, the traded goods index displays home bias: consumers in each region have a relative preference for the traded good that it produces and exports. Even though the law of one price (individual traded goods have the same price everywhere) holds, the price indexes for each region’s bundle of traded goods will differ across regions, because each depends on the region’s own consumption weighting of individual traded goods. This implies that an increase in a region’s income and expenditure improves its terms of trade, raising the price of its exports relative to that of its imports. The United States and Europe exhibit mirror symmetry in their preferences for each other’s traded good but place the same weight on the Asian traded good. Asia meanwhile weights the U.S. and the European traded goods the same, and the model allows the weight it places on those goods to be changed, reflecting changes in openness to trade. Whereas the weights on different goods thus differ across regions, elasticities of substitution among goods are assumed to be the same for all regions. The authors review a range of empirical studies to arrive at informed judgments about the size of these elasticities. In their baseline calculations the elasticity of substitution among tradables is assumed to be 2, implying that, ceteris paribus, a 10 percent change in the consumption of, say, an Asian import to the United States would be associated with a 5 percent change in its price relative to that of the U.S. traded good. The elasticity of substitution between nontraded goods and the index of consumption that aggregates the three traded goods is assumed to be 1, as in the simple example above. Given these preferences, the authors can solve for the prices that equate demand to supply for any global allocation of the six commodities. The bilateral terms of trade are simply the relative prices of any two regions’ traded goods. Given the assumption of CES utility, the authors compute exact price indexes for each region’s consumption bundle of traded goods and for its overall consumption. Ratios of the latter give the corresponding bilateral real exchange rates. As noted earlier, even though the law of one price holds, the price indexes for the bundle of traded goods differ across regions because of
xx
Brookings Papers on Economic Activity, 1:2005
differences in consumption weighting of the three traded goods. The authors assume that each region’s bundle of traded goods has a 0.25 weight in total consumption. This means that a change in a region’s bilateral real exchange rate is 0.25 of the change in the region’s relative price index for traded goods. That is, changes in the terms of trade, through their differing effects on regions’ price levels for traded goods, can be traced directly to real exchange rates. For example, if the price of the U.S. traded good falls relative to the price of Europe’s traded good—an improvement in Europe’s bilateral terms of trade—the relative price of the United States’ traded goods index will also fall. Hence there will be a real depreciation of the dollar relative to the euro. Most of the burden of reducing the U.S. current account deficit has to be borne by U.S. consumers reducing their consumption of traded goods, but part of the adjustment is accomplished through valuation effects. The United States is a net debtor, with its liabilities predominantly denominated in dollars and more than half of the foreign assets held by U.S. residents denominated in foreign currencies. Depreciation of the dollar therefore actually decreases U.S. net indebtedness. Although this decrease in U.S. net worth might be expected to affect demand gradually over time, it has an immediate effect on the current account. Since foreign-denominated U.S. assets exceed foreign-denominated U.S. liabilities, and interest payments are largely denominated in the same currency as the underlying asset, the dollar value of U.S. net interest receipts rises with a depreciation. To estimate the effect on the terms of trade and real exchange rates of reducing the U.S. current account deficit by 5 percent of GDP, the authors have to make assumptions about how the offsetting reduction in current account surpluses is distributed between Europe and Asia. They consider three scenarios: a global rebalancing scenario, where the current accounts of all three regions go to zero; a “Bretton Woods II” scenario, where Asia’s currencies remain pegged to the dollar (a hypothesis analyzed at length in the paper by Michael Dooley and Peter Garber in this volume); and a muted version of the latter, where Asia maintains its current account surplus so that a reduction in Europe’s surplus just balances the U.S. deficit reduction. Given the assumed baseline elasticities, the changes in consumption implied by global rebalancing imply very large real exchange rate changes. The euro appreciates in real terms by over 28 percent, and the Asian currencies by over 35 percent. The greater real appreciation for Asia reflects the fact that, initially, Asia has a much larger surplus than Europe, so that
William C. Brainard and George L. Perry
xxi
moving to balance requires a much larger increase in its consumption of traded goods. Although these may seem like large numbers, they are not so different from what would be expected if traded goods were perfect substitutes as in the earlier example. The fact that they are not, and that there is home bias, does result in a deterioration of the U.S. terms of trade with both Europe and Asia of about 14 percent; the result is a slightly larger real depreciation than would otherwise be required. The authors’ model is developed entirely in real terms, but they are able to translate the real exchange rate changes into nominal changes by making assumptions about how domestic price levels change. If, for example, each central bank targets stability in its region’s overall consumer price index, then real exchange rate changes are the only source of nominal exchange rate change. Stabilizing the GDP deflator, which has different weights than the consumer price index, gives much the same result. The substantial depreciation of the dollar predicted by the model has a large effect on the Asian net foreign asset position: because 80 percent of Asia’s foreign asset holdings, but only 34 percent of its foreign liabilities, are denominated in dollars, Asia’s net foreign asset position is reduced in value by 60 percent. Although this is a substantial wealth effect, it produces only a small decline in the current account, in turn only slightly reducing the required exchange rate adjustment. Europe is in a more balanced position and suffers a much smaller loss of wealth, with a negligible effect on the required dollar depreciation. The changes in consumption implied by the other two scenarios require quite different real exchange rate adjustments. Under the authors’ Bretton Woods II scenario, Asia raises its surplus in the process of pegging to the dollar, increasing the adjustment that Europe must make. Specifically, Asia increases its surplus as a percentage of U.S. traded-goods output from its current 15 percent to 31 percent, and Europe has to move from its current 5 percent surplus, measured the same way, to a 31 percent deficit. This adjustment requires an appreciation of the euro by roughly 60 percent against both the dollar and the Asian currencies. Europe’s terms of trade also rise dramatically, on the order of 25 percent. The authors conclude that sustaining Asia’s peg in the context of a substantial reduction in the U.S. current account deficit is likely to be politically unacceptable for Europeans. Furthermore, even though Europe’s net foreign asset position is much less sensitive than Asia’s to exchange rate changes, the required appreciation of the euro would be large enough to result in a significant loss in wealth.
xxii
Brookings Papers on Economic Activity, 1:2005
In the third scenario, Asia allows its currencies to rise against the dollar by roughly 20 percent, just enough to keep its current account surplus constant as the United States moves to current account balance, thus placing less of a burden on Europe than in the Bretton Woods II scenario. Although the euro still has to appreciate by nearly 45 percent against the dollar, Europe’s effective exchange rate is affected much less, by 32 percent rather than 60 percent, because of Europe’s substantial trade with Asia. Although the authors believe they have made fairly optimistic assumptions about elasticities, which, if anything, understate the required price adjustments, they also report results under some alternative assumptions. Raising the elasticity of substitution between nontraded and traded goods from 1 to 2 reduces the required depreciation of the dollar significantly. In the global rebalancing scenario the real dollar-euro rate rises by 19.3 percent rather than 28.6 percent, and the Asian currencies appreciate by 22.5 percent rather than 35.2 percent—still quite significant adjustments. Eliminating the changes in the terms of trade by assuming that traded goods are near-perfect substitutes for each other has similar quantitative effects, revealing that the terms-of-trade effects were responsible for about a third of the real dollar depreciation in the baseline model. As previously noted, because the United States’ foreign debts are mostly denominated in dollars and its foreign asset holdings mostly in foreign currencies, valuation effects dampen the depreciation of the dollar required to eliminate its current account deficit. The authors show that this effect is modest: in their baseline estimates, the depreciation in terms of the U.S. effective exchange rate is about 13 percent less than it would be in the absence of valuation effects, implying that improvements in the trade balance still have to do the heavy lifting. They also show the effect of the United States losing its historical ability to borrow at a low interest rate: the effect of putting the United States on a par with other debtors is of roughly the same magnitude as the valuation effects estimated above. The authors recognize that their model ignores some effects that might significantly change their estimates of needed dollar depreciation. In particular, they note that the realism of two of their key assumptions depends on the time horizon over which adjustments take place. In the short run, perhaps one or two years, the assumption of factor immobility appears reasonable, but the assumption of completely flexible prices seems less so. For the longer run, price flexibility seems more reasonable, but factor
William C. Brainard and George L. Perry
xxiii
immobility is implausible. The authors recognize that factor reallocation between sectors will dampen the expected real exchange rate adjustments compared with the adjustments estimated for the core model. Thus, if current account adjustments take place slowly and over many years, a smaller reduction in the real exchange rate will be required to achieve current account balance. Although they recognize that their model is incomplete in this and other ways, the authors believe that the framework they have developed for understanding needed relative price changes will be essential in any analysis of major adjustments in current accounts. The determinants of current account balances have been analyzed much more extensively than those of international capital flows. But the interaction of the capital and current accounts in determining exchange rates highlights that understanding capital flows is just as important as understanding trade flows. For emerging economies the role of such flows has received considerable attention and has been identified as an important factor in currency crises. However, for the U.S. dollar, which has been the world’s dominant reserve currency for over half a century, there is less empirical evidence and considerable uncertainty about how its exchange rate responds to U.S. current account deficits and the accumulation of dollar assets abroad. In the third article of this issue, Michael Dooley and Peter Garber expand on what they have elsewhere called the “Revived Bretton Woods” hypothesis, which stresses the willingness of foreign official sectors to accumulate dollar liabilities, and they marshal support for their prediction that U.S. current account deficits need not trigger a major dollar devaluation or currency crisis in the foreseeable future. Dooley and Garber first describe the key features of the global economy that underpin their Revived Bretton Woods view. The first is China’s ongoing transformation from a centrally planned to a nascent market economy, which has moved hundreds of millions of underemployed workers into the global market for labor. China’s continued economic development depends on employing such workers, and, in the authors’ view, pursuit of this goal will override the conventional pressures of trade imbalances in China’s exchange rate strategy. The second is that most successful emerging economies have been net exporters of capital, contradicting the usual assumption that successful development involves poor countries borrowing from rich ones. The authors relate this seeming anomaly to the emerging economies’ need for international capital by arguing that the export of sav-
xxiv
Brookings Papers on Economic Activity, 1:2005
ing supports two-way trade in financial assets that improves the productivity of their domestic capital formation. In China’s case access to needed international capital is currently inhibited by the country’s geopolitical past and its primitive financial system. The third key feature is that the large and growing current account deficit of the United States has been funded at low interest rates by foreign private and official lenders, suggesting that the large foreign holdings of U.S. assets have not diminished the demand for further accumulation. The motivation for emerging economies, and particularly China, in building foreign reserves is central to the authors’ argument. Export promotion has long been an accepted strategy for a developing economy, and the value of building reserves became apparent when foreign capital flight from East Asia and elsewhere led to the crises of the 1990s. However, these motives by themselves neither explain recent developments nor predict how far the reserve buildups will go. Export-led growth alone does not imply the need for a trade surplus and net export of capital, nor does the precautionary motive require an indefinite buildup of reserves. However, the authors hypothesize an additional motive for building reserves, which today applies most clearly to China. Growth requires efficient capital formation, yet the domestic financial system will, for a long time, not be up to the task of channeling China’s high rate of saving into a high rate of productive domestic investment. International financial intermediation can substitute for the inadequate domestic financial system, but potential foreign investors are put off by political risk. Dooley and Garber argue that China’s foreign reserves act as collateral that reduces this risk. They provide an extensive discussion of the role of private collateral arrangements and the uncertainties that inhibit financial investments in their absence, citing earlier work by Ricardo Caballero and Arvind Krishnamurthy on the role of international collateral for private financing in developing economies. The authors observe that the U.S. authorities are legally empowered to freeze or seize foreign-owned assets under a range of unusual circumstances, and they identify many occasions when this has been done. Although the conditions for taking such action are not well defined or even generally understood, market participants and other governments believe that the United States will take similar action in the future against foreign governments that expropriate private foreign assets. By holding dollar reserves that are vulnerable to seizure, a country thus provides effective collateral to potential investors.
William C. Brainard and George L. Perry
xxv
Dooley and Garber concede that the Chinese authorities may have stumbled on this role of foreign reserves inadvertently. But they argue that, having done so, the authorities now accept a continuing buildup of foreign reserves as support for continued growth in gross inward foreign investment. Hence the authors’ collateral hypothesis provides a connection between China’s current account surpluses (or net capital flows) and gross capital flows. In the present geopolitical climate, the collateral hypothesis would seem more relevant to China than to the more developed Asian economies, especially Japan. But Dooley and Garber note that Japan has managed its exchange rate for many years as a way of dealing with its own employment problem, and they see some of the other Asian economies as motivated to keep their currencies aligned with China’s. They also see little pressure from market forces that would cause Japan and the other economies to abandon their reserve buildups. History provides many examples of market forces overwhelming official attempts at intervention to support a weak currency, but the analysis of those cases does not necessarily apply to interventions to repress a strong currency. Nor have the constraints that often arise when undervaluation or intervention leads to excessive monetary expansion and overheating been a problem for China or for other developing Asian economies, and those constraints are clearly irrelevant for today’s cyclically depressed Japan. Having thus explained why, in their view, these buildups of dollar reserves abroad may continue, the authors turn to historical evidence of past episodes of buildups and how they have ended. From a sample of 115 developing and industrial countries for the period 1970–2004, they identify episodes in which a country ran current account surpluses for several consecutive years and the government increased its net foreign asset position by at least 25 percent of the change in national net foreign assets. They find several regularities in these episodes, one of which is that the typical episode of reserve buildup has a relatively benign ending. With few exceptions, current account surpluses grew during the period of reserve accumulation. When the accumulation stopped, current account surpluses declined on average by 2 percent of GDP in the first year, suggesting that the accumulation typically ended as a result of some shock to the previous situation. On average, a real appreciation occurred in the last three years of reserve accumulation, which in itself suggests a fundamental disequilibrium in the exchange rate and the current account. But rather than sub-
xxvi
Brookings Papers on Economic Activity, 1:2005
sequently appreciating, as might be expected from such a disequilibrium, currencies on average depreciated and economic growth moderated. From this analysis of the typical experience, the authors turn to a more detailed look at Japan, China, and Korea, which together accounted for 45 percent of global reserve holdings at the end of 2004. Since 1970 Japan has had three episodes of extended reserve accumulation: the first starting in 1986 and lasting three years, the second starting in 1992 and lasting five years, and the current episode, which started in 1999. Based on the pattern of the first two episodes, the authors project a moderate real appreciation of the yen and moderate economic growth in Japan in the immediate future, with reserve accumulation ending when the current account deteriorates, at which time the real value of the yen will fall. Korea experienced net reserve accumulation from 1986 to 1989 and again from 1998 to the present. The end of the first episode coincided with a decline in the current account surplus and a slowdown in GDP growth. The authors find nothing unusual about the present episode, with reserve accumulation roughly matching the current account surpluses and the won strengthening moderately in real terms. China’s episode of reserve accumulation is the longest in the entire sample, extending from 1990 to the present. Small current account surpluses were roughly matched by reserve accumulation from 1990 to 2001. Since then, however, China’s experience has been without modern precedent: the current account surplus has grown rapidly, and private capital inflows have been roughly as large; hence reserve accumulation has been about twice as large as the surplus. Because the authorities have been able to control inflation, the authors see no pressure to end the accumulation. However, they suggest that the buildup might end if an interruption of direct investment inflows or liberalization of capital outflows were to lead to a real depreciation of the renminbi. From these examinations of past episodes, the authors draw several generalizations. They find almost no support for the idea that reserve accumulations end with speculative attacks that force the currency to appreciate. Rather, they typically end when the current account surplus declines substantially or swings into deficit, and they are followed by a real depreciation and a modest downturn in the economy. One implication is that episodes of reserve buildups do not end with capital losses on the government’s reserves. Nor do they end with recessions generated by a sharp real appreciation. From this evidence the authors judge that there are no constraints or
William C. Brainard and George L. Perry
xxvii
obvious risks based on historical experience that would keep today’s current account surplus economies from continuing to finance U.S. current account deficits, as the Revived Bretton Woods hypothesis predicts. Dooley and Garber expect that China’s present financial repression of capital flows and distortion of the real exchange rate will end when the country’s industrial sector has grown sufficiently and the domestic financial system has become capable of efficient intermediation. But they also expect that this will take a long time. Many analysts and commentators in the business and financial press see the U.S. international net debtor position as a major risk on the economic horizon, not only for the United States but possibly also for the global economy. In the fourth article in this volume, Sebastian Edwards examines the recent history of the U.S. current and capital accounts, models some likely paths for them in the years ahead, and examines historical episodes of sustained deficits and growing foreign indebtedness in other countries for clues to how serious a risk the present U.S. situation entails. Edwards starts with a brief history of the U.S. international position in the three decades since exchange rates began floating in the early 1970s. Focusing on the real trade-weighted exchange rate of the dollar and the ratio of the U.S. current account balance to GDP, he identifies two extended episodes of major imbalance. The first began in the early 1980s, when the current account went deeply into deficit following a sharp real appreciation of the dollar. This episode resolved when a steep depreciation that began in 1985 returned the current account briefly to balance by 1991. The second is the present enlargement of the deficit to record levels, which started with a period of appreciation of the dollar from the mid-1990s to 2002 and has continued despite the real depreciation that followed. Each year’s current account deficit worsens the U.S. net international investment position (NIIP) by a corresponding amount. But the NIIP, which is measured in dollars, is also affected by changes in the valuation of assets held across borders. These valuation effects occur as exchange rate movements change the dollar value of foreign assets held by U.S. nationals. In the 1980s valuation effects that were predominantly positive partly offset the adverse effects of large current account deficits on the U.S. NIIP. Nevertheless, by 1986 the United States had become a net debtor, and the massive deficits of the current episode have increased the net debt position to about 30 percent of GDP. However, because the returns on U.S. assets held
xxviii
Brookings Papers on Economic Activity, 1:2005
by foreigners have been systematically lower than the returns on foreign assets held by U.S. nationals, the income component of the current account has to date remained positive. The entire current account deficit thus consists of an enormous deficit in goods and services trade and a modest deficit in transfers to foreigners. Edwards turns next to an analysis of where the U.S. current account, real exchange rate, and NIIP are likely to go from here. The three are, of course, interrelated, with the main linkages coming from the exchange rate affecting the trade balance, portfolio investments affecting the demand for dollars, and the balance between portfolio and trade flows affecting the exchange rate. Edwards captures these interrelationships in a model whose main features include elements of the models of Blanchard, Giavazzi, and Sa and of Obstfeld and Rogoff in their papers in this volume. Asset demands are driven by wealth, with a bias for home assets and exogenously determined portfolio shares. (The inclusion of demand by foreign central banks can be treated as a shift in this home bias.) Trade flows are driven by the real exchange rate, which affects the relative prices of traded and nontraded goods and services, and by fluctuations and growth in incomes; the magnitude of these effects is determined by price and income elasticities in the United States and abroad. With this model Edwards is able to analyze long-run equilibriums, that is, the eventual adjustments to real exchange rates and current accounts that can be expected in response to various shocks. With some simplifying assumptions, the sustainable ratio of the U.S. current account to GDP is proportional to the growth rate of U.S. nominal GDP, with the proportionality depending on the relative returns and riskiness of its assets and the degree of integration of capital markets— factors captured in the portfolio balance parameters. To go further and characterize the dynamic path of such adjustments to equilibrium, Edwards includes partial adjustments for asset holdings, which allow for imperfections in countries’ capital markets, and for the current account, which allow for consumption smoothing. As an example of the resulting rich dynamics, he shows that a decline in home bias in the rest of the world, which would increase the sustainable U.S. current account deficit, would lead initially to the deficit overshooting its new equilibrium level. Edwards applies his model to the current situation by calibrating its parameters using values from earlier studies and values for the dynamic adjustment terms that best explain the behavior of the U.S. current account since
William C. Brainard and George L. Perry
xxix
1996. He then uses the calibrated model to simulate the effect of shocks to portfolio choices, focusing on a specification in which the desired proportions of foreign and domestic assets remain fixed after the shock. The simulations also assume that annual economic growth rates in the United States and abroad average 3 percent and that the terms of trade do not change. In his base case, foreigners are assumed to gradually increase the desired proportion of dollar asset holdings in their portfolios from the present 30 percent to 40 percent in 2010, while U.S. nationals reduce their desired holdings of U.S. assets from 73 percent to 71 percent over the same period. These portfolio shifts have the effect of doubling foreigners’ net demand for U.S. assets to an amount equal to 60 percent of U.S. GDP by the end of the period. With this increase in demand for U.S. assets, the dollar appreciates in real terms for the first four years and then depreciates rapidly, eventually approaching a new equilibrium 19 percent below its initial 2005 level. The current account deficit initially continues to grow, peaking at 7.3 percent of GDP after four years. It declines sharply thereafter, approaching an equilibrium ratio of 3.2 percent of GDP after a few more years. The reversal of the trade deficit is even sharper and larger because the growing net debt position raises net income payments to foreigners. The main qualitative findings from this base case are robust under a range of alternative assumptions about the model’s parameters. Edwards simulates alternative assumptions about portfolio choice to test what difference they make to the outcome. If, after the initial five years, foreign investors gradually reduce their holdings of U.S. assets to 50 percent rather than 60 percent of U.S. GDP, the real depreciation and current account reversals are steeper and eventually greater. After three years the depreciation is 24 percent and the current account deficit has shrunk by 5.3 percent of GDP. Both changes continue for two more years, overshooting their eventual equilibrium values: an exchange rate about 23 percent below, and a current account deficit about 3.5 percent of GDP smaller than, 2004 values. Although the size of these changes is within U.S. historical experience, once the changes get under way, their abruptness, which comes from dollar accumulations abroad reaching an assumed limit, could be destabilizing. Edwards notes that different parameters for the adjustment process could produce less abrupt changes, but he regards the qualitative characteristics of the simulations as representative of the model’s dynamics. And he notes that all the reported simulations assume foreign demand for U.S. assets far exceeding today’s 30 percent of GDP.
xxx
Brookings Papers on Economic Activity, 1:2005
Large, abrupt swings in current account balances have often been accompanied by disruptions to employment and growth in the affected economies. Edwards looks to international experience with such reversals to see whether it offers any insight into what is in store if the United States undergoes the kind of changes predicted by his simulations. He defines two types of current account reversal, one in which the current account deficit declines by at least 6 percent of GDP within a three-year period, and one in which it declines by at least 4 percent in a single year. For the period 1971-2001 he finds that the first type of reversal occurs in 9.2 percent of all country-years, and the second type in 11.8 percent of all country-years. He reports a number of other interesting findings, including a close association of reversals and currency crises, a particular exchange rate pattern that typifies reversals, and a correlation of reversals with economic growth. However, the great majority of the reversals he finds are for small or less developed countries. The corresponding incidences of the two types of current account reversal for industrial countries are only 2.7 percent and 2.0 percent, and most of those reversals occurred in small countries. Among the larger industrial countries, only Italy (in 1975) and Canada (in 1982) experienced reversals in this thirty-year period. Thus, although Edwards’s rich analysis of historical reversals illuminates the structural and economic conditions—and the problems—typically associated with them, the relevance to the current U.S. situation is unclear. Edwards believes nonetheless that it is very likely that the United States will undergo a major adjustment in the not-too-distant future, which will modify the present global imbalances between the U.S. and rest-of-world current accounts. He identifies three main unresolved issues that will shape how that adjustment unfolds. One is how central banks conduct their reserves policy in a global economy with mostly flexible exchange rates. He notes that, in contrast to the argument made by Dooley and Garber in this volume, many observers believe that foreign central banks that have been accumulating dollar reserves will reduce their demand for dollar assets in the future, unleashing an abrupt collapse in the value of the dollar. Another issue is how world interest rates, which influence global investment, will be affected by a major adjustment to the U.S. current account. And the third is how private sector saving and government budget balances evolve, and whether, in tandem with interest rate adjustments, they will succeed in maintaining global economic growth as the correction of today’s current account imbalances works itself out.
William C. Brainard and George L. Perry
xxxi
Most informed observers forecast a substantial decline in the growth of the labor force in future decades. The baby-boom generation is approaching retirement, no new explosion of fertility is in sight, and the growth in female labor force participation that began in the 1960s is seen as largely complete. Some forecasters also predict a slowing of productivity growth in the longer run, although this is more controversial: the Social Security Administration’s 2005 trustees’ report projects that hours worked will grow by only 0.3 percent a year from 2015 to 2045, a slowdown of 1.2 percentage points from the average for this measure from 1958 to 2004. The trustees also predict that long-run productivity growth will moderate from its pace of the last fifteen years and that together these two factors will lead to a slowdown of GDP growth of between 1.6 and 2.2 percentage points a year. Along with increases in longevity, these projections are the major reason that the Social Security system in its present form will be unable to maintain current benefits into the indefinite future. Given this outlook, if equity investments earn as high a return as they have over the postwar period on average, then investing a portion of the Social Security trust fund in equities, or creating private accounts to allow individual workers to do so, seems an attractive and almost costless way to improve the system’s prospects. However, in the fifth article in this issue, Dean Baker, Bradford DeLong, and Paul Krugman question this reasoning, arguing that rates of return on equities are unlikely to match their historical levels if the pessimistic projections of labor force and productivity growth in the trustees’ report are correct. Standard economic growth models provide a relevant framework for analyzing the long-run effects of labor force and productivity growth on national income growth and the rate of return to capital. The authors begin by reviewing the predictions of the mainstay of growth analysis, the Solow model. In steady state the growth rate of national income is the sum of the growth rates of hours worked and labor-augmenting technical progress, and the capital stock grows at the same rate as income. At any income growth rate the net saving rate determines the capital-to-output ratio, which in turn determines wage rates and the rate of return to capital. It is easy to show that a change in the rate of growth of labor input or labor productivity results in a proportional change in the rate of return to capital, with the proportionality being the ratio of capital’s share of income (which is assumed constant) to the gross saving rate. The Solow model takes the saving rate as given, so that the output-tocapital ratio and the rate of return to capital fall with a reduction in growth.
xxxii
Brookings Papers on Economic Activity, 1:2005
But a sufficient decline in the saving rate will keep the return to capital constant. This leads the authors to analyze two canonical models that address the effect of changing demographics on saving and hence on the rate of return. The first is the Ramsey model, a highly stylized model that assumes that the representative household lives forever, maximizing utility over a consumption path into the indefinite future. Population growth in this model is captured by assuming that the size of the representative household grows over time. Household saving decisions maximize the welfare of this dynastic household, given the projected growth in household size. Usually it is assumed that the household’s utility in a given period is simply the sum of the utilities of the members present in that period, and that the household decisionmaker, contemplating the future, gives the same weight to the utility of the new members as to his or her own. (The authors call this “perfect familial altruism.”) With these assumptions and the assumption that utility is proportional to the logarithm of consumption (log utility), the steady-state rate of return rises one for one with labor productivity growth, as does the growth rate of consumption per worker. However, the rate of return is unaffected by population growth. The reason for this can most easily be seen by abstracting from productivity growth, so that consumption per capita is constant through time. Without population growth, the infinitely lived individual will want to accumulate capital to the point where the rate of return equals the rate of time preference. With population growth, the same condition will hold. The fact that a forgone unit of consumption by each household member today has to be divided among 1 + n members tomorrow is just balanced by the fact that there will be 1 + n fully weighted members tomorrow. So, as with a single individual, the rate of return will be driven to the rate of time preference. Why are these results different from those of the Solow model? In both models, in the steady state, each new member of the labor force has to be equipped with capital. In the Solow model the saving rate is constant, so that the capital-to-labor ratio is higher when there are fewer workers to equip with capital. In the Ramsey model the saving rate falls to keep the capital-to-labor ratio and the rate of return to capital unchanged. Baker, DeLong, and Krugman find the assumption of perfect familial altruism in the Ramsey model implausible, particularly when many of the future members of society are expected to be immigrants unrelated to today’s members. They show that if there is less than perfect altruism, so that current generations give less weight to future generations than to themselves, then, when
William C. Brainard and George L. Perry
xxxiii
population growth slows, saving does not fall enough to maintain the previous rate of return on capital. There are several reasons why the Ramsey model is ill suited for analyzing the effects of demographic change on saving and the rate of return. There is no meaningful way to analyze saving for retirement in a model where individuals live and work forever. Nor can the model analyze the effects on saving of changes in birth and death rates, the age of retirement, or uncertainty about the length of life. However, the second canonical model the authors examine, the Diamond overlapping-generations model, can readily incorporate such features. In this model, versions of which have been used by other authors to estimate empirically the effects of demographic change in the United States and elsewhere, individuals are assumed to go through a life cycle of earning, saving, and consuming. Each agent lives two periods, working and saving when young and consuming the returns on capital acquired through that saving when old. Generations all have the same preferences but differ in consumption opportunities as productivity grows over time. Individuals are assumed to maximize the present value of the utility of consumption over their two-period lifetimes, using log utility and with no bequest motive. Output is given by a CobbDouglas function combining the labor input of the young and the capital owned by the old. Even though it abstracts from some realistic features of the typical life cycle, the Diamond model shows the fundamental differences that arise from assuming finite rather than infinite horizons. First, the rate of return bears no necessary relationship to the pure rate of time preference. Households optimally allocate their income over two periods, and so determine individual household saving, but aggregate saving depends crucially on the demographic structure. With log utility, the fraction of income saved is independent of the rate of return—income and substitution effects just balance—making the analysis quite simple. In the absence of population or productivity growth, aggregate (net) saving would be zero: in every period the dissaving of retirees just balances the saving of the equally numerous young. With population growth there are more savers relative to dissavers, so that the capital stock grows along with the labor force in steady state. However, since the saving of one generation of workers is used by the more numerous next generation, the capital-to-labor ratio is lower and the rate of return higher than with a constant population. If saving per worker were fixed, the capital-to-
xxxiv
Brookings Papers on Economic Activity, 1:2005
labor ratio would fall at a rate proportionate to the growth in the number of workers. The resulting increase in the steady-state real rate of return would have no effect on saving with log utility, but the lower wage would reduce the saving of workers, resulting in an even greater increase in the rate of return than would otherwise be the case. And, if the utility function were less elastic than under log utility, as most analysts believe, the increase in the return to capital would in turn reduce saving, raising the rate of return even further. Productivity growth—an increase in the labor equivalence of a worker—divides one generation’s saving among a greater number of equivalent workers and has the same effect as labor force growth on the rate of return and wages. The authors conclude that there are good reasons to believe that the rate of return on capital will fall if population growth and productivity growth slow. Since capital is the underlying asset generating returns for the owners of firms, it is hard to construct a scenario where a permanent decline in the rate of return on capital does not imply a similar decline in equity returns. The authors illustrate this by examining the implications of the standard Gordon equation for equity prices. The rate of return on an equity claim is its current yield plus capital gains. In the absence of news that affects a company’s prospects, the price of its stock grows with its dividends. The Gordon equation simply shows that the price of a stock is equal to its current dividend divided by the expected rate of return minus the growth rate of dividends. Applying this equation to projections for the aggregate economy, the authors calculate that, if real GDP grows at the 1.5 percent annual rate consistent with the 2005 trustees’ report, and assuming a constant capital share, real earnings on capital will likewise grow at 1.5 percent a year, as will dividends in the absence of changes in firm’s debt-equity ratio or the dividend payout rate. With this growth rate and the current dividend yield, the Gordon equation implies an expected annual real rate of return on stocks of 4.4 percent, considerably less than the 6.5 percent annual real return averaged over the past half-century. Thus the authors conclude that this measure of market expectations is consistent with the fall in the rate of return on capital that they infer from their analysis of growth models. How might future stock market returns be higher than this calculation suggests? The authors discuss several possibilities but in the end are skeptical of their importance. The capital in the growth models corresponds to all productive assets in the economy, including, for example, those of unincorporated enterprises. Thus the rate of return to capital in traded firms
William C. Brainard and George L. Perry
xxxv
could be higher than the return to capital as a whole in the economy. But the authors see no reason to expect that this is so. A firm’s earnings and dividends typically go through a life cycle, so that the growth in dividends for the economy as a whole, reflecting the emergence of new firms, may differ from that of existing firms. But the authors suggest that this difference may mean lower rather than higher dividend growth for a broad stock index. Nor do the authors see much room to increase dividend payout rates. With a decline in the rate of return to capital, increased payouts to stockholders have to come either at the expense of bondholders or from a reduction in retained earnings. Although reducing leverage could temporarily raise the fraction of earnings paid to stockholders, such reductions could not continue indefinitely. As the authors show using the Solow model, a reduction in the saving rate could maintain a higher rate of return to capital by avoiding the increase in the capital-to-labor ratio. In that case, although growth in output would still fall as a result of the fall in growth rates of population and productivity, with growth in earnings and dividends following suit, the dividend yield would be higher, with less retained earnings and household saving required to grow capital at the lower rate. Because there would be less capital along the economy’s growth path, the rate of return would be maintained despite the lower growth rate. The most interesting possibility for maintaining a higher rate of return is a shift in the distribution of world investment away from the United States to regions where the labor force is growing faster and potential returns are higher. The authors reason that, if American companies were to increase their investment abroad, the growth of earnings of companies in the index could exceed the rate of growth of the domestic economy. However, they calculate that achieving the historical 6.5 percent return by this approach would require that companies increase their foreign investment by historically unprecedented proportions, unless that investment substituted for U.S. domestic investment. The authors also point out that, unless the U.S. trade balance changes, any such increase in U.S. firms’ investment abroad will have to be balanced by increased capital inflows of the same magnitude, reducing returns on domestic capital. Hence, if there is no change in U.S. saving, there will be no net effect on the growth of the domestic capital stock, and thus no effect on the rate of return in the United States. They do not address the possibility that any improvement in the trade balance, coming perhaps from dollar
xxxvi
Brookings Papers on Economic Activity, 1:2005
depreciation, would reduce domestic investment for a given level of saving. Restoring balance in U.S. trade would reduce net capital inflows, domestic investment, and growth in the domestic capital stock by the same amount. Since the current account deficit today is near 6 percent of GDP, a major fraction of the capital deepening associated with slowed population and productivity growth could be avoided in this way. Historically, equities have paid a significantly higher return than bonds, resulting in a large (and, many argue, excessive) risk premium on equities. If this equity premium can be counted on to persist, it would seem to provide a good reason for private investors, or the Social Security trust fund, to invest more heavily in equities and less heavily in bonds. But the authors observe that the reason for the high premium remains a puzzle, a fact that argues for caution in adopting any strategy to capitalize on the premium in the future. Insofar as the premium reflects a failure of markets to efficiently allocate risk among individuals, it could make sense for the government, the agent with the greatest ability to manage systematic risk, to take a direct position in equities. But if, as some believe, the growing sophistication of markets is already in the process of eliminating the equity premium, the gains from switching to equities from bonds will disappear. In that case any attempt to exploit the premium would fail in the long run. The authors acknowledge that much uncertainty remains about what the future holds for economic growth. But, they argue, the main inference of their analysis is that it is precisely in those cases when growth slows that returns to equities are likely to be lower than historical experience. Thus, if slower growth does contribute to the Social Security problem, investment in equities is likely to disappoint as a solution.
OLIVIER BLANCHARD Massachusetts Institute of Technology FRANCESCO GIAVAZZI Universitá Commerciale Luigi Bocconi FILIPA SA Massachusetts Institute of Technology
International Investors, the U.S. Current Account, and the Dollar TWO MAIN FORCES underlie the large U.S. current account deficits of the past decade. The first is an increase in U.S. demand for foreign goods, partly due to relatively faster U.S. growth and partly to shifts in demand away from U.S. goods toward foreign goods. The second is an increase in foreign demand for U.S. assets, starting with high foreign private demand for U.S. equities in the second half of the 1990s, and later shifting to foreign private and then central bank demand for U.S. bonds in the 2000s. Both forces have contributed to steadily increasing current account deficits since the mid-1990s, accompanied by a real dollar appreciation until late 2001 and a real depreciation since. The depreciation accelerated in late 2004, raising the issues of whether and how much more is to come and, if so, against which currencies: the euro, the yen, or the Chinese renminbi. We address these issues by developing a simple model of exchange rate and current account determination, which we then use to interpret the recent behavior of the U.S. current account and the dollar and explore what might happen in alternative future scenarios. The model’s central assumption is that there is imperfect substitutability not only between An earlier version of this paper was circulated as MIT working paper WP 05-02, January 2005. We thank Ben Bernanke, Ricardo Caballero, Menzie Chinn, William Cline, Guy Debelle, Kenneth Froot, Pierre-Olivier Gourinchas, Søren Harck, Maurice Obstfeld, Hélène Rey, Roberto Rigobon, Kenneth Rogoff, Nouriel Roubini, and the participants at the Brookings Panel conference for comments. We also thank Suman Basu, Nigel Gault, Brian Sack, Catherine Mann, Kenneth Matheny, Gian Maria Milesi-Ferretti, and Philip Lane for help with data.
1
2
Brookings Papers on Economic Activity, 1:2005
U.S. and foreign goods, but also between U.S. and foreign assets. This allows us to discuss the effects not only of shifts in the relative demand for goods, but also of shifts in the relative demand for assets. We show that increases in U.S. demand for foreign goods lead to an initial real dollar depreciation, followed by further, more gradual depreciation over time. Increases in foreign demand for U.S. assets lead instead to an initial appreciation, followed by depreciation over time, to a level lower than before the shift. The model provides a natural interpretation of the recent behavior of the U.S. current account and the dollar exchange rate. The initial net effect of the shifts in U.S. demand for foreign goods and in foreign demand for U.S. assets was a dollar appreciation. Both shifts, however, imply an eventual depreciation. The United States appears to have entered this depreciation phase. How much depreciation is to come, and at what rate, depends on how far the process has come and on future shifts in the demand for goods and the demand for assets. This raises two main issues. First, can one expect the deficit to largely reverse itself without changes in the exchange rate? If it does, the needed depreciation will obviously be smaller. Second, can one expect foreign demand for U.S. assets to continue to increase? If it does, the depreciation will be delayed, although it will still have to come eventually. Although there is substantial uncertainty about the answers, we conclude that neither scenario is likely. This leads us to anticipate, in the absence of surprises, more dollar depreciation to come at a slow but steady rate. Surprises will, however, take place; only their sign is unknown. We again use the model as a guide to discuss a number of alternative scenarios, from the abandonment of the renminbi’s peg against the dollar, to changes in the composition of reserves held by Asian central banks, to changes in U.S. interest rates. This leads us to the last part of the paper, where we ask how much of the dollar’s future depreciation is likely to take place against the euro, and how much against Asian currencies. We extend our model to allow for four “countries”: the United States, the euro area, Japan, and China. We conclude that, again absent surprises, the path of adjustment is likely to be associated primarily with an appreciation of the Asian currencies, but also with a further appreciation of the euro against the dollar.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
3
A Model of the Exchange Rate and the Current Account Much of economists’ intuition about joint movements in the exchange rate and the current account is based on the assumption of perfect substitutability between domestic and foreign assets. As we shall show, introducing imperfect substitutability changes the picture substantially. Obviously, it allows one to think about the dynamic effects of shifts in asset preferences. But it also modifies the dynamic effects of shifts in preferences with respect to goods. We are not the first to insist on the potential importance of imperfect substitutability. Indeed, the model we present builds on an older (largely and unjustly forgotten) set of papers by Paul Masson, Dale Henderson and Kenneth Rogoff, and, especially, Pentti Kouri.1 These papers relax the interest parity condition and instead assume imperfect substitutability of domestic and foreign assets. Masson and Henderson and Rogoff focus mainly on issues of stability; Kouri focuses on the effects of changes in portfolio preferences and the implications of imperfect substitutability between assets for shocks to the current account. The value added of this paper is in allowing for a richer description of gross asset positions. By doing this, we are able to incorporate into the analysis the “valuation effects” that have been at the center of recent empirical research on gross financial flows,2 and that play an important role in the context of U.S. current account deficits. Many of the themes we develop, including the roles of imperfect substitutability and valuation effects, have also been recently emphasized by Maurice Obstfeld.3 1. Masson (1981); Henderson and Rogoff (1982); Kouri (1983). The working paper version of the paper by Kouri dates from 1976. One could argue that there were two fundamental papers written that year, the first by Dornbusch (1976), who explored the implications of perfect substitutability, and the other by Kouri, who explored the implications of imperfect substitutability. The Dornbusch approach, with its powerful implications, has dominated research since then. But imperfect substitutability seems central to the issues we face today. Branson and Henderson (1985) provide a survey of this early literature. 2. See, in particular, Gourinchas and Rey (2005) and Lane and Milesi-Ferretti (2002, 2004). 3. Obstfeld (2004). We limit our analysis of valuation effects to those originating from exchange rate movements. Valuation effects can and do also arise from changes in asset prices, particularly stock prices. The empirical analysis of a much richer menu of possible valuation effects has recently become possible, thanks to the data on gross financial flows and gross asset positions assembled by Lane and Milesi-Ferretti.
4
Brookings Papers on Economic Activity, 1:2005
The Case of Perfect Substitutability To see how imperfect substitutability of assets matters, it is best to start from the well-understood case of perfect substitutability. Consider a world with two “countries”: the United States and a single foreign country comprising the rest of the world. We can think of the U.S. current account and exchange rate as being determined by two relations. The first is the uncovered interest parity condition:
(1 + r ) = (1 + r*) EE
e +1
,
where r and r* are U.S. and foreign real interest rates, respectively (asterisks denote foreign variables), E is the real exchange rate defined as the price of U.S. goods in terms of foreign goods (so that an increase in the exchange e is the expected real exrate denotes an appreciation of the dollar), and E+1 change rate in the next period. The condition states that expected returns on U.S. and foreign assets must be equal. The second relation is the equation giving net debt accumulation: F+1 = (1 + r ) F + D ( E+1 , z+1 ) , where D(E, z) is the trade deficit. The trade deficit is an increasing function of the real exchange rate (so that DE > 0). All other factors—changes in total U.S. or foreign spending, as well as changes in the composition of U.S. or foreign spending between foreign and domestic goods at a given exchange rate—are captured by the shift variable z. We define z such that an increase worsens the trade balance (DZ > 0). F is the net debt of the United States, denominated in terms of U.S. goods. The condition states that net debt in the next period is equal to net debt in the current period times 1 plus the interest rate, plus the trade deficit in the next period. Assume that the trade deficit is linear in E and z, so that D(E, z) = θE + z. Assume also, for convenience, that U.S. and foreign interest rates are equal (r* = r) and constant. From the interest parity condition, it follows that the expected exchange rate is constant and equal to the current exchange rate. The value of the exchange rate is obtained in turn by solving out the net debt accumulation forward and imposing the condition that net debt does not grow at a rate above the interest rate. Doing this gives
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
E = −
5
−i r 1 ∞ ( 1 + r ) z+e i . F−1 + ∑ 1+ r 0 θ
That is, the exchange rate depends negatively on the initial net debt position and on the sequence of current and expected shifts in the trade balance. Replacing the exchange rate in the net debt accumulation equation in turn gives −i r ∞ ( 1 + r ) z+e i . F − F−1 = z − ∑ 1+ r 0 That is, the change in the net debt position depends on the difference between the current shift and the present value of future shifts in the trade balance. For our purposes these two equations have one main implication. Consider an unexpected, permanent increase in z at time t—say, an increase in the U.S. demand for Chinese goods (at a given exchange rate)—by ∆z. Then, from the two equations above, E − E−1 = −
∆z ; θ
F − F−1 = 0.
In words: permanent shifts lead to a depreciation large enough to maintain current account balance. By a similar argument, shifts that are expected to be long lasting lead to a large depreciation and only a small current account deficit. As we argue later, this is not what has happened in the United States over the last ten years. The shift in z appears to be, if not permanent, at least long lasting. Yet it has not been offset by a large depreciation but has been reflected instead in a large current account deficit. This, we shall argue, is the result of two factors, both closely linked to imperfect substitutability. The first is that, under imperfect substitutability, the initial depreciation in response to an increase in z is more limited, and, by implication, the current account deficit is larger and longer lasting. The second is that, under imperfect substitutability, asset preferences matter. An increase in foreign demand for U.S. assets, for example—an event that obviously cannot be analyzed in the model with perfect substitutability we have just presented—leads to an initial appreciation and a current account deficit. And such a shift has indeed played an important role since the mid-1990s.
6
Brookings Papers on Economic Activity, 1:2005
Imperfect Substitutability and Portfolio Balance We now introduce imperfect substitutability between assets. Let W denote the wealth of U.S. investors, measured in units of U.S. goods. W is equal to the stock of U.S. assets, X, minus the net debt position of the United States, F: W = X − F. Similarly, let W* denote foreign wealth and X* denote foreign assets, both in terms of foreign goods. Then the wealth of foreign investors, expressed in terms of U.S. goods, is given by W* X* = + F. E E Let Re be the relative expected gross real rate of return on holding U.S. assets versus foreign assets: (1)
Re ≡
1 + r E+e1 . 1 + r* E
Under perfect substitutability, the case studied above, Re was always equal to 1; this need not be the case under imperfect substitutability.4 U.S. investors allocate their wealth W between U.S. and foreign assets. They allocate a share α to U.S. assets and, by implication, a share (1 − α) to foreign assets. Symmetrically, foreign investors invest a share α* of their wealth W* in foreign assets and a share (1 − α*) in U.S. assets. Assume that these shares are functions of the relative rate of return, so that α = α ( R e, s ) , α Re > 0, α s > 0
α* = α* ( R e, s ) , α*Re < 0, α*s < 0.
A higher relative rate of return on U.S. assets leads U.S. investors to increase the share they invest in U.S. assets, and foreign investors to decrease the share they invest in foreign assets. The variable s is a shift factor, standing for all the factors that shift portfolio shares for a given relative return. By convention, an increase in s leads both U.S. and foreign investors to increase the share of their portfolio in U.S. assets for a given relative rate of return. 4. One may wonder whether, even if many investors have strong asset preferences, the effects of these preferences on expected returns are not driven away by arbitrageurs, so that expected returns are equalized. The empirical work of Gourinchas and Rey (2005), which we discuss later, strongly suggests that this does not happen, and that financial assets denominated in different currencies are indeed imperfect substitutes.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
7
An important parameter in the model is the degree of home bias in U.S. and foreign portfolios. We assume that there is indeed home bias, and we capture it by assuming that the sum of portfolio shares falling on owncountry assets exceeds 1: α ( R e, s ) + α* ( R e, s ) > 1. Equilibrium in the market for U.S. assets (and, by implication, in the market for foreign assets) implies X = α ( R e, s )W + [1 − α* ( R e, s )](W */ E ). The supply of U.S. assets must be equal to U.S. demand plus foreign demand for those assets. Given the definition of F introduced earlier, this condition can be rewritten as (2)
X = α ( R e, s ) ( X − F ) + (1 − α* ( R e, s ))[( X */ E ) + F ] ,
where Re is given in turn by equation 1 and depends in particular on E and e E+1 . This gives us the first relation, which we refer to as the portfolio balance relation, between net debt, F, and the exchange rate, E. To see its implications most clearly, consider the limiting case where the degree of substitutability is zero, so that the shares α and α* do not depend on the relative rate of return. In this case —The portfolio balance condition fully determines the exchange rate as a function of the world distribution of wealth, (X − F) and [(X*/E ) +F)]. In sharp contrast to the case of perfect substitutability, news about current or future current account balances, such as a permanent shift in z, has no effect on the current exchange rate. —Over time, current account deficits lead to changes in F, and thus to changes in the exchange rate. The slope of the relation between the exchange rate and net debt is given by α + α* − 1 dE E =− < 0. ( 1 − α*) X */ E dF So, in the presence of home bias, an increase in net debt is associated with a lower exchange rate. The reason is that, as wealth is transferred from the United States to the rest of the world, home bias leads to a decrease in the demand for U.S. assets, which in turn requires a decrease in the exchange rate.
8
Brookings Papers on Economic Activity, 1:2005
Outside this limiting case, the portfolio balance determines a relation between net debt and the exchange rate for a given expected rate of depreciation. The exchange rate is no longer determined myopically. But the two insights from the limiting case remain: On the one hand, the exchange rate will respond less to news about the current account than it does under perfect substitutability. On the other, it will respond to changes either in the world distribution of wealth or in portfolio preferences. Imperfect Substitutability and Current Account Balance Assume, as before, that U.S. and foreign goods are imperfect substitutes and that the U.S. trade deficit, in terms of U.S. goods, is given by D = D ( E , z ) , DE > 0, Dz > 0. Turn now to the equation expressing the dynamics of the U.S. net debt position. Given our assumptions, U.S. net debt is given by F+1 = (1 − α*( R e, s ))
W* ( E 1 + r ) − (1 − α ( R e, s )) W (1 + r *) + D ( E+1 , z+1 ) . E E+1
In words, net debt in the next period is equal to the value of U.S. assets held by foreign investors next period, minus the value of foreign assets held by U.S. investors next period, plus the trade deficit next period: —The value of U.S. assets held by foreign investors next period is equal to their wealth in terms of U.S. goods this period times the share they invest in U.S. assets this period times the gross rate of return on U.S. assets in terms of U.S. goods. —The value of foreign assets held by U.S. investors next period is equal to U.S. wealth this period times the share they invest in foreign assets this period times the realized gross rate of return on foreign assets in terms of U.S. goods. The previous equation can be rewritten as 1+ r * E ( X − F ) + D ( E+1 , z+1 ). (3) F+1 = (1+ r ) F + (1− α ( R e, s )) (1+ r ) 1− 1+ r E+1 We shall call this the current account balance relation.5 5. This appears to give a special role to α rather than α*, but in fact this is not the case. A symmetrical expression can be derived with α* appearing instead of α. Put another way, F, α*, and α are not independent. F+1 can be expressed in terms of any two of the three.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
9
The first and last terms on the right-hand side of equation 3 are standard: next-period net debt is equal to this-period net debt times the gross rate of return, plus the trade deficit next period. The term in the middle reflects valuation effects, recently stressed by Pierre-Olivier Gourinchas and Hélène Rey and by Philip Lane and Gian Maria Milesi-Ferretti.6 Consider, for example, an unexpected decrease in the price of U.S. goods—that is, an unexpected decrease in E+1 relative to E. This dollar depreciation increases the dollar value of U.S. holdings of foreign assets, decreasing the U.S. net debt position. Putting things together, a depreciation improves the U.S. net debt position in two ways: the first, conventional way through the improvement in the trade balance, and a second way through asset revaluation. Note that —The strength of the valuation effects depends on gross rather than net positions and so on the share of the U.S. portfolio in foreign assets (1 − α) and on U.S. wealth (X − F ). It is present even if F = 0. —The strength of the valuation effects depends on our assumption that U.S. gross liabilities are denominated in dollars, so that their value in dollars is unaffected by a dollar depreciation. Valuation effects would obviously be very different when, as is typically the case for emerging market economies, gross positions are smaller and liabilities are denominated in foreign currency. Steady State and Dynamics Assume the stocks of assets X and X* and the shift variables z and s to be constant. Assume also r and r* to be constant and equal to each other. In this case the steady-state values of net debt F and E are characterized by two relations. The first is the portfolio balance relation (equation 2). Given the equality of interest rates and the constant exchange rate, Re = 1, the relation takes the form X = α (1, s ) ( X − F ) + (1 − α* (1, s ))[( X */ E ) + F ]. 6. Gourinchas and Rey (2005); Lane and Milesi-Ferretti (2004). As a matter of logic, one can have both perfect substitutability and valuation effects. (Following standard practice, we ignored valuation effects in the perfect substitutability model presented earlier by implicitly assuming that, if net debt was positive, U.S. investors did not hold foreign assets and net debt was therefore equal to the foreign holdings of dollar assets.) Under perfect substitutability, however, there is no guide as to what determines the shares, and therefore what determines the gross positions of U.S. and foreign investors.
10
Brookings Papers on Economic Activity, 1:2005
This first steady-state condition implies a negative relation between net debt and the exchange rate. As we showed earlier, in the presence of home bias, a larger U.S. net debt, which transfers wealth to foreign investors, shifts demand away from U.S. assets and thus lowers the exchange rate. The second relation is the current account balance relation (equation 3). Given the equality of interest rates, and given the constant exchange rate and net debt, the relation takes the form 0 = rF + D ( E , z ) . This second relation also implies a negative relation between net debt and the exchange rate. The larger the net debt, the larger the trade surplus required in steady state to finance interest payments on the debt, and thus the lower the exchange rate.7 This raises the question of the stability of the system. The system is (locally saddle point) stable if, as drawn in figure 1, the portfolio balance locus is steeper than the current account balance locus. (Appendix A characterizes the dynamics.) To understand this condition, consider an increase in U.S. net debt. This increase has two effects on the current account deficit, and thus on the change in net debt: it increases interest payments, but it also leads, through the portfolio balance relation, to a lower exchange rate and thus a decrease in the trade deficit. For stability to prevail, the net effect must be that the increase in net debt reduces the current account deficit. This condition appears to be satisfied for plausible parameter values (the next section explores this issue further), and we assume that it is satisfied here. In this case the path of adjustment— the saddle path—is downward sloping, as drawn in figure 1. The Effects of a Shift toward Foreign Goods We can now characterize the effects of shifts in preferences for goods or assets. Figure 2 shows the effect of an unexpected and permanent increase in z. One can think of this increase as coming either from an increase in U.S. activity relative to foreign activity, or from a shift in exports or imports at a given level of activity and a given exchange rate; we defer until 7. If we had allowed r and r* to differ, the relation would have an additional term and take the form 0 = rF + (1 − α)(r − r*)(X − F) + D(E, z). This additional term implies that if, for example, a country pays a lower rate of return on its liabilities than it receives on its assets, it may be able to combine positive net debt with positive net income payments from abroad—the situation in which the United States remains today.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
11
Figure 1. Determination of Exchange Rate and Net Debt in Steady State Exchange rate (E) Portfolio balance
Current account balance
Net debt (F) Source: Authors’ model described in the text.
later a discussion of the sources of the actual shift in z over the past decade in the United States. For any given level of net debt, current account balance requires a lower exchange rate: the current account balance locus shifts down. The new steady state is at point C, associated with a lower exchange rate and a larger net debt. Valuation effects imply that any unexpected depreciation leads to an unexpected decrease in the net debt position. If we denote by ∆E the unexpected change in the exchange rate at the time of the shift, it follows from equation 3 that the change in net debt at the time of the shift is given by (4)
∆F = (1 − α ) (1 + r*) ( X − F )
∆E . E
12
Brookings Papers on Economic Activity, 1:2005
Figure 2. Adjustment of Exchange Rate and Net Debt to an Increase in z Exchange rate (E) Portfolio balance
A
B
Current account balance
C
Net debt (F) Source: Authors’ model described in the text.
The economy jumps initially from point A to point B and then converges over time along the saddle path, from point B to point C. The shift in the trade deficit leads to an initial, unexpected depreciation, followed by further depreciation and net debt accumulation over time until the new steady state is reached. Note that the degree of substitutability between assets does not affect the steady state; more formally, the steady state depends on α(1, s) and α*(1, s), and so changes in αR and α*R that leave α(1, s) and α*(1, s) unchanged do not affect the steady state. In other words, the eventual depreciation is the same no matter how close substitutes U.S. and foreign assets are. But the degree of substitutability plays a central role in the dynamics of adjustment and in the relative roles of the initial unexpected depreciation and the anticipated depreciation thereafter. This is shown in figure 3, which shows the effects of three different values of αR and α*R on the path of adjustment. (The three simulations are based on values for the parameters introduced in
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
13
Figure 3. Responses of the Exchange Rate and Net Debt to a Shift in z Exchange rate Percent changea
–2
Low αR
–4 Medium αR –6 High αR
–8
Net debt Percentage-point change
Low αR
6
Medium αR
4
High αR
2 0
5
10
15
20
25 Years
30
35
40
45
Source: Authors’ calculations. a. All simulations are for a shift in z of 1 percent of U.S. GDP.
the next section. The purpose here is simply to show the qualitative properties of the paths. We return to the quantitative implications later.) The less substitutable U.S. and foreign assets are—that is, the smaller are αR and α*R—the smaller the initial depreciation and the higher the anticipated rate of depreciation thereafter. To understand why, consider the extreme case where the shares do not depend on rates of return: U.S. and foreign investors want to maintain constant shares, no matter what the relative rate of return is. In this case the portfolio balance relation (equation 2) implies that there will be no response of the exchange rate to the
14
Brookings Papers on Economic Activity, 1:2005
unexpected change in z at the time it happens: any movement in the exchange rate would be inconsistent with equilibrium in the market for U.S. assets. Only over time, as the deficit leads to an increase in net debt, will the exchange rate decline. Conversely, the more substitutable U.S. and foreign assets are, the larger will be the initial depreciation, the lower the anticipated rate of depreciation thereafter, and the longer the time taken to reach the new steady state. The limit of perfect substitutability—corresponding to the model discussed at the start—is actually degenerate: the initial depreciation is such as to maintain current account balance, and the economy does not move from there on, never reaching the new steady state (and so the anticipated rate of depreciation is equal to zero). To summarize: In contrast to the case of perfect substitutability between assets we saw earlier, an increase in U.S. demand for foreign goods leads to a limited depreciation initially, a potentially large and long-lasting current account deficit, and a steady depreciation over time. The Effects of a Shift toward U.S. Assets Figure 4 shows the effect of an unexpected and permanent increase in s, that is, an increase in the demand for U.S. assets. Again we defer to later a discussion of the potential factors behind such an increase. By assumption, the increase in s leads to an increase in α(1, s) and a decrease in α*(1, s). At a given level of net debt, portfolio balance requires an increase in the exchange rate. The portfolio balance locus shifts up. The new steady state is at point C, associated with a lower exchange rate and larger net debt. The dynamics are given by the path ABC. The initial adjustment of E and F must again satisfy the condition in equation 4. So the economy jumps from point A to point B and then converges over time from point B to point C. The dollar initially appreciates, triggering an increase in the trade deficit and a deterioration in the net debt position. Over time, net debt continues to increase and the dollar depreciates. In the new equilibrium the exchange rate is necessarily lower than before the shift: this reflects the need for a larger trade surplus to offset the interest payments on the nowlarger U.S. net debt. In the long run the favorable portfolio shift leads to a depreciation. Again the degree of substitutability between assets plays an important role in the adjustment. This is shown in figure 5, which plots the path of
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
15
Figure 4. Adjustment of Exchange Rate and Net Debt to an Increase in s Exchange rate (E)
Portfolio balance
D
B Current account balance
A C
Net debt (F) Source: Authors’ model described in the text.
adjustment for three different values of αR and α*R. The less substitutable are U.S. and foreign assets, the greater the initial appreciation and the higher the anticipated rate of depreciation thereafter. Although the depreciation is eventually the same in all cases (the steady state is invariant to the values of αR and α*R), the effect of portfolio shifts is more muted but longer lasting when the degree of substitutability is high. An Interpretation of the Past Looking at the effects of shifts in preferences for goods and for assets under imperfect asset substitutability suggests three main conclusions: —Shifts in preferences toward foreign goods lead to an initial depreciation, followed by a further anticipated depreciation. Shifts in preferences toward U.S. assets lead to an initial appreciation, followed by an anticipated depreciation.
16
Brookings Papers on Economic Activity, 1:2005
Figure 5. Responses of the Exchange Rates to a Shift in s Exchange rate Percent changea
20
Low αR
15
Medium αR
10
High αR
5
Net debt Percentage-point change
20
Low αR Medium αR
15 High αR
10 5
5
10
15
20
25 Years
30
35
40
45
Source: Authors’ calculations. a. All specifications are for a 5-percentage-point shift in s.
—The empirical evidence suggests that both types of shifts have been at work in the United States in the recent past. The first shift, by itself, would have implied a steady depreciation in line with increased trade deficits, whereas instead an initial appreciation was observed. The second shift can explain why the initial appreciation has been followed by a depreciation. But it attributes the increase in the trade deficit fully to the initial appreciation, whereas the evidence is of a large adverse shift in the trade balance even after controlling for the effects of the exchange rate. (This does not do justice to an alternative, and more conventional, monetary policy explana-
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
17
tion, in which high U.S. interest rates relative to foreign interest rates at the end of the 1990s led to an appreciation, followed since by a depreciation. The observed relative interest rate differentials seem too small, however, to explain the movement in exchange rates.) —Both shifts lead eventually to a steady depreciation, to a lower exchange rate than before the shift. This follows from the simple condition that a larger net debt, no matter what its origin, requires larger interest payments in steady state and thus a larger trade surplus. The lower the degree of substitutability between U.S. and foreign assets, the higher the expected rate of depreciation along the path of adjustment. The United States appears to have indeed entered this depreciation phase.
How Large a Depreciation? A Look at the Numbers The model is simple enough that one can insert some values for the parameters and draw the implications for the future. More generally, the model provides a way of looking at the data, and this is what we do in this section. Parameter Values Consider first what we know about portfolio shares: In 2003 U.S. financial wealth, W, was $34.1 trillion, or about three times U.S. GDP of $11 trillion.8 Non-U.S. world financial wealth is harder to assess. For the euro area financial wealth was about t15.5 trillion in 2003, compared with GDP of t7.5 trillion; Japanese financial wealth was about ¥1 quadrillion in 2004, compared with GDP of ¥500 trillion.9 If one extrapolates from a ratio of financial assets to GDP of about 2 for both Japan and Europe, and GDP for the non-U.S. world of approximately $18 trillion in 2003, a reasonable estimate for W*/E is $36 trillion—roughly the same as for the United States. The net U.S. debt position, F, measured at market value, was $2.7 trillion in 2003, up from approximate balance in the early 1990s.10 By implication, 8. Financial wealth data are from the Flow of Funds Accounts of the United States 1995–2003, table L100, Board of Governors of the Federal Reserve System, December 2004. 9. The figure for Europe is from ECB Bulletin, February 2005, table 3.1, and that for Japan from Bank of Japan, Flow of Funds (www.boj.or.jp/en/stat/stat_f.htm). 10. The source for the numbers in this and the next paragraph is Bureau of Economic Analysis, International Transactions, table 2, International Investment Position of the United States at Year End, 1976–2003, June 2004.
18
Brookings Papers on Economic Activity, 1:2005
U.S. assets, X, were W + F = $36.8 trillion ($34.1 trillion + $2.7 trillion), and foreign assets, X*/E, were W*/E − F = $33.3 trillion ($36.0 trillion − $2.7 trillion). Put another way, the ratio of U.S. net debt to U.S. assets, F/X, was 7.3 percent ($2.7 trillion ÷ $36.8 trillion); the ratio of U.S. net debt to U.S. GDP was 24.5 percent ($2.7 trillion ÷ $11.0 trillion). In 2003 gross U.S. holdings of foreign assets, at market value, were $7.9 trillion. Together with the value for W, this implies that the share of U.S. wealth in U.S. assets, α, was 1 − (7.9/34.1), or 0.77. Gross foreign holdings of U.S. assets, at market value, were $10.6 trillion. Together with the value of W*/E, this implies that the share of foreign wealth in foreign assets, α*, was equal to 1 − (10.6/36.0), or 0.71. To get a sense of the implications of these values for α and α*, note from equation 2 that a transfer of one dollar from U.S. wealth to foreign wealth implies a decrease in the demand for U.S. assets of (α + α* − 1) dollars, or 48 cents.11 To summarize: W W*/E X X*/E F α α*
= = = = = = =
$34.1 trillion $36.0 trillion $36.8 trillion $33.3 trillion $2.7 trillion 0.77 0.71.
We would like to know not only the values of the shares, but also their dependence on the relative rate of return—the values of the derivatives αR and α*R. Little is known about these values. Gourinchas and Rey provide indirect evidence of the relevance of imperfect substitutability by showing that a combination of the trade deficit and the net debt position helps predict a depreciation (we return to their results later);12 this would not be the case under perfect substitutability. However, it is difficult to back out estimates of αR and α*R from their results. Thus, when needed below, we derive results under alternative assumptions about these derivatives.
11. Note that this conclusion depends on the assumption we make in our model that marginal and average shares are equal. This may not be the case. 12. Gourinchas and Rey (2005).
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
19
The next important parameter in our model is θ, the effect of the exchange rate on the trade balance. The natural starting point here is the Marshall-Lerner relation: dD dE = [ ηimp − ηexp − 1] , Exports E where ηimp and ηexp are, respectively, the elasticities of imports and exports with respect to the real exchange rate. Estimates of the ηs based on estimated U.S. import and export equations range quite widely.13 In some cases the estimates imply that the MarshallLerner condition (the condition that the term in brackets be positive, so that a depreciation improves the trade balance) is barely satisfied. Estimates used in macroeconometric models imply a value for the term in brackets between 0.5 and 0.9. Put another way, together with the assumption that the ratio of U.S. exports to U.S. GDP is 10 percent, they imply that a reduction of the ratio of the trade deficit to GDP by 1 percentage point requires a depreciation of somewhere between 11 and 20 percent. One may believe, however, that measurement error, complex lag structures, and misspecification all bias these estimates downward. An alternative approach is to derive the elasticities from plausible specifications of utility and the pass-through behavior of firms. Using such an approach in a model with nontradable goods, domestic tradable goods, and foreign tradable goods, Obstfeld and Rogoff find that a 1-percentage-point decrease in the ratio of the trade deficit to GDP requires a decrease in the real exchange rate of somewhere between 7 and 10 percent—a smaller depreciation than implied by the macroeconometric models.14 Which value to use is obviously crucial in assessing the scope of the required exchange rate adjustment. We choose an estimate for the term in brackets in the Marshall-Lerner equation of 0.7—toward the high range of empirical estimates but lower than the Obstfeld-Rogoff elasticities. This estimate, together with an exports-to-GDP ratio of 10 percent, implies that a reduction in the ratio of the trade deficit to GDP of 1 percentage point requires a depreciation of 15 percent. 13. See the survey by Chinn (2004). 14. Obstfeld and Rogoff (2004).
20
Brookings Papers on Economic Activity, 1:2005
A Simple Exercise We have argued that a depreciation of the dollar has two effects: a conventional one through the trade balance, and another through valuation effects. To get a sense of their relative magnitudes, consider the effects of an unexpected depreciation in our model. More specifically, consider the effects of an unexpected 15 percent decrease in E+1 relative to E on net debt, F+1, in equation 3. The first effect of the depreciation is to improve the trade balance. Given our earlier discussion and assumptions, such a depreciation reduces the trade deficit by 1 percent of GDP (which is why we chose to look at a depreciation of 15 percent). The second effect is to increase the dollar value of U.S. holdings of foreign assets (and to reduce the foreign currency value of foreign holdings of U.S. assets) and thus reduce the U.S. net debt position. From equation 3 (with both sides divided by U.S. output, Y, to make the interpretation of the magnitudes easier), this effect is given by X − F dE dF = − (1 − α ) (1 + r *) . Y E Y From the earlier discussion, (1 − α) is equal to 0.23, and (X − F )/Y to 3. Assume that r* is equal to 4 percent. The effect of a 15 percent depreciation is then to reduce the ratio of net debt to GDP by 10 percentage points (0.23 × 1.04 × 3 × 0.15). This implies that, after the unexpected depreciation, interest payments are lower by 4 percent times 10 percent, or 0.4 percent of GDP.15 Putting things together, a 15 percent depreciation improves the current account balance by 1.4 percent of GDP, with roughly one-third of the improvement due to valuation effects.16 It is tempting here to ask how large an unexpected depreciation would have to occur to lead to a sustainable U.S. current account deficit today?17 15. This computation assumes that all foreign assets held by U.S. investors are denominated in foreign currency. In reality, some foreign bonds held by U.S. investors are denominated in dollars. This reduces the valuation effects. 16. Lane and Milesi-Ferretti (2004) give a similar computation for a number of countries, although not for the United States. 17. This is also the question taken up by Obstfeld and Rogoff in this volume. Their focus, relative to ours, is on the required adjustments in both the terms of trade and the real exchange rate, starting from a micro-founded model with nontraded goods, exportables, and importables.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
21
Take the actual current account deficit of about 6 percent. What the “sustainable” current account deficit is depends on the ratio of net debt to GDP that the United States is willing to sustain, and on the growth rate of GDP: if g is the growth rate of U.S. GDP, the United States can sustain a current account deficit of g(F/Y). Assuming, for example, a nominal GDP growth rate of 3 percent and a ratio of net debt to GDP of 25 percent (the ratio prevailing today, but one that has no particular claim to being the right one for this computation) implies that the United States can run a current account deficit of 0.75 percent while maintaining a constant ratio of net debt to GDP. In this case the depreciation required to shift from the actual to the sustainable current account deficit would be roughly 56 percent (6 percent − 0.75 percent) × (15 percent ÷ 1.4 percent). This is a large number, and despite the uncertainty attached to the underlying values of many of the parameters, it is a useful number to keep in mind. But one should be clear about the limitations of the computation: —The United States surely does not need to shift to sustainable current account balance right away. The rest of the world is still willing to lend to it, if perhaps not at the current rate. The longer the United States waits, however, the higher the ratio of net debt to GDP becomes, and thus the larger the eventual required depreciation. In this sense our computation gives a lower bound on the eventual depreciation. —The computation is based on the assumption that, at the current exchange rate, the trade deficit will remain as large as it is today. If, for example, we believed that part of the current trade deficit reflects the combined effect of recent depreciations and J-curve effects, the computation above would clearly overestimate the required depreciation. The rest of this section deals with these issues. First, by returning to dynamics, we try to get a sense of the eventual depreciation and of the rate at which it may be achieved. Second, we look at the evidence on the origins of the shifts in z and s. Returning to Dynamics How large is the effect of a given shift in z (or in s) on the accumulation of net debt and on the eventual exchange rate? And how long does it take to get there? The natural way to answer these questions is to simulate our model using the values of the parameters we derived earlier. This is indeed what the simulations presented in figures 3 and 5 did; we now look more closely at their quantitative implications.
22
Brookings Papers on Economic Activity, 1:2005
Both sets of simulations are based on the values of the parameters given above. Recognizing the presence of output growth (which we did not allow for in the model), and rewriting the equation for net debt as an equation for the ratio of net debt to output, we take the term in front of F in the current account balance relation (equation 3) to stand for the interest rate minus the growth rate. We choose an interest rate of 4 percent and a nominal growth rate of 3 percent, so that their difference is 1 percent. We write the portfolio shares as α ( R e, s ) = a + bR e + s, α*( R e, s ) = a* − bR e − s. The simulations show the results for three values (10, 1.0, and 0.1) of the parameter b. A value of 1 implies that an increase in the expected relative return on U.S. assets of 100 basis points increases the desired share in U.S. assets by 1 percentage point. Figure 3 showed the effect of an increase in z of 1 percent of U.S. GDP. Figure 5 showed the effect of an increase in s of 5 percentage points, leading to an increase in α and a decrease in α* of 5 percentage points at a given relative rate of return. Time is measured in years. Figure 3 leads to two main conclusions. First, the effect of a permanent increase in z by 1 percent is to eventually increase the ratio of net debt to GDP by 17 percentage points and require an eventual depreciation of 12.5 percent. (Recall that the long-run effects are independent of the degree of substitutability between assets—that is, independent of the value of b.) Second, it takes a long time to get there: the figure is truncated at fifty years, by which time the adjustment is still not complete. Figure 5 leads to similar conclusions. The initial effect of the increase in s is an appreciation of the dollar: by 23 percent if b = 0.1, and by 12 percent if b = 10. The long-run effect of the increase in s is an increase in the ratio of U.S. net debt to GDP of 35 percentage points and a depreciation of 15 percent. But even after fifty years the adjustment is far from complete, and the exchange rate is still above its initial level. What should one conclude from these exercises? We conclude that, under the following assumptions—that there are no anticipated changes in z or in α or α*, that investors have been and will be rational (the simulations are carried out under rational expectations), and that there are no surprises—the dollar will depreciate by a large amount, but at a steady and slow rate. There are good reasons to question each of these assumptions, and this we do next.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
23
A Closer Look at the Trade Deficit To think about the likely path of z, and thus of the path of the trade deficit at a given exchange rate, it is useful to write the trade deficit as the difference between the value of imports in terms of domestic goods, and exports: D ( E , z ) ≡ E imp ( E , Z , z ) − exp ( E , Z *, z*) We have decomposed z into two components: total U.S. spending, Z, and ˜z, which represents shifts in the relative U.S. demand for U.S. versus foreign goods, at a given level of spending and a given exchange rate. Similarly, z* is decomposed into Z* and ˜z *, the latter measuring shifts in the relative foreign demand for U.S. versus foreign goods. Most of the large current account fluctuations in developed countries of the last few decades have come from relative fluctuations in activity, that is, in Z relative to Z*.18 It has indeed been argued that the deterioration of the U.S. trade balance has come mostly from faster growth in the United States than in its trade partners, leading imports by the United States to increase faster than U.S. exports to the rest of the world. This appears, however, to have played a limited role. Europe and Japan indeed have had slower growth than the United States (U.S. output grew a cumulative 45 percent from 1990 to 2004, compared with 29 percent for the euro area and 25 percent for Japan), but these countries account for only 35 percent of U.S. exports, and meanwhile other U.S. trade partners have grown as fast as or faster than the United States. Indeed, a study by the International Monetary Fund finds nearly identical output growth rates for the United States and its export-weighted partners since the early 1990s.19 Some have argued that the deterioration in the trade balance reflects instead a combination of rapid growth both in the United States and abroad and a U.S. import elasticity with respect to domestic spending that is higher (1.5 or above) than the elasticity of U.S. exports with respect to foreign spending. In this view rapid U.S. growth has led to a more than proportional increase in imports and an increasing trade deficit. The debate about 18. For a review of current account deficits and adjustments for twenty-one countries over the last thirty years, and references to the literature, see Debelle and Galati (2005). 19. International Monetary Fund, Article IV United States Consultation—Staff Report, 2004. As the case of the United States indeed reminds us, output is not the same as domestic spending, but the differences in growth rates between the two over a decade are small.
24
Brookings Papers on Economic Activity, 1:2005
the correct value of the U.S. import elasticity is an old one, dating back to the estimates by Hendrik Houthakker and Stephen Magee; we tend to side with the recent conclusion by Jaime Marquez that the elasticity is close to 1.20 For our purposes, however, this discussion is not relevant. Whether the growth in the U.S. trade deficit is the result of a high import elasticity or of shifts in the ˜zs, there are no obvious reasons to expect either the shift to reverse or growth in the United States to drastically decrease in the future. One way of assessing the relative roles of shifts in spending, the exchange rate, and other factors is to look at the performance of import and export equations in detailed macroeconometric models. The numbers obtained using the macroeconometric model of Global Insight (formerly the Data Resources, Inc., or DRI, model) are as follows:21 The U.S. trade deficit in goods increased from $221 billion in the first quarter of 1998 (annualized) to $674 billion in the third quarter of 2004. Of this $453 billion increase, $126 billion was due to the increase in the value of oil imports, leaving $327 billion to be explained. When the export and import equations of the model are used, activity variables and exchange rates explain $202 billion, or about 60 percent of the increase. Unexplained time trends and residuals account for the remaining 40 percent, a substantial amount.22 Looking to the future, whether growth rate differentials, HouthakkerMagee effects, or unexplained shifts are behind the increase in the trade deficit is probably not essential. The slower growth in Europe and Japan reflects in large part structural factors, and neither Europe nor Japan is likely to make up much of the cumulative growth difference since 1995 over the next few years. One can still ask how much a given increase in growth in Europe and Japan would reduce the U.S. trade deficit. A simple computation is as follows. Suppose that Europe and Japan made up the roughly 20-percentage-point growth gap they have accumulated since 1990 vis-à-vis the United States—an unlikely scenario in the near future—so that U.S. exports to Europe and Japan increased by 20 percent. Given that U.S.
20. Houthakker and Magee (1969); Marquez (2000). 21. We thank Nigel Gault of Global Insight for communicating these results to us. 22. The model has a set of export and import equations disaggregated by product type. Most of the elasticities of the different components with respect to domestic or foreign spending are close to 1, indicating that Houthakker-Magee effects play a limited role (except for imports and exports of consumption goods, where the elasticity of imports with respect to consumption is 1.5 for the United States, but the elasticity of U.S. exports with respect to foreign GDP is an even higher 2.0).
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
25
exports to these countries are currently about $350 billion, the improvement would be 0.7 percent of U.S. GDP—not negligible, but not a major increase either. One other factor, however, may hold more hope for a reduction in the trade deficit, namely, the working out of the J-curve. Nominal depreciations increase import prices, but these decrease the volume of imports only with a lag. Thus, for a while, a depreciation can increase the value of imports and worsen the trade balance, before improving it later. One reason to think this may be important is the “dance of the dollar” and the movements of the dollar and the current account during the 1980s. From the first quarter of 1979 to the first quarter of 1985, the real exchange rate of the United States (measured by the trade-weighted major currencies index constructed by the Federal Reserve Board) increased by 41 percent (log percentage change). This appreciation was then followed by a sharp depreciation, with the dollar falling by 44 percent from the first quarter of 1985 to the first quarter of 1988. The appreciation was accompanied by a steady deterioration in the current account deficit, from rough balance in the early 1980s to a deficit of about 2.5 percent of GDP when the dollar reached its peak in early 1985. The current account continued to worsen, however, for more than two years, reaching a peak of 3.4 percent of GDP in 1987. The divergent paths of the exchange rate and the current account from 1985 to 1987 led a number of economists to explore the idea of hysteresis in trade:23 the notion that, once appreciation has led to a loss of market share, an equal depreciation may not be sufficient to reestablish trade balance. Just as the idea was taking hold, however, the current account position rapidly improved, and trade was roughly in balance by the end of the decade.24 The parallels with more recent developments are clear from figure 6, which plots the dollar exchange rate and the U.S. current account during both episodes, aligned in the figure so that the dollar peak of 1985:1 coincides with the dollar peak of 2001:2. The figure suggests two conclusions: —If the earlier episode is a reliable guide, and the lags today are similar to those that prevailed in the 1980s, the current account deficit may start 23. In particular, Baldwin and Krugman (1987). 24. These issues were discussed at length in the Brookings Papers at the time. Besides Baldwin and Krugman (1987), see, for example, Cooper (1986), Dornbusch (1987), and Sachs (1988), with post mortems by Lawrence (1990) and Krugman (1991). Another much-discussed issue, to which we return later, was the relative roles of fiscal deficit reduction and exchange rate adjustment in closing the deficit.
26
Brookings Papers on Economic Activity, 1:2005
Figure 6. Current Account Deficit and Effective Real Exchange Rate, 1978–93 and 1995–2004 Current account deficit Percent of GDP
1995–2004
5 4 3 2
1978–93 1 0 –1 Exchange ratea Index (March 1973 = 100) 1978–93 120 110 100
1995–2004
90 80
1980 1997
1982 1999
1984 2001
1986 2003 Year
1988 2005
1990
Source: Bureau of Economic Analysis, Table 1, U.S. International Transactions; Federal Reserve data. a. Price-adjusted Major Currencies index.
1992
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
27
to turn around soon. Today’s deficit, however, is much larger than the earlier deficit was at its peak in 1987 (approaching 6 percent of GDP versus 3.5 percent), and the depreciation so far has been more limited (23 percent from 2001:2 to 2004:4, compared with 33 percent over the equivalent period from 1985:1 to 1988:3).25 —Hence one can surely not conclude that the depreciation so far is enough to restore the current account deficit to sustainable levels. But it may be that, in our computation, the appropriate place to start is from a J-curve-adjusted ratio of the current account deficit to GDP of 4 or 5 percent instead of 6 percent.26 If we choose 4 percent—a very optimistic assumption—the remaining required depreciation is 34 percent (4 percent − 0.75 percent) × (15 percent ÷ 1.4 percent).27 A Closer Look at Portfolio Shares One striking aspect of the simulations presented above is how slow the depreciation is along the adjustment path. This is in contrast with some predictions of much more abrupt falls in the dollar in the near future.28 This raises two issues: Can the anticipated depreciation be greater than in these simulations? And are there possible surprises under which the depreciation might be much faster (or slower), and, if so, what are they? To answer the first question, we go back to the model. We noted earlier that the lower the degree of substitutability between assets, the higher the anticipated rate of depreciation. So, by assuming zero substitutability—that is, constant asset shares except for changes coming from shifts in s—we can
25. On the other hand, the gross positions, and thus the scope for valuation effects from dollar depreciation, are much larger now than they were then. In 1985 gross U.S. holdings of foreign assets were $1.5 trillion, compared with $8 trillion today. 26. Forecasts by Macroeconomic Advisers, LLC, are for an improvement in the trade balance of $75 billion, or less than 1 percent of GDP, over the next two years. (The forecast is based on a depreciation of the dollar of 4 percent over that period.) The residuals of the import price equations of the model, however, suggest an unusually low pass-through of the dollar decline to import prices over the recent past, and the forecast assumes that the low pass-through continues. If the pass-through were to return to its historical average, the improvement in the trade balance would be larger. 27. This number is surprisingly close to the 33 percent obtained by Obstfeld and Rogoff in this volume. 28. For example, by Roubini and Setser (2005).
28
Brookings Papers on Economic Activity, 1:2005
derive an upper bound on the anticipated rate of depreciation. Differentiating equation 2 gives
(α + α* −1) X dF ( X − F ) dα − ( X * E + F ) dα* dE . + = − (1 − α*) X * E X E (1 − α*) X * E In the absence of anticipated shifts in shares (so that the second term equals zero), the anticipated rate of depreciation depends on the change in the ratio of U.S. net debt to U.S. assets: the faster the increase in net debt, the faster the decrease in the relative demand for U.S. assets, and therefore the higher the rate of depreciation needed to maintain portfolio balance. Using the parameters we constructed earlier, this equation implies dE F = −1.8d + ( 3.5 dα − 3.7 dα*) . E X Suppose shares remain constant. If we take the annual increase in the ratio of net debt to U.S. GDP to be 5 percent and the ratio of U.S. GDP to U.S. assets to be one-third, this gives an anticipated annual rate of depreciation of 3 percent a year (1.8 × 0.05 ÷ 3).29 If, however, shares of U.S. assets in the portfolios of either domestic or foreign investors are expected to decline, the anticipated depreciation can clearly be much larger. If, for example, we anticipate that the share of U.S. assets in foreign portfolios will decline by 2 percent over the coming year, the anticipated depreciation is 8.7 percent (2.7 percent as calculated above, plus 3.0 times 2 percent). This is obviously an upper bound on the size of the anticipated depreciation, derived by assuming that private investors are willing to keep a constant share of their wealth in U.S. assets despite a high negative expected rate of return between now and then. (If, instead, anticipating this high negative rate of return, private investors decide to decrease their share of dollar assets, then some of the depreciation will take place now, rather than when the shift in portfolio composition occurs, and so the anticipated depreciation will be smaller.) Still, it implies that, under imperfect substitutability, and under the assumption that desired shares in U.S.
29. Although comparison is difficult, this rate appears lower than that implied by the estimates of Gourinchas and Rey (2005). Their results imply that a combination of net debt and trade deficits 2 standard deviations from the mean—a situation that would appear to characterize well the United States today—implies an anticipated annual rate of depreciation of about 5 percent over the following two years.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
29
assets will decrease, it is logically acceptable to predict a substantial depreciation of the dollar in the near future. Are there good reasons to expect these desired shares to decrease in the near future? This is the subject of a contentious debate. Some argue that the United States can continue to finance its current account deficits at today’s level for a long time to come at the same exchange rate. They argue that the poor development of financial markets in Asia and elsewhere, together with the need for Asian countries to accumulate international collateral, implies a steadily increasing relative demand for U.S. assets. They point to the latent demand for U.S. assets on the part of Chinese private investors, currently limited by capital controls. In short, they argue that foreign investors will be willing to further increase their holdings of U.S. assets for many years to come.30 Following this argument, we can ask what increase in shares—say, what increase in (1 − α*), the share of U.S. assets in foreign portfolios— would be needed to absorb the current increase in net debt at a given exchange rate. From the relation derived above, setting dE/E and dα equal to zero gives dα * = −
(α* + α − 1) X X* E + F
(Y X ) d F . Y
For the parameters we have constructed, a change of 5 percentage points in F/Y requires an increase in the share of U.S. assets in foreign portfolios of about 0.8 percentage point a year (0.47 × 5 percent ÷ 3).31 We find more plausible the argument that the relative demand for U.S. assets may actually decrease rather than increase in the future. This argument is based, in particular, on the fact that much of the recent accumulation of U.S. assets has taken the form of accumulation of reserves by the Japanese and Chinese central banks. Many worry that this will not last,
30. See, for example, Dooley, Folkerts-Landau, and Garber (2004) and Caballero, Farhi, and Hammour (2004). 31. A related argument is that, to the extent that the rest of the world is growing faster than the United States, an increase in the ratio of net debt to GDP in the United States is consistent with a constant share of U.S. assets in foreign portfolios. This argument falls quantitatively short: although some Asian countries are growing rapidly, their weight and their financial wealth are still far too small to absorb the U.S. current account deficit while maintaining constant shares of U.S. assets in their portfolios.
30
Brookings Papers on Economic Activity, 1:2005
that the pegging of the renminbi will come to an end, or that both central banks will want to change the composition of their reserves away from U.S. assets, leading to further depreciation of the dollar. Our model provides a simple way of discussing the issue and thinking about the numbers. Consider pegging first: the foreign central bank buys or sells dollar assets so as to keep E = E¯.32 Let B denote the reserves (U.S. assets) held by the foreign central bank, so that X = B + α (1) ( X − F ) + (1 − α* (1)) ( X * E + F ) . Figure 7 illustrates the resulting dynamics. Suppose that, in the absence of pegging, the steady state is given by point A and that the foreign central bank pegs the exchange rate at E¯. At that level the U.S. current account is in deficit, and so F increases over time. Wealth gets steadily transferred to the foreign country, and so the private demand for U.S. assets steadily decreases. To keep E unchanged, B must increase further over time. Pegging by the foreign central bank is thus equivalent to a continuous outward shift in the portfolio balance schedule: in effect, the foreign central bank is keeping world demand for U.S. assets unchanged by offsetting the fall in private demand. Pegging leads to a steady increase in U.S. net debt and a steady increase in the foreign central bank’s reserves, offsetting the steady decrease in private demand for U.S. assets (represented by the path DC in figure 7). What happens when the foreign central bank unexpectedly stops pegging? From point C just before the peg is abandoned, the economy jumps to point G (recall that valuation effects lead to a decrease in net debt, and therefore a capital loss for the foreign central bank, when there is an unexpected depreciation) and then adjusts along the saddle-point path GA′. The longer the peg lasts, the larger the initial and the eventual depreciation. In other words, an early end to the Chinese peg would obviously lead to a depreciation of the dollar (an appreciation of the renminbi). But the sooner it takes place, the smaller the required depreciation, both initially and in the long run. Put another way, the longer the Chinese wait to abandon the peg, the larger the eventual appreciation of the renminbi. The conclusions are very similar with respect to changes in the composition of reserves. We can think of such changes as changes in portfolio 32. Our two-country model has only one foreign central bank, and so we cannot discuss what happens if one foreign bank pegs its currency and the others do not. The issue is, however, relevant in thinking about the paths of the dollar-euro and the dollar-yen exchange rates. We discuss this further in the next section.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
31
Figure 7. Adjustment of Exchange Rate and Net Debt to Abandonment of Foreign Peg Exchange rate (E) Portfolio balance
D
C
E Current account balance A
G
A
Net debt (F) Source: Authors’ model described in the text.
preferences, this time not by private investors but by central banks, and so we can apply our earlier analysis directly. A shift away from U.S. assets will lead to an initial depreciation, leading in turn to a lower current account deficit, a smaller increase in net debt, and thus to a smaller depreciation in the long run. How large might these shifts be? Chinese reserves currently equal $610 billion, and Japanese reserves are $840 billion. Assuming that these reserves are now held mostly in dollars, if the People’s Bank of China and the Bank of Japan reduced their dollar holdings to half of their portfolio, this would represent a decrease in the share of U.S. assets in total foreign (private and central bank) portfolios, (1 − α*), from 30 percent to 28 percent. The computations we presented earlier suggest that this would be a substantial shift, leading to a decrease in the dollar exchange rate possibly as large as 8.7 percent.
32
Brookings Papers on Economic Activity, 1:2005
To summarize: Avoiding a depreciation of the dollar would require a steady and substantial increase in shares of U.S. assets in U.S. or foreign portfolios at a given exchange rate. This seems unlikely to hold for very long. A more likely scenario is the opposite, a decrease in shares, due in particular to diversification of reserves by central banks. If and when this happens, the dollar will depreciate. Note, however, that the larger the adverse shift, the larger the initial depreciation but the smaller the accumulation of debt thereafter, and therefore the smaller the eventual depreciation. “Bad news” on the dollar now may well be good news in the long run (and vice versa). The Path of Interest Rates Our model takes interest rates as given, and the discussion thus far has taken them as constant.33 Yield curves in the United States, Europe, and Japan indeed indicate little expected change in interest rates over the near and the medium term. However, it is easy to think of scenarios where changes in interest rates play an important role, and this leads us to discuss the role of budget deficit reduction in the adjustment process. First, however, we briefly show the effects of an increase in the U.S. interest rate in our model. Figure 8 shows the effects of an unexpected permanent increase in r over r*. (In contrast to the case of perfect substitutability, it is possible for the two interest rates to differ even in the steady state.) The portfolio balance locus shifts upward: At a given level of net debt, U.S. assets are more attractive, and so the exchange rate increases. The current account balance locus shifts down: the higher interest rate implies larger payments on foreign holdings of U.S. assets and thus requires a larger trade surplus, and in turn a lower exchange rate. The adjustment path is given by ABC. In response to the increase in r, the economy jumps from point A to point B and then moves over time from point B to point C. As drawn, there is an appreciation initially, but, in general, the initial effect on the exchange rate is ambiguous. If gross liabili33. Remember that, when financial assets are imperfect substitutes, the interest rate differential no longer directly reflects expected exchange rate changes. It is thus perfectly rational for the level of long-term interest rates in the United States and in other countries to be very similar, even as the market anticipates a depreciation of the dollar. Therefore, if we consider that financial assets denominated in different currencies can be imperfect substitutes, there is no “interest rate puzzle,” contrary to what is sometimes claimed in the financial press.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
33
Figure 8. Adjustment of Exchange Rate and Net Debt to an Increase in the Domestic Interest Rate Exchange rate (E) Portfolio balance
B A
Current account balance C
Net debt (F) Source: Authors’ model described in the text.
ties are large, for example, the effect of higher interest payments on the current account balance may dominate the more conventional “overshooting” effects of increased attractiveness and lead to an initial depreciation rather than an appreciation. In either case the steady-state effect is greater net debt accumulation, and thus a larger depreciation than if r had not increased. Thus, under the assumption that an increase in interest rates leads initially to an appreciation, an increase in U.S. interest rates beyond what is already implicit in the yield curve would delay the depreciation of the dollar, at the cost of greater net debt accumulation and a larger eventual depreciation. Interest rate changes, however, do not take place in a vacuum. It is more interesting to think about what may happen to interest rates as the dollar depreciates, either slowly along the saddle path or more sharply, in response,
34
Brookings Papers on Economic Activity, 1:2005
for example, to adverse portfolio shifts. As the dollar depreciates, relative demand shifts toward U.S. goods, reducing the trade deficit but also increasing total demand for U.S. goods. Suppose also that output is initially at its natural level (the level associated with the natural rate of unemployment), which appears to be a good description of the United States today. Three outcomes are possible: —Interest rates and fiscal policy remain unchanged. The increase in demand leads to an increase in output but also an increase in imports, which partly offsets the effect of the depreciation on the trade balance. (In terms of our model, it leads to an increase in domestic spending, Z, and thus to a shift in z.) —Interest rates remain unchanged, but fiscal policy is adjusted to offset the increase in demand and leave output at its natural level; in other words, the budget deficit is reduced so as to maintain internal balance. —Fiscal policy remains unchanged, but the Federal Reserve increases interest rates so as to maintain output at its natural level. In this case, higher U.S. interest rates limit the extent of the depreciation and mitigate the current account deficit reduction. In doing so, however, they lead to larger net debt accumulation and to a larger eventual depreciation. In short, an orderly reduction of the current account deficit—that is, one that occurs while maintaining internal balance—requires both a decrease in the exchange rate and a reduction in the budget deficit.34 The two are not substitutes: the depreciation is needed to achieve current account balance, and budget deficit reduction is needed to maintain internal balance at the natural level of output.35 (The frequently heard statement that deficit reduction would reduce the need for dollar depreciation leaves us puzzled.) If the decrease in the budget deficit is not accompanied by a depreciation, the result is likely to be lower demand and a recession. Although the recession 34. Many of the discussions at Brookings in the late 1980s were about the relative roles of budget deficit reduction and exchange rate adjustment. For example, Sachs (1988) argued that “the budget deficit is the most important source of the trade deficit. Reducing the budget deficit would help reduce the trade deficit . . . [while] an attempt to reduce the trade deficit by a depreciating exchange rate induced by easier monetary policy would produce inflation with little benefit on the current account,” a view consistent with the third scenario above. Cooper (1986), in a discussion of the policy package best suited to eliminate the U.S. imbalances, stated, “The drop in the dollar is an essential part of the policy package. The dollar’s decline will help offset the fiscal contraction through expansion of net exports and help maintain overall U.S. economic activity at a satisfactory level,” a view consistent with the second scenario. 35. Obstfeld and Rogoff (2004) emphasize a similar point.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
35
would reduce the current account deficit, this is hardly a desirable outcome. If the depreciation is not accompanied by a reduction in the budget deficit, one of two things can happen: demand will increase, and with it the risk that the economy will overheat, or, more likely, interest rates will increase so as to maintain internal balance. This increase would either limit or delay the depreciation of the dollar, but, as we have made clear, this would be a mixed blessing. Such a delay implies less depreciation in the short run but more net debt accumulation and more depreciation in the long run. The Euro, the Yen, and the Renminbi The depreciation of the dollar since the peak of 2002 has been very unevenly distributed: as of April 2005 the dollar had fallen 45 percent against the euro, 25 percent against the yen, and not at all against the renminbi. In this section we return to the questions asked in the introduction: if substantially more depreciation is indeed to come, against which currencies will the dollar fall? If China abandons its peg, or if Asian central banks diversify their reserves, how will the euro and the yen be affected? The basic answer is simple. Along the adjustment path, what matters— because of home bias in asset preferences—is the reallocation of wealth across countries, and thus the bilateral current account balances of the United States with its partners. Wealth transfers modify countries’ relative demands for assets, thus requiring corresponding exchange rate movements. Other things equal, countries with larger trade surpluses with the United States will see a larger appreciation of their currency. Other things may not be equal, however. Depending on portfolio preferences, a transfer of wealth from the United States to Japan, for example, may change the relative demand for euro assets and thus the euro exchange rate. In that context one can think of central banks as investors with different asset preferences. For example, a central bank that holds most of its reserves in dollars can be thought of as an investor with strong dollar preferences. Any increase in its reserves is likely to lead to an increase in the relative demand for dollar assets and thus an appreciation of the dollar. Any diversification of its reserves is likely to lead to a depreciation of the dollar. It is beyond the scope of this paper to construct and simulate a realistic multicountry portfolio model. But we can make some progress in thinking about mechanisms and magnitudes. The first step is to extend our model to allow for more countries.
36
Brookings Papers on Economic Activity, 1:2005
Extending the Portfolio Model to Four Regions In 2004 the U.S. trade deficit in goods (the only component of the current account for which a decomposition of the deficit by country is available) was $665 billion. Of this, $162 billion was with China, $77 billion with Japan, $85 billion with the euro area, and the remainder, $341 billion, with the rest of the world. We ignore the rest of the world here and think of the world as composed of four countries or regions: the United States, Europe, Japan, and China (indexed 1 through 4, respectively). We shall therefore think of China as accounting for roughly half the U.S. current account deficit, and Europe and Japan as accounting each for roughly one-fourth. We extend our portfolio model as follows. We assume that the share of asset j in the portfolio of country i is given by α ij(⋅) = aij + ∑ k β ijk R e , k
where Rek is the expected gross real rate of return, in dollars, from holding assets of country k (so that Rek denotes a rate of return, not a relative rate of return as in our two-country model). We assume further that βijk = βjk, so that the effect of the return on asset k on demand for asset j is the same for all investors, independent of the country of origin. This implies that differences in portfolio preferences across countries show up only as different constant terms, and derivatives with respect to rates of return are the same across countries. The following restrictions apply: From the budget constraint (the condition that the shares sum to 1, for any set of expected rates of return), it follows that Σj aij = 1 for all i, and Σj βjk = 0 for all k. The home bias assumption takes the form Σi aii > 1. The demand functions are assumed to be homogeneous of degree zero in expected gross rates of return, so that Σk βjk = 0 for all j. Domestic interest rates, in domestic currency, are assumed to be constant and all equal to r. Exchange rates, Ek, are defined as the price of U.S. goods in terms of foreign goods (so that E1 = 1, and an increase in E2, for example, indicates an appreciation of the dollar against the euro—or, equivalently, a depreciation of the euro against the dollar). It follows that the expected gross real rate of return, in dollars, from holding assets of country k is given by Rek = (1 + r)Ek/Ek+1. In steady state Rek = (1 + r), so that
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
37
Σk βjkRek = 0, and we can concentrate on the aij elements. The portfolio balance conditions, absent central bank intervention, are given by Xj X = ∑ i aij i − Fi , Ej Ei where Fi denotes the net foreign debt position of country i, so that Σi Fi = 0. So far we have treated all four countries symmetrically. China, however, is special in two respects: it enforces strict capital controls, and it pegs the renminbi to the dollar. We capture these two features as follows: —We formalize capital controls as the assumption that a4i = ai4 = 0 for all i ≠ 4; that is, capital controls prevent Chinese residents from investing in foreign assets but also prevent investors outside China from acquiring Chinese assets.36 —We assume that, to peg the renminbi-dollar exchange rate (E4 = 1), the People’s Bank of China passively acquires all dollars flowing into China: the wealth transfer from the United States to the euro area and Japan is thus the U.S. current account minus the fraction that is financed by the Chinese central bank: dF2 + dF3 = −dF1 − dF4. Some Simple Computations Consider now an increase in U.S. net debt equal to dF1. Assume that a share γ of the U.S. net debt is held by China. Assume that a fraction x of the remaining portion is held by the euro area and a fraction (1 − x) by Japan, so that the changes in net debt are given by dF2 = − x (1 − γ ) dF1 , dF3 = − (1 − x ) (1 − γ ) dF1 , dF4 = − γ dF1 . Assume further that China imposes capital controls and pegs the renminbi, that the other three economies are all the same size, and that the matrix of aij elements is symmetric in the following way: aii = a and aij = c = (1 − a)/ 2 < a for i ≠ j.37 In other words, investors want to put more than one-third 36. This ignores inflows of foreign direct investment into China, but since we are considering the financing of the U.S. current account deficit, this assumption is inconsequential for our analysis. 37. The assumption of countries of equal size allows us to specify the matrix in a simple and transparent way. Allowing countries to differ in size, as they obviously do, would lead to a more complex, size-adjusted matrix; but the results would be unaffected.
38
Brookings Papers on Economic Activity, 1:2005
of their portfolio into domestic assets (the conditions above imply a > 1⁄3) and allocate the rest of their portfolio equally among foreign assets. Under these assumptions, dE4 = 0 (because of pegging), and dE2 and dE3 are given by
( a − c ) (1 − γ )[ x (1 − a ) + c (1 − x )] dE2 cγ + =− 2 2 a−c − dF1 1 (1 − a ) − c ( a − c ) (1 − γ )[ xc + (1 − a ) (1 − x )] dE 3 cγ =− + . 2 dF1 1− a − c (1 − a ) − c 2 Consider first the effects of γ, the share of U.S. net debt held by China: —For γ = 0, dE2/dF1 and dE3/dF1 are both negative. Not surprisingly, an increase in U.S. net debt leads to a depreciation of the dollar against both the euro and the yen. —As γ increases, the depreciation of the dollar against the euro and the yen becomes smaller. This, too, is not surprising. What may be more surprising, however, is that, for high values of γ, the depreciation turns into an appreciation. For γ = 1, for example, the dollar appreciates against both the euro and the yen. The explanation is straightforward and is found in portfolio preferences: The transfer of wealth from the United States to China is a transfer of wealth from U.S. investors, who are willing to hold dollar, euro, and yen assets, to the People’s Bank of China, which holds only dollars. This transfer to an investor with extreme dollar preferences leads to a relative increase in the demand for dollars and hence an appreciation of the dollar against both the euro and the yen. Consider now the effects of x, the share of the U.S. net debt held by Europe, excluding the net debt held by China (for simplicity, we set γ equal to zero): —Consider first the case where x = 0, so that the accumulation of net debt is entirely vis-à-vis Japan. In this case, it follows that dE3/dF1 = 2 dE2/dF1. Both the yen and the euro appreciate against the dollar, with the yen appreciating twice as much as the euro. This result might again be surprising: why should a transfer of wealth from the United States to Japan lead to a change in the relative demand for euros? The answer is that it does not. The euro appreciates against the dollar but depreciates against the yen. The real effective exchange rate of the euro remains unchanged.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
39
—If x = 1⁄2 (which seems to correspond roughly to the ratio of trade deficits and thus to the relative accumulation of U.S. net debt today), then obviously the euro and the yen appreciate in the same proportion against the dollar. This simple framework also allows us to think about what would happen if China stopped pegging, or diversified its reserves away from dollars, or relaxed capital controls on Chinese and foreign investors, or any combination of these. Suppose China stopped pegging but maintained capital controls. Because the end of the peg, together with the assumption of maintained capital controls, implies a zero Chinese surplus, the renminbi would have to appreciate against the dollar. From then on, reserves of the Chinese central bank would remain constant. So, as the United States continued to accumulate net debt vis-à-vis Japan and Europe, relative net debt vis-à-vis China would decrease. In terms of our model, γ, the proportion of U.S. net debt held by China, would decrease.38 Building on our results, this would lead to a decrease in the role of an investor with extreme dollar preferences, the People’s Bank of China, and would lead to an appreciation of the euro and the yen. Suppose instead that China diversified its reserves away from dollars. Then, again, the demand for euros and for yen would increase, leading to an appreciation of both currencies against the dollar. To summarize: The trade deficits of the United States with Japan and the euro area imply an appreciation of both the yen and the euro against the dollar. For the time being, this effect is partly offset by the Chinese policies of pegging and keeping most of its reserves in dollars. If China were to give up its peg or to diversify its reserves, the euro and the yen would appreciate further against the dollar. This last argument is at odds with the often-heard statement that the Chinese peg has “increased the pressure on the euro-dollar exchange rate,” and that therefore the abandonment of the peg would remove some of the pressure, leading to a depreciation of the euro against the dollar. We do not understand the logic behind that statement. Two Simulations and a Look at Portfolios We have looked so far at equilibrium for a given distribution of Fs. This distribution is endogenous, however, in our model, determined by 38. Marginal γ, the proportion of the increase in U.S. net debt absorbed by China, would equal zero.
40
Brookings Papers on Economic Activity, 1:2005
trade deficits and portfolio preferences. We now report the results of two simulations of our extended model. In the first simulation we keep the symmetric portfolio assumptions introduced above. We take the three economies to be of the same size, and we use the values for the portfolio parameters introduced above of 0.70 for a and 0.15 for c. We consider a shift in the U.S. trade deficit, with half of the change in the deficit falling on China, one-fourth on Japan, and one-fourth on the euro area. We assume that each country trades only with the United States, so that we can focus on the bilateral balances with the United States. We perform this simulation under two alternative assumptions about Chinese policy. In both we assume capital controls, but in the first case we assume that China continues to peg the renminbi, and in the second we assume that the renminbi floats; together with the assumption of capital controls, this implies, as indicated above, a zero Chinese trade surplus. The top panel of figure 9 presents the results. Because of symmetry, the responses of the euro and the yen are identical and thus represented by the same line. The lower line shows the depreciation of the dollar against the euro and the yen when the renminbi floats. The higher locus shows the more limited depreciation of the dollar (and more limited appreciation of the euro and the yen) when the renminbi is pegged and the Chinese central bank accumulates dollars. One may wonder whether the preferences of private investors are really symmetric, however. Constructing portfolio shares for Japanese, European, and U.S. investors requires rather heroic assumptions. We have nevertheless given it a try, and the results are reported in table 1. Appendix B presents details of the construction. Note in table 1 the much larger share of dollar assets in European than in Japanese portfolios. Note also the small share of Japanese assets held by euro-area investors relative to the share of euro-area assets held by Japanese investors (the difference is much larger than the difference in relative size of the two economies). Portfolio preferences appear indeed to be asymmetric. To show what difference this asymmetry makes, the bottom panel of figure 9 presents results of a second simulation. This simulation is identical to that in the top panel but now takes into account the relative size of the three economies (the Xs) and uses the shares reported in table 1. The main conclusion we draw from the bottom panel is that it looks very similar to the top, except that the dollar depreciates initially a bit
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
41
Figure 9. Effects of a Shift in the U.S. Trade Deficit on Euro-Dollar and Yen-Dollar Exchange Rates, with and without Chinese Pega Symmetrical portfolio weights Percent change
–1 –2 Under peg –3 –4 Under float –5
Actual portfolio weights Percent change –1
Dollar-euro, under peg
–2 Dollar-yen, under peg
–3 –4
Dollar-euro, under float Dollar-yen, under float
–5 –6
5
10
15
20
25 Years
Source: Authors’ calculations. a. All simulations assume that China maintains capital controls.
30
35
40
45
42
Brookings Papers on Economic Activity, 1:2005
Table 1. Calculated Portfolio Shares by Investment Destinationa Investing country Destination United States Euro area Japan
United States
Euro area
Japan
0.77 0.15 0.08
0.42 0.53 0.05
0.22 0.15 0.63
Source: Authors’ calculations using data in appendix table B-1. a. Investment includes both portfolio investment and foreign direct investment.
more against the yen than against the euro. This difference is due to the larger share of dollar assets in European than in Japanese portfolios: a dollar transferred from the United States to Europe leads to a smaller decrease in the demand for U.S. assets than does a dollar transferred from the United States to Japan.
Summary and Conclusions We have argued that there have been two main forces behind the large U.S. current account deficits of the past ten years: an increase in the U.S. demand for foreign goods, and an increase in the foreign demand for U.S. assets. The path of the dollar since the late 1990s can be explained as the reaction to these forces. The shift in portfolio preferences toward U.S. assets manifested itself first, in the late 1990s, in the form of high private demand for U.S. equities, and more recently in the form of high central bank demand for U.S. bonds. The shift in demand away from U.S. goods is often attributed to more rapid growth in the United States than in its trading partners. This appears, however, to have played only a limited role: the performance of import and export equations in macroeconometric models shows that activity variables and exchange rates explain only about 60 percent of the increase in the U.S. trade deficit, with unexplained time trends and residuals accounting for the rest. We interpret this as evidence of a shift in the U.S. trade balance relation. Either shift could have induced the observed paths of the dollar and the U.S. current account only in a world where financial assets are imperfect substitutes. A shift in asset preferences could not account for these paths, because it would be meaningless in a world where assets are perfect sub-
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
43
stitutes. Nor can the shift in preferences for goods explain these paths, because with perfect substitutability such a shift—provided it were perceived as long lasting—would have induced a quicker and sharper depreciation of the exchange rate and a smaller increase in the current account than we have observed. As a way of organizing our thoughts about the U.S. current account deficit and the dollar, we have studied a simple model characterized by imperfect substitutability both among goods and among assets. The model allows for valuation effects, whose relevance has recently been emphasized in a number of papers. The explicit integration of valuation effects in a model of imperfect substitutability is, we believe, novel. We find that the degree of substitutability between assets does not affect the steady state. In other words, the eventual dollar depreciation induced by either shift is the same no matter how closely U.S. and foreign assets substitute for each other. But the degree of substitutability does play a central role in the dynamics of adjustment. In contrast to the case of perfect substitutability between assets, an increase in U.S. demand for foreign goods leads to a limited depreciation initially, a potentially large and long-lasting current account deficit, and a slow and steady depreciation over time. An increase in foreign demand for U.S. assets leads to an initial appreciation, followed by a slow and steady depreciation. The slow rate of dollar depreciation implied by imperfect substitutability contrasts with predictions by others of much more abrupt falls in the dollar in the near future. We show that, in the absence of anticipated portfolio shifts, the anticipated rate of depreciation depends on the change in the ratio of U.S. net debt to U.S. assets: the faster the increase in net debt, the faster the decrease in the relative demand for U.S. assets, and therefore the higher the rate of depreciation needed to maintain portfolio balance. If we take the annual increase in the ratio of net debt to U.S. GDP to be 5 percent, we derive an upper bound on the anticipated annual rate of depreciation of 2.7 percent a year. If, however, shares in U.S. assets in the portfolios of either U.S. or foreign investors are expected to decline, the anticipated depreciation can be much larger. If, for example, we anticipate that central banks will diversify their reserves away from dollars and, as a result, that the share of U.S. assets in foreign portfolios will decline by 2 percent over the coming year, then the anticipated depreciation may be as large as 8.7 percent. This is
44
Brookings Papers on Economic Activity, 1:2005
obviously an upper bound on the size of the depreciation, derived by assuming that private investors are willing to keep a constant share of their wealth in U.S. assets despite a high expected negative rate of return between now and then. (If, in anticipation of this high negative rate of return, private investors decide to decrease their share of dollar assets, then some of the depreciation will take place now, rather than at the time of the shift in composition of reserves, and so the anticipated depreciation will be smaller.) On the other hand, a further shift in investors’ preferences toward dollar assets would slow down, or even reverse, the path of dollar depreciation. The relief, however, would only be temporary. It would lead to an initial appreciation, but the accompanying loss of competitiveness would speed up the accumulation of foreign debt. The long-run value of the dollar would be even lower. The argument that the United States, thanks to the attractiveness of its assets, can keep running large current account deficits with no effect on the dollar appears to overlook the long-run consequences of a large accumulation of external liabilities. For basically the same reason, an increase in interest rates would be self-defeating. It might temporarily strengthen the dollar, but the depreciation eventually needed to restore equilibrium in the current account would be even larger—because (as in the case of a shift in portfolio preferences) the accumulation of foreign liabilities would accelerate, and eventually the United States would need to finance a larger flow of interest payments abroad. A better mix would be a decrease in interest rates and a reduction in budget deficits to avoid overheating. (To state the obvious: tighter fiscal policy is needed to reduce the current account deficit, but it is not a substitute for the dollar depreciation. Both are needed.) The same will happen so long as China keeps pegging the exchange rate. One should think of the People’s Bank of China as a special investor whose presence has the effect of raising the portfolio share of the world outside the United States invested in dollar assets. The longer the Chinese central bank intervenes, the larger this share. Sooner or later, however—as in the case of Korea in the late 1980s—the People’s Bank of China will find it increasingly difficult to sterilize the accumulation of reserves. Eventually, when the peg is abandoned, the depreciation of the dollar will be larger, the longer the peg will have lasted, because in the process the United States will have accumulated larger quantities of foreign liabilities. Thus, if China is worried about a loss of competitiveness, pegging may be a myopic choice.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
45
What would abandonment of the Chinese peg imply for the euro and the yen? Contrary to a commonly heard argument, if the renminbi were allowed to float, both currencies would be likely to appreciate further against the dollar. The reason is that, when the People’s Bank of China stops intervening, the market effectively loses an investor with extreme dollar preferences, to be replaced by private investors with less extreme preferences. A similar argument holds if the People’s Bank of China diversifies its reserves away from dollar assets. For Europe and Japan, however, what matter are effective exchange rates, and their currencies may well depreciate in effective terms even if they appreciate relative to the dollar in bilateral terms. We end with one more general remark. A large fall in the dollar would not by itself be a catastrophe for the United States. It would lead to higher demand for U.S. goods and higher output, and it would offer the opportunity to reduce budget deficits without triggering a recession. The danger is more serious for Japan and Europe, which suffer from slow growth already and have little room to use expansionary fiscal or monetary policy at this stage.
APPENDIX A
Dynamics of the Model THE DYNAMICS OF the system composed of equations 2 and 3 are more easily characterized by taking the continuous time limit. In continuous time the portfolio and current account balance equations become, respectively, E e X * E e X = α 1 + r − r * + , s ( X − F ) + 1 − α * 1 + r − r * + , s + F . E E E E e E e ( X − F ) + D ( E, z ). F = rF + 1 − α 1+ r − r * + , s E E Note the presence of both expected and actual appreciation in the current account balance equation. Expected appreciation determines the share of the U.S. portfolio invested in foreign assets; actual appreciation determines the change in the value of that portfolio, and in turn the change in the U.S. net debt position. We limit ourselves to a characterization of the equilibrium and local dynamics, using a phase diagram. (The global dynamics are more complex. The nonlinearities imbedded in the equations imply that the economy is
46
Brookings Papers on Economic Activity, 1:2005
likely to have two equilibriums, only one of which is potentially saddlepoint stable. This is the equilibrium we focus on.) We do so here under the additional assumption that r = r*. The extension to differences in interest rates, which we used to construct figure 8, is straightforward. The locus (E˙ = E˙e = 0) is obtained from the portfolio balance equation and is downward sloping. In the presence of home bias, an increase in net debt shifts wealth abroad, decreasing the demand for U.S. assets and requiring a depreciation. The locus (F˙ = 0) is obtained by assuming (E˙ e = E˙ ) in the current account balance equation and replacing (E˙ e) with its implied value from the portfolio balance equation. This locus is also downward sloping: a depreciation leads to a smaller trade deficit and thus allows for a larger net debt position consistent with current account balance. Note that the locus (F˙ = 0) is not the same as the current account balance locus in figure 1; that locus is derived under the assumption that both F˙ and E˙ are zero. Using that locus makes for a simple graphical characterization of the equilibrium but is not appropriate for studying stability or dynamics. The derivatives αR and α*R do not affect the slope of the locus (E˙ = 0) but do affect that of the locus (F˙ = 0). The smaller these derivatives are (that is, the lower the degree of substitutability between assets), the closer the locus (F˙ = 0) is to the locus (E˙ = 0). In the limit, if the degree of substitutability between U.S. and foreign assets is zero, the two loci coincide. The larger these derivatives are (that is, the higher the degree of substitutability between assets), the closer the (F˙ = 0) locus is to the current account balance locus, 0 = rF + D(E ,z). The condition for the equilibrium to be saddle-point stable is that the locus (E˙ = 0) be steeper than the locus (F˙ = 0); this turns out to be the same as the condition given in the text, that the portfolio balance locus be steeper than the current account balance locus. For this to hold, the following condition must be satisfied: r α + α* −1 . < (1 − α*) X * E EDE The interpretation of this condition was given in the text. It is more likely to be satisfied the lower the interest rate, the larger the home bias, and the larger the response of the trade balance to the exchange rate. If the condition is satisfied, the dynamics are as shown in figure A-1. The saddle path is downward sloping, implying that the adjustment to the steady state
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
47
Figure A-1. Adjustment of Exchange Rate and Net Debt in Continuous Time Exchange rate (E)
Current account balance rF + D(E,z) = 0 F=0
E=0
Net debt (F) Source: Authors’ model described in the text.
from below (in terms of F) is associated with an expected depreciation, and the adjustment from above with an expected appreciation. Valuation effects imply that unexpected shifts in z or s are associated with initial changes in F, according to ∆ F = (1 − α ) (1 + r *) ( X − F )
∆E . E
The effect of the degree of substitutability on the dynamics is as follows. The smaller are αR and α*R, the closer the locus (F˙ = 0) is to the locus (E˙ = 0), and so the closer the saddle-point path is to the locus (E˙ = 0). In the limit, if the degree of substitutability between U.S. and foreign assets is zero, the two loci and the saddle-point path coincide, and the economy remains on and adjusts along the (E˙ = 0) locus, the portfolio balance relation.
48
Brookings Papers on Economic Activity, 1:2005
The larger αR and α*R, the closer the (F˙ = 0) locus is to the locus given by 0 = rF + D(E ,z), and the closer the saddle-point path is to that locus as well. Also, the larger are αR and α*R, the slower is the adjustment of F and E over time. The slow adjustment of F comes from the fact that the current account is close to balance. The slow adjustment of E comes from the fact that, the larger the elasticities, the smaller is E˙ for a given distance from the E˙ = 0 locus. The limiting case of perfect substitutability is degenerate. The rate of adjustment to an unexpected, permanent shift in z goes to zero. The economy is then always on the locus 0 = rF + D(E,z). For any level of net debt, the exchange rate adjusts so that net debt remains constant, and, in the absence of shocks, the economy stays at that point. There is no unique steady state, and where the economy is depends on history.
APPENDIX B
Construction of Portfolio Shares DATA ON THE country allocation of gross portfolio investment are from the International Monetary Fund’s Coordinated Portfolio Survey for 2002. Data for the country allocation of direct investment are from the Organization for Economic Cooperation and Development and likewise refer to 2002. Financial wealth for the United States, the euro area, and Japan, which we need to compute the home bias of portfolios, are from official flow of funds data.39 From these data we construct the aij elements in two steps. First, we compute the geographical allocation of net foreign investment positions by weighting the shares of portfolio assets and foreign direct investment allocated to country j by the relative importance of portfolio ( pf ) and direct investment ( fdi) in country i’s total investment abroad. We then scale these shares by the share of total foreign investment (1 − aii), so that aij =
{[ pf ( pf i
i
+ fdii )] aij , p + [ fdii
( pf
i
+ fdii )] aij , fdi } × (1 − aii ) .
Table B-1 presents the results. 39. For the United States, see footnote 8. The source for Japan is the Bank of Japan flow of funds data (www.boj.or.jp/en/stat/sj/stat_f.htm), and that for the euro area is the ECB Economic Bulletin (released February, 2005 and available at www.ecb.int/pub/html/ index.en.html).
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
49
Table B-1. Calculated Portfolio Shares by Investment Destinationa Investing country Destination United States Euro area Japan Rest of the world
United States
Euro area
Japan
0.77 0.08 0.04 0.11
0.19 0.53 0.02 0.27
0.17 0.12 0.63 0.08
Sources: Authors’ calculations using data from the International Monetary Fund, the Organization for Economic Cooperation and Development, and national central banks. a. Investment includes both portfolio investment and foreign direct investment. Shares may not sum to 1.00 because of rounding.
To perform the simulation described in the text, we then allocate the shares invested in the “rest of the world” to foreign holdings so as to keep the relative shares in the remaining foreign assets the same. For the United States, for example, we increase the foreign shares in euro and yen assets to approximately 0.15 and 0.08, respectively. This gives us the numbers reported in table 1. The simulation presented in figure 9 uses these values, together with asset levels of $36.8 trillion for the United States, $23.0 trillion for the euro area, and $8.0 trillion for Japan. Trade is assumed to be bilateral between the United States and each of the other regions, with elasticities of the trade balance all being equal to the elasticity used in our earlier two-country model.
Comments and Discussion Ben S. Bernanke: Olivier Blanchard, Francesco Giavazzi, and Filipa Sa have produced a gem of a paper. They introduce a disarmingly simple model, which nevertheless provides a number of crucial insights about the joint dynamics of the current account and the exchange rate, in both the short and the long run. Their analysis will undoubtedly become a staple of graduate textbooks. The authors’ model has two features that deserve special emphasis. First, following an older and unjustly neglected literature, the model dispenses with the usual interest rate parity condition in favor of the assumption that financial assets may be imperfect substitutes in investors’ portfolios; that is, the model allows for the possibility that the demand for an asset may depend on features other than its rate of return, such as its liquidity or its usability as a component of international reserves. In focusing on imperfect asset substitutability and its implications, the authors identify an issue that has taken on great practical significance for policymakers in recent years. At least two contemporary policy debates turn in large part on the extent (or the existence) of imperfect asset substitutability. One is whether so-called nonstandard monetary policies—such as large purchases of government bonds or other assets by central banks—can stimulate the economy even when the policy interest rate has hit the zero lower bound. The other is whether sterilized foreign exchange interventions, like those recently undertaken on a massive scale by Japan and China, can persistently alter exchange rates and interest rates.1 The authors’ analysis explores yet another important implication of imperfect substitutability: that, if assets denominated in different currencies are imperfect substitutes, then agents may rationally anticipate the sustained depreciation of a currency even in the absence 1. Bernanke, Reinhart, and Sack (2004) present empirical evidence relevant to both of these debates.
50
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
51
of cross-currency interest rate differentials. Thus, by invoking imperfect substitutability, the authors are able to show that expected dollar depreciation is not necessarily inconsistent with the currently low level of U.S. long-term nominal interest rates and the evident willingness of foreigners to hold large quantities of U.S. assets. The assumption that financial assets of varying characteristics are imperfectly substitutable in investor portfolios seems quite reasonable. (Almost as I write these words, an announcement by the U.S. Treasury that it is contemplating the reinstatement of the thirty-year bond seems to have triggered a jump in long-term bond yields, suggestive of a supply effect on returns.) However, both the theoretical and the empirical literatures on asset substitutability are exceedingly thin, which is a problem for assessing the quantitative implications of the authors’ analysis. In particular, as they themselves note, in their model the speed of adjustment of the exchange rate and the current account depends importantly on the elasticities of foreign and domestic asset demands with respect to expected return differentials, numbers that are difficult to pin down with any confidence. Further complications arise if, as is plausibly the case, the degree of asset substitutability is not a constant but varies over time or across investors. For example, if private investors view assets denominated in different currencies as more substitutable than central banks do, which seems likely, then changes in the share of assets held by each type of investor will have implications for exchange rate dynamics. Finding satisfying microfoundations for the phenomenon of imperfect asset substitutability, and obtaining persuasive estimates of the degree of substitutability among various assets and for different types of investors, should be high on the profession’s research agenda. The second feature of the authors’ analysis worth special note is its attention to the long-run steady state. By integrating short-run and long-run analyses, the authors obtain some useful insights that a purely short-run approach does not deliver. Notably, they demonstrate that factors affecting the value of the dollar or the size of the U.S. current account deficit may have opposite effects in the short and in the long run. For example, an increased appetite for dollars on the part of foreign central banks is typically perceived by market participants as positive for the dollar in the short run, and the model supports this intuition. However, the authors show that, because the short-term appreciation of the dollar may delay necessary adjustment, in the long run the result of an increased preference for dollars may be more rather than less dollar depreciation. Thus developments
52
Brookings Papers on Economic Activity, 1:2005
that are “good news” for the currency in the short run may be “bad news” at a longer horizon. One point that I take from the paper’s analysis, however, is that the particular assumptions one makes about the nature of the steady state may significantly affect one’s predictions about short-run dynamics and the speed of adjustment. For example, the authors assume in most of their analysis that, in the long run, the U.S. current account must return to balance. One might reasonably assume instead that, in the long run, the current account will remain in deficit at levels consistent with long-run stability in the ratio of external debt to GDP. This apparently innocuous change in the steadystate assumption may have quantitatively important implications for the medium-term pace of adjustment. In particular, to the extent that foreigners are willing to accept a long-run U.S. debt-to-GDP ratio that is somewhat higher than the current level of about 25 percent, the authors’ model predicts that the period of current account adjustment could be extended for a number of years. Because we know little about the quantity of U.S. assets that foreigners may be willing to hold in the long run, the model suggests that one cannot forecast the speed of the adjustment process with any confidence. Although the authors’ model is extraordinarily useful, like any simple model it leaves out important factors. From my perspective, the model’s most important omissions are related to its treatment of asset values and interest rates. Except for the exchange rate itself, the model takes asset values and interest rates as exogenous, thereby excluding what surely must be an important source of current account dynamics, namely, the endogenous evolution of wealth and expected returns. For example, I doubt that the recent decline in U.S. household saving, a major factor (arithmetically at least) in the rise in the U.S. current account deficit, can reasonably be treated as exogenous, as is done in the paper. Instead, at least some part of the decline in saving likely reflects the substantial capital gains that U.S. households have enjoyed in the stock market (until 2000, and to some extent since 2003) and in the values of their homes. Capital gains have allowed Americans to feel wealthier without saving out of current income. Where did these capital gains come from? In my view an important driver of the rise in U.S. wealth is the rapid increase over the past decade or so in the global supply of saving, which in turn is the product of both the strong motivation to save on the part of other aging industrial societies and a reluctance of emerging economies to import capital since the financial crises of the 1990s. Increased global saving has produced a striking decline in real
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
53
interest rates around the world, a decline that has contributed to the increased valuation of stocks, housing, land, and other assets.2 Because of its openness to foreign capital, its financial sophistication, and its relatively strong economic performance, the United States has absorbed the lion’s share of this increment to global saving; however, other industrial countries (including France, Italy, Spain, and the United Kingdom) have also experienced increased asset values (house prices, for the most part), increased consumption, and corresponding movements in their current account balances toward deficit. An implication of this story is that an endogenous moderation of the U.S. current account deficit may be in store, even without major changes in exchange rates and interest rates, as a diminishing pace of capital gains slows U.S. consumption growth.3 This story, or any explanation that relies heavily on endogenous changes in asset prices and the ensuing wealth and spending dynamics, cannot be fully captured by the current version of the authors’ model. How might endogenous wealth dynamics change the authors’ conclusions? One way of developing an intuition about the effects of wealth dynamics in the context of their model is to use that model to consider the implications for the current account and the dollar of an exogenous change in the value of U.S. assets, X. Although this approach yields at best a simple approximation of the effect of making wealth endogenous, examining model outcomes when one drops the authors’ assumption of unchanging wealth should provide some insight. To carry out this exercise, I write the key equations of the model as follows: (1)
where R =
X* F+1 = [1 − α * ( R, s )] + F (1 + r ) E [1 − α ( R, s ) ( X − F )] ( − 1 + r ) + D ( E, X − F ) Rrealized (1 + r ) E+1 (1 + r ) E+e1 and Rrealized = ( (1 + r *) E 1 + r*)E
2. Bernanke (2005). 3. Recent experience in the United Kingdom shows that a stabilization of house prices after a period of rapid increases may damp consumer spending and increase saving rates.
54
(2)
Brookings Papers on Economic Activity, 1:2005
X* X = [ α ( R, s )] ( X − F ) + [1 − α * ( R, s )] + F . E
I use the authors’ notation, except that I find it useful to distinguish between the anticipated relative return on U.S. assets, R, and the realized relative return on U.S. assets, Rrealized. I also suppress the shock terms z and s, which I will not use here. Equation 1 is the current account equation, which describes the evolution of U.S. net foreign debt, F. The first term on the right-hand side of equation 1 captures the idea that, all else equal, foreign debt grows at the U.S. real rate of interest. The second term, which I have chosen to write in a slightly different form than the authors do, is the valuation effect associated with unanticipated changes in the exchange rate. In particular, when the value of the dollar is less than expected, Rrealized < R, and the dollar value of U.S. gross foreign assets rises. This valuation effect serves to reduce U.S. net dollar liabilities. The third term in equation 1 is the trade deficit, which adds directly to net foreign liabilities. I extend the authors’ model here by including U.S. domestic wealth, X − F, as a determinant of the trade deficit. I assume that the derivative of the trade deficit with respect to U.S. wealth is positive; higher wealth induces U.S. households to spend more, increasing the trade deficit. Equation 2, the portfolio balance equation, is the same as in the paper. This equation requires that the supply of U.S. assets X equal the sum of U.S. and foreign demands for those assets. The steady-state equations corresponding to equations 1 and 2 are (3)
rF = − D ( E, X − F )
(4)
X = [ α (1, s )] ( X − F ) + [1 − α * (1, s )] ( X * E + F ) .
Equation 3 is the steady-state version of the current account equation, modified to allow U.S. wealth to affect the trade balance. Here I retain the authors’ assumption that the current account must be in balance in the long run (as opposed to assuming a constant ratio of external debt to GDP in the long run). Equation 4 is the steady-state version of the portfolio balance, exactly as in the paper. Like the authors, I assume that the foreign real interest rate equals the domestic rate, so that R = Rrealized = 1 in the steady state. My figure 1, which is analogous to the figures in the paper, graphs the steady-state equations 3 and 4. Because foreign debt F is included as a
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
55
Figure 1. Adjustment of the Exchange Rate and the Net Debt Position to an Increase in U.S. Assets Exchange rate (E)
Portfolio balance
A B
Current account balance
C
Net debt (F) Source: Authors’ model described in the text.
determinant of the trade deficit (more foreign debt reduces U.S. wealth and thus the trade deficit), the current account line in my figure is flatter than its analogue in the authors’ model, all else equal; under reasonable assumptions, however, it is still downward sloping. The portfolio balance line is the same as in the authors’ analysis. Consider now the effects of an exogenous increase in X. A first issue is whether this increase is expected to be temporary or permanent. If consumers have a target wealth-to-income ratio, which is not an unreasonable supposition, the increase in X might be thought of as largely transitory. In this case it is straightforward to show that the steady state will be unaffected by the increase in U.S. assets, so that the current account and the exchange rate will return to their original values in the long run; that is, although it would imply a short-run depreciation, a temporary increase in
56
Brookings Papers on Economic Activity, 1:2005
the value of U.S. assets would have no lasting effect on the dollar or the U.S. net international position. Since this case, although possibly relevant, is not very interesting, I consider instead the case in which the increase in the value of U.S. assets is expected to be permanent. Figure 1 shows the graphical analysis of a permanent increase in U.S. assets. I assume that the economy is initially in the steady state defined by point A. Inspection of equation 3 shows that an increase in X shifts the current account line down, as greater U.S. wealth worsens the steady-state trade balance at any given exchange rate. Conceptually, this downward shift is analogous to the effect of an exogenous increase in the U.S. demand for foreign goods, as analyzed by the authors. Absent any change in the portfolio balance condition, this shift would imply both dollar depreciation and increased foreign debt in the long run, exactly as in the paper’s analysis of an exogenous shift in demand. However, the portfolio balance line is not unchanged in my scenario but instead is shifted downward by the increase in X, as foreigners are willing to hold their share of the increase in U.S. assets only if the dollar depreciates. (The depreciation implies an unanticipated reduction in the dollar share of foreigners’ portfolios, for which they are assumed to compensate by buying additional dollar assets.) With the shifts in both the current account and the portfolio balance relations taken into account, the new steady-state position is shown as point C in figure 1. As indicated, and under plausible assumptions, the economy adjusts by jumping immediately from point A to point B, as the dollar depreciates and U.S. net foreign debt declines. Over time the economy moves from point B to point C, as the dollar depreciates further and foreign debt accumulates. A key point is that, all else equal, the steady-state outcome described by point C involves less dollar depreciation and less accumulation of foreign debt than the scenario (analyzed by the authors) in which U.S. demand for foreign goods increases exogenously (that is, a scenario in which only the current account line shifts down). Economically, the unexpected depreciation induced by the requirement of portfolio balance assists the U.S. current account adjustment process in two ways: First, the depreciation reduces the initial dollar value of U.S. net foreign debt directly, by means of the valuation effect. Second, the early depreciation of the dollar associated with the portfolio balance requirement mitigates the trade impact of the rise in wealth. Note also that U.S. domestic wealth (that is, net of foreign liabilities) is very likely to be higher in the long run than initially, reflecting
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
57
the capital gains enjoyed at the beginning of the process. This analysis is overly simple, as already noted, but it suggests to me that inclusion of endogenous wealth dynamics might give different and possibly less worrisome predictions about U.S. current account adjustment than those presented in the paper. My final observations bear on the authors’ analysis of the case with more than two currencies. I found this part of the paper quite enlightening, particularly the discussion of the likely effects of a revaluation of the Chinese currency on the value of the euro. One occasionally hears the view expressed that yuan revaluation would “take the pressure off” the euro (that is, allow it to depreciate); the underlying intuition appears to be that the effective dollar exchange rate must fall by a certain amount, and so, if it cannot fall against the yuan, it will fall against the euro. The authors show that this intuition is likely misguided, in that a stronger yuan probably implies a stronger euro as well. Their argument can be understood either in terms of portfolio balance or in terms of trade balance. From a portfolio perspective, a yuan revaluation presumably would shift Chinese demand away from dollar assets and toward euro assets, strengthening the exchange value of the euro. From a trade perspective, if Chinese goods become more expensive for Americans, U.S. demand may shift toward euro-zone goods, again implying euro appreciation. I see much merit in this analysis but would note that these results may not generalize to cases with many countries and variable patterns of substitution and complementarity among goods and among currencies. To illustrate, suppose that Chinese goods and European goods are viewed as complements by potential buyers in other nations. Then, in the same way that a rise in the price of teacups lowers the price of saucers, a Chinese revaluation might reduce the global demand for European exports to an extent sufficient to cause the euro to depreciate. This example is probably not realistic (others could be given), but it shows that drawing general conclusions about how changes in the value of one currency affect that of another may be difficult. Even if a revaluation of the yuan did lead to an appreciation of the euro, however, one should not conclude that yuan revaluation is against the European interest. A yuan revaluation might well lead to both an increase in the demand for European exports (as U.S. demand is diverted from China) and a reduction in European interest rates (reflecting increased Chinese demand for euro assets). Yuan revaluation might therefore stimulate the European economy even though the euro appreciates.
58
Brookings Papers on Economic Activity, 1:2005
Hélène Rey: Olivier Blanchard, Francesco Giavazzi, and Filipa Sa have given us a very clear and elegant framework within which to discuss some complex and important questions. The U.S. current account deficit has been at the center of the economic policy debate for some time. The deficit stood at more than 6 percent of GDP in 2004, and in dollar terms it has reached historically unprecedented levels. A country can eliminate an external imbalance either by running trade surpluses, or by earning favorable returns on its net foreign asset portfolio, or both. The first of these, the trade channel of adjustment, has been traditionally emphasized in studies of current account sustainability. The valuation channel has received attention only lately, but with the recent upsurge in cross-border asset holdings, its quantitative significance has greatly increased. When the securities in which external assets and liabilities are held are imperfectly substitutable, any change in asset prices and, in particular, any change in the exchange rate create international wealth transfers, which can be sizable. These transfers significantly alter the dynamics of net foreign assets. The following example illustrates the power of the valuation channel to smooth the U.S. adjustment process. Following Cédric Tille,1 assume that U.S. external liabilities, which amounted to about $10.5 trillion in December 2003, are all denominated in dollars, whereas 70 percent of the $7.9 trillion in U.S. external assets are in foreign currency. Then a mere 10 percent depreciation of the dollar, by increasing the dollar value of the foreign-currency assets while leaving the dollar value of the liabilities constant, would create a wealth transfer from the rest of the world to the United States equal to 0.1 × 0.7 × 7 trillion, or about $553 billion, which is approximately 5 percent of U.S. GDP and on the order of the U.S. current account deficit in 2003. The exchange rate thus has a dual stabilizing role for the United States. A dollar depreciation helps improve the trade balance and increases the net foreign asset position, and this has to be taken into account when assessing the prospects of the U.S. external deficit and the future path of the dollar. The authors have set out to do just that. They use a portfolio balance model (drawing on the work of Pentti Kouri, Stanley Black, Dale Henderson and Kenneth Rogoff, and William Branson in the 1980s) to model jointly the dynamics of the current account and of the exchange rate, allowing for 1. Tille (2003).
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
59
imperfect substitutability between assets and for (some) valuation effects. In such a framework, a negative shock to preferences for U.S. goods, say, leads immediately to a depreciation of the dollar. This immediate, unexpected depreciation does not, however, fully offset the shock. If it did, there would be excess demand for U.S. assets, as the supply of those assets is taken to be fixed and the dollar value of the rest of the world’s wealth rises. Instead there is a less than fully offsetting drop in the dollar, and foreigners’ demand for U.S. assets is kept in check by a further, expected depreciation of the dollar toward its long-run steady-state value. Along the path of this depreciation, the United States accumulates more debt, so that the long-run level of the dollar will be below that which would have been needed to offset the entire negative shock immediately. The dollar is expected to depreciate at a decelerating rate in order for foreigners to keep accumulating U.S. assets. A remarkable prediction thus emerges from this simple model: foreigners continue to purchase U.S. assets and finance the U.S. current account deficit even though they expect a further dollar depreciation, which implies capital losses on their portfolio. This result stems entirely from the imperfect degree of substitutability between U.S. and foreign assets. If assets were perfect substitutes, the exchange rate would jump immediately to the steady-state level that would be compatible with the change in preferences for goods. Pierre-Olivier Gourinchas and I present strong evidence that assets are imperfect substitutes.2 We find that current external imbalances have substantial predictive power on net asset portfolio returns and, in particular, on exchange rates. Using a newly constructed database on U.S. external imbalances since 1952, we show that negative external imbalances imply future expected depreciations of the dollar. We find that a 1-standard-deviation increase in the imbalance leads to an expected annualized depreciation of around 4 percent over the next quarter. These empirical results are fully supportive of the portfolio balance approach and of Blanchard, Giavazzi, and Sa’s model. We also find, however, that the trade channel of adjustment kicks in at longer horizons, so that the valuation effects operate in the short to medium run whereas the trade balance effects operate in the longer run. In the authors’ model, in contrast, valuation and trade channels operate contemporaneously. There is no lag in the adjustment dynamics of the trade flows. If there were, the dynamics of the debt accumulation would be different. But I think 2. Gourinchas and Rey (2005).
60
Brookings Papers on Economic Activity, 1:2005
it is reasonable to conjecture that this would not change any of the qualitative results of the paper. A more important point is that the authors model rates of return using (exogenous) interest rates only. In reality, U.S. assets and liabilities include both equity and debt and indeed have a very asymmetric composition. The external assets of the United States consist mainly of foreign direct investment and equities, whereas U.S. external liabilities contain a larger share of bank loans and other debt. As a consequence, the returns on U.S. external assets and liabilities differ substantially. The United States, as the world’s banker, has traditionally enjoyed higher returns on its assets, which are dominated by long-term risky investments, than it has had to pay on its mostly liquid liabilities. (This explains in part why the income on U.S. net foreign assets is still positive even though the United States’ liabilities exceed its assets by about 30 percent.) Hence the net foreign asset dynamic is highly dependent on differences in relative returns on portfolio equity, FDI, and so forth, and is mischaracterized if one considers only the risk-free interest rate. The authors’ framework also ignores the joint determination of exchange rates, bond prices, and equity returns on asset markets. A more complete model would feature endogenous valuation effects on the stock of assets and liabilities, both in the current account equation and in the portfolio balance equation. This also means that the steady-state condition of the authors’ model, which equates the interest to be paid on the U.S. net foreign debt to the trade balance, may be significantly altered when one takes into account the composition of the net debt. If it is dominated by contingent claims such as equities, the equilibrium steady-state exchange rate necessary to generate the required trade balance may differ considerably from what their model assumes. The exogeneity of the rate of return (the interest rate) is a clear limitation. In principle, the interest rate should be determined by the reaction of the Federal Reserve and by endogenous changes in world supply and demand for capital. Proponents of the “global savings glut” theory see no mystery in persistently low long-term U.S. interest rates. As it stands, the model has nothing to say on these issues. The authors make a very natural extension of their model to a threecountry setting, and they demonstrate that putting pressure on China to introduce more flexibility in its exchange rate regime would be counterproductive if the objective is a less depreciated dollar against the euro. Indeed, by forcing China out of the business of buying dollars, one effectively bans from the market the agent with the stronger bias for dollars. Since the cur-
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
61
rency demand of the other agents is more diversified, this decreases the demand for dollars and increases the pressure on the euro to appreciate. Hence, at least in the short run, the dynamic is perverse. I think this is an excellent insight that should be discussed in policy circles. One of the messages of this very rich paper is that, as then-U.S. Treasury secretary John Connolly put it in the 1970s, “the dollar is our currency but your problem.” Indeed, the paper makes a very strong case that, to return to the steady state after a negative shock to the U.S. current account, one needs the dollar to depreciate in a predictable way at a moderate speed for a long period. Along the adjustment path, foreign investors incur capital losses as wealth is transferred to the United States. The adjustment is smooth and relatively painless for the U.S. economy, but the rest of the world suffers not only the capital loss but also a loss in competitiveness for the export sector (but increased purchasing power). I have two comments on this point. The first is that, within the model, the speed of the predicted depreciation of the dollar can be computed only with considerable uncertainty. It depends on several difficult-to-measure quantities such as world wealth, the degree of home bias in U.S. and foreign portfolios, and the future change in that bias. So it would not be surprising if the speed of depreciation turned out to be quicker than the upper bound of 2.7 percent a year (or even 8.4 percent a year) predicted by the authors. We just do not know. My second comment is that the assumptions implicit in these results are that bond prices are exogenous and that no run on dollar assets occurs. In the authors’ model, whatever happens to the exchange rate does not affect the U.S. interest rate. That is surely too extreme an assumption. Without making any predictions, I would like to suggest that a less rosy scenario be put on the table as well, in which turmoil occurs in both the bond and the foreign exchange markets simultaneously. One can imagine that some Asian central banks that are at least partly accountable to the citizens of their countries (such as the Korean central bank) might start diversifying out of dollar assets in order to decrease their exposure to exchange rate risk. To the extent that such a move creates jitters in financial markets and private investors follow suit, the U.S. interest rate could go up at the same time that the dollar is going down, which could lead to a further unwinding of positions. We had a small taste of such an event in early 2005, when the Korean central bank announced that it would diversify its future accumulations of reserves (that is, its flows, not even its stocks) out of the dollar, and U.S. interest rates rose sharply for a short period. This scenario could be particularly damaging if
62
Brookings Papers on Economic Activity, 1:2005
Asian central banks were to dump ten-year U.S. Treasuries, which constitute the backbone of the U.S. mortgage market. Since we do not have precise information on the maturity structure of the debt held by the Asian central banks, or precise estimates of the degree of substitutability between the tenyear bond and bonds at the short end of the yield curve, such a scenario would be sure to be full of surprises. In the end much would depend on the willingness of the Federal Reserve to tighten monetary policy aggressively. If U.S. interest rates jumped sharply, the whole world economy could be in for a hard landing. To conclude, this paper is a remarkable achievement, and I am sure it will prove to be an invaluable pedagogical tool. After almost three decades during which the portfolio balance approach was largely neglected, this paper and some other recent work point toward its renewed relevance. The authors provide a perfect example of how powerful it can be to gain clear insights on the very complex questions posed by the dynamics of the U.S. current account deficit and the dollar. The next, very important step in this line of research is to develop a more convincing model of asset prices and wealth dynamics. Until we endogenize international portfolio flows in different assets, the wealth dynamics, and the joint determination of the exchange rate, equity prices, and interest rates, we will not be able to fully comprehend the nature of the international adjustment process and will have to shy away from specific policy recommendations. General discussion: Gian Maria Milesi-Ferretti observed that foreigners own relatively little of U.S. housing wealth. As a consequence, any fall in home prices due to rising interest rates would have a relatively small valuation effect on foreign wealth, and therefore little effect on foreign demand for U.S. assets. By the same token, it would have a relatively large effect on U.S. wealth, saving, and the current account. He also pointed out that the large increase in world saving over the past decade has come mainly from China, where both saving and investment have risen spectacularly. Outside of China saving rates have mostly declined. Indeed, the rise in current account surpluses in other East Asian economies reflects a sharp decline in domestic investment rather than an increase in saving. Sebastian Edwards added that every region in the world outside North America, including Africa, has a current account surplus, and most emerging economies are purchasing U.S. assets. He reasoned that it will be difficult for these countries to grow rapidly if their saving continues to go abroad rather than into domestic investment.
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
63
Richard Cooper suggested that alternative assumptions about the character of financial markets and the distribution of world saving could alter some of the authors’ model results, quantitatively and possibly qualitatively. For example: Saving in the rest of the world is roughly three times U.S. saving, but more than half of the world’s easily marketable assets are located in the United States, making it the preferred destination for foreign investment. Even with home bias, as long as rest-of-world wealth is growing faster than U.S. wealth, net investment flows into U.S. assets are likely to continue, and with them U.S. current account deficits. William Nordhaus remarked that the situation Cooper described is changing: as Europe opens its capital markets, the large U.S. share of the world’s marketable assets should gradually fall. Michael Dooley argued that Cooper’s analysis, and the imperfect substitutability in the authors’ portfolio model, did not capture the growing risk that private agents would perceive as U.S. net indebtedness continues to grow. A counter to this constraint on private asset demand is provided when foreign official sectors invest in U.S. assets the way several Asian central banks are doing today. Peter Garber added that central banks of emerging economies are readily buying these assets because they provide the collateral that encourages outside investors to undertake gross investment flows into these economies. He believed the exchange rate movements of the past few years were mainly due to these official interventions, which underwrite the U.S. capital market at low interest rates. At these low rates, private sector investors have shifted their demand toward European securities, causing the euro to strengthen and reducing Europe’s current account surplus. Edmund Phelps explained that his own model projected a much lower dollar and a shift to U.S. current account surpluses, and he addressed the macroeconomic implications for the United States of such a move. He disagreed with the more optimistic experts who see such a transition as not affecting aggregate output and employment in any important way. On that scenario, the investment decline that accompanies lower business asset values in his model would be smoothly offset by rising exports and a move toward current account surpluses. Phelps, however, believed that the needed shift in resources would be incomplete to the extent that the investment-type activities are relatively labor intensive in production. He thus expected the needed adjustment to have a significant macroeconomic impact, and he saw the U.S. economy heading into a decade or more of slower growth and weakening employment.
64
Brookings Papers on Economic Activity, 1:2005
References Baldwin, Richard E., and Paul R. Krugman. 1987. “The Persistence of the U.S. Trade Deficit.” BPEA, no. 1: 1–43. Bernanke, Ben S. 2005. “The Global Saving Glut and the U.S. Current Account Deficit.” Homer Jones Lecture, Federal Reserve Bank of St. Louis, April 14. Available on the Internet at www.federalreserve.gov/boarddocs/speeches/2005/ 20050414/default.htm. Bernanke, Ben S., Vincent R. Reinhart, and Brian P. Sack. 2004. “Monetary Policy Alternatives at the Zero Bound: An Empirical Assessment.” BPEA, no. 2: 1–78. Branson, William H., and Dale W. Henderson. 1985. “The Specification and Influence of Asset Markets.” In Handbook of International Economics II, edited by Ronald W. Jones and Peter B. Kenen. Amsterdam: Elsevier Science Publishers. Caballero, Ricardo J., Emmanuel Farhi, and Mohamad L. Hammour. 2004. “Speculative Growth: Hints from the U.S. Economy.” Massachusetts Institute of Technology. Chinn, Menzie D. 2004. “Incomes, Exchange Rates and the U.S. Trade Deficit, Once Again.” International Finance 7, no. 3: 451–69. Cooper, Richard N. 1986. “Dealing with the Trade Deficit in a Floating Rate System.” BPEA, no. 1: 195–207. Debelle, Guy, and Gabriele Galati. 2005. “Current Account Adjustment and Capital Flows.” BIS Working Paper 169. Basel: Monetary and Economic Department, Bank for International Settlements. Dooley, Michael P., David Folkerts-Landau, and Peter M. Garber. 2004. “The U.S. Current Account Deficit and Economic Development: Collateral for a Total Return Swap.” Working Paper 10727. Cambridge, Mass.: National Bureau of Economic Research. Dornbusch, Rudiger. 1976. “Expectations and Exchange Rate Dynamics.” Journal of Political Economy 84: 1161–76. _________. 1987. “External Balance Correction: Depreciation or Protection?” BPEA, no. 1: 249–69. Gourinchas, Pierre-Olivier, and Hélène Rey. 2005. “International Financial Adjustment.” Working Paper 11155. Cambridge, Mass.: National Bureau of Economic Research (February). Henderson, Dale, and Kenneth Rogoff. 1982. “Negative Net Foreign Asset Positions and Stability in a World Portfolio Balance Model.” Journal of International Economics 13: 85–104. Houthakker, Hendrik S., and Stephen P. Magee. 1969. “Income and Price Elasticities in World Trade.” Review of Economics and Statistics 51, no. 2: 111–25. Kouri, Pentti. 1976. “Capital Flows and the Dynamics of the Exchange Rate.” Seminar Paper 67. Stockholm: Institute for International Economic Studies. _________. 1983. “Balance of Payments and the Foreign Exchange Market: A Dynamic Partial Equilibrium Model.” In Economic Interdependence and Flexible
Olivier Blanchard, Francesco Giavazzi, and Filipa Sa
65
Exchange Rates, edited by Jagdeep S. Bhandari and Bulford H. Putnam. MIT Press. Krugman, Paul R. 1991. “Introduction.” In International Adjustment and Financing, edited by C. Fred Bergsten and Paul R. Krugman. Washington: Institute for International Economics. Lane, Philip R., and Gian Maria Milesi-Ferretti. 2002. “Long-Term Capital Movements.” In NBER Macroeconomics Annual 2001, edited by Ben S. Bernanke and Kenneth S. Rogoff. MIT Press. _________. 2004. “Financial Globalization and Exchange Rates.” CEPR Discussion Paper 4745. London: Centre for Economic Policy Research. Lawrence, Robert Z. 1990. “U.S. Current Account Adjustment: An Appraisal.” BPEA, no. 2: 343–82. Marquez, Jaime. 2000. “The Puzzling Income Elasticity of U.S. Imports.” Washington: Federal Reserve Board. Masson, Paul. 1981. “Dynamic Stability of Portfolio Balance Models of the Exchange Rate.” Journal of International Economics 11: 467–77. Obstfeld, Maurice. 2004. “External Adjustment.” Review of World Economics 140, no. 4: 541–68. Obstfeld, Maurice, and Kenneth Rogoff. 2004. “The Unsustainable U.S. Current Account Position Revisited.” Working Paper 10869. Cambridge, Mass.: National Bureau of Economic Research. Roubini, Nouriel, and Brad Setser. 2005. “Will the Bretton Woods 2 Regime Unravel Soon? The Risk of a Hard Landing in 2005–2006.” Paper presented at a symposium on the “Revived Bretton Woods System: A New Paradigm for Asian Development?” organized by the Federal Reserve Bank of San Francisco and the University of California, Berkeley, San Francisco, February 4, 2005. Sachs, Jeffrey D. 1988. “Global Adjustments to a Shrinking U.S. Trade Deficit.” BPEA, no. 2: 639–67. Tille, Cédric. 2003. “The Impact of Exchange Rate Movements on U.S. Foreign Debt.” Issues in Economics and Finance 9, no. 1: 1–7.
MAURICE OBSTFELD University of California, Berkeley KENNETH S. ROGOFF Harvard University
Global Current Account Imbalances and Exchange Rate Adjustments THIS IS THE third in a series of papers we have written over the past five years about the growing U.S. current account deficit and the potentially sharp exchange rate movements any future adjustment toward current account balance might imply.1 The problem has hardly gone away in those five years. Indeed, the U.S. current account deficit today is running at around 6 percent of GDP, an all-time record. Incredibly, the U.S. deficit now soaks up about 75 percent of the combined current account surpluses of Germany, Japan, China, and all the world’s other surplus countries.2 To balance its current account simply through higher exports, the United States would have to increase export revenue by a staggering 58 percent over 2004 levels. And, as we argue in this paper, the speed at which the U.S. current account ultimately returns toward balance, the triggers that drive that adjustment, and the way in which the burden of adjustment is allocated across Europe Eyal Dvir, José Antonio Rodriguez-Lopez, and Jón Steinsson provided dedicated and excellent research assistance, for which we are extremely grateful. We also thank Philip Lane and Gian Maria Milesi-Ferretti for discussions and data. Jane Trahan’s technical support was outstanding. 1. See Obstfeld and Rogoff (2000a, 2004). From an accounting perspective, a country’s current account balance essentially adds net interest and dividend payments to its trade balance. As we discuss below, the United States presently receives about the same amount of income on its foreign assets as it pays out to foreign creditors. Hence, for the United States (and indeed many countries), the current account balance and the trade balance are quantitatively very similar. As we later emphasize, however, the current account does not include capital gains and losses on existing wealth. Thus the overall change in a country’s net foreign asset position can, in principle, be less than or greater than its current account deficit or surplus. 2. Calculated from the World Economic Outlook database of the International Monetary Fund, using current account data from 2004.
67
68
Brookings Papers on Economic Activity, 1:2005
and Asia all have enormous implications for global exchange rates. Each scenario for returning to balance poses, in turn, its own risks to financial markets and to general economic stability. Our assessment is that the risks of collateral damage—beyond the risks to exchange rate stability—have grown substantially over the five years since our first research paper on the topic, partly because the U.S. current account deficit itself has grown, but mainly because of a mix of other factors. These include, not least, the stunningly low U.S. personal saving rate (which, driven by unsustainable rates of housing appreciation and record low interest rates, fell to 1 percent of disposable personal income in 2004). But additional major risks are posed by the sharp deterioration in the U.S. federal government’s fiscal trajectory since 2000, rising energy prices, and the fact that the United States has become increasingly dependent on Asian central banks and politically unstable oil producers to finance its deficits. To these vulnerabilities must be added Europe’s conspicuously inflexible economy, Japan’s continuing dependence on export-driven growth, the susceptibility of emerging markets to any kind of global financial volatility, and the fact that, increasingly, the counterparties in international asset transactions are insurance companies, hedge funds, and other relatively unregulated nonbank financial entities. Perhaps above all, geopolitical risks and the threat of international terror have risen markedly since September 2001, confronting the United States with open-ended long-term costs for financing wars and homeland security. True, if some shock (such as a rise in foreign demand for U.S. exports) were to close up these global imbalances quickly without exposing any concomitant weaknesses, the damage might well be contained to exchange rates and to the collapse of a few large banks and financial firms—along with, perhaps, mild recession in Europe and Japan. But, given the broader risks, it seems prudent to try to find policies that will gradually reduce global imbalances now rather than later. Such policies would include finding ways to reverse the decline in U.S. saving, particularly by developing a more credible strategy to eliminate the structural federal budget deficit and to tackle the country’s actuarially insolvent old-age pension and medical benefit programs. More rapid productivity growth in the rest of the world would be particularly helpful in achieving a benign adjustment, but only, as the model we develop in this paper illustrates, if that growth is concentrated in nontraded (domestically produced and consumed) goods rather than the export sector, where such productivity growth could actually widen the U.S. trade deficit.
Maurice Obstfeld and Kenneth S. Rogoff
69
It is also essential that Asia, which now accounts for more than one-third of global output on a purchasing power parity basis, take responsibility for bearing its share of the burden of adjustment. Otherwise, if demand shifts caused the U.S. current account deficit to close even by half (from 6 percent to 3 percent of GDP), while Asian currencies remain fixed against the dollar, we find that European currencies would have to depreciate by roughly 29 percent. Not only would Europe potentially suffer a severe decline in export demand in that scenario; it would also incur huge losses on its net foreign asset position: Europe would lose about $1 trillion if the U.S. current account deficit were halved, and twice that sum if it went to zero. We do not regard our perspective as particularly alarmist. Nouriel Roubini and Brad Setser make the case that the situation is far grimmer than we suggest, with global interest rates set to skyrocket as the dollar loses its status as the premier reserve currency.3 Olivier Blanchard, Francesco Giavazzi, and Filipa Sa present an elegant and thoughtful analysis suggesting that prospective dollar exchange rate changes are even larger than those implied by our model.4 William Cline argues that an unsustainable U.S. fiscal policy has substantially elevated the risk of an adverse scenario.5 In our view, any sober policymaker or financial market analyst ought to regard the U.S. current account deficit as a sword of Damocles hanging over the global economy. Others, however, hold more Panglossian views. One leading benevolent interpretation, variously called the “Bretton Woods II” model or the “Deutsche Bank” view, focuses on China; that view is forcefully exposited in this volume by Michael Dooley and Peter Garber. This theory explains the large U.S. current account deficit as a consequence of the central problem now facing the Chinese authorities: how to maintain rapid economic growth so as to soak up surplus labor from the countryside. For China, a dollar peg (or near peg) helps preserve the international competitiveness
3. Roubini and Setser (2004). 4. Blanchard, Giavazzi, and Sa (this volume). 5. Cline (forthcoming). Mann (2005), although not alarmist, also points to risks in the adjustment process. Of course, similar discussions accompanied earlier U.S. adjustment episodes, but the present situation is quite different in both scale and setting. Krugman (1985, 1991) takes as dim a view as anyone on the sustainability of long-term twin (fiscal and current account) deficits. His views on the 1980s experience would seem to apply with even greater force to the current scene.
70
Brookings Papers on Economic Activity, 1:2005
of exports while attracting foreign direct investment and avoiding stress on the country’s fragile banking system. Is this argument plausible? Set aside the fact that China maintained its peg even through the Asian financial crisis of 1997–98 and as the dollar soared at the end of the 1990s (presumably making Chinese exports much less competitive), or that China risks a classic exchange rate crisis if its fortunes ever turn, say, because of political upheaval in the transition to a more democratic system. The real weakness in the Bretton Woods II theory is that the Chinese economy is still less than half the size of Japan’s, and less than three-quarters the size of Germany’s, at market exchange rates. So, while running surpluses of similar size to China’s relative to their GDP, Germany and Japan actually account for a much larger share of global surpluses in absolute terms. (After all, Germany, not China, is the world’s leading exporter.) And surplus labor is hardly the problem in these aging countries. U.S. Federal Reserve Chairman Alan Greenspan, in a 2003 speech at the Cato Institute and in many subsequent speeches, offers an intriguing argument.6 He agrees that the United States is unlikely to be able to continue borrowing such massive amounts relative to its income indefinitely, and he recognizes that the U.S. current account deficit will therefore narrow substantially someday. Greenspan argues, however, that increasing global financial integration is both what allows the United States to run such large deficits and the saving factor that will greatly cushion the process of unwinding those deficits. We completely agree that increasing global financial integration can explain larger current account deficits, particularly to the extent that greater trade integration helps underpin financial integration, as in our original analysis.7 Indeed, this was a major point of our first approaches to this problem. A narrowing of the U.S. current account deficit must ultimately be the result, however, of more balanced trade, because the trade account is overwhelmingly the main component of the current account. And, as seemingly open as the U.S. economy is to financial flows, international product markets remain quite imperfectly integrated. Thus any correction to the trade balance is likely to entail a very large change in the real effective dollar exchange rate: our baseline figure, which assumes a moderate speed of adjustment and that the world’s major 6. Greenspan (2004). 7. Obstfeld and Rogoff (2000a, 2000b).
Maurice Obstfeld and Kenneth S. Rogoff
71
regions all return to current account balance, is 33 percent. A much smaller dollar devaluation is possible only if the adjustment is stretched over a very long period (say, a decade), in which case labor and capital mobility across sectors and economies can significantly reduce the need for relative price changes. On the other hand, should adjustment take place very abruptly (say, because of a sudden collapse in U.S. housing prices leading to an increase in saving, or a dramatic reallocation of global central bank reserves toward the euro), the potential fall in the dollar is much larger than our baseline estimate of 33 percent, primarily because sticky nominal prices and incomplete exchange rate pass-through hamper adjustment. True, in a recent Federal Reserve study, Hilary Croke, Steven Kamin, and Sylvain Leduc argue that sustained current account imbalances in industrial countries have typically terminated in a relatively benign fashion.8 But their threshold for a current account “reversal”—the country must have run a deficit of at least 2 percent of GDP for three years, and must have improved its current account balance by at least 2 percent of GDP and a third of the total deficit—is a very low bar compared with where the United States stands today. (Croke, Kamin, and Leduc are forced to choose a low threshold, of course, because current account deficits of the size, relative to GDP, of the recent U.S. deficits, although far from unprecedented, are not the norm.) Most important, the United States accounts for over 75 percent of global deficits today, as we have noted, and so any comparison based on the experience of small countries, even small industrial countries, is of limited value. In addition to Chairman Greenspan, a number of academic researchers have emphasized how some important changes in the global financial system, particularly over the past ten years, have changed the nature of international financial adjustment. Philip Lane and Gian Maria Milesi-Ferretti, in a series of papers, have documented the explosion of gross asset flows.9 8. Croke, Kamin, and Leduc (2005). Freund and Warnock (2005) survey current account adjustment in industrial countries and find that deficits tend to be associated with real depreciations, which are larger for consumption-driven deficits. 9. See especially Lane and Milesi-Ferretti (2005a, 2005b). In line with this development, Cooper (2001) identifies ongoing international portfolio diversification as a driving force behind the U.S. deficit. Diversification does not, however, require any net capital flows: even with a balanced current account, foreigners and U.S. residents can still swap assets. According to preliminary estimates by the Bureau of Economic Analysis, for example, private foreign investors added $1.1 trillion in U.S. assets to their portfolios in 2004, far more than that year’s U.S. current account deficit of $666 billion.
72
Brookings Papers on Economic Activity, 1:2005
They and Cédric Tille, as well as Pierre-Olivier Gourinchas and Hélène Rey, have shown that asset revaluation effects from dollar depreciation can have a significant impact on U.S. net financial obligations to foreigners.10 Gourinchas and Rey point out, in fact, that the historical extent of such revaluations suggests that the United States might need to adjust its trade balance by only two-thirds of the amount that would be needed to fully repay its net external debt; even this, however, would still imply very large dollar movements. We agree that the size and composition of gross asset positions are increasingly important, and our model simulations in this paper explicitly take account of the revaluation channel. We find, however, that valuation effects mute the requisite exchange rate changes only modestly. The growing financial globalization that these authors and Chairman Greenspan emphasize is, moreover, a two-edged sword. Enhanced global financial integration may well facilitate gradual current account and exchange rate adjustment, but it might also promote the development of large, unbalanced financial positions that leave the world economy vulnerable to financial meltdown in the face of sharp exchange rate swings. The net foreign asset revaluation channel might help modestly, but a rise in U.S. interest rates could well wipe out the benefits. Because the United States borrows heavily in the form of low-risk bonds, while lending heavily in the form of equities and high-risk bonds, it is especially sensitive to even a modest rise in the interest rates it pays on its foreign debt. Indeed, we show that, in terms of exchange rate adjustments, the adverse effect of a 1.25-percentage-point rise in the interest rate that the United States pays on its short-term foreign debt is similar in magnitude to the benefits gained via the valuation channel, even with a 20 percent dollar depreciation. More generally, although increased global financial integration and leverage can indeed help countries diversify risk, they also expose the system to other vulnerabilities—such as counterparty risk—on a much larger scale than ever before. All in all, although we believe that growing financial globalization is largely a positive development, it does not justify excessive confidence in a benign adjustment process. This paper begins by trying to put the recent U.S. experience with current account imbalances in historical perspective. We hope this first section will provide a useful reference, although some readers will already be famil10. Tille (2004); Gourinchas and Rey (2005a, 2005b).
Maurice Obstfeld and Kenneth S. Rogoff
73
iar with the essential elements. One historical observation that is important for our later analysis is that the United States (so far) has had the remarkable ability to consistently pay a lower rate of interest on its liabilities than it earns on its assets. Some component of this differential in returns has been due to luck, another to huge central bank holdings of U.S. Treasury bills, another perhaps to the unique and central role of the dollar in international finance. Still another, which we have already emphasized, is the fact that Americans hold a much larger share of their foreign assets in equities and high-risk (equity-like) bonds than foreigners hold of U.S. assets (and thus benefit more from the equity premium). An open question is whether this advantage can continue in the face of large and persistent U.S. deficits. We then provide a nontechnical summary of our core three-region (Asia, Europe, and the United States) model. Readers interested in the technical details of our model can read the theoretical section that follows, and the most adventurous can venture into appendix A, where we fully lay out the structure. Our model simulations calibrate the requisite dollar decline against European and Asian currencies under various scenarios. Most of our analysis focuses on real exchange rates, but, by assuming that the regions’ central banks target GDP or consumption deflators (or sometimes, in the case of Asia, exchange rates against the dollar), we are able to extract nominal exchange rate predictions (relative to the initial position) as well. As noted earlier, our baseline simulation, in which Asia’s, Europe’s, and the United States’ current accounts all go to zero, implies that the dollar needs to depreciate in real effective terms by 33 percent (and in nominal terms by a similar amount). Because the trade balance responds to an exchange rate change only with a lag, this exercise slightly overstates the necessary depreciation relative to today’s exchange rates. However, our calibration assumes flexible prices and does not allow for possible exchange rate overshooting, which could significantly amplify the effect. A halving of the U.S. deficit, with counterpart surplus reductions shared by Asia and Europe in the same proportions as in the first simulation (arguably a more likely scenario over the short term) of complete current account adjustment, would lead to a depreciation of the real effective dollar of 17 percent. In our base case the real value of Asian currencies would need to rise by 35 percent and that of European currencies by 29 percent against the dollar. If, however, Asia sticks to its dollar exchange rate peg as the U.S. current account deficit narrows, the real effective value of the European currencies would have to rise by almost 60 percent. Indeed, to maintain its
74
Brookings Papers on Economic Activity, 1:2005
dollar peg in the face of global demand shifts that fully restore U.S. current account balance, Asia would actually have to better than double its already massive current account surplus. Even halving these numbers (corresponding, for example, to the case in which the U.S. current account deficit falls only by half), one can still appreciate the enormous protectionist pressures that are likely to emerge if Asia tries to stick to its dollar peg in the face of a significant pullback in the United States’ voracious borrowing. It is perhaps surprising that, despite Asia’s current account surplus being several times that of Europe (which we define broadly here to include the euro zone and the other largest non-Asian, non-U.S. economies), the required rise in the Asian currencies relative to the European currencies is not even larger in the global rebalancing scenario. As we shall see, a couple of factors drive this result: one is that Asia’s economies are relatively more open than Europe’s to the rest of the world, so that a given exchange rate change has a bigger impact on trade; the other is that a large, unanticipated dollar depreciation inflicts brutal damage on Asia’s net foreign asset position, a factor we explicitly incorporate in our calibrations. The analysis highlights two important but widely misunderstood points about the mechanism of U.S. current account deficit reduction. First, real dollar depreciation is not a substitute for policies that raise U.S. saving, such as reductions in the federal fiscal deficit. Instead, depreciation and saving increases are complements: exchange rate changes are needed to balance goods markets after a change in global consumption patterns, whereas dollar depreciation that is not accompanied by U.S. expenditure reduction will lead to inflationary pressures that, over time, will offset the initial gains in U.S. competitiveness. The second, and related, point is that it makes little sense to ask how much dollar depreciation is needed to reduce the current account deficit by 1 percent of GDP. Exchange rates and current account balances are jointly determined endogenous variables. As the simulations in this paper illustrate, there are numerous different scenarios in which the U.S. external deficit might be erased, all with different implications for the dollar’s foreign exchange value. Although our model is considerably richer than those previously advanced in the literature (including our own earlier studies), it remains subject to a wide range of qualifications and interpretations; we try to emphasize the most important ones. Nevertheless, we view the simulations as quite useful. The paper’s final section highlights the main conclusions that we draw from the technical analysis.
Maurice Obstfeld and Kenneth S. Rogoff
75
The U.S. Current Account and Foreign Wealth Position, 1970–2005 and Beyond The main analytical contribution of the paper is its modeling and numerical calibration of exchange rate and net foreign asset valuation adjustments under alternative scenarios for reducing the U.S. current account deficit. Our framework is intended as a tool for assessing risks and evaluating policy options. At some level, however, the exercise must entail an assessment of how unstable the current trajectory of external payments imbalances really is, along with the likelihood of adjustment taking place in the next few years. In order to think about this overarching issue, it is helpful to understand the history of the problem. Perspectives on the U.S. Deficit Figure 1 traces the U.S. current account balance as a percent of GDP from 1970 to the present. After fluctuating between +1 and −1 percent of GDP during the 1970s, the current account began to go into deep deficit during the mid-1980s, reaching 3.4 percent of GDP in 1987. After recovering temporarily at the end of the 1980s and actually attaining a slight surplus Figure 1. Current Account, 1970–2005a Percent of GDP 1 0 –1 –2 –3 –4 –5 –6 1972
1976
1980
1984
1988
1992
1996
2000
Source:– –Bureau of Economic Analysis, National Income and Product Accounts, International Transactions Accounts. a.––Data for 2005 are projections.
2004
76
Brookings Papers on Economic Activity, 1:2005
in 1991 (propped up by a large, one-time transfer from foreign governments to help pay for the Gulf War), the U.S. current account balance began a slow, steady deterioration throughout the 1990s, which continues today. As already noted, U.S. international borrowing in 2004 accounted for about 75 percent of the excess of national saving over investment of all the world’s current account surplus countries. What are the proximate causes of this profound deterioration in the U.S. external balance? That, of course, is the $666 billion (and rising) question. Since, in principle, the current account balances of all countries should add up to zero, the U.S. current account deficit—equal to the excess of U.S. investment over national saving—has to be viewed as the net result of the collective investment and saving decisions of the entire world. German demographics, OPEC oil revenue investment decisions, depressed investment in Asia—all these factors and many others impinge on global interest rates and exchange rates and, in turn, on U.S. investment and saving. We do not believe there is any simple answer. Nevertheless, U.S. fiscal policy clearly has played a dominant role in some episodes. The current account balance equals, by definition, the sum of government saving less investment plus private saving less investment. Because the Ricardian equivalence of public debt and taxes does not seem to hold in practice, the big Reagan tax cuts of the 1980s almost certainly played a role in the U.S. current account deficits of that era. Similarly, the Bush II tax cuts of the 2000s have likely played a role over the past few years, preventing the current account deficit from shrinking despite the post-2000 collapse in U.S. investment. Currency over- and undervaluations also loomed large in both episodes, usually operating with a lag of one to two years. For example, the peak of the U.S. current account deficit in 1987 lagged by two years the peak of the real trade-weighted dollar exchange rate (figure 2). The weak dollar of the mid-1990s was matched by a pause in the U.S. current account’s decline, and the dollar peak in early 2002 was followed again, with some lag, by a sharp worsening in the external balance. Admittedly, both correlations with the current account deficit— of fiscal deficits and dollar appreciation—are fairly loose. As figure 3 illustrates, U.S. fiscal deficits have expanded massively in recent years compared with those of the rest of the world. But, as the figure also illustrates, Japan has run even larger fiscal deficits relative to its GDP than the United States, yet at the same time it has consistently run the world’s largest current account surplus in absolute terms.
Maurice Obstfeld and Kenneth S. Rogoff
77
Figure 2. Real Effective Exchange Rate of the Dollar, 1973–2004a Index, March 1973 = 100 130 120 110 100 90 80 70 1975
1979
1983
1987
1991
1995
1999
2003
Source:–Federal Reserve data. a.–Broad currency index.
Figure 3. Fiscal Balances in Selected Major Economiesa Percent of GDP 2 0 –2 –4 –6 –8
1999–2001 2002–03 2004–05 Canada
United Kingdom
France
Germany
Source:–2005 OECD Factbook, OECD Economic Outlook No. 77. a.–The data are on a national accounts basis, averaged across years indicated.
Euro area
United States
Japan
78
Brookings Papers on Economic Activity, 1:2005
Figure 4. U.S. Current Account, Saving, and Investment, 1970–2004 Percent of GDP
Percent of GDP Private saving-investment balance (left scale) Government saving-investment balance (left scale)
6 4
15
Private investment (right scale)
2
20
10 0 5 –2 0
–4 Current account balance (left scale)
–6 1972
1976
1980
1984
1988
1992
–5
1996
2000
2004
Source:– –Bureau of Economic Analysis data.
Indeed, during the 1990s the major proximate drivers of the U.S. current account balance were a declining rate of private saving and rising rate of investment. The U.S. personal saving rate, which had been stable at around 10 percent of disposable personal income until 1985, has steadily declined since, reaching a mere 1 percent in 2004. The declining private saving rate has apparently been driven first by the stock price boom of the 1990s and then by the still-ongoing housing price boom.11 Were the U.S. personal saving rate simply to rise to 5 percent of disposable personal income, or halfway toward its level of two decades ago, more than half of today’s current account deficit could be eliminated. During the late 1990s U.S. investment was robust, as shown in figure 4, so that the United States’ high external borrowing really was, in principle, financing future growth. Today, however, the picture has changed. As figure 4 also shows, the main proximate driver of recent U.S. current account deficits has been low private and government saving rather than high
11. Obviously, if one measures saving taking into account capital gains and losses on wealth, the trend decline in saving is much less, although housing wealth is largely not internationally tradable and both housing and securities wealth can evaporate quickly.
Maurice Obstfeld and Kenneth S. Rogoff
79
investment. So much for the prominent view of former Treasury secretary Paul O’Neill, who argued that the U.S. external deficit was driven mainly by foreigners’ desire to invest in productive U.S. assets. The more sophisticated analysis of Jaume Ventura is also inconsistent with declining U.S. investment.12 Another important factor contributing to the U.S. current account deficit since the late 1990s has been the persistently low level of investment in Asia since the region’s 1997–98 financial crisis. Indeed, today, sluggish investment demand outside the United States, particularly in Europe and Japan but also in many emerging markets, is a major factor holding global interest rates down. Low global interest rates, in turn, are a major driver in home price appreciation, which, particularly in the United States with its deep, liquid home-equity loan markets, contributes to high consumption. International Assets, Liabilities, and Returns Naturally, this sustained string of current account deficits has led to a deterioration in the United States’ net foreign asset position, as illustrated in figure 5. In 1982 the United States held net foreign assets equal to just over 7 percent of GDP, whereas now the country has a net foreign debt amounting to about 25 percent of GDP. Accompanying this growth in net debt has been a stunning increase in gross international asset and liability positions, as figure 5 also shows. From 29.5 percent and 22.3 percent of GDP in 1982, U.S. gross foreign assets and liabilities, respectively, had risen to 71.5 percent and 95.6 percent of GDP by the end of 2003. This process of increasing international leverage—borrowing abroad in order to invest abroad—characterizes other industrial country portfolios and is in fact much further advanced for some smaller countries such as the Netherlands and primary financial hubs such as the United Kingdom; see table 1 for some illustrative comparative data.13 The implications of the reduction in U.S. net foreign wealth would be darker but for the fact that the United States has long enjoyed much better 12. Ventura (2001). 13. See also Lane and Milesi-Ferretti (2005a, 2005b) and Obstfeld (2004). The BEA applies market valuation to foreign direct investment holdings starting only in 1982. Gourinchas and Rey (2005b) construct U.S. international position data going back to 1952. In 1976, with foreign direct investment valued at current cost rather than at market value, U.S. gross foreign assets amounted to 25 percent of GDP, and gross foreign liabilities were 12.6 percent of GDP.
80
Brookings Papers on Economic Activity, 1:2005
Figure 5. U.S. Foreign Assets and Liabilities, 1982–2003 Percent of GDP
80 Liabilities 60
Assets
40 20 0 Net foreign assets
–20
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
Source:– –Bureau of Economic Analysis data.
investment performance on its foreign assets than have foreign residents on their U.S. assets. This rate-of-return advantage, coupled with the expansion in foreign leverage documented in figure 5, has so far allowed the United States to maintain a generally positive balance of net international investment income even as its net international investment position has become increasingly negative. Figure 6 shows two measures of U.S. Table 1. International Investment Positions of Selected Industrial Countries, 2003 a Percent of GDP Country Canada Euro area France Germany Italy Japan Switzerland United Kingdom
Gross assets
Gross liabilities
Net position
75 107 179 148 95 87 503 326
93 118 172 141 100 48 367 329
−18 −10 7 6 −5 39 135 −2
Source: International Monetary Fund, International Financial Statistics. a. Gross assets may differ from the sum of gross liabilities and the net position because of rounding.
Maurice Obstfeld and Kenneth S. Rogoff
81
Figure 6. U.S. Net Foreign Investment Income and Total Net Return on Foreign Assets, 1983–2003 Billions of dollars 500
Total net return on foreign assets
400 300 200 100 0 –100
Net foreign investment income
–200 –300 1985
1987
1989
1991
1993
1995
1997
1999
2001
Source: Bureau of Economic Analysis, National Income and Product Accounts, International Transactions Accounts.
net international investment income.14 The first, net foreign investment income (income receipts on U.S. assets owned abroad less income payments on foreign-owned assets in the United States), is taken from the U.S. balance of payments accounts and comprises transactions data only, that is, actual income earned on assets. Interestingly, this balance has not yet entered negative territory, although it could do so soon. Over 1983– 2003 the income return on U.S.-owned assets exceeded that on U.S. liabilities by 1.2 percentage points a year on average. A more comprehensive investment income measure adds the capital gains on foreign assets and liabilities, reflecting price changes that could be due to either asset price movements (such as stock price changes) or exchange rate changes. The Bureau of Economic Analysis (BEA) incorporates estimates 14. Gourinchas and Rey (2005b) present a similar graph covering a much longer period. The estimates in the text are consistent with those found by Obstfeld and Taylor (forthcoming) using a different methodology. For a complementary discussion of returns on foreign assets and liabilities, see Lane and Milesi-Ferretti (2005b).
82
Brookings Papers on Economic Activity, 1:2005
of these gains into its updates of the U.S. international investment position, although they do not appear in the international transactions or national income accounts. As one would expect, figure 6 shows this net income measure to be much more volatile than that based on investment income alone. Although it is negative in some years, cumulatively this balance is even more favorable for the United States than the smoother transactions measure. On average over 1983–2003, the total return on the United States’ foreign investment, inclusive of capital gains, exceeded that on U.S. liabilities to foreigners by a remarkable 3.1 percentage points a year.15 To understand better the implications of the U.S. rate-of-return advantage, let rW be the rate of return on foreign assets, rU the rate of return on liabilities, F the stock of net foreign assets, and L gross liabilities. Then the net total return on the international portfolio is rW F + (rW − rU)L. This expression shows that, even when F < 0 as it is for the United States, total investment inflows can still easily be positive when rW > rU and the stock of gross liabilities is sufficiently large. The expression also reveals, however, that the leveraging mechanism generating the U.S. surplus on investment returns also heightens the risk associated with a possible reversal. An unresolved but critical question is whether the United States’ favorable position in international markets will be sustained in the face of a large and growing external debt. Should the United States at some point be forced to pay a higher rate on its liabilities, the negative income effect will be proportional to the extent of leverage, L. Part of the historical U.S. international investment advantage is a matter of chance and circumstance. Japanese investors famously bought trophy properties like Pebble Beach golf club, Rockefeller Center, and Columbia Pictures at premium prices, only to see those investments sour. Europeans poured money into the U.S. stock market only at the end of the 1990s, just as the technology bubble was about to burst. However, a deeper reason why the United States’ net debt position has accumulated only relatively slowly is that Americans hold a considerably larger fraction of their foreign assets 15. The broad rate-of-return measures for gross assets and liabilities are constructed by adding to the investment income flow the total capital gain on the previous end-of-period assets (or liabilities) and then dividing this total return by the previous value of assets (or liabilities). Thus, in 2003, a year in which the dollar depreciated, the rate of return on U.S. foreign assets was 19 percent, and that on liabilities 8.4 percent. Total capital gains are calculated by subtracting the change in U.S.-owned assets abroad (change in foreign-owned assets in the United States), as reported in the financial account, from the change in U.S. foreign assets (liabilities) at market value, as reported in the BEA international position data.
Maurice Obstfeld and Kenneth S. Rogoff
83
Figure 7. Foreign Exchange Reserves, Selected Countries, Various Years Billions of dollars Dec 1997 Dec 2001 May 2005
800 700 600 500 400 300 200 100 Japan
China
India
NIEsa
ASEAN4b
Sources:–The Economist, Ministry of Finance of Japan. a.–Newly industrializing economies (Brazil, Hong Kong, Korea, Singapore, Taiwan). b.–Four members of the Association of Southeast Asian Nations (Indonesia, Malaysia, Philippines, Thailand).
in equities (both portfolio equity and foreign direct investment) than do foreigners of their U.S. assets. At the end of 2003, Americans held almost $7.9 trillion in foreign assets, of which 60 percent was in equities, either foreign stocks or foreign direct investment (here measured at market value). Foreigners, by contrast, held only 38 percent of their $10.5 trillion in U.S. assets in the form of equity. Given that equity has, over long periods, consistently paid a significant premium over bonds, it is not surprising that U.S. residents have remained net recipients of investment returns even though the United States apparently crossed the line to being a net debtor in the late 1980s. A major reason why foreigners hold relatively more U.S. bonds than Americans hold foreign bonds is that the dollar remains the world’s main reserve and vehicle currency. Indeed, of the $3.8 trillion in international reserves held by central banks worldwide, a very large share is in dollars, and much of it is in short-term instruments.16 Figure 7 illustrates the burgeoning reserves of Asia, now in excess of $2 trillion. According to the 16. See the survey in Central Banking, “The Rise of Reserve Management,” March 2005, p. 14.
84
Brookings Papers on Economic Activity, 1:2005
BEA, over 45 percent of the $700 billion stock of dollar currency is held abroad, and this is probably an underestimate.17 (Note that, when one speaks of the United States enjoying rents or seigniorage from issuing a reserve currency, the main effects may come from foreigners’ relative willingness to hold cash or liquid short-term Treasury debt, rather than from any substantial inherent U.S. interest rate advantage.) In any event, our empirical analysis will take account of the systematically lower return on U.S. liabilities than on assets elsewhere, and will ask what might happen should that advantage suddenly disappear in the process of current account reversal.18 At present, as we have noted, the net U.S. foreign debt equals about 25 percent of GDP. This ratio already roughly equals the previous peak of 26 percent, reached in 1894. A simple calculation shows that if U.S. nominal GDP grows at 6 percent a year and the current account deficit remains at 6 percent of nominal GDP, the ratio of U.S. net foreign debt to GDP will asymptotically approach 100 percent. Few countries have ever reached anywhere near that level of indebtedness without having a crisis of some sort.19 If large, sudden exchange rate movements are possible, the greater depth of today’s international financial markets becomes a potential source of systemic stress. As we have documented, the volume of international asset trading is now vast. Although many participants believe themselves to be hedged against exchange rate and interest rate risks, the wide range of lightly regulated or unregulated nonbank counterparties now operating in the markets raises a real risk of cascading financial collapse. In a world where a country’s current account may adjust abruptly, bringing with it large changes in international relative prices, a persistently large U.S. deficit constitutes an overhanging systemic threat. A sober assessment of present global imbalances suggests the need for a quantitative analysis of how a U.S. current account adjustment would affect exchange rates. We take this up next.
17. See Porter and Judson (1996). 18. Of course, multinationals’ practice of income shifting in response to differing national tax rates on profits distorts reported investment income flows, making an accurate picture of the true flows difficult to obtain. See, for example, Grubert, Goodspeed, and Swenson (1993) and Harris and others (1993). The expansion of gross international positions over the past decade may have worsened this problem. 19. Obstfeld and Rogoff (2000a).
Maurice Obstfeld and Kenneth S. Rogoff
85
Summary of the Analytical Framework Here we summarize the main features and mechanisms in our analysis. After reading this section, readers who are primarily interested in our exchange rate predictions can skip the following section, which presents the details of the model, and proceed directly to the discussion of our numerical findings. We work within a three-region model of a world economy consisting of the United States, Europe, and Asia. These regions are linked by trade and by a matrix of international asset and liability positions. Each region produces a distinctive export good, which its residents consume along with imports from the other two regions. In addition, each region produces nontraded goods, which its residents alone consume. A key but realistic assumption is that each country’s residents have a substantial relative preference for the traded good that is produced at home and exported; that is, consumption of traded goods is intensive in the home export, creating a home bias in traded goods consumption. This feature builds in a “transfer effect” on the terms of trade, which provides one of the key mechanisms through which changes in the international pattern of current account balances change real and nominal exchange rates. A reduction in the U.S. current account deficit, if driven by a fall in U.S. spending and a matching rise in U.S. saving, represents a shift in world demand toward foreign traded goods, which depresses the price of U.S. exports relative to that of imports from both Asia and Europe. (The international terms of trade of the United States deteriorate.) Because the U.S.-produced export good has a larger weight in the U.S. consumer price index (CPI) than that of foreign imports, whereas foreign export goods similarly have larger weights in their home countries’ CPIs, the result is both a real and a nominal depreciation of the dollar. This terms-of-trade effect of current account adjustment has been prominent in the literature, but it is potentially less important quantitatively than is a second real exchange rate effect captured in our model. That effect is the impact of current account adjustment on the prices of nontraded goods. The CPI can be viewed as made up of individual sub-CPIs for traded and nontraded goods, with the latter empirically having about three times the weight of the former in the overall CPI, given the importance of nontraded service inputs into the delivery even of traded products to consumers. The real exchange rate between two currencies is the ratio of the
86
Brookings Papers on Economic Activity, 1:2005
issuing countries’ overall CPIs, both expressed in a common currency. Thus a fall in a country’s prices for nontraded goods, relative to the samecurrency price of nontraded goods abroad, will depress its relative price level just as a terms-of-trade setback does, causing both a real and a nominal depreciation of its currency. Because nontraded goods are so important a component of the CPI, ignoring effects involving their prices would omit much of the effect of current account adjustment on exchange rates. Hence this additional mechanism, absent from much of the policy discussion, is critical to include. When the U.S. external deficit falls as a result of a cut in domestic consumption, part of the reduction in demand falls on traded goods (exports as well as imports), but much of it falls on U.S. nontraded goods. The consequent fall in the nontraded goods’ prices reinforces the effect of weaker terms of trade in causing the dollar to depreciate against the currencies of Europe and Asia. As noted, in our calibration this second effect receives more than twice the weight that terms-of-trade effects receive in explaining exchange rate movements. We consider several scenarios for U.S. current account adjustment, involving different degrees of burden sharing by Europe and Asia and the resulting effect on those regions’ bilateral and effective exchange rates. For example, if Europe’s deficit rises to offset a fall in America’s deficit, while Asia’s surplus remains constant, the dollar will depreciate more against Europe’s currencies, and less against Asia’s, than if Asia and Europe shared in the burden of accommodating the U.S. return to external balance. In terms of its trade-weighted effective exchange rate, the dollar depreciates more under the second of these two scenarios. Because Asia trades more with the United States than Europe does, bilateral depreciation against Asia’s currencies plays the more important role in determining the effective depreciation. We also consider the effect of dollar exchange rate changes in revaluing gross foreign asset positions, thus redistributing the burden of international indebtedness, as well as the possibility that the adjustment process, especially if disorderly, could entail higher interest payments abroad on U.S. short-term foreign obligations. Finally, key parameters in our model govern the substitutability in consumption among various traded goods and between traded and nontraded goods. In general, the lower these substitution elasticities, the greater the relative price changes caused by current account adjustment and the greater, therefore, the resulting terms-of-trade and exchange rate responses. Because the values of these elasticities are
Maurice Obstfeld and Kenneth S. Rogoff
87
quite uncertain and can differ between the short and the long run, we quantitatively examine their role in generating our numerical estimates.
The Model The three-country endowment model we develop here extends our earlier small-country and two-country frameworks.20 We label the three countries (or regions), whose sizes can be flexibly calibrated, U (for the United States), E (for Europe), and A (for Asia). The model distinguishes both between home- and foreign-produced traded goods and between traded and nontraded goods (with the latter margin, largely ignored in many discussions of the U.S. current account deficit, turning out to be the more important of the two quantitatively in our simulations). Our focus here will be on articulating the new insights that can be gained by going from two countries to three, particularly in understanding different scenarios of real exchange rate adjustment across regions as the current account deficit of the United States falls to a sustainable level. Four features of our model are of particular interest. First, by assuming that endowments are given exogenously for the various types of outputs, we implicitly assume that capital and labor are not mobile between sectors in the short run. To the extent that global imbalances close only slowly over long periods (which experience suggests is not the most likely case), factor mobility across sectors will mute any real exchange rate effects.21 Second, we do not allow for changes in the mix of traded goods produced or for the endogenous determination of the range of nontraded goods, two factors that would operate over the longer run and could also mute the effects on real exchange rates of current account movements. Third, our main analysis assumes that nominal prices are completely flexible. That assumption—in contrast to our assumption on factor mobility— almost surely leads us to understate the likely real exchange rate effects of a current account reversal. As we discuss later, with nominal rigidities and imperfect pass-through from exchange rates to prices, the exchange rate will need to move more, and perhaps much more, than in our base
20. See Obstfeld and Rogoff (2000a and 2004, respectively). 21. Obstfeld and Rogoff (1996).
88
Brookings Papers on Economic Activity, 1:2005
case in order to maintain employment stability. Fourth, we do not explicitly model the intertemporal allocation of consumption, but rather focus on the intratemporal price consequences of alternative patterns of productionconsumption imbalances. The Core Model Although notationally intricate, our core three-region model is conceptually quite simple. We assume that consumers in each of the three regions allocate their spending between traded and nontraded goods. Within the category of traded goods, they choose among goods produced in each of the three regions. The equilibrium terms of trade and the relative price of traded and nontraded goods (and thus both bilateral and effective real exchange rates) are determined endogenously. Given assumptions about central bank policy (depending, for example, on whether the central bank aims to stabilize the CPI deflator, the GDP deflator, or a bilateral exchange rate), the model can also generate nominal exchange rates. We begin by defining Cij ≡ country i consumption of good (or good category) j. The comprehensive country i consumption index depends on U.S., European, and Asian traded goods consumption (T), as well as consumption of domestic nontraded goods (N). It is written in the following nested form: (1)
C i = γ (CTi ) 1 θ
θ −1 θ
θ θ −1
+ (1 − γ ) (C Ni ) , i = U , E , A, θ −1 θ
1 θ
with (2)
CTU = α (CUU )
η −1 η
+ (β − α ) (CEU )
CTE = α (CEE )
η −1 η
+ (β − α ) (CUE )
1 η
1 η
CTA = δ (C AA ) 1 η
η −1 η
1 η
1 η
1 η
1 − δ (CEA ) + 2
η −1 η
+ (1 − β ) (C UA )
η −1 η
+ (1 − β ) (C AE )
η −1 η
η −1 η
1 η
η −1 η
1 η
η η −1
η η −1
η η −1
1 − δ (CUA ) . + 2 1 η
η −1 η
We do not assume identical preferences in the three countries. On the contrary, we wish to allow, both in defining real exchange rates and in
Maurice Obstfeld and Kenneth S. Rogoff
89
assessing the effects of shocks, for a realistic home bias in traded goods consumption, such that each country has a substantial relative preference for the traded good that it produces and exports abroad.22 Home consumption bias gives rise to a “transfer effect,” whereby an increase in relative national expenditure improves a country’s terms of trade, that is, raises the price of its exports relative to that of its imports. In the equations above, the United States and Europe are “mirror symmetric” in their preferences for each other’s goods, but each attaches the same weight to Asian goods. Asia weights U.S. and European imports equally but may differ in openness from the United States and Europe. Specifically, we assume that 1 > β > α > 1⁄2. We also assume that δ > 1⁄2. For example, if β = 0.8 and α = 0.7, then the U.S. traded goods consumption basket has a weight of 0.7 on U.S. exports, 0.1 on European exports, and 0.2 on Asian exports. (A very similar—and for many exercises isomorphic— model arises if one assumes that all countries have identical preferences, but that international trading costs are higher than domestic trading costs.)23 The values of the two parameters θ and η are critical in our analysis. Parameter θ is the (constant) elasticity of substitution between traded and nontraded goods. Parameter η is the (constant) elasticity of substitution between domestically produced traded goods and imports from either foreign region. The two parameters are important because they underlie the magnitudes of price responses to quantity adjustments. Lower substitution elasticities imply that sharper price changes are needed to accommodate a given change in quantities consumed. Price Indexes and Real Exchange Rates Using standard methods, we derive exact consumption-based price indexes.24 Define Pji ≡ the country i exact price index for consumption category j. The corresponding overall CPIs, in dollars, are (3)
P = γ ( PTi ) i C
1− θ
+ (1 − γ ) ( P
i N
)
1− θ
22. Warnock (2003) also takes this approach. 23. Obstfeld and Rogoff (2000b). 24. See, for example, Obstfeld and Rogoff (1996).
1 1− θ
, i = U , E , A,
90
Brookings Papers on Economic Activity, 1:2005
where subscript C denotes the comprehensive consumption basket. (Our main analysis is in terms of real prices and exchange rates, so all prices can be expressed in terms of the common numeraire.) In equation 3, (4)
PTU = [ αPU1− η + (β − α ) PE1− η + (1 − β ) PA1− η ] PTE = [ αPE1− η + (β − α ) PU1− η + (1 − β ) PA1− η ]
1 1− η
1 1− η 1 1− η
1 − δ 1− η 1 − δ 1− η P = δPA1− η + P + P . 2 U 2 E A T
Here Pi, i = U, E, A, is just the price of the differentiated traded good produced by country i. We assume the law of one price for traded goods, so that the price of any given country’s traded good is the same in all regions. (In practice, of course, the law of one price holds mainly in the breach, partly because of the difficulties in separating out the truly tradable component of “traded” goods.) Because of the home export consumption bias we have assumed, the price indexes for traded goods PTi can differ across countries even when the law of one price holds, reflecting the asymmetric consumption weightings. As a result, changes in the terms of trade, through their differential effects on countries’ price levels for traded goods, affect real exchange rates. There are three bilateral terms of trade, three bilateral real exchange rates, and three real effective exchange rates. The terms of trade are (5)
τU , E =
τ PE P P , τU , A = A , τ E , A = A = U , A . τU , E PU PU PE
Here, for example, a rise in τU,E is a rise in the price of European traded goods in terms of U.S. traded goods, that is, a deterioration in the U.S. terms of trade. Bilateral real exchange rates are (6)
qU ,E =
q PCE PCA PCA , q = , q = = U ,A . E ,A U ,A E U U qU ,E PC PC PC
A rise in qU,E, for example, is a rise in the price of the European consumption basket in terms of the U.S. consumption basket, that is, a real depreciation of the dollar. As we have noted, asymmetric preferences over traded goods allow the terms of trade to affect traded goods price indexes. The United States’ price index places a comparatively high weight on U.S. exports, whereas
Maurice Obstfeld and Kenneth S. Rogoff
91
Europe’s does the same for its own exports. Thus the U.S. traded goods price index falls relative to Europe’s when Europe’s bilateral terms of trade against the United States improve. Denoting a percent change with a caret, we can logarithmically approximate the evolution of the relative European-to-American traded goods price ratio as (7)
PˆTE − PˆTU = ( 2α − β ) τˆ U ,E .
(Exact formulas for relative price indexes, which we use to generate the numerical results reported below, are given in appendix A.) This expression equates the difference between European and U.S. price inflation in traded goods to the European consumption weight on its own exports, α, less the U.S. consumption weight on imports from Europe, β − α, all multiplied by the percentage increase in Europe’s terms of trade against the United States. Observe that the terms of trade against Asia do not enter this expression. Given the bilateral Europe-U.S. terms of trade, changes in the terms of trade against Asia enter the European and U.S. traded goods price indexes symmetrically (that is, with identical consumption weights of 1 − β) and therefore drop out in computing their log-difference change. Similarly, the evolution of the Asian price level for traded goods relative to that of the United States also reflects terms-of-trade movements. But because, under our assumptions, Asia trades more extensively with Europe than the United States does, the prices of European exports have a relatively bigger impact on Asia’s average import prices. This is shown by the following logarithmic approximation: (8)
1 − δ PˆTA − PˆTU = [ δ − (1 − β )] τˆ U , A + − (β − α ) τˆ U ,E . 2
The weights on the terms-of-trade changes here simply reflect relative consumption weights, as before. Now, however, given the bilateral Asia-U.S. terms of trade, an improvement in Europe’s terms of trade vis-à-vis the United States raises Asia’s price index for traded goods relative to that in the United States when, as we assume in our simulations, the Asian consumption weight on European imports, (1 − δ)/2, exceeds the weight attached by U.S. consumers, β − α. Such third-country asymmetries cannot be captured, of course, in a two-country framework. Bilateral real exchange rate movements follow immediately from the expressions above. For Europe and the United States, for example, the log
92
Brookings Papers on Economic Activity, 1:2005
change in the bilateral real exchange rate is simply the consumption weight on traded goods times the log change in relative traded goods price indexes, plus the consumption weight on nontraded goods times the log change in relative nontraded goods price indexes: (9)
qˆU ,E = γ ( 2α − β ) τˆ U ,E + (1 − γ ) ( PˆNE − PˆNU ) .
Analogously, between the United States and Asia we have (10)
1 − δ qˆU , A = γ [ δ − (1 − β )] τˆ U , A + γ − (β − α ) τˆ U ,E 2 A U ˆ ˆ ( ) ( + 1 − γ P − P ). N
N
We emphasize one key aspect of these expressions. The weight on nontraded goods is likely to be quite large because of the large component of nontradable services included in the consumer prices of goods generally classified as entirely tradable. In our simulations we therefore take the weight on nontraded goods above, 1 − γ, to be 0.75. An implication is that, although the terms of trade certainly are an empirically important factor in real exchange rate determination given home consumption bias, relative prices for nontraded goods potentially play an even larger quantitative role. Solution Methodology The methodology we use to calculate the effects of current account shifts on relative prices is essentially the same as that in our earlier papers, extended to a three-region setting.25 Given fixed output endowments, an assumed initial pattern of current account imbalances, an assumed initial pattern of international indebtedness, and a global interest rate, relative prices are determined by the equality of supply and demand in all goods markets. Changes in the international pattern of external imbalances, whether due to consumption shifts or other changes (including changes in productivity), shift the supply and demand curves in the various markets, resulting in a new set of equilibrium prices. These are the price changes we report below, under a variety of current account adjustment scenarios. (The global sums of external imbalances and of net international asset positions are both constrained to be zero.) 25. The methodology is specified in appendix A and further online at www.economics. harvard.edu/faculty/rogoff/papers/BPEA2005.pdf.
Maurice Obstfeld and Kenneth S. Rogoff
93
There are six market-clearing conditions, covering the three regional nontraded goods markets and the three global markets for traded goods (although one of these is redundant by Walras’ Law). The five independent equilibrium conditions allow solutions for —the U.S. terms of trade against Europe, τU,E —the U.S. terms of trade against Asia, τU,A —the price of nontraded goods in terms of traded goods in the United States, PNU/PUT —the price of nontraded goods in terms of traded goods in Europe, PNE/PET —the price of nontraded goods in terms of traded goods in Asia, PNA/PTA. One can then calculate the three bilateral real exchange rates, for which these five relative prices are the critical inputs. Because of the asymmetric preferences over traded goods, there is, as we have noted, a transfer effect in the model (wealth transfers feed into the terms of trade and through that channel into real exchange rates), although it is more complex than would be the case with only two countries in the world. Finally, we will also want to define and analyze real effective (loosely speaking, trade-weighted) exchange rates: (11)
qU = q = E
q = A
β−α 1− α
(P ) (P ) E C
A C
1− β 1− α
PCU β−α 1− α
(P ) (P ) U C
A C
1− β 1− α
PCE
(P ) (P ) U C
1 2
PCA
E C
1 2
.
Three extensions to the analysis add to its relevance and realism.26 First, we ask how real exchange rate changes translate into nominal exchange rate changes; this depends on central bank policy. In general, this turns out not to be a critical issue empirically; the other two extensions are potentially far more important. One of these is to take into account how exchange rate changes affect the net foreign asset positions of the different regions, because of currency mismatches between gross 26. Details can be found in appendix A and online at www.economics.harvard.edu/ faculty/rogoff/papers/BPEA2005.pdf.
94
Brookings Papers on Economic Activity, 1:2005
assets and liabilities.27 This valuation effect is significant, but its impact on aggregate demand is of secondary importance compared with the primary demand shifts emphasized in our preceding analysis. Finally, our third extension takes into account the effect of a rise in relative U.S. interest rates (due, say, to concern about government deficits or erosion of the dollar’s reserve currency status). This effect, which works to worsen rather than ease the adjustment problem, is also significant, although again it is less important (at least over the range of interest rates we consider) than the primary effects of a rebalancing of global demand. Model Predictions With these critical behavioral parameters in hand, we are now ready to explore the model’s quantitative predictions for global exchange rates and the terms of trade under various scenarios for rebalancing the U.S. current account. We first need to think about parametrizing the model. Choosing Parameters As we have already observed, the critical parameters in the model are θ, the elasticity of substitution in consumption between traded and nontraded goods, and η, the elasticity of substitution in consumption among the traded goods produced by the three regions. The lower are these elasticities, the greater the exchange rate and price adjustments needed to accommodate any interregional shifts in aggregate demand. Most of our simulations will be based on a value of θ = 1, which is high relative to some estimates suggested in the literature.28 We will also report results, however, for an even higher elasticity of θ = 2. Our baseline choice of η = 2 as a representative aggregate trade elasticity is a compromise between two sets of evidence. Estimates based on trade flows within disaggregated product categories cover a wide range 27. As noted above, this effect has recently been emphasized by Tille (2004), Lane and Milesi-Ferretti (2005a, 2005b), and Gourinchas and Rey (2005a, 2005b). 28. Mendoza’s (1991) point estimate is 0.74, Ostry and Reinhart (1992) report estimates in the range 0.66 to 1.28 for a sample of developing countries, and Stockman and Tesar (1995) use an estimate of 0.44. Using a different approach, Lane and Milesi-Ferretti (2004) derive estimates as low as 0.5. Indeed, for larger and relatively closed economies (such as the United States, Europe, and Japan), they suggest that the value should be even lower.
Maurice Obstfeld and Kenneth S. Rogoff
95
but typically include many values much higher than η = 2.29 On the other hand, conventionally estimated aggregate trade equations, as well as calibrations of dynamic general equilibrium models, tend to indicate much smaller values for η, typically 1 or even lower. A number of mechanisms have been suggested to explain this discrepancy, some echoing Guy Orcutt’s classic skepticism about the low elasticities seemingly implied by macro-level estimators.30 Aggregation bias lowers estimated macroelasticities because the price movements of lowelasticity goods tend to dominate overall movements in import and export price indexes.31 Another issue is that macroeconomic estimates of businesscycle frequency correlations tend to confound permanent and temporary price movements, in contrast to micro-level cross-sectional or panel studies centered on trade liberalization episodes.32 In taking η = 2, we try, in a crude way, to address these biases while also recognizing the empirically inspired rules of thumb that inform policymakers’ forecasts. We also include an illustrative simulation of the case η = 100 (in which all traded goods are essentially perfect substitutes). That simulation shuts down the termsof-trade effects and thereby shows how large a role is being played by substitution between traded and nontraded goods, the channel we have emphasized elsewhere.33 We set both α and δ equal to 0.7; these are the consumption weights that Americans and Europeans, on the one hand, and Asians, on the other, attach to their own domestic products within their traded goods consumption baskets. That choice is plausible based on our discussion in an earlier 29. Examples are the estimates of Feenstra (1994) and the more recent figures of Broda and Weinstein (2004). 30. Orcutt (1950). 31. For an excellent example of this bias in action, see Hooper, Johnson, and Marquez (2000), who report that, because oil and tourism demand are relatively price-inelastic, trade equations based on aggregates that include oil and services imply apparently much lower price elasticities than equations for nonoil manufactures only. For the Group of Seven countries, Hooper, Johnson, and Marquez report short-run price elasticities for imports and exports (including oil and services) that in most cases do not satisfy the Marshall-Lerner condition. We view the elasticities implied even by aggregated estimates that exclude oil and services as unreasonably low; but, if they are accurate, they imply larger terms-of-trade and real exchange rate effects of international spending shifts. 32. See Ruhl (2003). Our model omits not only dynamics of the type suggested by Ruhl, but also those resulting from the introduction of new product varieties, which would act over the longer run to dampen the extent to which a rise in a country’s relative productivity lowers its terms of trade. See, for example, Krugman (1989) and Gagnon (2003). 33. Obstfeld and Rogoff (2000a).
96
Brookings Papers on Economic Activity, 1:2005
paper.34 We set β = 0.8, implying that Europe and the United States alike place weights of β − α = 0.1 on each other’s traded goods, and twice that weight (0.2) on Asian goods. Asia, by assumption, distributes its demand evenly across the other two regions (placing a weight of 0.15 on the exports of each). So, in our model, Europe and the United States both trade more with Asia than with each other. We assume that all three regions produce the same number of units of tradable goods output. Appendix A discusses in detail our assumptions regarding gross liabilities and assets for each region, as well as the currencies of denomination of these stocks. The point we stress here is that, to a first approximation, the United States is a net debtor (to the tune of 25 percent of its GDP, or 100 percent of its exportable GDP), and greater Europe has approximately a zero net international position. Our model’s third region, Asia, therefore is left as a net international creditor in an amount equal to 100 percent of U.S. tradable GDP. U.S. gross foreign liabilities are almost all in dollars, but U.S. gross foreign assets are only about 40 percent in dollars. We assume that greater Asia’s gross liabilities are equally divided among U.S., European, and Asian currencies (because Japan borrows in yen), whereas Asian gross foreign assets are 80 percent in dollars and 20 percent in European currencies. For Europe we assume that gross foreign assets are 32 percent in dollars, 11 percent in Asian currencies, and 57 percent in European currencies. In our model, 80 percent of European gross liabilities are denominated in European currencies, and the balance in dollars. These numbers are very rough approximations, based in some cases on fragmentary or impressionistic data, but portfolio shares can shift sharply over time, and so there is little point in trying too hard to refine the estimates. As we shall see, these shares do imply large potential international redistributions of wealth due to exchange rate changes, but those redistributions themselves have only a secondary impact on the exchange rate implications of current account adjustment. For nominal interest rates we take a baseline value of 3.75 percent a year for U.S. liabilities but 5 percent a year for all other countries’ liabilities. This assumption captures the “exorbitant privilege” the United States has long enjoyed of borrowing from the world more cheaply than it lends.35 34. Obstfeld and Rogoff (2000b). 35. The phrase “exorbitant privilege” is commonly but wrongly attributed to French president Charles de Gaulle. For its true origin, see the interesting historical note provided by Gourinchas and Rey (2005b).
Maurice Obstfeld and Kenneth S. Rogoff
97
Turning to current accounts, we place the U.S. external deficit at 20 percent of U.S. tradable GDP.36 This is consistent with a U.S. current account deficit of 5 percent of total GDP, a reasonable baseline if part of the 2004 deficit is due to temporarily high oil prices. Because we find our simulation results to be approximately linear within the parameter space we are considering, it is easy to adjust the prediction to the case in which the 2004 deficit of 6 percent of GDP persists. In any event, what matters most for our calibration is how much the current account balance adjusts (for example, from 6 to 3 percent of GDP). We assume an initial position with Europe’s current account surplus at 5 percent of U.S. tradable GDP and Asia’s at 15 percent.37 A final benchmark to establish is our initial reference value for measuring subsequent exchange rate adjustments. This issue was less critical in our earlier two papers, because trade-weighted effective exchange rates move more slowly than the bilateral exchange rates that we consider below. In our basic model prices are flexible and economic responses to them are immediate. In practice, however, there are considerable lags: Michael Mussa, for example, posits the rule of thumb that the U.S. trade balance responds with a two-year lag to dollar exchange rate changes.38 In that case, if today’s current account balances reflect averages of exchange rates over the past two years, it would be more accurate to think of our simulations as giving exchange rate changes relative to two-year average reference rates rather than current rates. Table 2 presents some resulting reference exchange rates. (The Chinese and Malaysian currencies have been pegged over the past two years, and so their current and average rates are the same.) Simulations With the model and our parameter assumptions in hand, we are ready to consider alternative simulations. Underlying much of our analysis is the assumption that demand shocks (such as a rise in U.S. saving) are driving the redistribution of global imbalances. This seems by far the most realistic assumption, given the magnitude of the external financing gaps.
36. As noted earlier, we estimate tradable GDP to be at most 25 percent of total GDP. 37. It would be interesting and useful to extend the model to include emerging markets and OPEC as a composite fourth region, as suggested by our discussant T. N. Srinivasan. 38. Mussa (2005).
98
Brookings Papers on Economic Activity, 1:2005
Table 2. Recent and Two-Year-Average Exchange Rates of Selected Currencies Currency units per dollar except where noted otherwise Exchange rate Currency U.K. pound sterling Canadian dollar Euroa Korean won New Taiwan dollar Singapore dollar Japanese yen
a
As of June 1, 2005
Two-year average
1.81 1.25 1.22 1,010 31.30 1.67 108.4
1.79 1.23 1.23 1,129 33.21 1.69 109.3
Source: Federal Reserve data. a. In dollars per indicated currency unit.
Tables 3 through 6 lay out the results of three scenarios under which the U.S. current account balance might improve by 20 percent of tradable GDP or, equivalently, 5 percent of total GDP. (All simulations include the effect of exchange rate changes in revaluing the regions’ foreign assets and liabilities.) In the “global rebalancing” scenario (the first column in each table), all regions’ current account balances go to zero (with trade balances adjusting as needed to service interest flows on the endogenously determined stocks of net foreign assets). Looking first at bilateral real exchange rates, in table 3, we see that Asia’s exchange rate with the United States rises by 35.2 percent, and Europe’s rises by 28.6 percent (we define the real exchange rate such that these changes indicate real depreciations of the dollar). Europe sees an improvement in its terms of trade against the United States (a rise in the price of Europe’s exports relative to its U.S. imports) of 14.0 percent, and Asia sees an improvement of 14.5 percent. What are the implications for nominal exchange rates? To answer this question we must specify monetary policies. We consider two possibilities: that central banks stabilize the domestic CPI, and that they stabilize the domestic GDP deflator. Table 4 reports the results. Under CPI targeting, the monetary authorities hold overall price levels constant, so that the only source of real exchange rate change is nominal exchange rate change. As a result, nominal and real exchange rate changes are equal, as can be seen by comparing table 4 with table 3.39 Because none of the three regions is extremely open to trade, movements in CPIs and in GDP deflators are 39. We provide a detailed account of nominal exchange rate determination under GDP deflator targeting at www.economics.harvard.edu/faculty/rogoff/papers/BPEA2005.pdf.
Maurice Obstfeld and Kenneth S. Rogoff
99
Table 3. Changes in Real Exchange Rates and Terms of Trade Following U.S. Current Account Adjustment under Baseline Assumptionsa Log change × 100 Adjustment scenario Real exchange rate or terms of trade
Global rebalancingb
Real exchange rate United States/Europe United States/Asia Europe/Asia Terms of trade United States/Europe United States/Asia Europe/Asia
Bretton Woods IIc
Europe and United States trade placesd
28.6 35.2 6.7
58.5 −0.5 −59.0
44.6 19.4 −25.2
14.0 14.5 0.5
29.4 7.2 −22.2
22.0 11.1 −10.8
Source: Authors’ calculations using model described in the text. a. Exchange rates are defined such that an increase represents a real depreciation of the first region’s currency against the second’s; terms of trade are defined such that an increase represents a deterioration for the first region (that is, a fall in the price of the first region’s export good against the second). Assumed parameter values are as follows: substitution elasticity between traded and nontraded goods θ = 1; substitution elasticity between traded goods of different regions η = 2; share of traded goods in total consumption γ = 0.25. b. Current account balances of all three regions go to zero. c. Asia’s current account surplus rises to keep its exchange rate with the dollar fixed. Europe’s current account absorbs all changes in the U.S. and Asian current accounts. d. Europe absorbs the entire improvement in the U.S. current account balance while Asia’s current account balance remains unchanged.
Table 4. Changes in Nominal Exchange Rates Following U.S. Current Account Adjustment under Alternative Inflation Targetsa Log change × 100 Adjustment scenario Nominal exchange rate Target is consumer price indexb United States/Europe United States/Asia Europe/Asia Target is GDP deflator United States/Europe United States/Asia Europe/Asia
Global rebalancing
Bretton Woods II
Europe and United States trade places
28.6 35.2 6.7
58.5 −0.5 −59.0
44.6 19.4 −25.2
30.0 36.9 6.9
61.4 0.0 −61.4
46.8 20.6 −26.3
Source: Authors’ calculations using model described in the text. a. See table 3 for definitions of exchange rates, scenarios, and parameter assumptions. b. With flexible prices and CPI targeting by central banks, nominal exchange rate changes are equal to the real exchange rate changes reported in table 3.
100
Brookings Papers on Economic Activity, 1:2005
Table 5. Changes in Real and Nominal Effective (Trade-Weighted) Exchange Rates Following U.S. Current Account Adjustment under Baseline Assumptionsa Log change × 100 Adjustment scenario Effective exchange rateb U.S. real U.S. nominal Europe real Europe nominal Asia real Asia nominal
Global rebalancing −33.0 −34.6 5.1 5.4 20.9 21.9
Bretton Woods II −19.1 −20.5 58.9 61.4 −29.8 −30.7
Europe and United States trade places −27.8 −29.3 31.7 33.1 −2.9 −2.9
Source: Authors’ calculations using model described in the text. a. See table 3 for definitions of scenarios and parameter assumptions. An increase is an appreciation of the indicated currency against foreign currencies. b. Nominal exchange rate changes are calculated under the assumption of GDP deflator targeting; see appendix A for details.
fairly close, and, as a result, nominal exchange rate changes when the GDP deflator is stabilized differ very little from those under CPI stabilization. The appreciation of Europe’s currencies against the dollar is smaller than that of Asia’s under the first scenario, because Asia starts out in our simulation with a much larger external surplus than Europe does, and so it has more adjusting to do. But the Asian currencies’ appreciation against the dollar is mitigated somewhat by the fact that Asia trades more with the United States than Europe does.40 We see in table 5 that Europe’s real effective currency appreciation—represented, as is traditional for such multilateral indexes, by a positive number—is much smaller than Asia’s: only 5.1 percent versus 20.9 percent. Again, this reflects the greater weight of the dollar in Asia’s trade-weighted real exchange rate than in Europe’s. Notice that, as in table 4, nominal (under GDP deflator targeting) and real effective exchange rate changes are again quite close numerically. Another factor underlying the equilibrium exchange rate responses is that dollar depreciation implies a much bigger reduction in Asia’s net foreign asset position than in Europe’s. (Table 6 shows the impacts under GDP 40. Indeed, if one recalibrates the model so that β = 0.85 (in which case all countries’ preferences are completely symmetric, so that Europeans and Americans no longer prefer Asian goods to each other’s), then, in the global rebalancing scenario, Asia’s currency appreciates in real terms against the dollar by 37.8 percent and against European currencies by 12.2 percent. These numbers exceed the 35.2 percent and 6.7 percent reported in table 3.
Maurice Obstfeld and Kenneth S. Rogoff
101
Table 6. Net Foreign Assets by Region Following U.S. Current Account Adjustmenta Ratio to value of U.S. traded goods output Adjustment scenariob
Region United States Europe Asia
Baseline net foreign asset position
Global rebalancing
Bretton Woods II
Europe and United States trade places
−1.0 0.0 1.0
−0.3 −0.1 0.4
−0.1 −0.7 0.8
−0.2 −0.4 0.6
Source: Authors’ calculations using model described in the text. a. See table 3 for definitions of exchange rates, scenarios, and parameter assumptions. b. Net asset positions taking into account valuation effects of changes in nominal exchange rates under GDP deflator targeting.
deflator targeting.) Asia has 80 percent of its assets, but only 34 percent of its liabilities, in dollars. Thus, under the global rebalancing scenario, dollar depreciation raises Asia’s gross liabilities relative to its gross assets, pushing its net foreign assets down (as a fraction of U.S. tradable GDP) by 60 percent. Europe, by contrast, has only 32 percent of its assets and 20 percent of its liabilities in dollars. The fact that Asia loses so much on the asset side implies that its trade surplus shrinks by less than its current account surplus does. Because trade surpluses are what drive the constellation of real exchange rates, the real appreciation of the Asian currencies is mitigated. In sum, thanks to Asia’s greater openness and to the fact that Asia suffers particularly large capital losses on foreign assets when the dollar falls, Asian exchange rates do not need to change quite as much as a model-free, back-of-the-envelope calculation might suggest. The tables cover two other possible scenarios. The second column in tables 3 through 6 analyzes a “Bretton Woods II” scenario, in which Asia clings to its dollar peg.41 We calibrate this case by setting the U.S. current account balance to zero and endogenously varying Asia’s and Europe’s current account balances in a way that both maintains Asia’s bilateral nominal exchange rate with the United States (assuming GDP deflator targeting) and absorbs the fall in U.S. borrowing. (Of course, nonmonetary policy instruments such as fiscal policy would have to be used to attain just the right constellation of current account balances.) In this case the bilateral real exchange rates of the European currencies against the dollar 41. Dooley, Folkerts-Landau, and Garber (2004a, 2000b).
102
Brookings Papers on Economic Activity, 1:2005
must rise spectacularly, by 58.5 percent, and they would rise against the Asian currencies by 59 percent. This result also is approximately linear in the change in the U.S. current account balance. Thus, under the Bretton Woods II scenario, eliminating only half the U.S. current account deficit would raise the real value of the European currencies against the dollar by as much as would occur in a global rebalancing scenario that eliminates the U.S. current account deficit entirely. For Asia to maintain its nominal exchange rate peg in the face of a balanced U.S. current account, it must drive its own current account balance significantly further into surplus, from 15 percent to 31 percent of U.S. tradable GDP. And Europe would have to move from a surplus equal to 5 percent of U.S. tradable GDP to a 31 percent deficit! (See the footnotes to table 3.) When Asia pegs its currencies to a falling dollar, its own traded goods become more competitive and its imports more expensive relative to domestic nontraded goods. Both factors shift world demand away from Europe, which, by assumption, is passively absorbing the blow, and toward Asia. These calibrations make patently clear why sustaining Asia’s dollar peg is likely to be politically unpalatable for many of its trading partners if the U.S. current account deficit ever shrinks. Asia would be extremely vulnerable to a protectionist backlash. As table 6 shows, the sharp appreciation of Europe’s currencies in the Bretton Woods II scenario also decimates its external asset position, which declines from balance to −70 percent of the value of U.S. tradable production. Asia suffers somewhat, and the U.S. net asset position is the major beneficiary, because U.S.-owned foreign assets are concentrated in European currencies. Europe is thus hammered both by a sharp decline in its competitiveness and by a loss on its net foreign assets of about $2 trillion. The third scenario reported in tables 3 through 6 is a muted version of the Bretton Woods II scenario. Here, instead of maintaining its dollar currency peg, Asia maintains its current account surplus unchanged in the face of U.S. adjustment to a balanced position. That is, rather than increasing its current account surplus, it allows enough exchange rate adjustment to keep the surplus constant. In this case, as table 5 shows, Europe’s real effective exchange rate rises by much less than in the Bretton Woods II scenario (31.7 percent versus 58.9 percent), and the Asian currencies experience a real effective depreciation of only 2.9 percent, versus 29.8 percent in Bretton Woods II. Still, because the U.S. current account balance
Maurice Obstfeld and Kenneth S. Rogoff
103
improves dramatically while Asia’s holds steady, the Asian currencies rise in real terms by 19.4 percent against the dollar (table 3). This exercise reveals a fallacy in the argument that Asia cannot allow its dollar peg to move without losing the ability to absorb its surplus labor. To the extent that European demand increases, Asia can retain its external surplus while releasing its dollar peg. In table 7 we revisit the global rebalancing scenario but vary the critical substitution elasticities in the model. (Only real exchange rate changes, which equal nominal changes under CPI inflation targeting, are listed.) In the first column we assume an elasticity of substitution between traded and nontraded goods, θ, of 2 instead of 1. As we have already argued, the limited evidence in the empirical macroeconomics literature suggests that this estimate is well on the high side, but it allows us to incorporate a more conservative range of potential exchange rate adjustments alongside our baseline estimates. Under this assumption the real dollar exchange rate with the European currencies rises by only 19.3 percent, instead of 28.6 percent as in the first column of table 3, and the Asian currencies rise against the dollar by 22.5 percent instead of 35.2 percent. The dollar falls in real effective terms (results not shown) by 21.5 percent rather than 33 percent. These calculations show that, even with a relatively high value for θ, the required adjustment of exchange rates is quite significant even if, as here, prices are flexible. Table 7. Changes in Real Exchange Rates and Terms of Trade in the Global Rebalancing Scenario under Alternative Calibrationsa Log change × 100
Real exchange rate or terms of trade Real exchange rate United States/Europe United States/Asia Europe/Asia Terms of trade United States/Europe United States/Asia Europe/Asia
Higher elasticity of substitution between traded and nontraded goods (θ = 2, η = 2)
Very high elasticity of substitution between regions’ traded goods (θ = 1, η = 100)
19.3 22.5 3.3
16.5 23.5 7.0
14.6 15.1 0.5
0.0 0.0 0.0
Source: Authors’ calculations using model described in the text. a. In the global rebalancing scenario all regions’ current account balances go to zero. See table 3 for definitions of exchange rates and other parameter assumptions.
104
Brookings Papers on Economic Activity, 1:2005
The second column in table 7 examines the case in which θ = 1 but η = 100, so that the various countries’ tradable outputs are almost perfect substitutes. This exercise, which essentially eliminates terms-of-trade adjustments as a factor in moving real exchange rates, allows us to see how much of the change in exchange rates is due to within-country substitution between traded and nontraded goods. This variation mutes the exchange rate changes by an amount roughly similar to those found in the previous exercise. The real effective dollar exchange rate (again not shown) falls by 21 percent. According to this calibration, roughly two-thirds of the needed dollar adjustment is driven by substitution between traded and nontraded goods, and only one-third is driven by the terms-of-trade channel typically emphasized in the literature. This should not be surprising, given that (according to our previously cited calibration) roughly 75 percent of GDP is nontraded. With more conservative assumptions about international trade, however (either greater home bias in consumption or lower substitutability of countries’ traded outputs, such that η = 1), the terms-of-trade channel would become more important. At present the United States is absorbing traded goods (domestic and foreign) equivalent to roughly 30 percent of its GDP. This demand needs to adjust downward while avoiding a reduction in nontraded goods absorption if full employment is to be maintained; such a shift will therefore require a significant change in the relative price of nontraded goods. Still, termsof-trade changes do account for about one-third of the overall adjustment, a proportion slightly larger than that found in our two-country model, where we did not allow for trade or terms-of-trade adjustments between non-U.S. economies. Given the United States’ leveraged international portfolio, with gross debts mostly in dollars and assets significantly in foreign currencies, an unexpected dollar depreciation reduces the U.S. net foreign debt. The first two columns of table 8 report the results of simulations, within the global rebalancing scenario, that illustrate the quantitative importance of such asset valuation effects. Gourinchas and Rey have recently estimated that nearly one-third of the settlement of the U.S. net foreign debt has historically been effected by valuation changes, with the remaining two-thirds covered by higher net exports.42 The first column in table 8 shows results
42. Gourinchas and Rey (2005a).
Maurice Obstfeld and Kenneth S. Rogoff
105
Table 8. Changes in Real Exchange Rates and Terms of Trade in the Global Rebalancing Scenario with and without Valuation and Interest Rate Effectsa Log change × 100 Real exchange rate or terms of trade Real exchange rate United States/Europe United States/Asia Europe/Asia Terms of trade United States/Europe United States/Asia Europe/Asia
With valuation effects and without interest rate effectsb
Without valuation effects or interest rate effects
With valuation effects and interest rate effectsc
28.6 35.2 6.7
33.7 40.7 7.0
30.1 37.2 6.3
14.0 14.5 0.5
16.5 16.5 0.0
15.1 15.3 0.2
Source: Authors’ calculations using the model described in the text. a. In the global rebalancing scenario all regions’ current account balances go to zero. See table 3 for definitions of exchange rates and other parameter assumptions. b. Same as the baseline scenario reported in first column of table 3. c. Interest rates on U.S. short-term liabilities held by foreigners are assumed to rise 1.25 percentage points, to the same level as the return earned by U.S. residents abroad.
for the global rebalancing scenario with valuation effects taken into account (identical to the first column in table 3). The second column shows the changes in bilateral exchange rates that would be required if there were no valuation effects (or, equivalently, if exchange rate changes were accurately anticipated and nominal returns adjusted fully to compensate). All relative price changes against the United States are larger in this case, because the United States does not get the benefit of a sharp reduction in its net dollar liabilities. Correspondingly, the U.S. trade balance needs to adjust more for any given adjustment in the current account deficit. The real exchange rate between the dollar and the European currencies needs to move by 33.7 percent, rather than 28.6 percent when valuation effects are taken into account, and the real value of the Asian currencies needs to rise by 40.7 percent against the dollar instead of 35.2 percent. The real effective dollar exchange rate falls by 37.8 percent instead of 33.0 percent (results not shown). According to these numbers, asset revaluation effects will mute the required movement in exchange rates as the U.S. current account closes up, but the trade balance has to do the heavy lifting, since 87 percent (33.0 ÷ 37.8) of the necessary real exchange rate adjustment remains. That valuation effects have only a secondary effect on equilibrium relative price changes is not surprising: big valuation effects can only come from big exchange rate movements.
106
Brookings Papers on Economic Activity, 1:2005
Our calculations so far do not take into account the likelihood of an accompanying rise in global interest rates, which would hurt the United States (a net debtor) and help Asia (a net creditor). A broad range of scenarios are possible here; we examine only a single very simple one. (Appendix A gives details of the calculation.) In the third column of table 8, we assume that annual interest rates on short-term U.S. debt rise from 3.75 percent to 5 percent, the same level assumed for all other liabilities. In other words, perhaps because of heightened risk perceptions, the United States simply loses its historical low borrowing rate and is put on a par with other debtors. This change wipes out a good deal of the effect of the valuation changes (and would wipe out even more if it applied to all U.S. external liabilities, not just the roughly 30 percent consisting of short-maturity debt). As our introductory discussion suggested, the United States, as an important issuer of bonds relative to equity, is extremely vulnerable to increases in interest rates, even when all global bond rates rise together. Until now we have been concentrating on demand shocks. Productivity shocks may make the adjustment process more or less difficult, depending on their source. Higher productivity in foreign traded goods production can actually result in an even greater real depreciation of the dollar as equilibrium is reestablished in world markets. If, on the other hand, it is nontraded goods productivity in Asia and Europe that rises, the exchange rate effects of global rebalancing will be muted. As table 9 illustrates, a Table 9. Changes in Real Exchange Rates and Terms of Trade in Global Rebalancing Scenario with Higher Productivity in Non-U.S. Nontraded Goodsa Log change × 100 Real exchange rate or terms of trade Real exchange rate United States/Europe United States/Asia Europe/Asia Terms of trade United States/Europe United States/Asia Europe/Asia
Without increase in productivityb
With 20 percent increase in productivity in European and Asian nontraded goods
28.6 35.2 6.7
17.0 23.6 6.6
14.0 14.5 0.5
15.0 15.3 0.2
Source: Authors’ calculations using model described in the text. a. In the global rebalancing scenario all regions’ current account balances go to zero. See table 3 for definitions of exchange rates and other parameter assumptions. b. Same as the baseline scenario reported in the first column of table 3.
Maurice Obstfeld and Kenneth S. Rogoff
107
20 percent rise in nontraded goods productivity outside the United States implies notably smaller real exchange rate changes, although the terms-oftrade shifts are similar. A large rise in U.S. traded goods productivity would also facilitate a softer landing. In this case, however, although the extent of real dollar depreciation is somewhat reduced, the U.S. terms of trade fall much more sharply (results not reported).
Some Further Considerations We believe our model offers many useful insights, but of course there are many caveats to its interpretation. Some of these suggest that our results understate the dollar’s potential decline, and some that they overstate it. Intersectoral Factor Mobility A critical implicit assumption of our model is that capital and labor cannot quickly migrate across sectors, so that prices rather than quantities must bear the burden of adjustment in response to any sudden change in relative demands for different goods. This assumption seems entirely reasonable if global current account adjustment (full or partial) takes place moderately quickly, say, over one to two years. In the short run, workers cannot change location easily, worker retraining is expensive, and a great deal of capital is sector-specific. Over much longer periods, however (say, ten to twelve years), factor mobility is considerable. If, for example, prices rise dramatically in the U.S. traded goods sector, new investment will be skewed toward that sector, as will new employment. Thus, in principle, a gradual closing of the U.S. current account deficit would facilitate much smoother adjustment with less exchange rate volatility. Unfortunately, our model is not explicitly dynamic.43 One can, however, artificially approximate gradual current account adjustment by allowing for progressively higher elasticities of substitution. We do this in table 10, where we reconsider our central scenario (which assumed θ = 1 and η = 2) by comparing it with two cases in which substitution elasticities are much higher. As the table shows, in the case with “gradual” unwinding (proxied
43. For an example of a dynamic approach see the small-country q-model analysis in Obstfeld and Rogoff (1996, chapter 4).
108
Brookings Papers on Economic Activity, 1:2005
Table 10. Changes in Real Exchange Rates under Alternative Assumed Speeds of Global Rebalancinga Log change × 100 Speedb Real exchange rate United States/Europe United States/Asia Europe/Asia
Moderate (1–2 years)c
Gradual (5–7 years)
Very gradual (10–12 years)
28.6 35.2 6.7
13.4 17.3 3.9
6.5 8.5 2.0
Source: Authors’ calculations using model described in the text. a. In the global rebalancing scenario all regions’ current account balances go to zero. See table 3 for definitions of exchange rates and other parameter assumptions. b. Proxied by varying elasticities of substitution: moderate, θ = 1, η = 2; gradual, θ = 2, η = 4; very gradual, θ = 4, η = 8. c. Same as the baseline scenario reported in the first column of table 3.
by θ = 2 and η = 4), which we loosely take to capture a five- to seven-year adjustment horizon, the bilateral exchange rate changes involving the dollar are only about half as big as in our central global rebalancing scenario. For a “very gradual” unwinding (which we take to occur over ten to twelve years, with θ = 4 and η = 8), the same real exchange rate changes are less than a quarter as large as in the central scenario. Sticky Prices Factor mobility kicks in to smooth current account adjustment if the adjustment is slow and relatively well anticipated. If, on the other hand, current account imbalances have to close up very quickly (say, because of a collapse in U.S. housing prices), the bias in our estimates would point in the other direction. Nominal rigidities in prices would then play a large role, and actual exchange rate movements would likely be two or more times as large as in our central scenario, for several reasons.44 For one thing, our model assumes that the law of one price holds for traded goods, whereas in fact at most half of an exchange rate adjustment typically passes through to traded goods prices even after one year.45 Thus, in order to balance supply and demand for the different categories of goods
44. See the discussion in Obstfeld and Rogoff (2000a). 45. P. Goldberg and Knetter (1997); Campa and L. Goldberg (2002). For recent evidence suggesting a substantial decline in pass-through to U.S. import prices, see Marazzi and others (2005).
Maurice Obstfeld and Kenneth S. Rogoff
109
while maintaining full employment, central banks would have to allow much larger exchange rate movements—possibly double those suggested by the model. These larger movements would be “overshoots” in the sense that they would unwind over time as domestic prices adjust. The nominal prices of nontraded goods are typically even stickier than those of traded goods; this further amplifies the overshooting effect. In general, both sticky prices and slow factor mobility point toward the likelihood that a slow unwinding of the U.S. current account deficit will lead to smaller changes in real exchange rates than would a relatively abrupt correction. Rising U.S. Interest Rates and the Dollar Another qualification to our results is that our model does not account for financial factors, and in particular for the possibility of temporarily high real interest rates in the United States muting the dollar’s decline. Using the Federal Reserve’s macroeconomic model, David Reifschneider, Robert Tetlow, and John Williams estimate that a 1-percentage-point rise in the federal funds rate (presumably unmatched by the rest of the world) leads to a 2.2 percent appreciation of the dollar after one year, and a 4.9 percent appreciation after two years.46 Therefore the fact that, over the past year, U.S. short-term interest rates have been rising relative to Europe’s is a countervailing consideration to those discussed above (although our calculations suggest that it is likely to be far less important quantitatively). In addition, Europe and Asia can always choose to lower their interest rates to further mute the dollar’s decline. Of course, interest rate policy can only affect the dollar’s real value temporarily, and so long-term global rebalancing will still require a combination of real exchange rate adjustment and factor reallocation across sectors. The Fundamental Unpredictability of Exchange Rates Our model suggests that the gaping U.S. current account deficit is a very large negative factor in assessing the future prospects of the dollar. It
46. Reifschneider, Tetlow, and Williams (1999). A back-of-the-envelope calculation based on the Dornbusch overshooting model (Dornbusch, 1976) yields a similar result.
110
Brookings Papers on Economic Activity, 1:2005
is well known, however, that it is extremely difficult to explain exchange rate swings between major currencies, much less forecast them, at least at horizons up to eighteen months.47 Although a number of small qualifications must be made to this result,48 it remains broadly true. How, then, can one be concerned about a medium-term dollar decline if a rise is equally likely? There are two broad answers to this question. First, even the most cheery U.S. current account optimist would have to concede that an abrupt reversal is a potential risk, particularly while federal government deficits remain less than fully tamed. Reversal need not result from what Guillermo Calvo, in the context of emerging markets, has called a “sudden stop” of capital inflows;49 as we have noted, it could follow, for example, from a rise in U.S. saving due to a purely domestic asset price collapse. Our calibrations are useful in laying out the exchange rate consequences and in illuminating how the burden of adjustment might be shared among the major economies. Second, and more fundamentally, there is some evidence that nonlinearities are also important, so that, when exchange rates are particularly far out of line with one or more fundamentals, some predictability emerges. Obstfeld and Alan Taylor, for example, argue that convergence to purchasing power parity is much more important quantitatively when a currency is relatively heavily over- or undervalued compared with its long-term real exchange rate.50 Gourinchas and Rey argue that, contrary to the canonical Meese-Rogoff result, there is a forecastable component to tradeweighted dollar exchange rate movements when net foreign assets or debts are large relative to the United States’ net export base.51 Their work supports much earlier work by Peter Hooper and John Morton suggesting that net foreign assets may be important in explaining dollar movements.52 As we argued in the introduction, the U.S. current account deficit today is so large and unprecedented that it is difficult to project its future path and the consequences thereof simply by extrapolating from past data.
47. 48. 49. 50. 51. 52.
Meese and Rogoff (1983). See the survey in Frankel and Rose (1995), for example. Calvo (1998). Obstfeld and Taylor (1997). Gourinchas and Rey (2005a). Hooper and Morton (1982).
Maurice Obstfeld and Kenneth S. Rogoff
111
Conclusions We have developed a simple stylized model that can be used to calibrate exchange rate changes in response to various scenarios under which the U.S. current account deficit might be reduced from its unprecedented current level. Aside from its quantitative predictions, the model yields a number of important qualitative insights. First, Asia’s greater openness to trade implies that the requisite exchange rate adjustments for that region are not all that much greater than Europe’s. This appears true despite the fact that Asia starts from a much larger current account surplus than Europe. Second, we find that, if Asia tries to stick to its dollar peg in the face of, say, a rise in the U.S. saving rate that closes up the U.S. current account gap even partly, Asia will actually have to run significantly larger surpluses than it does now. Europe would bear the brunt of this policy, ending up with a current account deficit even larger than that of the United States today, while at the same time suffering a huge loss on its net foreign assets. Third, although dollar depreciation does tend to improve the U.S. net foreign asset position (because virtually all of its gross foreign liabilities, but less than half of its gross foreign assets, are denominated in dollars), this effect only slightly mitigates the requisite exchange rate change. Valuation effects will not rescue the United States from a huge trade balance adjustment. Indeed, if relative interest rates on U.S. short-term debt rise even moderately during the adjustment process, this adverse effect could easily cancel out any gain due to valuation effects. Fourth, our model suggests that the need for deficit countries to shift demand toward nontraded goods (and for surplus countries to shift demand away from them) is roughly twice as important quantitatively as the much more commonly stressed terms-of-trade channel (which involves substitution between the traded goods produced by different countries). The importance of the terms of trade would be greater with lower international trade elasticities than we have assumed, or with a greater degree of home bias in consumption. We have only scratched the surface of the possible questions that can be asked within our framework. To that end, we have tried to make our approach as transparent as possible so that other researchers can easily investigate alternative scenarios using the model. Clearly, it would be interesting to extend the model in many dimensions, in particular to allow for sticky
112
Brookings Papers on Economic Activity, 1:2005
prices and for dynamic adjustments, such as factor movement across sectors. It would also be interesting to extend the framework to allow for more regions of the world economy, for example, oil producers, non-Asian emerging markets, and Asian subregions. Nonetheless, in a literature that is often long on polemics and short on analysis, we hope it is useful to have a concrete model on which to base policy evaluation.
APPENDIX A
Equilibrium Prices, Revaluation Effects, and Interest Rate Effects Equilibrium Prices Here we show how real exchange rates depend on equilibrium relative prices, and we spell out the relevant equilibrium conditions for our threeregion world economy. By definition, real exchange rates depend on relative international prices for both traded and nontraded goods. We take up relative traded goods prices first. As the text noted, notwithstanding the law of one price, the assumed internationally asymmetric preferences over tradables permit relative regional price indexes for tradable consumption to vary over time. Instead of being fixed at unity, these ratios are given in our model by PTE [ατ1U−,Eη + (β − α ) + (1 − β ) τ1U−,ηA ] = PTU [α + (β − α ) τU1−,Eη + (1 − β ) τU1−,ηA ]
1 1− η
( A1)
1 1− η
1− η 1 − δ 1 − δ 1− η δτU , A + 2 + 2 τU ,E PTA = PTU [α + (β − α ) τ1U−,Eη + (1 − β ) τ1U−,ηA ]
1 1− η
1 1− η
1− η 1 − δ 1 − δ 1− η δτU , A + 2 + 2 τU ,E PTA = PTE [ατ1U−,Eη + (β − α ) + (1 − β ) τ1U−,ηA ]
1 1− η
1 1− η
.
Maurice Obstfeld and Kenneth S. Rogoff
113
Thus shifts in interregional real exchange rates q reflect both shifts in the relative prices of traded and nontraded goods and shifts in the relative prices of exports and imports: γ + (1 − γ ) ( PNE PTE ) PE = TU × 1− θ PT γ + (1 − γ ) ( PNU PTU ) 1− θ
( A2)
qU ,E
1 1− θ
1 1− θ
[ατ + (β − α ) + (1 − β ) τ ] = [α + (β − α ) τ + (1 − β ) τ ] 1− η U ,E
1− η U ,A
1− η U ,E
1− η U ,A
γ + (1 − γ ) ( PNE PTE ) × 1− θ γ + (1 − γ ) ( PNU PTU ) 1− θ
1 1− η
1 1− θ
1 1− θ
γ + (1 − γ ) ( PNA PTA ) PA = TU × 1− θ PT γ + (1 − γ ) ( PNU PTU ) 1− θ
qU , A
1 1− η
1 1− θ
1 1− θ
1− η 1 − δ 1 − δ 1− η δτU , A + 2 + 2 τU ,E = [α + (β − α ) τ1U−,Eη + (1 − β ) τ1U−,ηA ]
1 1− η
1 1− η
γ + (1 − γ ) ( PNA PTA ) × 1− θ γ + (1 − γ ) ( PNU PTU ) 1− θ
1 1− θ
1 1−θ
.
Having defined relative price indexes, one can easily derive global marketclearing conditions for each region’s tradable output, again using very standard techniques for constant elasticity of substitution models such as the one we have here.53 For real U.S. tradable goods output, the marketclearing condition is given by −η
−θ
−η
−θ
P PU P PE ( A3) YTU = γ α UU TU C U + γ (β − α ) UE TE C E PT PC PT PC −η −θ A 1 − δ PU PT CA, + γ 2 PTA PCA 53. As illustrated, for example, in Obstfeld and Rogoff (1996).
114
Brookings Papers on Economic Activity, 1:2005
and that for real European traded goods output is given by −η
−θ
−η
−θ
P PU P PE ( A4) YTE = γ α EE TE C E + γ (β − α ) EU TU C U PT PC PT PC −η
−θ
1 − δ PE PTA CA. + γ 2 PTA PCA Walras’ Law implies that the condition for Asian traded goods equilibrium is superfluous, given the two others. One can similarly derive the market-clearing condition for U.S. nontraded goods as −θ
PU Y = (1 − γ ) NU C U PC
( A5)
U N
(which depends, of course, only on U.S. demand), as well as the two corresponding conditions for European and Asian nontraded goods. We take output endowments as given, and we then use the marketequilibrium conditions just stated to solve for relative prices as functions of current account balances and initial net foreign asset positions. (In our simulations we allow for currency revaluation effects on foreign assets and liabilities, and for the feedback to trade balances needed to sustain any given constellation of current accounts.) To proceed, we first rewrite the equilibrium condition for the U.S. export good’s market as −η
−η
−η
P P 1 − δ PU ( A6) YTU = α UU CTU + (β − α ) UE CTE + CTA , PT PT 2 PTA or, in nominal terms, as ( A7)
P PU Y = α UU PT U T
1− η
P P C + (β − α ) UE PT U T
U T
1 − δ PU + 2 P A
1− η
PTE CTE
1− η
PTACTA .
T
If trade were balanced and international debts zero, then, of course, the value of U.S. traded goods consumption would have to equal that of U.S.
Maurice Obstfeld and Kenneth S. Rogoff
115
traded goods production. Here we want to allow for international debt as well as for trade and current account imbalances (which are the same in the model except for net factor payments). The U.S. current account surplus in dollars is given by CAU = PU YTU + rF U − PTU CTU ,
(A8)
where FU is the stock of U.S. net foreign assets (in dollars) and r is the nominal (dollar) rate of interest. Similarly, for Europe (and again measuring in dollars), CA E = PE YTE + rF E − PTE CTE .
(A9)
In the aggregate, of course (in theory if not in the actual data), CAU + CAE + CA A = 0.
( A10) Similarly,
F U + F E + F A = 0.
( A11) Thus, ( A12)
CA A = − (CAU + CAE ) = PAYTA − r ( F U + F E ) − PTACTA .
In this framework one can consider the effects of a variety of shocks that change the current nexus of global current account imbalances into one where, say, CAU = 0. (Other external balance benchmarks can be analyzed just as easily.) To do so, we use the above current account equations (and the implied trade balances) to substitute for dollar values of consumption of traded goods in the goods-market equilibrium conditions. The results are ( A13)
P PU Y = α UU PT U T
1− η
(P Y U
U T
+ rF U − CAU )
P + (β − α ) UE PT
1− η
1 − δ PU + 2 PTA
1− η
(P Y E
E T
[P Y A
A T
+ rF E − CAE ) − r ( F U + F E ) + CAU + CAE ]
116
Brookings Papers on Economic Activity, 1:2005
P PEYTE = α EE PT
1− η
(P Y E
E T
P + (β − α ) EU PT
+ rF E − CAE )
1− η
1 − δ PE + 2 P A
U
1− η
+ rF U − CAU )
(P Y
U T
[P Y A
− r ( F U + F E ) + CAU + CAE ].
A T
T
Critically, current account imbalances also spill over into relative prices for nontraded goods, to a degree that depends on the elasticity of substitution between traded and nontraded goods. For the three nontraded goods markets, one can show that ( A14)
1 − γ PNU γ PTU
1− θ
PNU YNU =
1 − γ PNU γ PTU
1− θ
=
1 − γ PNE PY = γ PTE
1− θ
1 − γ PNA PY = γ PTA
1− θ
E N
A N
PTU CTU + rF U − CAU )
(P Y
+ rF E − CAE )
[P Y
− r ( F U + F E ) + CAU + CAE ].
U T
U
E N
A N
(P Y E
A
E T
A T
Revaluation of Gross Asset Stocks through Exchange Rate Changes A key variable in the simulation analysis is f i, which is the ratio of net foreign assets (in dollars), Fi, divided by the dollar traded goods income of the United States, PUY TU. In reality, a country’s gross assets and liabilities are often denominated in different currencies, so that focusing only on the net position misses important revaluation effects that can occur as the exchange rate changes. Here we show how we have modified our simulation analysis to take into account both the normalization of dollar net foreign assets and the revaluation effects of exchange rate changes.54 54. Details can be found online at www.economics.harvard.edu/faculty/rogoff/papers/ BPEA2005.pdf.
Maurice Obstfeld and Kenneth S. Rogoff
117
Let Hi equal the gross assets of country i and Li its gross liabilities, measured in dollars. Then F i = H i − Li
(A15) and
fi =
( A16)
H i − Li . PU YTU
One can show that, under a monetary policy that targets the GDP deflator, (A17)
PU PU = NU PT
γ −1
[α + (β − α ) τ
1− η U ,E
+ (1 − β ) τ1U−, ηA ] . γ −1 1− η
The first step is to substitute this formula for PU into the denominators of f U, f E, and f A. The second step is to consider how exchange rate changes affect the numerators. Let ωij be the share of region i gross foreign assets denominated in the currency of region j, j = U, E, A, where the European and (especially) the Asian regional currencies are composites. Similarly, define the portfolio currency shares λij on the liability side. We will assume that central banks target GDP deflators and that EU, j denotes the (nominal) dollar price of currency j ( j = E, A) under the monetary rule. Then, after a change in exchange rates, the new dollar values of net foreign assets (with values after the change denoted by primes) are ( A18)
E´ − EU ,E U U (ω H − λUE LU ) F U´ = F U + U ,E EU ,E E E´ − EU , A U U (ω H − λUA LU ) + U ,A EU , A A E´ − EU ,E E E (ω H − λ EE LE ) F E´ = F E + U ,E EU ,E E E´ − EU , A E E (ω H − λ EA LE ) + U ,A EU , A A E´ − EU ,E A A (ω H − λ EA LA ) F A´ = F A + U ,E EU ,E E E´ − EU , A A A (ω H − λ AA LA ) . + U ,A EU , A A
118
Brookings Papers on Economic Activity, 1:2005
Note that the following two constraints must hold in a closed system: ( A19)
ω UE H U + ω EE H E + ω EA H A = λUE LU + λ EE LE + λ EA LA ω UA H U + ω EA H E + ω AA H A = λUA LU + λ EA LE + λ AA LA .
So we can eliminate the European asset shares by writing the preceding as post-change net asset values: ( A20)
E´ − EU ,E U U (ω H − λUE LU ) F U´ = F U + U ,E EU ,E E E´ − EU , A U U (ω H − λUA LU ) + U ,A EU , A A E´ − EU ,E U U ( λ L + λ EA LA − ωUE H U − ω EA H A ) F E´ = F E + U ,E EU ,E E E´ − EU , A U U ( λ L + λ AA LA − ωUA H U − ω AA H A ) + U ,A EU , A A E´ − EU ,E A A (ω H − λ EA LA ) F A´ = F A + U ,E EU ,E E E´ − EU , A A A (ω H − λ AA LA ) . + U ,A EU , A A
We also know that (A21)
H U + H E + H A = LU + LE + LA .
For our numerical findings we must posit estimated values for nominal assets and liabilities. Given the well-known measurement problems, any numbers are bound to be loose approximations at best. For the United States, the numbers we use are for end-2003 (from the 2005 Economic Report of the President) and show foreign-owned assets in the United States to be $10.5 trillion and U.S.-owned assets abroad to be $7.9 trillion. We take the current values to be $11 trillion and $8.25 trillion, respectively, for purposes of our simulations. To a first approximation, essentially all U.S. foreign liabilities are denominated in dollars, but only about 40 percent of U.S. foreign assets are. (In principle, foreign assets such as stocks and land are real, but in practice the dollar returns on these
Maurice Obstfeld and Kenneth S. Rogoff
119
assets are highly correlated with dollar exchange rate movements.) Of the remaining 60 percent, we take 41 percent to be in European currencies and 19 percent in Asian currencies. Following Tille (2004), and including Canada, the United Kingdom, and Switzerland in region E, the United States does have a very small share of its liabilities in foreign currencies. The exact portfolio weights that we assume for the United States are ( A22)
ω UE = 0.405, ω UA = 0.193, λUE = 0.03, λUA = 0.006.
Drawing on the work of Lane and Milesi-Ferretti (but taking into account the adding-up constraints that need to hold in our theoretical model), we take Asia’s assets to be $11 trillion and its liabilities to be $8.25 trillion.55 As for portfolio shares, on the asset side, data from the International Monetary Fund’s 2001 Coordinated Portfolio Investment Survey suggest that most Asian countries hold predominantly U.S. dollars (and some yen), but that Japan’s foreign assets are more evenly balanced between dollar and euro holdings. If we assume that Japan owns about 40 percent of the region’s gross foreign assets, we have the following approximation: ( A23)
ω EA = 0.2, ω AA = 0.
On the liabilities side, Japan borrows in yen, but the other Asian economies have equity liabilities (including foreign direct investment) in local currencies, and extraregional debt liabilities predominantly in dollars and euros (or sterling). We assume that ( A24)
λ EA = 0.33, λ AA = 0.33.
We again base our portfolio estimates for the E zone in our model on the latest data from Lane and Milesi-Ferretti, which indicate that assets and liabilities at the end of 2003 were both approximately $11 trillion. Thus we take HE = LE = $11 trillion. Most of greater Europe’s liabilities are in domestic currencies; here we assume the share is 80 percent. We take the remaining 20 percent to be entirely in U.S. dollars. On the asset side, however, we derive from equation A19 that 32 percent of Europe’s holdings are in dollar assets, and 11 percent in assets denominated in
55. Lane and Milesi-Ferretti (forthcoming).
120
Brookings Papers on Economic Activity, 1:2005
Asian currencies, with the remaining 57 percent in assets denominated in European currencies.56 In our simulations we take PUY TU = $(11/4) trillion, based on Obstfeld and Rogoff (2000b), who argue that roughly one-quarter of U.S. GDP may be regarded as traded. Given our assumptions on each region’s gross assets and liabilities and their currencies of denomination, our analysis will also tell us how net foreign assets change across various scenarios for the current account and the exchange rate, as well as allow for the feedback effect on interest payments. We will see that, given the large size of gross stocks, large changes in exchange rates can translate into large changes in net foreign asset positions. Indeed, for many short-run and medium-run issues, knowing the gross asset and liability positions is at least as important as understanding the net positions. This conclusion is very much in line with the empirical findings of Gourinchas and Rey (2005a) for the United States.
Effects of Changing Interest Rates It seems plausible that, in the process of U.S. current account adjustment, global interest rates will shift. Such changes could come about simply as a result of the reequilibration of the global capital market, or they could also reflect a shift in the portfolio preferences of foreign investors such that, given the exchange rate of the dollar, higher dollar interest rates are necessary to persuade them to maintain their existing dollardenominated portfolio shares. We adopt the latter perspective, allowing the interest rate on U.S. short-term debt liabilities to rise as the dollar adjusts, without a corresponding increase in the earnings on U.S. foreign assets. Capital market shifts of this nature are likely to be quantitatively more important for the dollar than more generalized, synchronized increases in world interest rates (although the United States, as a debtor, would naturally lose while its creditors would gain). To illustrate this channel, we first, for simplicity, abstract from the effects of nominal exchange rate changes on asset stocks (for the purpose of our 56. The European position assumptions are not needed to implement equation A20, but they are necessary for assessing the effects of interest rate changes below.
Maurice Obstfeld and Kenneth S. Rogoff
121
simulations, this case is only a computation check). We focus on the scenario under which, as the United States adjusts, it faces a sharp increase in its borrowing rates. Thus there are two interest rates in the world economy: the rate rU that the United States pays on its liabilities, and the rate rW > rU that all other countries pay on their liabilities and that all countries, including the United States, earn on assets outside the United States. We focus on the implications of rU rising when the United States adjusts; the increase in rU may itself have an effect on U.S. adjustment, although that possibility does not affect our calculation. There is also a long-run versus short-run distinction: in the short run only U.S. short-term liabilities will pay higher interest (as these are rolled over). According to U.S. Treasury data for September 2004 (from www.treas. gov/tic/debta904.html), U.S. short-term liabilities were about 30 percent of total liabilities (and thus about 30 percent of U.S. GDP). If the United States were required to pay, for example, 200 basis points more on this liability base, the result would be an additional drain of about 0.02 × 0.3 = 0.6 percent of total GDP. ∼ i represent the share of country i gross foreign assets invested in Let ω j country j. To make the previous modeling consistent, we replace rFi everywhere (for the United States, Europe, and Asia, respectively) by r W H U − r U LU
( A25)
[ω r [ω r E U
A U
U
U
+ (1 − ω UE ) r W ] H E − r W LE + (1 − ω UA ) r W ] H A − r W LA .
From estimates described in the last subsection, we have the dollar values of Hi and Li. Asian currency shares probably exceed the Asian country shares, because of Asian claims on offshore Eurodollars; we might assume ∼ A = 0.6. Since total U.S. liabilities equal the claims on the United that ω U States of Europe and Asia, ( A26)
ω UE H E + ω UA H A = LU ,
and so, with HE, HA, and LU each equal to $11 trillion, we must have ∼ E = 0.4. ω U We now turn to the calibration of interest rates (or, rather, nominal rates of return on asset and liability portfolios). We know that, for the United States currently, rWHU − rULU ≈ 0. Since, also, HU/LU ≈ 0.75, rU/rW ≈ 0.75.
122
Brookings Papers on Economic Activity, 1:2005
So we take rU = 3.75 percent initially,57 but we maintain the earlier baseline assumption that rW = 5 percent. We ultimately wish to consider alternative increases in rU, for example, of 125 basis points or more. These possibilities range from a scenario in which the United States simply loses its privilege of borrowing at a favorable rate, to some in which there is an element of loss of confidence in U.S. solvency absent ongoing dollar depreciation. We will also assume that only the interest rate on short-term liabilities rises in the short run. Suppose the share σ of short-term liabilities in total U.S. foreign liabilities is 30 percent, or σ = 0.3. Then the investment income account of the United States and the other two regions would change as follows: ( A27)
r W H U − r U LU → r W H U − ( r U + σ∆r U ) LU
[ω
E U
r U + (1 − ω UE ) r W ] H E − r W LE →
[ω (r + σ∆r + (1 − ω ) r ] H [ω (r + σ∆rr E U
[ω
A U
rU
U
A U
W
U
A
) + (1 − ω UE ) r W ] H E − r W LE − r W LA →
) + (1 − ω UA ) r W ] H A − r W LA . ∼ A ≈ 1 and that Europe ∼E + ω The last two changes assume that, empirically, ω A U
U
U
U
U
and Asia hold equal proportions of short-term U.S. liabilities. One might also consider a formulation where ∆rU = f (∆CAU), f ′ > 0. In this case adjustment could be quite painful if the f function is too rapidly increasing, LU is too big, or σ is too big (or any combination of these three). We leave this possibility for future research. Synthesis of Interest Rate Changes and Asset Revaluations We are now ready to illustrate the techniques used to calculate the results in the third column in table 8, in which asset revaluations and interest rate changes occur simultaneously and interactively. We proceed as in the last section but add the following equations:
57. This number is in line with the estimate given above of the excess return of U.S. foreign assets over U.S. liabilities to foreigners.
Maurice Obstfeld and Kenneth S. Rogoff
( A28)
123
E ´ − EU ,E U U EU´, A − EU , A U U ω H H U´ = H U + U ,E ω H + EU ,E E EU , A A E ´ − EU ,E E E EU´, A − EU , A E E H E´ = H E + U ,E ω H + ω H EU , A A EU ,E E E ´ − EU ,E A A EU´, A − EU , A A A H A´ = H A + U ,E ω H + ω H EU , A A EU ,E E
and ( A29)
E´ − EU ,E U U EU´ , A − EU , A U U LU´ = LU + U ,E λ L λ L + EU ,E E EU , A A E´ − EU ,E E E EU´ , A − EU , A E E LE´ = LE + U ,E λ L + λ L EU , A A EU ,E E E´ − EU ,E A A EU´ , A − EU , A A A LA´ = LA + U ,E λ L + λ L. EU , A A EU ,E E
These equations, rather than the equations for net positions used in the simpler revaluation exercise in which interest rates do not change, become necessary because assets and liabilities can now pay different rates of interest and therefore must be tracked separately.
Comments and Discussion Richard N. Cooper: This paper by Maurice Obstfeld and Kenneth Rogoff is very much a “what if” exercise. What if demand behaves according to constant elasticity of substitution functions? What if consumption is fixed, apart from terms-of-trade effects? What if the U.S. current account deficit is eliminated (or, in one of the authors’ simulations, halved) by an appropriate increase in the U.S. saving rate? Then we learn from the authors’ model what the implied changes in exchange rates and the terms of trade must be. The authors’ calibration is necessarily arbitrary, but it seems reasonable. A major contribution of their model is to provide a general equilibrium framework that includes two stylized regions outside the United States. It therefore permits an exploration of differing effects by region, and that seems very useful. The model also includes asset revaluation effects and not just trade effects. The model assumes flexible prices. As the authors note, it probably understates the exchange rate changes that would be required in a sticky-price regime. The authors also usefully note the potentially ambiguous impact of faster European or Japanese economic growth, which many people see as a potential partial solution to the correction of the U.S. current account deficit. The source of the growth makes a difference, and one should not assume that more-rapid growth abroad will ease the problem. Productivity increases in traded goods, which is where such increases have typically occurred, especially in Japan, could aggravate rather than mitigate the imbalances. No doubt it is interesting to see how large are the exchange rate changes required to close the U.S. current account deficit, but all the adjustment here is done through prices, including the asset revaluation effect. Whether that is helpful to policy is not at all clear to me. There is no explicit treatment of output or employment in the model. The simulations are driven 124
Maurice Obstfeld and Kenneth S. Rogoff
125
by a postulated increase in U.S. saving and a corresponding decline in or, in most of the simulations, elimination of the U.S. current account deficit, so that presumably U.S. output is unaffected. The paper’s simulations are, however, unclear about what is happening to saving in the other two regions of the world as the U.S. external deficit is reduced. Their current account surpluses at the outset reflect excessive saving in those two regions, and they disappear in the simulation in which all three current accounts go to zero. But how do they disappear? Consumption is assumed to be fixed, except for the terms-of-trade effect. The fall in saving in the paper’s simulated Europe and Asia therefore implies a corresponding fall in output and income in order to get these results. But a decline in output will surely affect employment, hence income, hence consumption and saving, raising the question of how the initial level of consumption is sustained in Asia and Europe. That in turn leaves me wondering whether the paper provides any useful lessons for addressing the issue of global imbalances in, say, the coming decade. Let me offer, in sketchy terms, my own view of the issue. The discussion of the U.S. current account deficit has focused largely on how much adjustment must occur in the United States. The authors’ model is properly a general equilibrium model, and so it includes the rest of the world. I will focus on adjustment in the rest of the world. In 2004 the large current account surpluses in the world were in Japan, at $172 billion, and in Germany and the Netherlands (which can be considered a satellite of the German economy) at $116 billion. Thus these three countries together account for nearly half the U.S. current account deficit. Russia and China together add another $130 billion. These are where the big numbers are. Adding up all of the rest of East Asia accounts for another $110 billion, and OPEC another $100 billion. Then there is the statistical discrepancy, which has grown to about $200 billion. There is, as always in ex post accounting, a problem of attribution, but if we stipulate that the U.S. current account deficit is to be eliminated, we need to ask where the impact will fall in the rest of the world, and it has to fall largely where the big surpluses are. (The present surpluses of emerging economies could also shrink, or their deficits grow, but that would require a willingness on the part of savers around the world to invest in those countries on the required scale, which is not a given. If all the adjustment occurred there, their current account deficits would have to rise far in excess of generally accepted levels.)
126
Brookings Papers on Economic Activity, 1:2005
Like the authors, I will put oil to one side. The increase in oil prices over the past two years has added roughly $100 billion to the U.S. current account deficit. I assume that, one way or another, either through a decline in oil prices or through an increase in absorption by the oil-exporting countries, their surpluses will decline significantly in the next few years. Instead I will focus on Germany and Japan, which, again, are where the really big surpluses are, and then I will comment on China. Germany and Japan are rapidly aging, high-saving societies with limited domestic investment. Saving rates have declined in Japan, but saving in the corporate sector remains quite high. What has fallen in Japan is investment, which remains low even after a revival in the last year. A big absorber of capital in rich countries is the residential sector. Investing in housing does not look very attractive in rapidly aging societies, with very low birth rates and low new household formation, which is the case in both of these countries. If anything, Germany and Japan have a surplus of housing in the aggregate, although it may not all be in quite the right places. Housing construction is down essentially to replacement plus a little bit to allow for mobility. Meanwhile rates of return on industrial investment are low and, of course, very sensitive to what is happening to the export sector. I will now make some sweeping (perhaps too sweeping) national generalizations. For reasons having to do with their defeat in World War II, a key question for the Germans and the Japanese was how to rebuild their national self-esteem. Both countries built it on export performance. That legacy continues sixty years later. The national psyche in both Germany and Japan is heavily influenced by export performance. If exports are not doing well, people feel badly about the economy and society. In my view, that influences their saving behavior. If the economy is not performing well, precautionary saving rises in these now-rich countries. Given the aging of their society, as the Japanese have been saying for some years, Japanese saving should decline and eventually become negative. That may be so, but it has been a much slower process than the life cycle advocates forecast twenty years ago. Saving remains remarkably high given Japan’s demographic structure, and the same is true of Germany. That syndrome, in which German and Japanese saving is sensitive to perceived economic performance, which in turn is remarkably sensitive to export performance, is important when it comes to correcting the U.S.
Maurice Obstfeld and Kenneth S. Rogoff
127
current account. If, as Obstfeld and Rogoff suggest, there will be big changes in exchange rates and big declines in the export competitiveness of key surplus countries, we are likely to see an increase, not a reduction, in the propensity to save in those countries. Whether that increase gets translated into actual additional saving depends, of course, on what happens to income. The conditions just described are, after all, the conditions under which a recession could occur. An increase in the propensity to save with no obvious vehicle for that saving leads to a fall in output and income. In the textbooks the adjustment mechanism in this process is the interest rate, which is assumed to reconcile ex ante differences in saving and investment. Suppose the long-term nominal interest rate is only 2 percent, as it has been in Japan for several years, and not much higher in Germany. The question then becomes, What sort of investment in Japan and Germany will be stimulated by a 2 percent interest rate, given the demographics, in the presence or even with the prospect of a significantly stronger yen and euro? The sector most responsive to low interest rates in rich economies generally is the housing sector, not industrial investment. Firms will not invest in increased capacity if they see poor sales prospects, no matter how low the interest rate. Yet, for the demographic reasons already noted, demand for housing will be limited, even at low long-term interest rates. Hence I do not see the interest rate as being an effective adjuster here. With a large appreciation of these surplus countries’ currencies, the adjuster is more likely to be economic activity. Economic activity will decline, except insofar as the authorities become so concerned about it that the Europeans break all the rules they have imposed on themselves, through the Stability and Growth Pact’s constraints on fiscal policy and the European Central Bank’s primary focus on price stability, and pursue an aggressively stimulative policy. I therefore see a big problem with substantial current account adjustment, mainly for Europe but also for Japan. Both already have large budget deficits: the Japanese budget deficit is roughly 7 percent of GDP, and France, Germany, and Italy, the core of Europe, have fiscal deficits expected to exceed the 3 percent limit under the Pact. Excess saving in these big rich countries manifests itself in budget deficits and current account surpluses. Savers directly or indirectly buy claims on their governments or claims on foreigners. In my judgment, further reductions in the long-term interest rate are not going to produce
128
Brookings Papers on Economic Activity, 1:2005
enough domestic investment to substitute for those two, particularly in the face of a decline in competitiveness brought about through large appreciations of the currency. Japan and Germany are perhaps unusual because of their peculiar history and their dependence on export performance. But I do not see how the currency appreciations that the paper simulates will produce the changes in saving required to eliminate, or even greatly reduce, the current account surpluses of Asia and Europe. A decline in Asian and European output in turn is likely to reduce the value of U.S. equity claims on those countries, weakening and possibly even reversing the valuation effect arising from dollar depreciation. China is more complicated, and I will not discuss it in detail. Although I do not subscribe to the whole of the Dooley-Garber thesis, I am sympathetic to one of its main thrusts. China is a very-high-saving, high-investment country. I believe there is a pent-up, latent demand in China for foreign assets, which cannot be realized because violation of the foreign currency rules, especially those regarding the export of resident funds, is severely punished. China has a very weak capital market, and the central bank of China is, in effect, making the foreign investments that the public is prevented from making. Tighter fiscal policy, which might be called for at the moment on domestic grounds, would actually increase national saving in China. That moves in the wrong direction. To conclude, I believe that the United States has comparative advantage at producing marketable assets. We sell these marketable assets to the rest of the world. As long as Americans use the proceeds of the sale of those marketable assets productively—and that is an important qualification, bearing on the desirability of the fiscal deficit—I do not see why that process cannot go on indefinitely. I am not saying the U.S. current account deficit will go on forever. My crystal ball fades rapidly after fifteen or twenty years. But, for the next decade, I do not see why the process whereby the United States generates marketable assets and sells them to foreigners who are eager to buy them cannot continue on the current scale, that is, roughly half a trillion dollars a year. Will there be a dollar crisis? I don’t have any idea. It depends in large measure on how markets react to debates such as the one we are having here. Expectations in financial markets can be very fragile. Currency markets could start to run away rapidly from the dollar. My main point is that there need not be a crisis.
Maurice Obstfeld and Kenneth S. Rogoff
129
T. N. Srinivasan:1 Had Yogi Berra, the great baseball player and savant, been the discussant of this paper, he likely would have begun with his famous words, “It is déjà vu all over again!” Fifteen years ago Stefan Gerlach and Peter Petri published a collection of essays with a pompous title, The Economics of the Dollar Cycle.2 They viewed the movements of the external value of the dollar as cyclical: having appreciated by more than 40 percent between late 1979 and February 1985, the dollar had then collapsed to a new low by 1987, only to stabilize and fluctuate narrowly around the bottom of the range experienced during the 1980s. Gerlach and Petri also made the following astounding claim: Unlike narrowly focused studies in a technical specialty, this book explores the subject simultaneously from the viewpoints of exchange rate economics, empirical trade analysis, the economics of international financial markets, and macroeconomic policy-making in the United States, Japan, Europe, and the developing countries.3
Here we are, fifteen years later, exploring the same issues, except that the buzzword of that day—“newly industrializing countries” or NICs—has been replaced by another, “emerging markets.” Evsey Domar, when asked by a graduate student at MIT why the questions in the macroeconomics examination do not change from year to year, is said to have replied, “Ah—but the answers do!” Domar did not claim that the answers got better over time, but one can hope that the papers in this volume will provide better answers to the same questions covered in the Gerlach-Petri volume. In preparing this comment, I found particularly useful the contributions to that volume by my late colleague James Tobin, and the comments on Tobin’s paper by his Yale student Ralph Bryant and my dear departed friend Rudiger Dornbusch. In 1988, as now, the U.S. current account was in deficit, albeit at a little more than 2 percent of GDP rather than 6 percent as now. Then as now, the United States was running a fiscal deficit of around 3.7 percent of GDP, similar to today’s 4 percent.4 In 1988 nominal interest rates in the United States had fallen from the dizzy Volckerian heights of over 19 percent in the early 1980s to a low of less than 7 percent and
1. I thank Benjamin Friedman and Robert Solow for their comments. 2. Interestingly, some of the contributors to the Gerlach-Petri volume were participants at this Brookings Panel meeting. 3. Gerlach and Petri (1990, p.2 ). 4. See the authors’ figure 4.
130
Brookings Papers on Economic Activity, 1:2005
had again begun to rise. Nominal rates again reached a low of around 1 percent (in terms of the federal funds rate) in mid-2004 and again have begun to rise. On the financing of the U.S. current account deficit in those days, Tobin remarked, “We are being warned incessantly that we depend on foreigners— mainly Japanese banks, insurance companies and pension funds—to buy US Treasury bonds and other dollar assets. . . . Should they decide not to buy dollar securities, we are told, [the result] would be calamitous.”5 Now, besides the Japanese, the financiers are the Asian central banks, particularly those of China, India, and Korea. (Even then, as Dornbusch noted in his comment, “Central banks rather than private savers have been financing the US current account.”)6 But the dire warnings are being repeated. Then, as Dornbusch put it, there was a “sharp shift in trade with the NICs. The United States has experienced a $60 billion shift in its manufactures trade with these countries since 1980.”7 Now China figures prominently in U.S. and world manufacturing trade, not to mention the prominence of India and other countries in services trade through offshoring. Related to this shift in trade was the issue of the domestic price consequences of dollar depreciation. To quote Tobin again, with any further depreciation of the dollar, “certainly imports from Japan and Europe will be more costly in dollars. So will imports from Asian ‘NICs’ if we induce them to let their currencies rise against the dollar.”8 Today the inducement takes the form of a demand by the secretary of the Treasury of the United States that the Chinese revalue their renminbi by at least 10 percent. At that time some held that the dollar was correctly valued, and hence there was no need for policy-induced corrections.9 This view was rationalized in three ways. First was the J-curve story: the response of the economy to the fall in the value of the dollar that had already taken place had yet to manifest itself fully and would surely do so soon. Second, U.S. current account deficits can be financed indefinitely by selling U.S. assets, since, in the portfolio of the rest of the world, the share of U.S. assets was
5. Tobin (1990, p. 34). 6. Dornbusch (1990, p. 53). 7. Dornbusch (1990, p. 53). 8. Tobin (1990, p. 30, emphasis added). 9. In the Gerlach-Petri volume, these views are summarized by Dornbusch (1990, pp. 53–54).
Maurice Obstfeld and Kenneth S. Rogoff
131
probably below its long-run equilibrium value. The contemporary version of this argument holds that China (and probably other Asian countries, including India, as well) undervalues its currency to sustain its export-led growth, so as to provide employment to its huge stock of underemployed labor. Until this stock is exhausted, undervaluation of its currency may continue, with the result that any possible appreciation from capital inflows is prevented by accumulation of reserves held in dollar-denominated assets. This argument is forcefully put forward by Dooley, Folkerts-Landau, and Garber.10 Third, the United States can wipe out the value of dollardenominated assets held by the rest of the world through inflation. Then as now, the question was raised whether whatever transitional adjustments (policy induced or otherwise) would inevitably take place to restore the dollar to its long-run equilibrium would be orderly and smooth (a soft landing), or delayed such that, when they eventually do take place, they would be abrupt and large (a hard landing). Brookings also held a workshop in that earlier period, in January 1987, to discuss these issues, just as it is doing now. Then, however, it assembled a group of multicountry-macroeconometric modelers and gave them alternative, but commonly specified, scenarios for the future. Starting from the common actual history of U.S. and foreign prices and exchange rates, the modelers (five in all) were asked to investigate the causes of the burgeoning external deficit of the United States during 1980–86 and to study the likely path of the deficit for the period 1987–91. Bryant, in his comment on Tobin’s paper in the Gerlach-Petri volume, reported on his updating of the conclusions of that workshop. His qualitative and quantitative findings regarding policy are not that much different from those of Obstfeld and Rogoff in the paper under discussion here, once one adjusts for differences in initial conditions. Briefly, he found that, first, in the short run (that is, until 1989), it was plausible to expect a substantial reduction in the U.S. external deficit.11 Second, in the medium and long run (1990 and after), the prospects were much less encouraging, with the improvement in the constant-price deficit leveling off after 1989 and the current-price deficit (the deficit measured in nominal dollars) ceasing to improve well before it reached an acceptably low level. Third, a fall in
10. Dooley, Folkerts-Landau, and Garber (2004a, 2004b). 11. Bryant (1990, pp. 42–43).
132
Brookings Papers on Economic Activity, 1:2005
the dollar’s real exchange value can play a powerful role in reducing the external deficit. Fourth, the deficit in nominal dollar terms was unlikely to decline to an acceptable level ($30 billion to $40 billion, or around 0.75 percent of GDP—one-fifth of its value in 1987) without either a somewhat further depreciation of the dollar or markedly slower growth in the United States than abroad. Bryant’s own estimate of the real dollar depreciation needed to bring the deficit down to an acceptable rate was between 7 percent and 15 percent; he viewed a 20 percent depreciation as excessive. Obstfeld and Rogoff estimate the depreciation in terms of the real trade-weighted exchange rate required in order to eliminate the present U.S. deficit of about 5 percent of GDP to be between 19 and 28 percent, in a scenario in which Asia neither adjusts its exchange rate nor reduces its current account surplus (see their table 4). Bringing the deficit down to 1 percent of GDP, or one-fifth of its initial value (if one can make a linear interpolation from the authors’ estimates), would call for a dollar depreciation in the range of 15 to 22 percent, very close to the estimate of 20 percent, for a similar proportional reduction of the deficit in 1987, that Bryant cited but found excessive. Tobin’s contribution to the Gerlach-Petri volume is aptly titled, “Eight Myths about the Dollar.” He encountered these myths regarding what should be done to eliminate the U.S. external deficit “all too often in contemporary public and, yes, professional discussion.”12 In exploding the myths, he also provided an analytical framework for thinking about policies for eliminating the deficit. Some of the myths seem to be still going around, and, more important, Tobin’s analytical framework remains relevant today. I therefore briefly summarize Tobin’s contribution in the next section. TOBIN’S ANALYTICAL FRAMEWORK AND THE EIGHT MYTHS. Unsurprisingly, Tobin found that, “just as [Hicks’s] IS-LM [model], for all the hard knocks it has received from theorists, remains a good general first approximation, so its international application, Mundell-Fleming, has been a good guide.”13 In the basic, two-country version of this model, there are two goods, and each country specializes in producing and exporting part of its output of one of the goods. There are two real assets, consisting of
12. Tobin (1990, p. 28). 13. Tobin (1990, p. 28).
Maurice Obstfeld and Kenneth S. Rogoff
133
the real money stocks of the two countries. Each country holds some of the other country’s asset. The simplest way of incorporating a trade (or current account) deficit in equilibrium in this model is through a capital transfer from one country to another that allows the receiving country to spend more on the two goods than its income, by the amount of the transfer. With one country’s good as the numeraire, the price of the second country’s good (that is, its relative price in terms of the first country’s good) is the real exchange rate in the model. Given the amount of the transfer in numeraire terms, goods market equilibrium determines the real exchange rate. The income of each country includes, besides the value of its goods output, its asset income, which in this simple framework equals the interest income on the part of its domestic real money stock held at home at its domestic interest rate and the interest income (in numeraire terms) on the foreign asset held by domestic residents at the foreign interest rate. Asset market equilibrium requires that, given the portfolio choices (in which capital transfers from one country to the other are incorporated), the demand for each country’s asset equal its exogenous supply. With free capital mobility, asset market equilibrium implies that the difference between the domestic and the foreign interest rate satisfies the uncovered interest party condition: in other words, that it equals the rate of expected real depreciation. Under perfect foresight (rational expectations) the expected rate of depreciation is zero in equilibrium, so that the model solves for two prices: the real exchange rate and the common interest rate. Obviously, this real model cannot determine the nominal exchange rate or any other nominal variables. Trivially, one could introduce nominal variables by viewing each country’s asset as its nominal currency stock. By choosing units of measurement of the two goods such that the price of each country’s good in its own currency is unity, nominal and real exchange rates can be made to equal each other. Any other price normalization rule will lead to a different nominal exchange rate corresponding to a given real exchange rate, but has no consequence for the determination of equilibrium real values. Clearly, away from equilibrium and assuming away unwanted inventory accumulation (for example, assuming that the two commodities are perishable), the excess of expenditure over income by one country will equal the capital transfer from the other country ex post—this is the identity by which the current account deficit is matched by an equivalent capital inflow. However, the excess expenditure and the corresponding capital inflows
134
Brookings Papers on Economic Activity, 1:2005
are not equilibrium amounts. Thus, as Rachel McCulloch correctly pointed out, from the ex post identities that national dissaving (the excess of expenditure over income or production) equals capital surplus in the balance of payments, which in turn equals the current account deficit, no causal relationship among the variables can be inferred.14 Put another way, no inference about the policy or behavioral changes needed to restore equilibrium can be drawn from ex post identities. Indeed, some of Tobin’s eight myths illustrate this proposition very clearly. Simply put, policies that claim to restore equilibrium in both markets by operating only in one market cannot, in general, succeed. In other words, both the real exchange rate and the interest rate will have to change to restore equilibrium. Tobin’s first myth, that “eliminating the federal budget deficit will automatically eliminate the deficit in the U.S. external current account,” and the second myth, its corollary, that “correction of the federal budget would solve the problem of external imbalance without further depreciation of the dollar,”15 illustrate this proposition. I cannot resist referring also to Tobin’s sixth myth: “depreciation will be counterproductive for the United States because it will cause recessions in Europe and Japan and diminish their demands for US goods and services.”16 As Tobin rightly said, even an undergraduate should be ashamed to fall for the chain of arguments that lead to this myth, yet “it has been advanced with straight faces by high financial officials”17—a statement that remains true today. This palpably false reasoning follows, in effect, by reading causality from ex post accounting identities. Myths three, four, five, and seven illustrate other but similarly faulty reasoning. Myth eight, on the naïveté of faith in macroeconomic policy coordination among major economies, lives on. At every meeting of the governors of the International Monetary Fund or the summit of the G-7 or G-8, such coordination for addressing “global imbalances” is advocated. Tobin was right: attempts at coordination on a possibly wrong policy can be very harmful, and such an event cannot be ruled out as unlikely, given the differences among analysts as to what constitutes the right policy in the first place.
14. 15. 16. 17.
McCulloch (1990). Tobin (1990, pp. 28–29). Tobin (1990, p. 33). Tobin (1990, p. 33).
Maurice Obstfeld and Kenneth S. Rogoff
135
THE OBSTFELD-ROGOFF CALIBRATED MODEL AND ITS SIMULATIONS. The Obstfeld-Rogoff model is, in some aspects, a more elaborate version of the basic real model described above, and less elaborate in others. Each country produces two goods, an internationally traded and a nontraded good, instead of one in the basic model; each country consumes three goods— the traded good that it produces, that produced by the other country, and, of course, its own nontraded good—instead of two. However, theirs is a static endowment model: each country has an exogenously determined endowment of its two goods; there is no production (and there are no factors of production) or investment, only consumption and trade. There are no factor or asset markets. Appearances to the contrary, in the ObstfeldRogoff model nominal exchange rates, the currency denomination of assets held by U.S. and foreign residents, their valuation, and interest rate effects are all add-ons: they do not form part of the behavioral specification of the model. The model determines the equilibrium real exchange rates from which Obstfeld and Rogoff then derive nominal exchange rates under alternative assumptions of the policy objectives of the central bank: to stabilize the CPI deflator, the GDP deflator, or a bilateral exchange rate. These assumptions, being unrelated to the behavioral variables of the model, do not influence its equilibrium determination. Investment decisions, interest rates, and portfolio choices obviously do not arise in an endowments model. Obstfeld and Rogoff add the return from net asset holdings at an exogenously specified interest rate to the value of endowment on the income side. Although asset valuation effects and interest rates do influence the equilibrium through their income effects, they are still add-ons, since gross values of assets and liabilities, portfolio weights, and interest rates are not endogenously determined by the model, and the fact that gross values are in nominal terms does not matter, since nominal exchange rates are mechanically linked to real rates, given the assumed behavior of the central bank. When I say that add-on assumptions are not part of the behavioral structure model, I do not mean to imply that the results of the authors’ numerical simulations, such as the magnitudes for nominal exchange rates, are not plausible. They might well be, but their plausibility or otherwise cannot be inferred from the plausibility of the assumptions alone. I will not, therefore, comment on the paper’s numerical results. Obstfeld and Rogoff simulate the impact of specified changes in the international pattern of external imbalances, those changes being the consequence
136
Brookings Papers on Economic Activity, 1:2005
of unspecified shocks to demand. In one scenario these shocks are assumed to bring down initial current account deficits and surpluses in all three regions to zero. In another the U.S. current account is set at zero, Asia pegs its nominal exchange rate to the dollar, and so on. The model then determines the comparative-static changes in equilibrium real exchange rates and other endogenous variables associated with the change in external imbalances. Since, in an endowment model, the only way to influence international balances is by affecting demand, analysis of changes in international balances brought about through shocks to supply, investment, or saving behavior is ruled out. Policies are not explicitly modeled—they are whatever is needed to induce the unspecified shocks to demand. For example, all fiscal policy combinations that, through shocks to consumption, result in the specified changes in external imbalances are equivalent from the perspective of the model. The focus of the analysis is entirely on the real exchange rate implications of the specified changes in the pattern of initial external imbalances, and not on the policies that are behind those changes in imbalances. Those policy changes are outside the model and cannot be evaluated through the model. Thus the various policy proposals currently being made, including a policy of neglect, benign or otherwise, of the initial imbalances, cannot be evaluated. Also, the policy-relevant question of whether the prevailing U.S. current account deficit of 6 percent of GDP is indefinitely sustainable cannot be meaningfully posed, let alone answered, by the model. The implications for U.S. interest rates of a shift away from dollar-denominated assets in the portfolios of foreigners, including central banks, or a reduction in saving propensities abroad, or a rise in saving propensities in the United States, cannot be examined. Such questions as whether the end to global imbalances will come smoothly, predictably, and at a modest cost, or abruptly, unexpectedly, and at a heavy cost, cannot be analyzed either. The analytics of the model are easily illustrated in a slightly simpler version in which there are two countries, home and foreign, each producing two traded goods. Each traded good produced by one country is a perfect substitute for the corresponding good produced by the other country. This simplification, of course, rules out home bias, as in the Obstfeld-Rogoff model, in the consumption of traded goods, and it rules out having more than one real exchange rate. However, it goes beyond the Obstfeld-Rogoff model by allowing a production (supply) response to changes in relative prices. Allowing both demand and supply responses to changes in global
Maurice Obstfeld and Kenneth S. Rogoff
137
balances will, in general, attenuate the change in relative prices required to restore equilibrium. By distinguishing the short from the long run, such that flexibility to shift resources from producing one good to producing the other is limited in the short run relative to the long run, one can allow for possible overshooting in equilibrium relative prices in the short run without having to appeal to price rigidities. The model collapses to the endowment model if there is complete inflexibility in the short run. In a special case of the model, there is no change in the equilibrium real exchange rate from its initial value as the economy adjusts to the elimination of the trade deficit. The model is static, and there are no investment or capital flows. The current account deficit is modeled as a pure income transfer from one country to the other. For simplicity, preferences over three commodities (the two traded goods and the nontraded good) in each country are assumed to be homothetic. For simplicity, assume that the economy running a trade deficit financed by an income transfer is a small open economy trading with a large rest of the world, so that it is a price taker in the world market for its traded goods. (This is the so-called dependent economy of W. Salter and T. Swan,18 in which the relative price of one traded good in terms of the other is set by the world market.) This means that the two traded goods can be aggregated into a single composite traded good for the small open economy, as long as conditions in the world market do not change. Suppose there are two factors of production (say, capital and labor), and suppose the technology of production of all three goods exhibits constant returns to scale and factor intensity (capital-labor ratio) nonreversal, so that the ranking of the goods in terms of cost-minimizing factor intensities is independent of factor prices. Then, given the relative price of traded goods, the factor prices can be uniquely solved from the two expressions equating price to unit cost, if both goods are produced in positive amounts. Given the unique factor prices, the minimum unit cost of production of the nontraded good and hence its price, given that a positive amount of it is produced, are determined. Suppose the factor endowments are such that the production possibility frontier (PPF) includes a nonempty subset S in which all three goods are produced in positive amounts. In S the marginal rate of transformation between the traded composite good (T ) and the nontraded good is constant. 18. Salter (1959); Swan (1963).
138
Brookings Papers on Economic Activity, 1:2005
Figure 1. Small Open Economy Model
Traded composite D
Q0 A E Q'0
Q1 P0 F
B P1
O
C
Nontraded good
Source: Author’s model described in the text.
In figure 1 above, the PPF is depicted as ABC, where the linear stretch AB (except at points A and B) corresponds to S. The consumption possibility frontier (CPF), given a transfer AD (in units of the traded composite good) from the rest of the world, is depicted by DEFC. It is simply a vertical shift of the PPF by the distance AD. Assume that preferences are represented by a homothetic, quasi-concave utility function. Maximizing utility subject to the CPF leads to the initial equilibrium consumption at Q0 and production at P0 vertically below it. At Q0 an indifference curve touches the PPF so that the common slope of the two is the slope of DE, which equals the relative price of the traded composite good, that is, the real exchange rate. Suppose now that the transfer is withdrawn so that the CPF coincides with the PPF. If the real exchange rate does not change, by virtue of homothetic preferences, consumption shifts to Q′0 on AB, where the ray OQ0 from the origin intersects AB. If production remains at P0, there will be an excess supply of the nontraded good. However, since the marginal rate of transformation between traded and nontraded goods is constant along AB, the
Maurice Obstfeld and Kenneth S. Rogoff
139
production point shifts to Q′0 and the excess supply is eliminated. Thus the economy adjusts to the elimination of the trade deficit by a pure quantity adjustment with no change in the real exchange rate. One could depict an endowment economy by reinterpreting the PPF as just a single point P0. With income transfer P0Q0, the economy consumes at Q0. Let DE be the tangent to the indifference curve at Q0, so that it is the real exchange rate from a consumption perspective. The withdrawal of the transfers requires that the consumption point move to P0. But if the consumption real exchange rate does change, it will move to Q′0, once again creating an excess demand for traded goods. To eliminate this excess demand, the real exchange rate for consumption has to change to the slope of the indifference curve through P0. This means (under the standard assumption of convexity of preferences and both goods being normal) there has to be a rise in the relative price of the traded good in terms of the nontraded good, or a real depreciation, since at P0 and Q0 the consumption of the nontraded good is the same, whereas that of the traded good composite is lower at P0. Returning to the production economy, what if the initial consumption point is on EF, such as Q1? By construction, because each point on EF is at the same vertical distance (of AD) above the point on the stretch BC vertically below it, the slope of the CPF at Q1 is the same as the slope of the PPF at P1, and their common slope is equal to the equilibrium real exchange rate from both a consumption and a production perspective. Now, with the withdrawal of the income transfer, the CPF coincides with the PPF, and production will remain at P1. At unchanged real exchange rates, consumption will move to a point (not shown) to the left of P1 on the straight line that is tangent to the PPF at P1, thus creating excess demand for traded goods. To eliminate this excess demand, a real depreciation has to occur, with the new equilibrium point lying to the left of P1 on the PPF, where an indifference curve touches the PPF (not shown). Short-run inflexibility and long-run flexibility in shifting resources starting from the production point P1 can be easily illustrated. The short-run PPF touches the long-run PPF at the initial production point P1 but is below it otherwise, with the vertical distance between the two increasing as the production of the nontraded good increasingly deviates from its level at P1 in either direction. Under these assumptions, which seem natural for depicting short-run inflexibility, it is clear that the short-run equilibrium point where an indifference curve touches the short-run PPF will imply a
140
Brookings Papers on Economic Activity, 1:2005
larger depreciation in terms of the real exchange rate than its long-run value at the point where an indifference curve touches the long-run PPF. In other words there is overshooting of the real exchange rate in the short run. If we add on a dynamic adjustment of the short-run PPF to the long-run PPF over time, it follows that, after overshooting, the real exchange rate will converge to its long-run value. The essential features of the adjustment will remain in its extension to a multicountry general equilibrium setup. However, its diagrammatic exposition will not. The reason is that the convenient device of a Hicksian composite traded good depends on the relative price of traded goods not changing. This cannot hold in general in the general equilibrium setup, because the relative prices are endogenous. Although the Obstfeld-Rogoff model is a three-country general equilibrium model, it replicates the essential qualitative conclusion of adjustment in a small open economy, namely, that a real depreciation is generally (though not necessarily always) needed to eliminate global imbalances. For their purpose, which is to arrive at a quantitative estimate of the extent of the real depreciation needed to eliminate global imbalances, Obstfeld and Rogoff have had to calibrate their general equilibrium model. Let me therefore conclude with a couple of comments on the calibration.19 Obstfeld and Rogoff relate their choice of values for the two crucial parameters, θ (the elasticity of substitution in consumption between the traded aggregate and the nontraded good) and η (the so-called Armington elasticity of substitution between the domestic and foreign traded goods in the traded goods aggregate), to econometric estimates in the literature. There are several problems with this procedure. First, although θ is arguably a “deep” parameter in the Lucas sense, since it relates to preferences, η is not. As such, any estimate of η will depend on the trade policy regime and therefore cannot be stably estimated econometrically. Second, setting aside the policy dependence of parameter values, since the Obstfeld-Rogoff model involves aggregates, alternative schemes of aggregation will influence parameter values, and whether the estimates in the literature are all comparable and correspond to the implied aggregation of the Obstfeld-Rogoff model is not obvious. To be fair, the authors are certainly aware of these issues, and their simulations cover a range of values for the two parameters. Perhaps they should cover an even broader range of values, particularly for η. 19. For a more detailed discussion, see Dawkins, Srinivasan, and Whalley (2001).
Maurice Obstfeld and Kenneth S. Rogoff
141
General discussion: William Nordhaus agreed with the authors that productivity’s effect on the trade deficit depends critically on whether the change in productivity occurs in the traded goods or the nontraded goods sector. He reported his own recent findings showing that, whereas the productivity slowdown in the United States during the 1970s had occurred in both sectors, the acceleration of the 1990s was mainly in traded goods. Although some estimates attribute almost all the acceleration to computers and associated industries, Nordhaus estimated that only somewhere between a half and two-thirds came from that source. The most recent data also suggest that productivity has accelerated, but he cautioned that this may in part reflect mismeasurement of productivity in the retail sector. Noting that Jack Triplett had found the United States to be on the frontier of improved measurement techniques that have tended to raise estimates of productivity growth, Nordhaus speculated that Europe’s productivity performance might look more like the United States’ if the same measurement techniques were used for both. Gian Maria Milesi-Ferretti noted that the April 2005 issue of the International Monetary Fund’s World Economic Outlook (WEO) contains a paper using a four-region global economic model that is similar to the authors’ three-region model but allows for changes in production. As one would expect, the WEO model finds that allowing for a production response leads to a smaller, but still quite substantial, real depreciation of the dollar. Sebastian Edwards observed that, in the authors’ Bretton Woods II scenario, Asia’s surplus increases from 15 percent of U.S. traded-goods GDP to 25 percent. The authors’ real model is unable to consider the monetary consequences of this increase, but Edwards suggested that in the real world it would create enormous pressure to expand the money supply in China and the other Asian countries, requiring an extraordinary amount of sterilization to avoid inflation. Indeed, the latest data already show an increase in inflation in China. Edwards also observed that the results of the authors’ global rebalancing scenario do not differ significantly from those of an earlier two-region model of theirs; the introduction of the third region does not appear to make a significant difference to the results. Edmund Phelps reminded the panel of a paper he had co-written in 1986, which argued that the expansionary U.S. fiscal policy of that era would result in a boom in the United States while causing world real interest rates to increase, leading to a recession in Europe. This analysis suggests that if the United States were now to adopt fiscal austerity, world real interest
142
Brookings Papers on Economic Activity, 1:2005
rates would decrease. This in turn would likely lead to an increase in asset values in Europe and Asia, and thus an increase, rather than a decrease as Richard Cooper had predicted in his comment, in output in those countries. In the United States the shadow prices of business assets would fall, causing a decline in production and investment. Peter Garber argued that although Japan’s and Germany’s populations are aging, and their populations growing slowly, there is significant underemployment in their economies, especially in the nontradable services industries. Underemployment is the reason that Japan has dramatically increased its monetary base in order to maintain a high yen-dollar exchange rate. Garber suggested that Germany is facing the same problem but is incapable of making a similar intervention, and its difficulties are exerting pressure on the European Central Bank to lower interest rates.
Maurice Obstfeld and Kenneth S. Rogoff
143
References Broda, Christian, and David E. Weinstein. 2004. “Globalization and the Gains from Variety.” Working Paper 10314. Cambridge, Mass.: National Bureau of Economic Research (February). Bryant, Ralph C. 1990. “Comment on ‘Eight Myths About the Dollar.’ ” In The Economics of the Dollar Cycle, edited by Stefan Gerlach and Peter A. Petri. MIT Press. Calvo, Guillermo A. 1998. “Capital Flows and Capital-Market Crises: The Simple Economics of Sudden Stops.” Journal of Applied Economics 1: 35–54. Campa, José Manuel, and Linda S. Goldberg. 2002. “Exchange Rate Pass-Through into Import Prices: A Macro or a Micro Phenomenon?” Working Paper 8934. Cambridge, Mass.: National Bureau of Economic Research (May). Cline, William R. Forthcoming. The United States as a Debtor Nation: Risks and Policy Reform. Washington: Institute for International Economics. Cooper, Richard N. 2001. “Is the U.S. Current Account Sustainable? Will It Be Sustained?” BPEA, no. 1: 217–26. Croke, Hilary, Steven B. Kamin, and Sylvain Leduc. 2005. “Financial Market Developments and Economic Activity during Current Account Adjustments in Industrial Economies.” International Finance Discussion Papers 827. Washington: Board of Governors of the Federal Reserve System (February). Dawkins, Christina, T. N. Srinivasan, and John Whalley. 2001. “Calibration.” In Handbook of Econometrics, vol. 5, edited by James J. Heckman and Edward Leamer. Amsterdam: Elsevier. Dooley, Michael P., David Folkerts-Landau, and Peter M. Garber. 2004a. “The Revived Bretton Woods System: The Effects of Periphery Intervention and Reserve Management on Interest Rates and Exchange Rates in Center Countries.” Working Paper 10332. Cambridge, Mass.: National Bureau of Economic Research (March). ________. 2004b. “Direct Investment, Rising Real Wages, and the Absorption of Excess Labor in the Periphery.” Working Paper 10626. Cambridge, Mass.: National Bureau of Economic Research (September). Dornbusch, Rudiger. 1976. “Expectations and Exchange Rate Dynamics.” Journal of Political Economy 84, no. 6: 1161–76. ________. 1990. Comment on Tobin, “Eight Myths about the Dollar.” In The Economics of the Dollar Cycle, edited by Stefan Gerlach and Peter A. Petri. MIT Press. Edwards, Sebastian. 2004. “Thirty Years of Current Account Imbalances, Current Account Reversals, and Sudden Stops.” International Monetary Fund Staff Papers 51(special issue): 1–49. Feenstra, Robert C. 1994. “New Product Varieties and the Measurement of International Prices.” American Economic Review 84, no. 1: 157–77.
144
Brookings Papers on Economic Activity, 1:2005
Frankel, Jeffrey A., and Andrew K. Rose. 1995. “Empirical Research on Nominal Exchange Rates.” In Handbook of International Economics, vol. 3, edited by Gene M. Grossman and Kenneth Rogoff. Amsterdam: Elsevier. Freund, Caroline, and Frank Warnock. 2005. “Current Account Deficits in Industrial Countries: The Bigger They Are, The Harder They Fall?” World Bank and Darden Graduate School of Business, University of Virginia (May). Gagnon, Joseph E. 2003. “Productive Capacity, Product Varieties, and the Elasticities Approach to the Trade Balance.” International Finance Discussion Paper 781. Washington: Board of Governors of the Federal Reserve System (October). Gerlach, Stefan, and Peter A. Petri. 1990. The Economics of the Dollar Cycle. MIT Press. Goldberg, Pinelopi Koujianou, and Michael M. Knetter. 1997. “Goods Prices and Exchange Rates: What Have We Learned?” Journal of Economic Literature 35, no. 3: 1243–72. Gourinchas, Pierre-Olivier, and Hélène Rey. 2005a. “International Financial Adjustment.” Working Paper 11155. Cambridge, Mass.: National Bureau of Economic Research (February). ________. 2005b. “U.S. External Adjustment: The Exorbitant Privilege.” University of California, Berkeley, and Princeton University (April). Greenspan, Alan. 2004. “The Evolving U.S. Payments Imbalance and Its Impact on Europe and the Rest of the World.” Cato Journal 24 (Spring-Summer): 1–11. Grubert, Harry, Timothy Goodspeed, and Deborah Swenson. 1993. “Explaining the Low Taxable Income of Foreign-Controlled Companies in the United States.” In Studies in International Taxation, edited by Alberto Giovannini, R. Glenn Hubbard, and Joel Slemrod. University of Chicago Press. Harris, David, Randall Morck, Joel Slemrod, and Bernard Yeung. 1993. “Income Shifting in U.S. Multinational Corporations.” In Studies in International Taxation, edited by Alberto Giovannini, R. Glenn Hubbard, and Joel Slemrod. University of Chicago Press. Hooper, Peter, Karen Johnson, and Jaime Marquez. 2000. Trade Elasticities for the G-7 Countries. Princeton Studies in International Economics 87. Princeton University (August). Hooper, Peter, and John Morton. 1982. “Fluctuations in the Dollar: A Model of Nominal and Real Exchange Rate Determination.” Journal of International Money and Finance 1, no. 1: 39–56. Krugman, Paul R. 1985. “Is the Strong Dollar Sustainable?” In The U.S. Dollar— Recent Developments, Outlook, and Policy Options. Kansas City, Mo.: Federal Reserve Bank of Kansas City. ________. 1989. “Differences in Income Elasticities and Trends in Real Exchange Rates.” European Economic Review 33, no. 5: 1031–54. ________. 1991. Has the Adjustment Process Worked? Policy Analyses in International Economics 34. Washington: Institute for International Economics.
Maurice Obstfeld and Kenneth S. Rogoff
145
Lane, Philip R., and Gian Maria Milesi-Ferretti. 2004. “The Transfer Problem Revisited: Net Foreign Assets and Real Exchange Rates.” Review of Economics and Statistics 86, no. 4: 841–57. ________. 2005a. “Financial Globalization and Exchange Rates.” Working Paper 05/3 (January). Washington: International Monetary Fund. ________. 2005b. “A Global Perspective on External Positions.” Trinity College, Dublin, and International Monetary Fund (May). ________. Forthcoming. “The External Wealth of Nations Mark II: Revised and Extended Estimates of Foreign Assets and Liabilities, 1970–2003.” Trinity College, Dublin, and International Monetary Fund. Mann, Catherine L. 2005. “Breaking Up Is Hard to Do: Global Co-Dependency, Collective Action, and the Challenges of Global Adjustment.” CESifo Forum 6, no. 1: 16–23. Marazzi, Mario, and others. 2005. “Exchange Rate Pass-through to U.S. Import Prices: Some New Evidence.” International Finance Discussion Paper 2005-833. Federal Reserve Board (April). McCulloch, Rachel. 1990. “Macroeconomic Policy, Trade and the Dollar: Remarkable Events of the 1980s.” In The Economics of the Dollar Cycle, edited by Stefan Gerlach and Peter A. Petri. MIT Press. Meese, Richard, and Kenneth Rogoff. 1983. “Empirical Exchange Rate Models of the Seventies: Do They Fit Out of Sample?” Journal of International Economics 14 (February): 3–24. Mendoza, Enrique G. 1991. “Real Business Cycles in a Small Open Economy.” American Economic Review 81, no. 4: 797–818. Mussa, Michael. 2005. “Sustaining Global Growth while Reducing External Imbalances.” In The United States and the World Economy: Foreign and Economic Policy for the Next Decade, edited by C. Fred Bergsten. Washington: Institute for International Economics. Obstfeld, Maurice. 2004. “External Adjustment.” Review of World Economics 140, no. 4: 541–68. Obstfeld, Maurice, and Kenneth Rogoff. 1996. Foundations of International Macroeconomics. MIT Press. ________. 2000a. “Perspectives on OECD Economic Integration: Implications for U.S. Current Account Adjustment.” In Global Economic Integration: Opportunities and Challenges. Kansas City, Mo.: Federal Reserve Bank of Kansas City (August). ________. 2000b. “The Six Major Puzzles in International Macroeconomics: Is There a Common Cause?” In NBER Macroeconomics Annual 2000, edited by Ben S. Bernanke and Kenneth Rogoff. MIT Press. ________. 2004. “The Unsustainable US Current Account Position Revisited.” Working Paper 10869. Cambridge, Mass.: National Bureau of Economic Research (November).
146
Brookings Papers on Economic Activity, 1:2005
Obstfeld, Maurice, and Alan M. Taylor. 1997. “Nonlinear Aspects of GoodsMarket Arbitrage and Adjustment: Heckscher’s Commodity Points Revisited.” Journal of the Japanese and International Economies 11, no. 4: 441–79. ________. Forthcoming. “Sources of America’s ‘Exorbitant Privilege.’ ” University of California. Orcutt, Guy H. 1950. “Measurement of Price Elasticities in International Trade.” Review of Economics and Statistics 32, no. 2: 117–32. Ostry, Jonathan D., and Carmen M. Reinhart. 1992. “Private Saving and Terms of Trade Shocks: Evidence from Developing Countries.” International Monetary Fund Staff Papers 39, no. 3: 495–517. Porter, Richard D., and Ruth A. Judson. 1996. “The Location of U.S. Currency: How Much Is Abroad?” Federal Reserve Bulletin 82 (October): 883–903. Reifschneider, David, Robert Tetlow, and John Williams. 1999. “Aggregate Disturbances, Monetary Policy, and the Macroeconomy: The FRB/US Perspective.” Federal Reserve Bulletin 85, no. 1: 1–19. Roubini, Nouriel, and Brad Setser. 2004. “The U.S. as a Net Debtor: The Sustainability of U.S. External Imbalances.” Stern School of Business, New York University, and University College, Oxford (August). Ruhl, Kim. 2003. “Solving the Elasticity Puzzle in International Economics.” Federal Reserve Bank of Minneapolis (November). Salter, W. E. G. 1959. “Internal and External Balance—The Role of Price and Expenditure Effects.” Economic Record 35: 226–38. Stockman, Alan C., and Linda L. Tesar. 1995. “Tastes and Technology in a TwoCountry Model of the Business Cycle: Explaining International Comovements.” American Economic Review 85, no. 1: 168–85. Swan, T. 1963. “Longer Run Problems of the Balance of Payments.” In The Australian Economy, edited by H. W. Arndt and W. M. Corden. Melbourne: Cheshire. Tille, Cédric. 2004. “Financial Integration and the Wealth Effect of Exchange Rate Fluctuations.” Federal Reserve Bank of New York (August). Tobin, James. 1990. “Eight Myths about the Dollar.” In The Economics of the Dollar Cycle, edited by Stefan Gerlach and Peter A. Petri. MIT Press. Ventura, Jaume. 2001. “A Portfolio View of the U.S. Current Account Deficit.” BPEA, no. 1: 241–53. Warnock, Francis E. 2003. “Exchange Rate Dynamics and the Welfare Effects of Monetary Policy in a Two-Country Model with Home-Product Bias.” Journal of International Money and Finance 22, no. 3 (June): 343–63.
MICHAEL DOOLEY University of California, Santa Cruz PETER GARBER Deutsche Bank
Is It 1958 or 1968? Three Notes on the Longevity of the Revived Bretton Woods System IT IS NOW widely accepted that the broad outlines of the current international monetary system are as we described them almost two years ago and labeled “the Revived Bretton Woods system.” This system’s main features are —the emergence of a macroeconomically important group of economies that manage their currencies vis-à-vis the dollar to support export-driven growth —the United States as center and reserve currency country, providing financial intermediation services for foreign, and particularly Asian, saving through its national balance sheet, and willing to accept large current account imbalances —a group of poorer economies implementing export-led development policies and exporting large amounts of capital to richer economies, mostly the United States —unusually low and even falling short- and long-term real interest rates as a result of this glut of mobile global savings, and —a group of industrial and emerging economies with floating exchange rates, whose currencies are under incessant pressure to appreciate. Not agreed and under vigorous discussion is how long this system can last. Will it be a meteoric flash with a spectacular end soon to come? Or
We are grateful to Daniel Riera-Crichton for his able and diligent research assistance in producing the empirical results of this paper.
147
148
Brookings Papers on Economic Activity, 1:2005
will it last for the reasonably foreseeable future? We package these questions here in an analogous question: Is the Revived Bretton Woods system at the point in its development where the original Bretton Woods was in 1958 or 1968 or 1971? In a series of publications we have provided a fundamental underpinning for why we believe the system will last—that the situation today is more like 1958.1 We argue that the gains to the players from continuing their actions outweigh the costs that many have argued will arise in an endgame asset price shift or in unexploited benefits of portfolio diversification. Rather than characterize the situation with geopolitically charged rhetoric like “balance of financial terror,”2 we think it more valid to think in the familiar economic terms of “mutually beneficial gains from trade,” such as might exist between any borrower and lender or between any purveyor of goods and its customer. Here we further develop our argument in the form of three notes addressing particular issues that have cropped up in critiques of the Revived Bretton Woods view. These notes both respond to the critiques and continue to expand our ideas. The first note explains how we think about what is driving capital flows to the United States and keeping interest rates low. We view the fact of unusually low long-term real interest rates for this stage of the business cycle as a direct challenge to those who, exaggerating the importance of rumors about central bank reserve management practices, claim that the end is near. The second note seeks to provide some information about the experience of those emerging economies with chronic current account surpluses since the breakdown of the first Bretton Woods system. A very large empirical literature evaluates the experience of emerging economies that have run chronic deficits, and the costs and frequency of associated financial crises. But we are not aware of any similar evaluations of the durability and stability of those foreign exchange regimes that have resulted in unusual sequences of current account surpluses and accumulations of international reserves. The widespread view that the surplus regimes at the core of the
1. Eichengreen (2004), in contrast, seems to favor 1968, that is, to allow the system a few years more to run, whereas Frankel (2005a) favors 1971. Roubini and Setser (2004) call for something even more immediate and apocalyptic, yet they acknowledge that the day of reckoning may be as long as two years off. 2. Summers (2004a, 2004b).
Michael Dooley and Peter Garber
149
Revived Bretton Woods system will come to a quick and costly end has likely been inferred in part from the recent experience of debtor emerging economies. Our interpretation of the experience of these surplus regimes is that they have been and may well remain durable and immune from financial crises. The third note addresses an issue that has been raised frequently in criticisms of our comparing the current system to the Bretton Woods system, namely, that the United States is running large current account deficits now, but it was not then. Of course, many aspects of the current system are different from what they were in the heyday of Bretton Woods: Konrad Adenauer is no longer chancellor of Germany, Charles de Gaulle is dead, the United States no longer guarantees gold convertibility, and there is now a serious pretender to reserve currency status. Our first reaction was that this difference was as superficial as these others and not at the heart of the comparison we wanted to draw. But the United States did have a major balance of payments deficit during the Bretton Woods era, which was the proximate driver of the deterioration of the system. So we relate the U.S. balance of payments deficits under Bretton Woods to the U.S. current account deficits under the Revived Bretton Woods to show that there is a close analogy. This is something more than an exercise in the history of economic ideas, because it plays into our view that collateral is the key to opening sizable gross cross-border trade in assets in a system that is short on trust.
Real Interest Rates Say It Is 1958 Why is the real interest rate in the United States so low and falling today, in the growth phase of the U.S. and global business cycles, even as the U.S. current account deficit reaches record levels? At the end of June 2002, about when the euro began its recent appreciation, realized ten-year real annual interest rates on U.S. Treasury securities were 3.70 percent on nominal notes and 3.07 percent on inflation-protected securities. The corresponding numbers at the end of December 2003 were 2.35 percent and 1.95 percent, respectively. One year later the respective numbers were 1.17 percent and 1.63 percent, and as we write in mid-May 2005, they are 1.02 percent and 1.65 percent. This fall in rates has come in a period when the media swirls daily with stories about foreigners losing confidence,
150
Brookings Papers on Economic Activity, 1:2005
foreign exchange reserve managers diversifying portfolios, and imminent collapse as everyone tries to be the first out the door. If all this is true, the bond and credit markets have not noticed. Three broad realities underpin our view of this global phenomenon, all of which we expect to continue into the foreseeable future: Reality 1: About fifteen years ago, hundreds of millions of underemployed workers joined the world’s market economies. They had no capital to speak of, but they had a desire to work in industry and to get rich. One might expect that such an increase in the global supply of labor would drive real interest rates up, but these workers came with an enormously high saving rate and lived under the yoke of a dead financial system, which had served them in the past as a capital destroyer, as it does to this day. They lacked modern technology and management. Theirs was a communist society that was and is problematic geopolitically, which might, in turn, make their access to cross-border credit problematic. This created a profound global disequilibrium for the industrial world, equal in magnitude to the global unemployment problem of the Great Depression although much more concentrated geographically. The industrial world’s economic system has to resolve this fundamental imbalance over the course of time by absorbing these workers. To focus today on trade imbalances when in fact there is an enormous labor market imbalance is to make the same mistake that economists and policymakers made in the 1930s. Reality 2: The emerging economies that have developed most successfully are those that export capital on net. Joshua Aizenman, Brian Pinto, and Artur Radziwill demonstrate, in a sample of forty-seven developing countries from 1981 to 2001, that the net exporters of domestic savings among them had significantly higher growth rates.3 They conclude that “a rise in the self-financing ratio [the stock of tangible capital supported only by past national saving, divided by the actual stock of capital] from 1 to 1.1 is associated with an increase in the growth rate from 2.8% to 4.4%. Further, reducing the self-financing ratio from 1 to 0.9 is associated with a drop in the growth rate from 2.8% to 2.2%.”4 These estimates control for differences in institutional quality as well as in trade and financial openness. They clearly contradict the usual assumption that developing countries 3. Aizenman, Pinto, and Radziwill (2004). 4. Aizenman, Pinto, and Radziwill (2004, p. 9).
Michael Dooley and Peter Garber
151
Figure 1. Market Equilibrium for Loanable Funds Real interest rate
U.S. demand
Private sector supply
Quantity of funds Source: Authors’ model described in the text.
have been successful in using net foreign saving to augment capital formation and economic growth. We argue below that their results are consistent with the idea that net exports of saving from poor countries support two-way trade in private financial assets that improves the quality and productivity of domestic capital formation. Reality 3: The United States has a large and growing current account deficit, funded at this moment by the foreign private and official sectors at low and falling real interest rates. In recent years the official sector has taken up a large share of this deficit. Let us focus on reality 3 for a moment. We like to think about the United States’ external deficit problem in a simple loanable funds flow framework. After netting U.S. public and private investment demand from U.S. saving, the United States has a demand for saving from the rest of the world that is downward sloping when plotted against the real interest rate, as in figure 1: the lower the real interest rate that it faces, the less the United States wants to save and the more it wants to invest. Given a real interest rate, then, we
152
Brookings Papers on Economic Activity, 1:2005
Figure 2. Effect of Expansionary Fiscal Policy Real interest rate
U.S. demand
Private sector supply
Quantity of funds Source: Authors’ model described in the text.
can read off the U.S. current account deficit. Meanwhile there is an upwardsloping supply of saving coming from the foreign private sector (we introduce foreign official flows below), perhaps from asset managers looking only at Sharp ratios and benchmarks, or perhaps from foreign industrial corporations interested in return on capital. The higher the real interest rate available in the United States, the more of this private foreign saving flows in. The intersection of these two curves determines the global real interest rate, the U.S. current account deficit, and the rest of the world’s current account surplus. A looser fiscal policy might shift the demand for foreign saving upward as in figure 2. This would bring in more foreign saving or, equivalently, increase the current account deficit. And it would cause the real interest rate to rise, as in the Reagan-era deficits of the early 1980s. Some of this may be going on today, but it is clearly not dominant. Since 2002, mar-
Michael Dooley and Peter Garber
153
Table 1. Marketable U.S. Government Debt, Interest Rates, and Fiscal Deficits, 2002–05 Item Marketable debt (billions of dollars)a Average interest rate paid (percent a year)a Fiscal deficit (percent of GDP)b
2002
2003
2004
2005
3,020 3,317 3,721 4,085 4.99 4.11 3.61 3.94 1.5 3.4 3.6 3.0
Source: Bloomberg data. a. As of March 31. b. In preceding fiscal year.
ketable U.S. debt has increased by more than one-third while nominal and real interest rates have declined (table 1). Moreover, relative to those of other industrial countries, the U.S. budget deficit is not unusually large (figure 3); it is hard to see why this factor alone would increase the U.S. current account deficit relative to the deficits or surpluses of other industrial countries, especially given the much more rapid GDP growth rate (real and nominal) in the United States. Instead consider figure 4, which adds a vertical official sector supply curve, reflecting the fact that policymakers have objectives other than a narrow risk-return calculus localized to this small portion of national saving. Their development goals require the export of domestic saving, and they will accept whatever interest rate the market determines. Adding, horizontally, this new public sector supply of foreign saving to the private sector supply shifts total supply rightward and brings down the interest rate that clears this global market for savings. So, if a falling U.S. real interest rate is observed alongside a rising U.S. current account deficit, it can only mean that official capital is being pushed into the United States and private capital is being pushed out, but by a smaller amount than the official capital coming in. On net, capital is not being pulled in by U.S. demand shifts. This is the combination of facts that shows us that the United States is passive and that the foreign official sector is the active player in global imbalances. This means that the typical denunciation of U.S. “profligacy” is worse than useless for understanding the situation: it is actually misleading. Usually, this rhetoric includes a reference to the role of the U.S. fiscal deficit in reducing net U.S. saving, but a larger fiscal deficit should increase the interest rate. Whatever the size of this effect, it has clearly been more than overcome by the effects of foreign official capital pushing in.
154
Brookings Papers on Economic Activity, 1:2005
Figure 3. Fiscal Deficits, Growth, and Inflation, Selected Countries, 2004a Percent of GDP or percent a year Germany U.S. France Italy Euroland b Japan
7 6 5 4 3 2 1 0 –1 Deficit–GDP ratio
Nominal GDP growth
Inflation
GDP growth
Source:–Bloomberg. a. Budget deficit data are for fiscal 2004; all other data are for calendar 2004. b. Countries that have adopted the euro.
Figure 4. Effect of Adding an Inelastic Official Sector Supply Real interest rate Pre-intervention equilibrium
Private sector supply
Official sector supply
U.S. demand Official + private sector supply
Quantity of funds Source: Authors’ model described in the text.
Michael Dooley and Peter Garber
155
One often hears that private saving flows to the United States are falling because of increased risks, stemming perhaps from the worsening U.S. international investment position. This would show up as a shift of the private supply curve in figures 1, 2, and 4 upward and to the left, and it would put further upward pressure on real interest rates. But this is exactly the opposite of what we observe. Rather, the evidence is far more consistent with a downward slide along a given private supply curve after the public sector supply is added. To be sure, foreign private saving is financing far less of the U.S. current account deficit than it did, say, five years ago. But the reason is that private investors are being driven out by official sector flows willing to replace them at much lower interest rates. One should beware of making too much of a rising or falling fraction of official sector finance in any given quarter or year. A steadily growing flow from the foreign official sector to the United States year in and year out is not necessary to maintain the system. Official flows are necessary only when the foreign central bank must intervene to keep its currency undervalued. As in a target zone exchange rate regime, when the private sector is confident that the regime is durable and will be sustained by future interventions as the need arises, private inflows are sufficient to provide the deficit financing. So far, so good. But it means that today’s low real interest rate is a momentary flow effect that will evaporate should official sector lending to the United States dry up permanently. If this were to happen, the picture would snap back from figure 3 to figure 1, and interest rates would jump. If this is what the market expects, we should today see low short-term real interest rates and much higher long-term rates. But we do not see this. Longterm real rates are low—hence the conundrum that we are studying now. Implicit in the real yield curve is that the equilibrium of figure 3 should last a long time. It follows that even a hint that Asian governments might reduce their flow demand for dollar assets will generate an immediate jump in the ten-year rate in the United States. Indeed, many observers doubt that foreign official interests in funding the U.S. current account deficit are sustainable. Back to Reality 1 The stakes are high indeed. Why should Asian authorities remain willing to increase their claims against the United States? To answer this, we have
156
Brookings Papers on Economic Activity, 1:2005
to focus on the strategy that Asia (notably China) has chosen to solve the development and unemployment problems that are part of reality 1. The problem for China is to mobilize its existing enormous domestic saving to create a growing, internationally competitive capital stock that can rapidly employ hundreds of millions of workers in productive activity. A serious constraint is the lack of a domestic financial system capable of channeling this saving into productive capital, technology, and management skills. The solution, perhaps stumbled upon inadvertently, has been to engage in export-led growth, thereby providing an immediate global quality check on the goods produced. This avoids falling off the cliff of another Great Leap Forward. To get export markets open, part of the policy has been to offer a large incentive to potential industrial exporters, both domestic and foreign-based, in the form of low dollar wages and the expectation that wages will rise only slowly toward world levels. Slowly rising dollar wages could be associated with a gradual nominal revaluation of the renminbi or a slightly higher rate of inflation than in China’s trading partners. For example, a 3 to 5 percent revaluation of the renmimbi later this year and the adoption of a carefully controlled float of the exchange rate would not signal the end or even a material change in the development strategy we have described.5 The typical problem in emerging economies is how not to offer too high an industrial wage relative to wages elsewhere in the economy: too-rapid industrialization could drive industrial wages sharply above agricultural wages, deterring investment in industry and triggering a flood of migration to the cities. By keeping wages low and relatively uniform, an initially low but rising currency helps both to induce resource transfers to industry and to restrain migration to a rate consistent with capital formation in the industrial sector.6 Foreign direct investors have been encouraged because they bring the discipline of international financial intermediation. Additional benefits include technology transfer and the proven political clout to keep export 5. In Dooley, Folkerts-Landau, and Garber (2004b), we treat the initial stock of labor as an exhaustible resource. In that context it is optimal for the government to absorb labor more rapidly at the beginning of the regime. It follows that dollar wages are initially set at a low level but rise over time to the world wage when the last worker is absorbed. See Salant (1976) for a more general discussion. 6. We thank Vincent Reinhart for this insight.
Michael Dooley and Peter Garber
157
markets open. The importance of direct investment in keeping U.S. export markets open has been questioned on a variety of grounds. Eswar Prasad and Shang-Jin Wei argue that most foreign direct investment into China comes from outside the United States and is not likely a significant factor in keeping U.S. markets open to Chinese exports.7 Leaving aside the inherent difficulty in determining the nationality of direct investors, we would point out that Asian direct investors in China also have an enviable record in penetrating U.S. markets and in dealing with the threat of U.S. protection. It seems likely to us that the building political pressure in the United States to do something about the bilateral trade deficit with China would be more effective if U.S. and other multinational corporations were not active in direct investment in China. We are more than willing, however, to base our forecast of the durability of the system on the lasting inability of domestic credit markets in emerging economies to efficiently intermediate domestic saving. The difficulty in reforming financial markets in these economies, and the frequency and costs associated with crises in economies that have not been successfully developed, are in our view the primary lesson provided by the failures of development in Latin America. But why does the need for international financial intermediation (twoway trade in financial assets) create saving-investment imbalances and a flood of net capital exports in the first place? After all, an export-based development policy need not imply a net export of capital. All that is needed is export growth, and this can just as well be balanced by import growth as not. In general, the successful emerging economies have not needed net foreign saving; such inflows are generally small and unreliable relative to domestic saving (reality 2). Nevertheless, other things equal, even a small addition of net foreign saving should contribute to investment and growth in poor countries. A positive argument in favor of net exports of saving requires that some other important ingredient to growth not be available in equal measure. Our hypothesis is that net exports of domestic saving are necessary to earn the collateral required for efficient international intermediation of domestic saving. Asian emerging economies do not need net foreign saving, but they do need efficient financial intermediation. We have emphasized 7. Prasad and Wei (2005).
158
Brookings Papers on Economic Activity, 1:2005
foreign direct investment and other types of international financial intermediation because we are not optimistic about the rapid development of domestic credit markets. That is, residents of these countries can avoid these markets by placing some of their assets off shore. These will return if international investors are protected from political risk—especially important when private capital is flowing to and from a geopolitically problematic country in large amounts. The government can relax this credit constraint by keeping its balance sheet very strong versus the rest of the world, that is, by building net reserves. The government’s net reserves then provide protection for private international financial intermediation against various geopolitical risks. In effect, the emerging economy’s government promises to stay on the sidelines by becoming a net creditor to the rest of the world. Note that a government cannot borrow this credibility; it has to earn it by placing goods and services in the rest of the world on net. And placing more goods and services in the rest of the world than one is taking in means a current account surplus. Imagine that foreign direct investment flows are matched by official sector reserve growth in the balance of payments accounts and that the current account is balanced. Then the capital account is balanced in terms of both net and gross flows. But the country sending the foreign direct investment is taking an unbalanced risk position, effectively buying equity and borrowing in fixed-interest securities. Usually, in private markets, this requires some collateral from the lesser credit. The way an emerging economy delivers collateral is by running a current account surplus. The faster the gross positions in the capital account grow, the faster must the current account imbalance grow to support the unbalanced risk positions. We believe that this view of current account surpluses as collateral provides a first explanation of the connection between net and gross capital flows, currently a noteworthy lacuna in models of international finance, which ignore gross trade in assets. Other Asia and Japan A reasonable objection to our argument is that it does not fit the more developed countries in Asia, especially Japan, that have been the most eager buyers of U.S. assets. In fact, it is useful to consider China and Japan as spanning the problems facing Asia. Both Japan and China have an employment problem, but in Japan it is the result of a very long cyclical downturn, whereas in China it is a long-term development problem. Both governments
Michael Dooley and Peter Garber
159
look to export growth as a solution to this problem, and both have a long history of managing the exchange rate. For quite different reasons, both countries have been able to sterilize very large reserve accumulations. In deflationary Asian countries, notably Japan, it is difficult to understand why there might be some limit on the ability or motivation of the authorities to create yen in stemming an attack on the currency. With interest rates at zero, it is costless to create as much yen cash as is demanded, whereas dollar reserves produce a positive yield. Normally, a limit on foreign exchange acquisition is reached when the resulting monetary expansion causes excessive overheating and inflation. But such an expansion is still not in sight for Japan and would not, in any case, be the appropriate monetary policy. The lessons of attacks on weak currencies and fixed exchange rate regimes seem to be the ones being applied by the global private financial sector here. The authorities in such regimes face a limit on reserves or credit or the amount of pain they are willing to put the economy through, and so each attack on the currency is simultaneously a ratcheting up of the probability that the currency will indeed collapse. Some observers seem to be holding a case study of a typical speculative attack against a mirror and thinking that private capital inflows likewise ratchet up the pain in Japan. Yet quite the opposite is true in deflationary Japan. Japan has ceased its massive intervention since the first quarter of 2004, and the yen has actually depreciated somewhat against the dollar. Our expectation is that the authorities will return to the market if private flows to the United States again decline and the yen again appreciates, especially if it is tested in another attack. In China financial repression has allowed the authorities to place domestic assets generated by sterilization without much increasing domestic interest rates, and it has been very successful in containing inflation. The People’s Bank of China currently places three-year domestic currency debt in the banks at an annual interest rate of about 3 percent and is experiencing positive carry on its foreign exchange.8 Other emerging economies in Asia with relatively open capital markets have followed a middle course of trying to stay competitive with China but allowing some appreciation of their currencies against the dollar, although still with heavy currency management and accumulation of reserves. The success and durability of these efforts 8. This in contradiction to the continual alarmist statements from Goldstein and Lardy (2004), among others.
160
Brookings Papers on Economic Activity, 1:2005
are a matter of intense debate, but we doubt there is much to be gained from continuing the debate at the theoretical level. We turn to the empirical debate in the next section.
How Do Episodes of Reserve Accumulation End? In a series of papers,9 we have argued the case for a meaningful distinction between countries that allow private international investment decisions to determine important macroeconomic variables such as the real exchange rate and the current account balance, and countries for which government investment decisions determine these magnitudes. We have referred to these as “capital account countries” and “trade account countries,” respectively. Trade account countries repress private financial flows and overwhelm with official flows those that slip through the repression. Capital account countries, in contrast, do not block cross-border flows or significantly intervene in foreign exchange markets. It is often assumed that the conventional analytical framework developed to understand the behavior of capital account countries applies also to trade account countries, because capital and foreign exchange controls are mostly ineffective. In our view this is entirely an unresolved empirical issue. The opinion that the U.S. current account deficit is unsustainable flows from a conviction that private international investors will be unwilling to continue to accumulate net claims on the United States. In this view, moreover, either the official capital flows that have partly financed the U.S. current account deficit will be overwhelmed by private sector flows, or governments will come to their senses in time to avoid a crisis. The usual dark warning is that the longer it takes the official sector to realize the inevitable truth, the harsher will be the consequences. The phase diagrams of the speculative attack models dance in our collective heads. We fully agree with half of this prediction. Two years ago we predicted that private investors would become more reluctant to finance the U.S. current account deficit as official sector capital flowed in.10 We also predicted the very large appreciation of the euro and other currencies whose trade in foreign exchange markets is dominated by private capital flows. This was 9. Dooley, Folkerts-Landau, and Garber (2003, 2004a, 2004b, 2004c). 10. Dooley, Folkerts-Landau, and Garber (2003).
Michael Dooley and Peter Garber
161
not unusual in itself. But we also argued that governments of a group of what we called “trade account countries” (countries where repression of private financial flows determine the real exchange rate and the current account balance) had good reasons to continue to invest in the United States for an extended period and that this would keep U.S. interest rates low, contrary to then-prevailing opinion.11 The length of this period is derived from an optimal rate of absorption of those countries’ unemployed labor. In our view of the real forces behind this system, this suggests a decade at least. So it is important to understand why nonresidents are supplying net saving to the United States at very low expected yields and why this may or may not continue. To us, it is irrelevant to the overall picture whether the net foreign investment is in Treasury securities, agency securities, private fixed-income securities, equity, or something else. It is irrelevant whether private or official foreigners take larger or smaller shares of the foreign investment in the United States in any given year. It is mostly irrelevant how the spreads across different classes of financial instruments in the United States might be affected. This is not a discussion of investment strategy or asset allocation; it is entirely directional. Historical Evidence One way to begin to evaluate the durability of the Revived Bretton Woods system and the likely consequences of its demise is to study the experience of economies that have had unusually long sequences of current account surpluses and accumulations of official reserves. Doubts about the durability of the system have generally centered on the ability and willingness of surplus economies to maintain an undervalued currency for an extended period. Does historical experience suggest that periods of reserve accumulation are followed by speculative attacks that generate a real appreciation (through either inflation or a nominal appreciation), losses on dollar reserves, and painful recessions as resources are transferred from traded goods industries? The experience of emerging economies with chronic current account surpluses since the breakdown of the original Bretton Woods system in 1971 11. The logic was that an increasingly indebted United States, with an interest rate effectively underwritten by the trade account countries, would attract less private funding from other (capital account) countries. Smaller net capital flows meant smaller net current account flows, which would be accomplished through appreciation against the dollar.
162
Brookings Papers on Economic Activity, 1:2005
has not attracted much attention, perhaps because until recently they have been quantitatively unimportant. An alternative possibility is that observers assume that such regimes cannot last for long and will end badly, because of the evidence provided by emerging economies with chronic deficits. Assumptions are necessary because past empirical work on crises and current account reversals has considered only episodes identified by large depreciations or swings in current accounts from deficit to surplus.12 The theoretical symmetry between speculative attacks on undervalued currencies and those on overvalued currencies is well known.13 In an attack on a strong currency, anticipated capital gains generate private capital inflows when speculators believe the regime can be overwhelmed. Intervention to limit nominal appreciation takes either of two forms, both with unfavorable side effects: an increase in the monetary base, which raises the domestic price level, or sterilization, which increases reserve assets and the government’s domestic currency liabilities. The regime can appear to be stable for a time, but the government’s tolerance for inflation or reserve accumulation is limited, and a speculative attack will bring the regime to an end. Data Methods: Identifying Precedents To identify historical precedents for today’s surplus economies, we first identify sequences of reserve accumulation that might provide a typical pattern for emerging economies that accumulate net reserves for an extended period. We then examine the behavior of other variables in the years during and after the accumulation sequence. For a sample of 115 developing and industrial economies, we examine yearly data from 1970 to 2004. We first identify sequences of consecutive years in which the economy experienced current account surpluses on average and the government increased its net foreign asset position. For surplus economies the change in the government’s net foreign asset position is usually dominated by changes in international reserve assets, but our measure of net reserve accumulation also includes changes in government debt and other official sector capital flows. We are interested in the consolidated government contribution to financing the change in national net foreign assets or its mirror image the current account balance. To further restrict attention to episodes in which the government was an important 12. Frankel and Rose (1996); Razin and Milesi-Ferretti (1998). 13. Grilli (1986).
Michael Dooley and Peter Garber
163
Figure 5. Real Exchange Rate and Current Account Balances during and after an Episodea Real exchange rate
Indexb 160 140 120 100 80 60 40 20
102.23
100.94
104.09
102.89
101.14
97.54
Current account–GDP ratio
Percent 10 5 0.47
0.73
0
0.18
0.05 –0.08
–2.06
–5 –10 –15 –3
–2
–1 Year
0
1
2
Source: International Monetary Fund, International Financial Statistics; International Institute of Finance; World Bank, World Development Indicators; Organization for Economic Cooperation and Development. a. Year zero refers to the first year after the end of an episode. An episode is defined as a sequence of years (three or more) in which the official sector increases its stock of international assets, runs a current account surplus (on average), and generates more than 25 percent of the change in national net foreign assets. Each observation denotes the real exchange rate or current account ratio of an economy at the specified point in its episode. Numbers in italic are means. b. The index equals 100 three years before the end of an episode.
participant in international financial markets, we exclude episodes during which the government generated less than 25 percent of the change in national net foreign assets. The typical experience of surplus economies during the three years before the end of a sequence of net reserve accumulations and the three years that followed is summarized in figure 5. Definitions and data sources
164
Brookings Papers on Economic Activity, 1:2005
are reported in appendix A, and detailed data for all accumulation episodes in appendix C. During periods of reserve accumulation, several regularities stand out. First, with very few exceptions, the current account begins in surplus, and the surplus increases during the period of net reserve accumulation. The average increase in the surplus was 0.73 percent of GDP in year t − 3, 0.47 percent in t − 2, and 0.18 percent in t − 1, the final year of net reserve accumulation. In t − 0, the first year following the sequence of reserve accumulation, the current account surplus declines by an average of 2.06 percent of GDP. This certainly suggests that some important shock has occurred. Moreover, the shock typically persists, with little further change in the current account balance in the two years that follow. In the final three years of net reserve accumulation, the currency typically appreciates in real terms each year; the average cumulative increase (we define the real exchange rate such that an increase represents an appreciation) is about 4 percent. The behavior of the exchange rate before the official sector leaves the market is not surprising and is fully consistent with the conjecture that the growing current account surplus, appreciation of the currency, and reserve accumulation reflect a growing fundamental disequilibrium in the real exchange rate and the current account. This sequence is supposed to end with a jump (appreciation) in the real exchange rate and a gradual decline in the current account balance. Instead, in the average case, the government retreats from the market, the currency depreciates in real terms in the following year by 1.2 percent,14 and the depreciation continues for two more years. The real exchange rate ends more than 3 percent below its level five years earlier, and the government enjoys a substantial capital gain on its reserve accumulation. The behavior of the macroeconomic variables when the government stops accumulating net reserves clearly does not fit with the standard model of a speculative attack on a strong currency. With the government out of the market, there is a “sudden start” of private capital inflows, as the model predicts. But these inflows are associated with a persistent real depreciation of the currency, not an appreciation. Economic growth is above trend 14. At the end of the original Bretton Woods system, Germany experienced a threeyear episode of reserve accumulation and currency appreciation followed by depreciation in 1974. In this case, however, the cumulative appreciation was larger than the depreciation in the following and subsequent two years. The usual assumption that reserve accumulation ends with large capital losses on reserves is probably influenced by this episode.
Michael Dooley and Peter Garber
165
during the reserve accumulation episode and generally moderates thereafter. In most cases growth remains positive and recovers in a year or two. A variety of shocks to the world economy could account for these empirical regularities. For example, a decline in foreign demand for the country’s exports could explain the swing in the current account and the decline in growth. The deterioration in the national net foreign asset position is consistent with a decline in the real exchange rate. Indeed, such a decline may create the expected yield differentials necessary for the sudden start in capital inflows. The deterioration of the national net foreign asset position would be consistent with a real depreciation. Another possibility is that intended or unintended financial liberalization allows residents to diversify away from domestic assets toward international assets. For example, if China suddenly opened its capital markets, residents’ desire to diversify into foreign currency assets would suggest a depreciation of the renmimbi rather than the appreciation currently expected. In this context it would be fully rational for the authorities to build a stock of dollars now in anticipation of private demand when financial markets are liberalized. Moreover, it makes no sense to allow the renmimbi to appreciate now only to depreciate sharply later. China, Japan, and Korea Seven economies accounted for two-thirds of worldwide international reserve holdings at the end of 2004 and for three-quarters of the $600 billion growth in international reserve assets in that year. Three economies— Japan, China (excluding Hong Kong), and Korea—held 45 percent of the global total and acquired 60 percent of the 2004 increase. The general sequence of current account imbalances, reserve gains, growth, and real exchange rate changes described above holds even more clearly for this group of economies. Several are in the midst of a stretch of reserve accumulation today and have experienced two or three similar episodes in the past. All seven have in the past experienced unusually long episodes of net reserve accumulation relative to our complete sample: the champion to date is Singapore with its twenty-seven-year run from 1974 through 2000. It seems particularly relevant, in evaluating how the current episode of reserve accumulation might end, to look at the previous experiences of these seven economies. In this section we review the recent experience of the “big three.” For each we present charts comparing, for each reserve
166
Brookings Papers on Economic Activity, 1:2005
accumulation episode, current account deficits and net reserve accumulation on the left-hand side, and economic growth and the real effective exchange rate on the right-hand side. Appendix B provides similar figures for the other four economies: Hong Kong, India, Singapore, and Russia. Japan has experienced three extended episodes of net reserve accumulation in the post–Bretton Woods era: a three-year sequence from 1986 to 1988, a five-year sequence from 1992 to 1996, and a six-year sequence that began in 1999 and continued through 2004 (figure 6). What might the first two episodes suggest for the current episode in Japan? We look first at the current account. In each episode the current account surplus (in billions of dollars) is growing before the net reserve accumulation begins. The surplus moderates during the accumulation, then declines in the year before and the year in which the accumulation ends. In the first episode reserve accumulation accounted for about 28 percent of the current account surplus during the period of reserve growth, and in the second about 26 percent. In the present episode the current account surplus has continued to grow. Reserve accumulation absorbed about half of the surplus through 2004. Considering just these data, one might expect a moderation in the rate of reserve accumulation going forward, but not an end to the sequence of net reserve gains. In the two previous episodes, the real effective exchange rate and the growth rate behaved as described above for the typical experience. The exchange rate rose before and during the reserve accumulation but then fell sharply for two or three years. In the first episode the real exchange rate fell in the year following the end of the accumulation and in 1986, the year the authorities withdrew from the foreign exchange market. As is typical for the larger sample, in both episodes the GDP growth rate rose during the accumulation sequence and then turned down, for three years in the first episode and two years in the second. During the recent episode both the real exchange rate and the growth rate have departed from the norm. Growth increased in the first year of the recent episode, and the real effective exchange rate rose as would be expected, but growth collapsed in 2001 and 2002, and the real exchange rate fell. Since then there has been a recovery in output and a small rise in the real exchange rate. If history is a reliable guide, reserve accumulation in this episode will moderate relative to the current account surplus but will continue until there is a significant decline in the surplus. Meanwhile the real exchange rate will continue to rise, but at a moderate rate, and output growth will
Michael Dooley and Peter Garber
167
Figure 6. Japan: Current Account Balance, Net Official Assets, Exchange Rates, and GDP Growth in Three Episodes 1986–88 Percent
Billions of dollars
Index
Current account
100
110
Exchange rate (right scale)
8
150
90
6
70
4 50
GDP growth (left scale)
2
0
0
Net official assets 1983
1985
1987
1989
50 30 10
1983
1985
1987
1989
1992–96
150
Current account
8
Exchange rate (right scale)
110 90
6
100
70
4 50
50
2
0
Net official assets 1991
1991
1997
1995
1993
30 GDP growth (left scale)
0 1993
1995
10
1997
1999–2004
150
Exchange rate (right scale)
8 Current account
6
100 Net official assets
0
2 0
1998
2000
2002
2004
90
70 GDP growth 50 (left scale) 30
4 50
110
10 1998
2000
2002
2004
Source: International Monetary Fund, International Financial Statistics; International Institute of Finance; World Bank, World Development Indicators; Organization for Economic Cooperation and Development. a. Year in bold refers to the first year after the end of an episode. See footnote a of figure 5 for the definition of an episode.
168
Brookings Papers on Economic Activity, 1:2005
Figure 7. China: Current Account Balance, Net Official Assets, Exchange Rates, and GDP Growth in One Episode Billions of dollars 200
Percent
Net official assets
150 100 50 Current account
0 1998
2000
2002
2004
12 10 8 6 4 2 0 –2
Index 130
GDP growth (left scale)
120 110 100
Exchange rate (right scale)
1998
2000
2002
90 2004
Source: International Monetary Fund, International Financial Statistics; International Institute of Finance; World Bank, World Development Indicators; Organization for Economic Cooperation and Development. a. See footnote a of figure 5 for the definition of an episode.
continue to improve slowly. When the current account deteriorates, reserve accumulation will end and the real value of the yen will fall. China has the longest continuing sequence of net reserve accumulation in our sample. From 1990 to 2001 small current account surpluses were roughly matched by reserve accumulation, with little participation by the domestic private sector in international financial markets (figure 7). Since then the current account surplus has grown rapidly, and reserve accumulation has consistently been about double the surplus, as large net inflows of direct investment have been matched by reserve accumulation. Clearly, the reserve buildup since 2002 is unusual by historical measures. We have not seen a sequence of private capital inflows financing reserve accumulation on anything like this scale before. Nor, as the right-hand panel shows, have we yet seen any of the predicted precursors of a successful speculative attack. The real exchange rate fell until 1999, when the nominal rate was fixed. Since then the rate has moved with the dollar, falling by about 4 percent from 1999 through 2003. Recall that this is a real effective rate, so that the standard model would predict a gradual erosion of the authorities’ ability to control inflation. We have seen no evidence of this to date. Finally, high growth rates during the sequence of current account surpluses are clearly a feature of this history. As in the case of Japan, our
Michael Dooley and Peter Garber
169
Figure 8. Korea: Current Account Balance, Net Official Assets, Exchange Rates, and GDP Growth in Two Episodes 1986–89 Percent
Billions of dollars 40
Index
10
150
8
140
30 20
Current account
130
6
10
120
GDP growth 4 (left scale)
0 Net official assets
–10 –20
110
2
Exchange rate (right scale)
0
–30 1982
1984
1986
1988
1990
1982
1984
1986
1988
100 90
1990
1998–2004 40 30 20 10 0 –10 –20 –30
150
10 GDP growth (left scale)
8
130
6 Current account
120 4
110
2 Net official assets 1998
2000
100 Exchange rate (right scale)
0 2002
2004
140
1998
2000
2002
90
2004
Source: International Monetary Fund, International Financial Statistics; International Institute of Finance; World Bank, World Development Indicators; Organization for Economic Cooperation and Development. a. Year in bold refers to the first year after the end of an episode. See footnote a of figure 5 for the definition of an episode.
reading of history is that the reserve accumulation will continue until the current account surplus turns around for other reasons. In China’s case an interruption of direct investment inflows or liberalization of capital outflows might generate a real depreciation and an end to the sequence of reserve accumulations. Korea experienced one sequence of net reserve accumulations that meets our criteria from 1986 to 1989, and a second episode started in 1998 and continues today (figure 8). During the earlier episode, reserve accumulations
170
Brookings Papers on Economic Activity, 1:2005
roughly matched an increasing current account surplus and came to an end when the surplus declined. The real exchange rate rose in the final two years of the episode and declined the year following and for the next two years. In the familiar pattern, growth slowed in the final year of accumulation but then rebounded for the next two years. The episode that started in 1998 is not unusual. Reserve accumulation has approximately matched a U-shaped sequence of current account surpluses, and the real exchange rate has risen. Summary of Findings To conclude, we have looked at a large body of data to evaluate the relevance of the standard model for understanding developments in emerging economies with chronic current account surpluses since 1970. We find almost no support for the standard model, which predicts an eventual speculative attack on a strong currency. Episodes of net reserve accumulation coincide with growing current account surpluses. Reserve accumulations end when the current account surplus declines or (as often happens) swings all the way into deficit. Most important, the real exchange rate weakens at the end of accumulation episodes, and there is generally a small downturn in economic activity. Such a sequence is consistent with a variety of real and financial shocks to the surplus economy. But a real depreciation following the authorities’ decision to stop accumulating reserves is not consistent with a speculative capital inflow or a successful speculative attack. Recall that, in the standard model, the regime ends with a burst of inflation or a forced nominal appreciation of the currency, either of which would be associated with a real appreciation. We do observe “sudden starts” of private capital inflows to finance a current account deficit, but these are associated with a falling real value for the currency, presumably to generate increases in expected yields that draw private capital into the economy. Let us reemphasize what we did not find in the data. We did not find sequences of reserve accumulation followed by revaluations that generated capital losses for the government.15 We did not find sequences of reserve accumulation followed by recessions generated by a real appreciation of
15. Calculations of such book losses have become a central arithmetical exercise among those issuing dire warnings and calling for an end to the system. See, for example, Roubini and Setser (2004) and Eichengreen (forthcoming).
Michael Dooley and Peter Garber
171
the currency. This history suggests to us that the contemporary pattern of current account surpluses can continue in these economies until there is a major negative shock to demand for their exports. A cyclical downturn in the United States might be a likely candidate.
Nothing Lasts Forever The historical record we have presented suggests that most current account surplus regimes have not been terminated by speculative attacks. If this interpretation is correct, there is no obvious constraint on the ability of existing surplus regimes to continue to finance a current account deficit in the center country. A common theme in international finance is that repressed systems do not last forever. We agree they last for no more than twenty years and probably less, but the important point is that they are effective for substantial periods. Of course, what countries can do tells us nothing about what they will want to accomplish. They could listen to the eminent advice and join the Washington consensus and the international finance textbooks by importing capital and developing internally. Our Revived Bretton Woods argument suggests that they will want to do just the opposite. That is, the governments of trade account countries will want to lend to the rest of the world and, in particular, to the center country. And they will counter efforts by the domestic private sector to export capital, through controls and sterilized intervention. An important part of our story is that the real exchange rate distortion will decline over time and vanish at the end of the adjustment period. So the big speculative incentive is front-loaded, and the beginning of a reserve accumulation episode is precisely the time in an emerging economy’s history when financial repression is most likely to be effective. An important constraint on capital inflows into China is the underdeveloped and bankrupt domestic financial market. As the industrial sector grows and that sector lobbies for a better domestic financial system, the whole fabric of financial repression will unravel. But this takes time. In our framework the Chinese government is not accumulating reserves because of a mindless infatuation with a fixed nominal exchange rate. It is instead using a real undervaluation of its currency to limit urban migration and to subsidize rapid industrialization and absorption of unemployed labor. So, at the end of the process, the government anticipates holding a
172
Brookings Papers on Economic Activity, 1:2005
stock of dollar reserves that may or may not generate a capital loss. Clearly, if, as is typical, the renminbi depreciates in real terms, there is no capital loss in the endgame. In any case the government anticipates having by that time a physical capital stock that is larger and more productive than today’s and a labor force that is employed and paying taxes. The one is the prerequisite for the other. The government’s portfolio of interests includes the domestic capital stock as well as foreign exchange reserves; the value of that portfolio should not be maximized locally over its individual subcomponents.
That Old-Time Religion, It’s Good Enough for Me: It Is 1958 The financial press and several widely quoted experts have argued that our comparison of the current international monetary system to the Bretton Woods system is problematic. In particular, they point out that the United States did not finance a large and persistent current account deficit under Bretton Woods, and indeed the mere forecast of such a deficit in the late 1960s was enough to bring the system to a painful end. In addition, unlike in the original Bretton Woods system, there is now a viable alternative reserve currency, the euro, and there are no formal arrangements to prevent reserve diversification. We argue below that this is a misreading of the nature of the system then and now and of the forces that brought the Bretton Woods system to an end. The Old-Time Religion: Balance of Payments Deficits Are Not Current Account Deficits During the Bretton Woods years, the United States did not run large current account deficits, the measure of external imbalance that most draws our attention today. But, in the reckoning of the day, it did run large and persistent balance of payments deficits. The definition of an external deficit that was natural to economists and policymakers at the time seems today to have been forgotten or to be treated as a curious and outmoded accounting convention. Almost all the old-timers focused on a liquidity definition of the balance of payments, which Ragnar Nurkse explained as follows: A country with a deficit in its balance of payments can cover the deficit either by an outflow of gold or an inflow of foreign short-term funds. . . . These funds are equivalent to a loan by foreigners and should be regarded as a draft on the
Michael Dooley and Peter Garber
173
recipient countries stock of international reserves. . . . The foreign short term funds are a liability, can be withdrawn at any moment, and must be treated as a negative gold reserve.16
Notice that this definition implicitly adds elements of the capital account, namely, the balance of trade in longer-term assets, to the current account in order to define a payments imbalance. It emphasizes strictly net flows of gold and short-term claims, that is, liquidity, in defining the balance of payments. Two generations of students of international economics have been kept in the dark about this concept or, at most, trained to think of it as an odd creation of the old-timers, mentally straitjacketed by the completely controlled economies of their day. Yet there it is in the literature: they harped continually about the growing U.S. balance of payments deficits. For instance, in his valedictory on the old international monetary system, French president Charles de Gaulle said . . . . But in addition, the fact that a large number of countries accept, out of principle, dollars in the same way as gold to compensate, when appropriate, any deficits that arise to their advantage from the American balance of payments, leads the United States to become voluntarily indebted to foreign countries. . . . instead of paying them totally in gold, the value of which is real, that you can only possess if you have earned it and that you cannot transfer to others without risk and without sacrifice. . . . The United States, for want of having necessarily to pay in gold, at least totally, for their negative balances of payment in accordance with the old rules, that required countries to take the required steps, sometimes rigorously, to remedy their imbalance, is suffering year after year from a deficit balance. No less because the total of their commercial exchanges is to their disadvantage. Quite the opposite! Their material exports always exceed their imports. But that is also the case for dollars, exports of which are always in excess of imports. In other words, capital sums are being built up in America, by means of what should really be called inflation, which, in the form of dollar loans granted to countries or to private individuals, are being exported. As, in the United States itself, the increase in currency circulation that results from this makes investments within the country less remunerative, there is an increasing trend there to invest abroad. This leads, for certain countries, to a sort of expropriation of some of their companies. . . . But circumstances are such today that we can even wonder how far the problem would go if the countries that hold dollars wanted, sooner or later, to change them into gold? Although such a general movement would never take place, it is still the fact that there is an imbalance that is, to a certain extent, fundamental.17 16. Nurkse (1945, p. 3). 17. Press conference at the Palais de l’Elysée, February 4, 1965.
174
Brookings Papers on Economic Activity, 1:2005
The old fundamentalists said there was a balance of payments problem. The modern secularists say there was not, because the current account was in surplus. So what brought about this change? A change in definition. The Modern Secular View: Yes, They Are The intertemporal maximization model of the international monetary system found in most modern textbooks assumes that the system is based entirely on trust and freely flowing capital. Private international capital transactions dominate and therefore undo official interference. Such transactions are based on the assumption that debtors willingly repay creditors, and those who suffer capital losses willingly repay those who enjoy capital gains without the imposition of infrastructure to secure this result. Observed net and gross capital transfers are interpreted as private intertemporal trade in goods and services. Boiled down to this dimension, goods and services should flow, on net, from high-income, slow-growing economies to lowincome, fast-growing economies so that consumption can be smoothed over time. This flow imbalance can be sustained for a long time and reach high levels because it can be repaid later with surpluses that come from rapid growth. Trust is all that is needed. That this theory generates more puzzles than insights is problematic but has not hindered its dominance. For example, an inconvenient parallel literature on sovereign debt has difficulty concluding that anyone should repay international debt, yet we somehow reconcile ourselves to this contradiction in two basic traditions in international finance. Which Is More Realistic, Collateral or Trust? A unifying conceptual basis for both the original and our Revived Bretton Woods system is the idea that the international monetary system was and is based on collateral, not on trust. Nurkse and his contemporaries believed the international monetary system depended on countries’ willingness and ability to deliver gold on demand. A country’s ability to deliver gold could be instantly reduced by calling its short-term credits. It follows that the liquidity balance was the natural measure of the change in the position of governments, including the government of the center country. It is our contention that the current system also runs on collateral, not on trust. International net saving transfers are too small (except to the United States) because no one trusts a net debtor (the Feldstein-Horioka puzzle).
Michael Dooley and Peter Garber
175
Gross two-way trade in assets is too small because no one trusts a potential loser (the home bias puzzle). In the original Bretton Woods system, the United States was able to provide intermediation services to the world because it posted a stock of collateral in the only form that was acceptable at that time, that is, gold.18 Nurkse was right that the ability of the United States and other countries to participate in international markets was limited by the stock and distribution of gold. A similar implication of our view of collateral in the Revived Bretton Woods system is that a country that wants to participate in private international intermediation has to post collateral. In 1949 the United States had, as it were, the only triple-A credit rating in the system, and so it could hold its own collateral. As de Gaulle pointed out, however, this was no longer the case in 1965, when liquid claims on its collateral were substantial. The key idea in our analysis of the current system is that “earned” U.S. dollar reserve assets have replaced gold as the ultimate reserve asset. The only collateral “asset” that everyone trusts are goods already delivered to the United States by other countries. These goods come to the United States via U.S. current account deficits. Everyone trusts the United States to keep these goods or, what is the same thing, to “default” on U.S. official liabilities to selected foreign governments if those governments steal the private assets of U.S. residents or others, especially in the context of a geopolitical bump. In this sort of default, the Treasury does not cease paying on its own obligations owned by the problematic foreign government. In practice, it has in the past frozen assets, converting them from liquid to completely illiquid claims, placed service payments into blocked accounts, forced longterm rollovers at Treasury bill rates, and redefined the ultimate claimants and recipients of these payments in legal cases, which may emanate from ex post legislation. Moreover, as in Nurkse’s explanation above, a country cannot usefully borrow reserves. It would then have nothing to lose, since it could simply default on its liability.19 In our view reserves and other official or even private foreign-held assets are collateral only if they have been earned by net
18. The United States did sit on its own gold reserve, and so there was some trust even in this arrangement. It also sat on a large chunk of everyone else’s gold. 19. That is why Argentina’s reserves are not collateral, but rather loans from the International Monetary Fund. To seize them would be to seize the IMF’s capital.
176
Brookings Papers on Economic Activity, 1:2005
exports of goods and services. If they are not so earned, they are by definition borrowed. If borrowed, there are no already-delivered goods for the United States to keep, and there is hence no collateral. Critics of the collateral approach argue that the U.S. Treasury would never damage its reputation by defaulting on an official reserve liability. We have two reactions. First, the Treasury has frequently done so in the past. Several such actions are described below, along with a more detailed history of a recent case. Second, we argue that transferring collateral to the rightful owner in circumstances envisioned in the collateral relationship preserves the reputation of the U.S. government both as a debtor and as an impartial and reliable enforcer of collateral arrangements. In delivering its liability to the injured party, the United States is not defaulting on its obligations. It is honoring both its promise to pay and its promise to pay the rightful owner of its obligation. The identity of the rightful owner is conditioned by the terms of the collateral arrangement. Both reputations contribute to the demand for U.S. international reserves. But an important implication of our approach is that the second of the dual roles, that of enforcer of collateral arrangements, is the only unique function of an international reserve currency. In a private collateral arrangement, the rights and obligations of the participants are clear and explicit. The rights and obligations of governments in the collateral arrangement we have described are implicit and necessarily less clear. For example, the event that would trigger transfer of ownership of U.S. official liabilities is not defined, as it would be in a private collateral arrangement. But historical precedents exist. The United States has transferred ownership following major geopolitical incidents such as wars, invasions, revolutions, hostage takings, and nationalizations of foreign investment. That there is uncertainty about what set of events would trigger transfer of collateral does not mean that there are no such events or that private investors do not value the protection offered by collateral in those circumstances. There is also uncertainty concerning what set of creditors to a country would actually benefit from collateral arrangements. But even a random distribution among creditors would be a significant disincentive for a sovereign on the international periphery considering whether to seize assets, provided it had enough collateral at risk. Uncertainty about what events will trigger transfer of collateral and uncertainty about the distribution of
Michael Dooley and Peter Garber
177
the transfer make governments’ collateral less powerful than private collateral. Our conclusion is that more of it is needed to support a given scale of financial intermediation. Ricardo Caballero and A. Krishnamurthy have similarly argued that international collateral is necessary to support private financial intermediation within advanced and emerging economies.20 They also emphasize that an important market failure in emerging economies is the inability to produce assets that can be used as collateral, making it necessary to import such assets. Caballero elsewhere relates this to the private financing of the U.S. current account deficit as follows: “There is an enormous demand for saving instruments in the world, and the US is the most efficient producer of such instruments. No other place combines the volume from new opportunities and ability to generate trustworthy saving instruments from each unit of physical investment put on the ground.”21 An important aspect of their analysis is that financial crises can reduce the supply of collateral assets in emerging economies, and that this might constitute the real costs of such crises. Moreover, even developed financial markets can lose their ability to produce safe assets following a severe financial crisis like that which has plagued Japan in recent years. We are just beginning to explore the economic significance of private and official holdings of international collateral and how the two might interact. Is private collateral a substitute for official holdings of safe assets? Is official collateral necessary for the credibility of cross-border private collateral arrangements? Our framework is based on the idea that official collateral is required because, when trouble comes, private international credit arrangements are enforced, if at all, by governments. There is, of course, ample room for clarification and improvement of our understanding of these mechanisms. But two things seem to us clear. The United States is a source of safe assets that cannot be produced locally in most of the rest of the world. And, since borrowed collateral is an oxymoron, most of the rest of the world has to earn these assets by delivering goods to the United States. Could Europe, offering the euro as an alternative reserve currency, replace the United States as the preferred custodian of collateral? Clearly this is possible. As many observers have recently pointed out, the European 20. Caballero and Krishnamurthy (2001, 2003). 21. Caballero (2004).
178
Brookings Papers on Economic Activity, 1:2005
Union already provides euro-denominated government debt that is a credible promise to pay. Moreover, some diversification from dollars to euros might make sense in terms of a narrow risk-return calculus. But our conjecture is that the dollar will remain the dominant reserve currency as long as the European Union is less willing or less able than the United States to enforce collateral arrangements. Since the European Union has no track record in this regard, it seems unlikely that the euro will soon challenge the position of the dollar in the international monetary system. For this to change would require markets to come to expect that some European governments would be willing to accept large current account deficits and to block the movement of euro reserves as a way of punishing a country (possibly an aggressive one) that expropriates foreign assets. In our view both expectations are unlikely. At this point in history, substituting euros for dollars places collateral out of the reach of creditors and therefore considerably reduces its usefulness. Our approach is based on the view that there is little trust between key countries in the international monetary system. In such a system, everyone sees tremendous benefits from international financial intermediation, but no one can afford the risk of letting another country owe them substantial amounts of goods. The best risk is the central reserve country. Put another way, without trust, the stock of net financial indebtedness must always be less than the stock of collateral that can be seized. In domestic financial markets the stock of real capital that can be pledged as collateral is large relative to credit balances. Although collateral is a universal feature of domestic credit relationships, it is seldom a binding constraint, at least in the aggregate. In international finance just the opposite is the case. Huge stocks of national wealth exist but are useless in creating incentives for repayment, because mass default is often generated by government via the domestic legal system. Some Evidence on the Durability of Reserve Currency Status The International Emergency Economic Powers Act of 1977 (IEEPA, which supplanted the Trading with the Enemy Act of 1917) empowers the president of the United States to freeze foreign-owned assets under U.S. control. The IEEPA authorizes the use of sanctions when the president sees an “unusual and extraordinary threat” to the “national security, foreign policy, or economy” of the United States and declares a national
Michael Dooley and Peter Garber
179
emergency.22 The word “emergency” allows the window to be slammed shut if, for example, a foreign country threatens to launch a financial attack by withdrawing funds or to pull out a substantial amount of funds out in order to prevent their seizure. As described by the U.S. Information Agency, a freeze on foreign-owned assets can be applied selectively to a particular country, or to a group of countries, in time of war or in response to a national emergency. . . . The procedure can be used to serve three purposes: —to deny authorities in blocked countries access to assets that might be used against the US —to protect the true owners of the assets from illegal attempts to seize their property —to create a pool of assets for possible use in settling US claims against blocked countries, or for use as a bargaining chip in negotiating an eventual return to normal relations.23
During World War II, assets owned by Germany, Japan, and Italy were blocked and eventually used in settling war claims against them. Similarly, assets of Hungary, Romania, Latvia, Lithuania, Estonia, Bulgaria, and Czechoslovakia were blocked after these countries fell under Soviet domination. Asset blockings were subsequently imposed against North Korea and China in 1950, Cuba in 1963, North Vietnam in 1964, Rhodesia (now Zimbabwe) in 1965, Kampuchea (Cambodia) in 1975, Iran in 1979, Libya in 1986, Panama in 1988, the Federal Republic of Yugoslavia (Serbia and Montenegro) in 1992, and Afghanistan in 1999. In 1990 the United States blocked $30 billion in assets belonging to Iraq and Kuwait. In 1979 it blocked $12 billion of Iran’s assets, including $5 billion in offshore branches of U.S. banks; part of this was used to pay off syndicated loans by U.S. banks to Iran, and $1.4 billion was sent to the Bank of England to cover claims in the United Kingdom. Another $1 billion was held against awards from the Iran-U.S. claims tribunal.24 These asset freezes have occurred under a variety of circumstances. Some of the asset blockings were aimed at adversaries in a declared or undeclared war (Germany, Japan, Italy, China, North Korea, and Iraq). Some were aimed 22. See the International Emergency Economic Powers Act (IEEPA), United States Code (www.treas.gov/offices/enforcement/ofac/legal/statutes/ieepa.pdf). Assets frozen under the IEEPA are administered by the Treasury’s Office of Foreign Asset Control (OFAC). 23. U.S. Information Agency, “Freeze of Iraq, Kuwait Assets Has Many Precedents,” August 28, 1990 (www.fas.org/news/iraq/1990/900828-152460.htm). 24. U.S. Information Agency, “Freeze of Iraq, Kuwait Assets Has Many Precedents.”
180
Brookings Papers on Economic Activity, 1:2005
at friendly countries that had been occupied, with the aim of preserving the assets pending the restoration of a government recognized by the United States (Latvia, Lithuania, Estonia, and Kuwait). Some countries saw their assets blocked when they opposed the United States geopolitically or became hostile without war breaking out (Cuba, Iran). Some freezes were implemented as part of a global imposition of sanctions (F.R. Yugoslavia, Rhodesia). The differences in circumstances notwithstanding, this history shows that the center country can repeatedly “default” on official liabilities and still remain the only important provider of reserves.
Conclusion The international monetary system must create collateral in order to support international capital transactions. In the industrial countries, the lack of such collateral might account for the relatively small net and gross capital flows among them. Collateral is expensive, and the benefits of trade in financial assets among similar countries are probably not great, even though the legal and expropriation risks are relatively small. For emerging economies, in contrast, the benefits of trade in financial assets are very large. The irony here is that, to accumulate collateral (or “net reserves” to the more traditional among our readers), the emerging economy must export national saving. This is bad from the modern secularist perspective, but it is orthodoxy in the old-time religion. The benefits of two-way trade in financial assets are potentially enormous for countries that have high saving rates but waste the resources thus generated when they are channeled through inept domestic financial systems. These countries need to run the modern version of a liquidity surplus. Some observers have taken a too-legalistic interpretation of our definition of international collateral. We do not argue that any set of private investors in an emerging economy would benefit or would expect to benefit from the collection of collateral by the United States, and in the case of China we have in mind much more the sort of expropriation that might result from a geopolitical clash. Nevertheless, both U.S. and non-U.S. private (portfolio and direct) investors know that an emerging economy that is an international creditor has something to lose from confiscation of its investments abroad. It seems clear to us that European direct investors in Argentina, for example, would have fared much better in recent years if
Michael Dooley and Peter Garber
181
the government of Argentina had owned net assets in the United States. In a general sense our argument is that the government of an emerging economy needs a strong incentive to stay out of the way of private international financial intermediation. Building a positive net international asset position seems to us the obvious way for it to create that incentive. The real potential for globalization of international finance lies in governments of emerging economies posting collateral in the United States to support private two-way trade in financial assets. The current general move in emerging economies, in both Asia and Latin America, toward reducing sovereign debt and building international reserves may be based on an implicit understanding of how the system really works.
APPENDIX A
Data Sources and Methodological Notes Episodes of Official Asset Accumulation We define an episode as a period of three or more years where —the official sector increases its stock of international assets —on average the official sector entirely or partly finances the current account, and —the official sector generates more than 25 percent of the change in national net foreign assets. The second part of the definition is equivalent to the country running current account surpluses during the episode. As described above, the second requirement binds only on average; it is possible to find one or more observations where the country runs current account deficits, although in the data this appears very rarely. Net Official Assets Net official assets are defined as the sum of the following items in the “general government” and “monetary authority” accounts in the balance of payments (all on a net basis): capital transfers, portfolio investment assets (equity and debt, the latter including bonds, notes, and money market
182
Brookings Papers on Economic Activity, 1:2005
instruments), financial derivatives, other investment (trade credits, loans, currency and deposits), and reserve assets. For the following countries, partial quarterly estimations have been calculated for 2004: Japan, Pakistan, Russia, and Ukraine (two-quarter estimations); Denmark, Indonesia, and Korea (three-quarter estimations). Current Account Current account data were obtained from the International Financial Statistics (IFS) of the International Monetary Fund and from the International Institute of Finance (IIF) dataset. All forecasts for 2005 and 2006 come from the IIF dataset. The current account data are expressed as a flow variable in millions of dollars from the end of one fiscal period to the beginning of the next. GDP Growth GDP growth data were obtained from the World Bank’s World Development Indicators and the IIF dataset. All forecasts for 2005 and 2006 come from the IIF dataset. Real Effective Exchange Rates The commonly used definition of the real effective exchange rate is REER = Π i [( e ei ) ( P Pi )] , t
wi
where e is the exchange rate of the subject currency against the dollar (in dollars per subject currency unit), in index form; ei is the exchange rate of currency i against the dollar (in dollars per currency unit), in index form; wi is the weight attached to currency i; P is the consumer price index (CPI) of the subject country; and Pi is the consumer price index of country i. REER data were retrieved from the IFS, IIF, and Organization for Economic Cooperation and Development datasets. Data are reported as index values, where an increase indicates an appreciation of the local currency. Coverage The list of countries in the sample is available from the authors. The initial sample of 164 countries was reduced to 115 because of unavailability of data.
APPENDIX B
Reserve Accumulation Episodes in Hong Kong, India, Russia, and Singapore Hong Kong, 1999–2001 Percent
Billions of dollars 40 30 20 10 0 –10 –20
Current account Net official assets
10 8 6 4 2 0 –2 –4
GDP growth
1996 1998 2000 2002
1996 1998 2000 2002
India, 1976–79 8 6 4 2 0 –2
Current account
Net official assets
10 8 6 4 2 0 –2 –4
GDP growth
1974 1976 1978 1980
1974 1976 1978 1980
Russia, 1999–2004 Index 40 30 20 10 0 –10 –20
Current account
Net official assets
GDP growth (left scale)
10 8 6 4 2 0 –2 –4
1998 2000 2002 2004
Exchange rate (right scale)
150 140 130 120 110 100 90
1998 2000 2002 2004
Singapore, 1974–2000 40 30 20 10 0 –10 –20
Current account Net official assets
1991 1993 1995 1997 1999 2001
10 8 6 4 2 0 –2 –4
Exchange rate (right scale) GDP growth (left scale)
110 90 70 50 30 10
1991 1993 1995 1997 1999 2001
Source: International Monetary Fund, International Financial Statistics; International Institute of Finance; World Bank, World Development Indicators; Organization for Economic Cooperation and Development. a. Year in bold refers to the first year after the end of an episode. See footnote a of figure 5 for the definition of an episode.
184
Brookings Papers on Economic Activity, 1:2005
APPENDIX C
Real Exchange Rate Changes, Reserves, Current Account Balances, and GDP Growth in Reserve Accumulation Episodes Average real depreciation (percent a year)
Cumulative change in reserves (millions of dollars)
Duration (years)
During episode
First year after episode
During episode
First year after episode
4 4 6 15+ 5
6.7 2 −0.2 −4.5 n.a.
−23.9 −26.9 3.7 ... 4.6
8,810 1,053 19,926 410,674 4,287
−2,598 −885 −185 ... −21
Colombia, 1989–94 Croatia, 1993–95 Denmark, 2001–present Egypt, 1990–94 Finland, 1997–2002
6 3 4+ 5 6
2 14.3 3.3 −2.9 −1.2
6.3 0.3 ... 4 4.2
4,728 1,652 13,537 15,615 3,261
−4 533 ... 409 −508
France, 1977–80 France, 1983–86 Germany, 1971–73 Germany, 1976–78 Hong Kong, 1999–2001
4 4 3 3 3
2.3 1.4 4.2 1.2 n.a.
−3.7 −1.1 −1.7 −4.3 −4.4
11,796 10,417 18,312 21,011 24,755
−3,588 −2,602 −523 −3,590 −2,377
Hungary, 1990–92 Hungary, 1995–97 India, 1976–79 Indonesia, 2002–present Ireland, 1993–95
3 3 4 3+ 3
7.9 1.2 n.a. 6.4 −2.2
1.1 0.5 n.a. ... −1
2,905 2,983 6,578 8,245 4,823
2,575 791 −624 ... −52
Italy, 1987–89 Japan, 1986–88 Japan, 1992–96 Japan, 1999–2004 Jordan, 1999–2003
3 3 5 6+ 5
1.4 12.1 1.8 0.4 −0.3
1.3 −4.4 1.6 ... −0.2
25,245 70,747 147,110 398,985 3,411
11,623 −13,058 6,567 ... n.a.
Korea, 1976–78 Korea, 1986–89 Korea, 1998–present Malaysia, 1976–80 Malaysia, 1985–93
3 4 7+ 5 9
−13.5 2.9 −0.6 −3 −2.2
−16.9 −3 ... 3.7 0.4
3,385 13,036 122,894 2,652 25,398
749 −1,208 ... −235 −3,160
Malaysia, 2001–03 Morocco, 1996–2002
3 7
−3 0.3
−3 0.3
14,838 6,782
n.a. 1,649
Episode Argentina, 1976–79 Bulgaria, 1982–85 Canada, 1996–2001 China, 1990–present Colombia, 1976–80
Michael Dooley and Peter Garber
185
Cumulative change in current account balancea (millions of dollars) Real GDP growth (percent a year) During episode
First year after episode
3,120 612 24,902 315,476 1,029
Years of deficit after episode
During episode
Average for three years after episode
−4,774 −951 14,447 ... −1,961
3+ 2 0 0 3+
10 3 4.2 9.3 5.4
−1.5 5.1 2.6 ... 6
−2,185 −413 22,033 −2,937 49,845
−4,527 −1,049 ... −1,000 9,295
3+ 3+ 0 3+ 0
3.8 1.6 1 3.6 3.4
4.4 5.1 ... 5.1 1.9
7,592 −3,646 4,627 16,929 27,309
−4,811 −4,446 9,085 −5,387 12,596
3+ 3+ 0 3+ 0
2.9 1.7 3 3.7 4.7
1.8 3.8 2.3 1.8 2.6
1,134 −3,451 4,441 18,676 5,063
−4,262 −2,228 −1,785 ... 2,049
3+ 3+ 3+ 0 0
−6.2 2.5 2.4 4.1 6.1
1.3 4.7 5.6 ... 9.3
−22,628 249,000 551,000 743,000 1984
−16,479 63,000 97,000 ... 208
3+ 0 0 0 0
3.3 4.5 1.5 1.4 3.9
1.4 4.6 0.3 ... 5.2
−1383 34,634 130,483 130 −6,326
−4,151 −2,003 ... 941 −4,521
3+ 3+ 0 0 3+
10 9.8 4.2 8.6 8.1
3.8 8.1 ... 6.4 9.7
27,885 2,253
14,400 1,593
0 0
3.3 4
5.8 3.7 (continued )
186
Brookings Papers on Economic Activity, 1:2005 Average real depreciation (percent a year)
Cumulative change in reserves (millions of dollars)
Duration (years)
During episode
First year after episode
During episode
First year after episode
Netherlands, 1970–77 Netherlands, 1992–95 Norway, 1980–85 Norway, 1993–97 Norway, 1999–2003
8 4 6 5 5
1.5 2.2 1 −0.5 1.7
−1.3 −1.8 1 −1.7 −4.8
3,902 11,349 9,442 14,353 13,393
−770 −5,695 −3,211 −6,384 n.a.
Oman, 1989–92 Oman, 1999–2001 Pakistan, 2001–present Portugal, 1985–92 Romania, 1986–89
4 3 4+ 8 4
−1.4 −1.7 −0.3 2.2 −2.7
−6.6 −4.9 ... −0.4 −31.1
1,302 3,483 10,244 16,235 1,614
−1,058 309 ... −2,848 −1,494
−0.1 1.6 −0.3 2.9 −0.4
7.2 −3.3 −2.1 −1 0.2
63,731 11,432 81,135 2,030 1,499
n.a. −1,509 −861 503 −753
−2.7 2.1 −4.7 11.3
−0.1 −0.3 ... −5.9
5,306 10,189 5,383 7,839
−1,048 −2,382 3,092 −8,160
Episode
Russia, 1999–2004 Saudi Arabia, 1979–82 Singapore, 1974–2000 South Africa, 1989–91 Sweden, 1970–73 Sweden, 1998–2000 Switzerland, 1982–87 Ukraine, 1999–present Venezuela, 1979–81
6 4 27 3 4 3 6 6+ 3
Sources: International Monetary Fund, International Financial Statistics; International Institute of Finance; World Bank, World Development Indicators; Organization for Economic Cooperation and Development. a. A positive number indicates a change in the direction of surplus. n.a., data not available; . . . , not applicable.
Michael Dooley and Peter Garber
187
Cumulative change in current account balancea (millions of dollars) Real GDP growth (percent a year) During episode
First year after episode
16,476 63,117 11,736 33,520 113,812
−904 21,502 −4,551 6 n.a.
1 0 3+ 0 0
3.6 2 3.4 4.6 1.9
2 3.7 1.9 2.5 n.a.
563 5 9,277 −14 9,874
−1,190 2 ... 233 −3,254
3+ 0 0 0 3+
6.7 4.3 5.3 4.6 −0.8
4.6 3.2 ... 1.1 −9.1
199,007 98,912 116,349 5,734 2,149
40,800 −16,852 16,137 1,967 −529
0 3+ 0 0 3+
6.8 2.5 7.5 0.4 3.4
5.2 −5.2 0.7 0.8 2.3
29,648 26,869 10,606 9,078
8,531 8,846 ... −4,246
0 0 0 1
4.2 1.9 8.3 −1.3
1.5 3.7 ... −1.5
Years of deficit after episode
During episode
Average for three years after episode
Comments and Discussion Barry Eichengreen: The first rule of forecasting is, “Give them a forecast or give them a date, but never give them both.”1 Michael Dooley and Peter Garber have given us a forecast, namely, that the dollar will fall and U.S. Treasury yields will rise. Bravely, they have also given us a date. Unfortunately for those of us interested in the future, that date is 1971. Like Dooley and Garber, I agree that what cannot go on forever generally will not. But unlike them I do not believe that recent events in financial markets can help us pin down the timing. The middle of March, just before the Brookings Panel meeting, saw an increase in noise about the possibility that foreign central banks might diversify out of dollars. The governor of the Bank of Korea made some widely reported comments about the need for more-active reserve management. Prime Minister Junichiro Koizumi of Japan told a parliamentary committee that reserve diversification was “necessary.”2 Y. V. Reddy, governor of the Reserve Bank of India, said that the diversification of reserves was under active discussion.3 Ukrainian economy minister Sergiy Teriokhin argued publicly that the country should diversify its reserves out of dollars and into euros.4 This upsurge in noise was associated with an eight-month high in Treasury yields, reinforcing the belief that reserve diversification could eventually force the dollar down and Treasury yields up. At the same time, that eight-month high in Treasury yields was not all that high. I would acknowledge that this is a troubling point. I am not alone, of 1. Attributed to Edgar R. Fiedler by Dickson (1978). 2. Steve Johnson, “Dollar Wobbles on Japan Diversification Talk,” Financial Times, March 10, 2005. 3. Reuters, “Asian Foreign Exchange Reserves: A $2.46 Trillion Question,” March 11, 2005, 11:36 AM. 4. Reuters, March 16, 2005, 12:50PM.
188
Michael Dooley and Peter Garber
189
course: Federal Reserve Chairman Alan Greenspan has commented on this issue extensively, to the point where it is now known as the Greenspan conundrum. Factors invoked to help explain it include the relatively short supply of new long-term Treasury debt coming onto the market as the debt managers at the U.S. Treasury shorten maturities, and the inelastic demands of various institutional investors for government securities. In Dooley and Garber’s view, the proper interpretation is that financial market participants are attaching a positive probability to Asian central banks continuing to support the dollar by making massive purchases of Treasury bonds. This is the substance of the first of the authors’ three notes. Who am I to second-guess the markets, much less to second-guess our esteemed authors? Well, I’m an economic historian who can recall a substantial number of previous episodes where major imbalances leading to sharp changes in exchange rates were not obviously factored into financial markets until immediately before the event. For example, the JanuaryMarch 1933 run on the dollar, a suggestive precedent, had virtually no discernible impact on interest rate differentials or forward exchange rates until almost immediately before it occurred, despite the fact that the possibility had been actively discussed for the better part of a year.5 The 1992 attacks on the pound sterling, a currency that commentators regularly cited as ready for a fall, were similarly not preceded by the emergence of noticeable interest rate differentials or a forward discount in the foreign exchange market until a couple of weeks before the denouement.6 Particularly interesting, given the context, is that in 1968–71, in the run-up to the collapse of the Bretton Woods system, the forward discount on the dollar was very modest, as was the interest rate differential between the United States and Germany.7 Then, in the summer of 1971, the forward discount jumped upward. Although one can always ascribe such behavior to the arrival of new information, it is not as if people failed to see the collapse of the Bretton Woods system and a substantial devaluation of the dollar coming. To the contrary, there was an immense contemporary literature warn-
5. See Hsieh and Romer (2001). 6. See Eichengreen and Wyplosz (1993). 7. See Obstfeld (1993). Imperfect capital mobility complicates the drawing of inferences from these data, although, as Obstfeld notes, capital mobility was rising strongly toward the end of the Bretton Woods period. Imperfect capital mobility also complicates drawing inferences from interest rates today, as Dooley and Garber themselves note.
190
Brookings Papers on Economic Activity, 1:2005
ing that the system would dissolve and that the dollar would have to fall substantially. Yet there seemed to be a striking reluctance to take a position on this basis until one minute before the clock struck midnight. This behavior may be hard to reconcile with perfect foresight, but, if it exists, asset prices will not be telling us when it is 11:58.8 Dooley and Garber’s second note is an analysis of the end of several episodes of large current account surpluses accompanied by reserve accumulations. The authors’ intriguing finding is that many such episodes come to an unhappy end with a sharp real and nominal depreciation, which is not how most observers of China expect the current situation to play out. I would simply observe that how ancillary variables like the real exchange rate will react when China’s accumulation of reserves ends will depend on why it ends. If it ends because the People’s Bank of China and the government, observing robust economic growth and mounting inflationary pressure, choose to tighten monetary policy in order to cool fears of overheating, the currency will appreciate. But if they wait until domestic financial excesses further infect the banking system, creating a crisis of confidence, there may instead be a scramble to get out, causing the currency to crash. The wisdom of moving away from the peg while confidence is strong, capital is still flowing in, and reserves are still being accumulated is the central lesson of the literature on exit strategies.9 It provides the strongest argument for why China should abandon its peg to the dollar now. Dooley and Garber’s third note extends an earlier paper of theirs that characterizes the role of the United States in the current international system as providing financial intermediation services to the rest of the world.10 The United States borrows short, indeed increasingly short given the shorter and shorter tenor of Treasury debt, and lends long in the form
8. It is worth recalling that, in the original Krugman-Flood-Garber model, interest rates remain perfectly stable until the moment the exchange rate collapses, despite the fact that everyone has perfect foresight and can see the end coming. I understand, of course, that the interest rate in their model is the instantaneous rate, not the long-term rate, and that the result depends on some restrictive assumptions. But add to this the well-known fact that the markets appear to attach a heavier weight to short-term rates than the term-structure hypothesis would lead one to predict, and it is possible to reconcile the conviction that the end is near with the observed low level of long-term rates. 9. See, for example, Eichengreen and Masson and others (1998) and, for an application to China, Eichengreen (forthcoming). 10. Dooley, Folkerts-Landau, and Garber (2004b).
Michael Dooley and Peter Garber
191
of foreign direct investment (FDI). It thus provides liquidity transformation services like those of a bank. It enhances the efficiency of resource allocation, also like a bank. Again, there is a suggestive analogy with the Bretton Woods system, there having been an argument in that period that the United States was acting as banker to the world, borrowing short and lending long. It was similarly argued that the U.S. deficit on liquidity account was not a problem because foreigners were simply the happy recipients of these intermediation services.11 But there is a difference between the Bretton Woods era and today, namely, that the present situation occurs against the backdrop of large, ongoing current account deficits for the country that is banker to the world. In principle, there is no reason why the country with the most efficient financial system, which is therefore providing intermediation services to the rest of the world, cannot run a balanced current account or, for that matter, a surplus. There is no reason why importing short-term capital and exporting long-term capital should also require it to run a current account deficit, as the United States is doing. The United States ran current account surpluses following World War II, even after contemporary economists stopped referring to the dollar gap. Britain ran persistent current account surpluses before World War I, when it was similarly acting as banker to the world. Being an international financial center and providing maturity transformation services to the rest of the world does not doom a country to current account deficits. So China must be buying something else through its bilateral surpluses and heavy investment in U.S. Treasury bonds. According to Dooley and Garber, it is buying custodial services. The United States, in their view, is now custodian to the world (this is not a comment on the postindustrial economy). In other words, the United States holds the collateral against which countries like China are able to borrow. China can then attract FDI from abroad, because if it ever decided to nationalize U.S. or other private assets, the United States would then default on its official liabilities held by the Chinese government. This is a provocative hypothesis whose validity is highly questionable. For one thing, the story is specific to China, whereas the accumulation of 11. See Despres, Kindleberger, and Salant (1966). Others objected to this view on the grounds that the capital inflow was really the temporary balancing item that offset a U.S. current account surplus that was too small to fully finance U.S. FDI abroad.
192
Brookings Papers on Economic Activity, 1:2005
reserves and chronic surpluses in trade with the United States is a pan-Asian phenomenon. No one worries that Japan, Korea, or Taiwan will expropriate U.S. investments, yet they, too, hold massive claims on the United States. I am not aware of any U.S. corporate executives pointing to China’s large dollar reserves as a form of collateral justifying their decision to invest there. Nor am I aware of statements by Chinese officials explaining that they are accumulating Treasuries as a way of posting collateral for FDI inflows. Dooley and Garber rightly emphasize the importance of looking at the current international financial system through the eyes and statements of Asian officials. Here is an instance where their point works against them. Moreover, the timing is wrong: U.S. FDI in China began to rise around 1992, yet the massive reserve accumulation started nearly a decade later. Then there is the fact that the United States accounts for only a small fraction of FDI in China: Morris Goldstein and Nicholas Lardy find that it accounts for less than 10 percent.12 Thus one must assume that the United States would be willing to go to bat not just on behalf of U.S. private foreign investors but also on behalf of investors from other countries. In addition, the way foreign investments in China have been expropriated historically is through the surreptitious stripping of assets by Chinese managers and joint-venture partners. It is hard to imagine that the U.S. government would risk tarnishing its public credit in response to such behavior. Rather, one has to assume a major geopolitical blow-up between the United States and China, a decision by Beijing to freeze all U.S. investments there, and the U.S. government retaliating by freezing all U.S. Treasury bonds held by China in custody in the United States. Such events are not beyond the realm of possibility, but they do not exactly strike me as an obvious way of explaining the current pattern of global imbalances. This suggests testing the hypothesis with a systematic analysis of the impact of inward FDI and property rights protection on the demand for reserves. Specifically, I am imagining a regression of the level of reserves on FDI, where the coefficient on the latter is expected to be larger in economies where property rights are less secure, the government is communist, and the country is not politically allied with the United States. Of course, one would have to control for the other standard determinants of the demand for reserves and correct for the endogeneity of FDI. But this
12. Goldstein and Lardy (2005).
Michael Dooley and Peter Garber
193
would be a much more direct way of marshalling evidence for Dooley and Garber’s hypothesis. If one accepts that the collateral story is implausible or at best unsubstantiated, how then to explain the Chinese authorities’ insistence on keeping their currency down against the dollar? We are forced to fall back on the traditional rationale for export-led growth. The export sector, in this account, is the locus of knowledge spillovers and productivity growth in a developing economy. Distortions affecting the economy justify the imposition of another distortion in the form of an undervalued currency, which pushes more resources into the export sector than would occur under the unfettered operation of market forces. The original distortion might be the fact that the productivity effects associated with producing for export are external to the firm, providing an inadequate incentive for private investors to shift resources into the sector absent other interventions. Or it might be an inefficient financial system that prevents saving in the developing economy from underwriting adequate investment in the export sector. Or it might be a shortage of organizational knowledge that is strongly complementary with exports and can only be augmented by export-linked FDI that imports this organizational knowledge from abroad. Which of these distortions provides the primary motivation for pursuing the export-led growth strategy matters importantly for how quickly one should expect the government to move away from current arrangements. I have the sense that Chinese managers and entrepreneurs are rather quickly gaining the organizational knowledge necessary to run a modern, exportoriented manufacturing firm. I also have the sense that the productivity effects from learning by exporting are internal as well as external to the firm, much as they are in other countries. In other words, the time may not be very long in coming when these justifications for keeping the exchange rate artificially low are no stronger in China than in a variety of other middle-income countries. The strongest argument in favor of the indefinite maintenance of the status quo is that the export sector, where productivity is higher than in the rest of the economy, is being starved of funds by an inefficient Chinese banking system and that an undervalued renminbi is needed to offset this distortion, perhaps by artificially boosting the prices of traded goods relative to nontraded goods and thus enhancing the profitability of investing in the traded goods sector. But this would boost the prices of traded goods across the board, whether manufactures or agricultural products, and whether
194
Brookings Papers on Economic Activity, 1:2005
produced by state enterprises or by private firms. This hardly looks like a sensible strategy for enhancing efficiency.13 Alternatively, an undervalued renminbi could lead to China accumulating Treasuries and thus encouraging efficiently allocated FDI in China. This is the reasoning that forces Dooley and Garber into the logical corner in which they now find themselves. It also brings us back to my earlier objections to the collateral-and-selectivedefault story. If one rejects, as I do, their collateral argument, one must hang one’s hat on the first two distortions: learning spillovers external to the firm, and shortages of organizational knowledge. It is then hard to resist the conclusion that these justifications for an undervalued currency will grow less compelling rather quickly, for the reasons I have just enumerated. As the authors point out, capital losses on dollar-denominated reserves should be counterbalanced against the returns, and these returns are likely to be strongly declining. In addition, the liability side of the equation should include also the other costs of undervaluation, such as the limited extent of monetary control, the incentive under current conditions for large amounts of credit to flow into the property market (heightening financial fragility), and the additional difficulty that this poses for efforts to raise lending standards and otherwise strengthen the banking system. These feature not at all in Dooley and Garber’s analysis. These are all reasons, then, why Asian governments are likely to move away from the strategy of export-led growth through undervaluation before long. (But recall the first rule of forecasting.) Once countries see their neighbors doing so, and thereby lending less support to the dollar, there will be an obvious incentive not to be late in jumping off the bandwagon.14 At that point the familiar models of speculative attacks, which the authors helped to pioneer, will not just be “dancing in our heads.”
13. There is also the question of whether undervaluing relative to the dollar is an appropriate strategy for boosting exports in general. The United States takes about a third of China’s exports (including exports that go via Hong Kong). Even if one aggregates the other dollar peggers, the share of the “dollar area” is still only 40 percent. Europe meanwhile takes 25 percent of Chinese exports. Over the last ten years the dollar has risen as well as fallen against the euro. Between 1994 and 2002 it rose very substantially. Was Chinese policymakers’ preference then for increasing overvaluation on an effective basis? 14. As emphasized in Eichengreen (2004).
Michael Dooley and Peter Garber
195
Jeffrey A. Frankel: In a recent series of papers that have deservedly received a lot of attention, Michael Dooley and Peter Garber (usually with David Folkerts-Landau) have put forward the view that the essential elements of the current international monetary situation bear a strong resemblance to the old Bretton Woods system, with the Asian central banks today playing the role of dollar accumulators that Europe played in the 1960s. There is a lot of insight in their views, which are all the more useful for their unconventional perspective and language. This is not just another recitation of familiar positions in a fixed-versus-floating debate. For example, I think the authors are right that “The problem for China is to mobilize its existing enormous domestic saving to create a growing, internationally competitive capital stock that can rapidly employ hundreds of millions of workers in productive activity. A serious constraint is the lack of a domestic financial system capable of channeling this saving into productive capital, technology, and management skills.” To link these issues to the U.S. balance of payments situation was an original, imaginative, and thought-provoking leap. It is fair to claim, as the authors do, that much of their analysis has now been generally accepted. This is not to say that one has to credit them for the insight that Asian central banks are now financing much of the U.S. current account deficit; that much is obvious. And, at the other end of the argument, I think the authors have been forced to clarify that they are not claiming that the current system can continue indefinitely—that the dollar will never have to depreciate—which many of us thought we heard them saying at first.1 But I do credit them with the useful insight that the willingness of Asia, and especially China, to pile up unheard-of quantities of dollars is not simple myopia or mercantilism, but rather part of a deliberate strategy of export-led growth, which may not be foolish given the structural limitations of their financial and corporate governance systems. I will begin by restating the authors’ case. It is not always easy to understand everything they say, perhaps because of the novelty of much of it. I will try to be helpful in defending them against some of the critiques that have demanded their response. Indeed, I may carry the parallel with
1. See, for example, Eichengreen (2004). Dooley, Folkerts-Landau, and Garber (2003, abstract) wrote, “there is a line of countries waiting [to follow the development strategy of keeping their currencies undervalued against the dollar] sufficient to keep the system intact for the foreseeable future.”
196
Brookings Papers on Economic Activity, 1:2005
the Bretton Woods period further than they themselves intended or would like. I will then, in some respects, try to suggest some different (and perhaps less unconventional) language to make what seems to me the same argument. On the key question posed by the title of their paper, however, which amounts to asking whether we are at the beginning of the revived Bretton Woods system or the end, I come down on the latter side, or, more precisely, that we are closer to the end than to the beginning. One could argue that we are actually at the end, 1971, since the dollar depreciated substantially against most major currencies during 2002–04, as so many of us predicted it would when the United States shifted macroeconomic gears four years ago. But the system may still have a little ways to go in other respects, such as the much-anticipated end of the peg of the renminbi to the dollar. I shall argue that in one respect we may be at 1967. IS THE BRETTON WOODS ANALOGY APPROPRIATE? The five features of the current system that the authors enumerate at the beginning of their paper in fact bear little resemblance to the Bretton Woods system. Certainly, with no mention of pegged exchange rates or gold, their list bears little relation to the system that was agreed upon at Bretton Woods, New Hampshire, in 1944. But I think readers are meant to take “Bretton Woods” to refer to the de facto system under which all other currencies were pegged to the dollar, and all other central banks held dollars as the reserve asset because the dollar was convertible into gold.2 Specifically, the “Bretton Woods” period in this sense refers to the years from 1958, when the European countries restored convertibility as envisaged in the International Monetary Fund Articles of Agreement, to 1971, when the dollar was devalued, or perhaps just to 1968, when the United States ceased to allow foreign citizens to convert dollars to gold. The ten to thirteen years that this de facto system lasted is not very long in the broad scheme of things. One might even question whether it was a “system,” given that it was already breaking down throughout much of that period, under strain from the U.S. balance of payments deficit. But almost everyone has long understood these points. That the dollar is still the main international reserve currency today, that some Asian central banks see it in their interest to pile up large quantities of dollars, and that their doing so finances a large U.S. balance of payments deficit like those of the
2. See, for example, McKinnon (1996, especially p. 41).
Michael Dooley and Peter Garber
197
1960s, together probably suffice to justify the analogy with Bretton Woods. The picture is indeed that of a “system,” in the sense that it shows how the two halves, the United States and the periphery, fit together into a whole— a mode of analysis that has largely died out from international macroeconomics since 1971, as Barry Eichengreen points out in his comment. True, only China and a couple of other Asian economies (Hong Kong and Malaysia) have currencies that are fixed to the dollar today, even de facto. But dollar purchases by foreign central banks, especially in Asia, seeking to prevent their currencies from appreciating against the dollar have nonetheless been very large recently. They have financed an increasing share of the U.S. current account deficit, which itself has been widening sharply, over the last four years. In the third of the three “notes” that constitute the Dooley-Garber paper, the authors take pains to defend themselves against the criticism that the Bretton Woods analogy is invalid because the United States was not running a current account deficit in the 1960s as it is today. I agree with them on this. In the first place, the long-term trend in the U.S. current account balance during the Bretton Woods period was negative, even if the balance was usually greater than zero. The United States had run substantial surpluses in the late 1940s, because it had emerged from World War II with its productive capacity intact. (This was called the period of “dollar shortage.”) The surpluses diminished in the 1950s. True, they then recovered a bit, peaking in 1964. But from then on the general trend was downward until the system fell apart in 1971. In the second place, more-comprehensive measures of the balance of payments, starting with the basic balance (which includes foreign direct investment and other long-term capital flows) and the liquidity balance (which adds short-term, nonliquid flows and errors and omissions), did turn negative. The overall U.S. balance of payments went into deficit in 1958, which is presumably why the authors single out that year for their Bretton Woods analogy. These deficits defined the 1960s as a period of excess supply of dollars.3 Both in Europe in the 1960s and in China today, rising foreign direct investment (FDI)—bilateral FDI as well as overall outward FDI from the 3. The United States was losing reserves throughout 1958–67, and in large amounts during 1970–71, which forced the devaluation and the closing of the gold window. Foreign central banks were also piling up dollars, and in ever-larger magnitudes, through most of the 1960s, and especially starting in 1970. The trend in 1968–69 was actually in the other direction, perhaps because the United States had begun to make an effort to tighten fiscal policy with a tax surcharge.
198
Brookings Papers on Economic Activity, 1:2005
United States and overall inward FDI to the partners in question—was an important part of the story, indeed just as important as the trade balance. One difference is that in Europe the voices decrying “the American challenge” were louder than those favoring the entry of foreign multinational companies as part of an intelligent growth strategy, whereas China seems more welcoming to FDI. When the French complained about the United States’ “exorbitant privilege”—the ability to trade pieces of paper for items of more tangible value—they had in mind Americans acquiring French factories as much as, or more than, French goods. So I think the authors are on firm ground in focusing on a broader measure of the balance of payments, one that includes direct investment, such as the basic balance or the liquidity balance. TERMINOLOGY. It may be worth spending a moment on language choices, so that we are all sure we are talking about the same things. I have some translating to propose. The authors use the phrase “trade account country” to refer to a country that offsets current account imbalances with large changes in official reserves. They particularly have in mind China, which has been running a trade account surplus and offsetting it with increases in reserves. They use the phrase “capital account country” to refer to one that offsets its current account deficits with capital inflows. (Then there is the third set of countries—Europe and a few others—that are floaters.) In standard models the first group would be characterized by low capital mobility and a high propensity for the authorities to intervene in foreign exchange markets to stabilize the exchange rate, and the second group by high capital mobility and less of a tendency to intervene. I am afraid that I don’t care for the terms “trade account country” and “capital account country.” Each time I read them, I have to stop and remind myself which one is which in the authors’ scheme. I think it would be much more intuitive to call Asia the “exporting countries” and the United States the “consuming country.” AMERICA AS THE WORLD’S BANKER. I learned from Charles Kindleberger thirty years ago that one way to think of the chronic willingness of other countries to absorb dollars is that the United States acts as the world’s banker, taking short-term deposits and investing in longer-term, higher-risk, higher-return assets such as FDI.4 Thus I accept the idea that
4. Kindleberger (1965).
Michael Dooley and Peter Garber
199
both parties stand to gain from an exchange of gross capital flows, whereby U.S. companies undertake direct investment in China and China acquires dollar bonds. Notwithstanding that U.S. Treasury bills tend to pay a much lower rate of return on average than the United States earns on its investments in other countries, I accept the argument that this can be part of a useful development strategy. (I leave it to others to reconcile this with the fact that the recent surges of capital into China have been portfolio investment, not FDI, and that even the FDI has for the most part not come from the United States.5) I am still, however, mulling over the notion that FDI is, as the authors put it, the “lesser credit.” This must mean that the danger of expropriation by China’s government is greater than the danger that the United States will depreciate away the value of the bonds, and that therefore China has to offer collateral to the United States, not vice versa, and that supplying exports up front constitutes this collateral. It is an interesting notion. But while I am thinking about this, I have a question. Where do private short-term liquid portfolio flows belong in the story? Are we sure it is the liquidity balance that matters, as the authors say, and not the official settlements balance (or official reserve transactions balance), which is after all the most comprehensive measure of the balance of payments? Why not also include short-term portfolio capital flows, even those that are liquid? In other words, why not draw the line so that all private transactions appear above the line, and nothing besides official reserve transactions appears below it? Isn’t the main point that the United States can run deficits on the overall balance of payments, and that other countries are forced to run corresponding surpluses in order to earn foreign exchange reserves? Wasn’t that the point of the two-country version of the monetary approach to the balance of payments in the 1960s?6 Don’t the arguments about how developing countries are forced to post “collateral” against capital inflows apply just as much—actually, more—to their inflows of shortterm portfolio capital as to FDI? If it is just the U.S. monetary authorities who are playing the role of world banker, the answer is that what matters is the official settlements balance. If it is the entire U.S. commercial banking sector, what matters is the liquidity balance. The language of private versus public domination of 5. See, for example, Prasad and Wei (2005). 6. See, for example, Mundell (1971) and Dornbusch (1973).
200
Brookings Papers on Economic Activity, 1:2005
the capital account makes it sound as if the official settlements balance were key, that one should draw the line below all private transactions, even liquid short-term banking flows. Consistent with this, the biggest increase in China’s balance of payments over the last few years has been neither in the current account nor in FDI, but rather in unmeasured inflows that are generally considered to be speculative and short-term: it is Chinese citizens bringing back onshore liquid investments that they had previously managed to accumulate offshore. Nor is the circumstance unique to China of short-term capital flowing from the United States to the emerging market, that is, in the same direction as the FDI. This has been a feature of other emerging markets, both during the most recent boom phase of the international debt cycle (2002–05) and during the preceding one (1990–96). Hence all the talk about how capital inflows are more likely to lead to crises if their composition is tilted in the direction of short-term loans, unless they are fully offset by reserves. One reason to skip directly to the official settlements balance is that the liquidity balance and the others are no longer computed. The United States stopped trying to calculate these statistics long ago, in part because the capital account statistics are not reliable, in part because it is difficult even in principle to define what is a short-term and what is a long-term flow, and in part because the entire exercise of distinguishing between autonomous and accommodating transactions became obsolete with the end of the Bretton Woods system. (The “basic balance,” however, is still reported for some countries, such as Japan.) LONG-TERM INTEREST RATES. The authors cite low real interest rates as evidence that the world is experiencing a glut of saving—that the relationship is being pushed by the Asian desire to save, rather than pulled by the U.S. desire to consume. It is true, as they say, that “the fact of unusually low long-term real interest rates for this stage of the business cycle [is] a direct challenge to those who . . . claim that the end is near.” My view is that one of the ways economists can most usefully contribute to real-time analysis of the economy is by pointing out when some market price seems to be out of line with historical relationships, the implication being that it is likely to correct itself within a couple of years. (Examples include the undervalued U.S. stock market in 1980, the overvalued dollar in 1985, the overvalued yen in 1995, the overvalued stock market in 2000, and the undervalued euro in 2002.) Perhaps those observers are right who say that the recent coexistence of low U.S. interest rates with large U.S. budget
Michael Dooley and Peter Garber
201
deficits shows that there has been a major fundamental shift in the willingness of global investors to hold U.S. debt. But I am willing to bet against it. Here is one testable disagreement. My view is that the bond market was buoyed over 2001–04 by three factors, each of which is already starting to come to an end: —easy monetary policy in the United States, that is, the purchases of U.S. Treasury securities by the Federal Reserve (which began to reverse in the summer of 2004) —easy monetary policy in Asia, that is, the purchases of U.S. securities by Asian central banks (who get center stage in the Dooley-Garber story), and —the fact that investors appear still to be putting some weight on official government projections of declining future U.S. budget deficits, despite all the reasons to disbelieve them. As these three factors come to an end, nominal long-term interest rates should rise from their current levels, near 4 percent at the time of this writing, to above 6 percent. I calculate this as approximately 21⁄2 percent inflation plus 2 percent for a normal real short-term rate plus 1 percent for a normal term premium plus 1 percent as an extra term premium for an expected path of a rising debt-GDP ratio. This does not even count the proposed dumping of huge quantities of new U.S. Treasury bonds on the market to fund a transition to privatized Social Security accounts. Nor does it count possible unforeseen factors such as further instability coming from the Middle East or new oil price increases. It seems to me that a crash is more likely to come in the bond market than anywhere else. But time will tell. I would also guess that the dollar will resume its depreciation long before all unemployed or badly employed labor in China is reallocated to worldclass production (which the authors say will take at least ten years). It is now in China’s own interest to move away from the peg. It has more reserves than it can use.7 But the Chinese authorities do not want to be pushed into revaluing the renminbi against the dollar.8 It is safer to bet 7. Frankel (2005b). 8. Li Ruogu, deputy governor of the People’s Bank of China, may have best captured the Chinese perspective when he said in May 2004, “I think those who call for a fixed exchange rate are right in the short run. And those who call for a floating exchange rate are right in the long run. How long is the short run, you ask? You must understand. China is 8,000 years old. So when I say, short run, it could be 100 years” (author’s paraphrase).
202
Brookings Papers on Economic Activity, 1:2005
against the U.S. bond market: I think it will probably fall before the dollar does, or at worst at the same time. Perhaps my answer to the question of whether it is now 1958 or 1968 is that it is 1967. In that year the seriousness of the U.S. balance of payments imbalance had become clear, yet large increases in government spending were still coming out of Washington—much of that spending, then as now, on a foreign military adventure—with no evidence of any willingness to pay for it by raising taxes.9 Perhaps if it had not been for the Vietnam War and the Great Society programs—the determination to have both the guns and the butter—and the associated fiscal and monetary profligacy, the Triffin dilemma would have taken decades to work itself out fully.10 But the excessively expansionary U.S. macroeconomic policy accelerated the process, so that the end came in 1971. To my way of thinking, a fundamental systemic structure that produces rising U.S. balance of payments deficits, combined with excessively expansionary macroeconomic policies that accelerate the trend, sounds like the situation today. It is in this respect that I think the current period resembles the 1960s even more, perhaps, than Dooley and Garber think. But it is also for this reason that I think the situation is unlikely to be sustainable for ten or twenty years. Incidentally, easy monetary policy kept real interest rates fairly low in 1967 as well: the federal funds rate was 4.2 percent and the ten-year yield on government bonds was 5.1 percent, compared with consumer price inflation at 3.0 percent. Thus the Johnson deficits seem to be a better precedent for the Bush deficits than are the Reagan deficits, which were not initially accommodated by monetary policy. THE EURO AS A RIVAL FOR THE DOLLAR. I both agree and disagree with the authors’ rejection of recent media reports that central banks are jumping on a bandwagon of diversification of their reserves out of dollars. Where I agree is that there is an element of hysteria to such reports, which come in highly predictable waves every time the dollar has been depreciating for a couple of years in a row. The most recent previous cycle occurred in 1994–95, when articles suggested that the dollar might lose its status as the unrivaled international reserve currency, that it was in danger of “going
9. See, for example, Solomon (1977, pp. 102–04). In 1968 Congress did pass an income tax surcharge to help pay for the spending. It was too little, or too late, or both, to head off rising fiscal deficits and inflation. 10. Triffin (1960).
Michael Dooley and Peter Garber
203
the way of sterling, the guilder, the ducat and the bezant.”11 The recent cycle of alarmist articles in the financial media was not hard to predict. The earlier cycle of alarms was wrong. In the first place, shares of reserve holdings in central banks and other measures of international currency use change only slowly over time. In the second place, these measures in the 1990s actually showed that the dollar’s share had reversed the downward trend of the 1970s and 1980s.12 But this time may be different, and this is where I am less confident than the authors. The share of the dollar in international reserves has again resumed its downward trend. And there are two or three available explanations. One is that there is now an obvious rival, the euro, which is more credible as an alternative than the deutsche mark or the yen ever were. Another is that the United States now has a net international debt that is large and rapidly rising. Some argue that, because central banks in Asia hold so many dollars already, they will be reluctant to diversify into other assets for fear of precipitating a sharp depreciation of the dollar, in which the value of their holdings will be the hardest hit. No doubt individual central banks have this concern. But, as often, I agree with Eichengreen: when each individual participant decides that it stands to lose more by holding pat than by joining the run, it will act in its own self-interest.13 If narrow economic self-interest is not sufficient to stop the slide, will enlightened geopolitical calculation do it? In the meantime, the United States has lost popular sympathy and political support in much of the rest of the world. Our past deficits due to imperial overstretch were manageable when others paid the bills for our troops overseas: Germany and Japan during the Cold War, Kuwait and Saudi Arabia in 1991. Now the hegemon has lost its claim to legitimacy in the eyes of many. The next time the United States asks other central banks to bail out the dollar, will they be as willing to do so as Europe was in the 1960s, or as Japan was in the late 1980s after the Louvre Accord? It seems unlikely. Menzie Chinn and I have documented econometrically some of the commonly hypothesized determinants of reserve currency shares: size of the home economy, rates of return, stability of the currency, historical inertia,
11. Kindleberger (1995, p. 6). 12. Frankel (1995). 13. Eichengreen (2004).
204
Brookings Papers on Economic Activity, 1:2005
and a tipping phenomenon.14 We have found each of these to be statistically significant during the period 1973–98. We found surprisingly little support for another hypothesized variable, the net international investment position of the home country. We use the admittedly brief period of European Economic and Monetary Union (EMU), 1999–2003, as a crude test of the out-of-sample performance of the estimated equation. The equation correctly predicts a small increase in the euro share at the expense of the dollar during this period. We then use the parameter estimates to project into the future. The projections are naturally very sensitive to what one assumes about the explanatory variables, particularly the size of Euroland. Our preliminary finding is that, if some Central and Eastern European countries join EMU by 2010, and if Sweden, Denmark, and (most important) the United Kingdom join by 2015, so that Euroland becomes larger economically than the United States, the tipping phenomenon could set in soon thereafter. The euro could definitively pass the dollar, perhaps in the subsequent decade. Even if some of the EU countries stay out of Euroland, as is likely, the tipping could result in response to a future trend depreciation of the dollar that continues at the same pace as in the past. It is still true that measures of international currency use such as reserve shares change slowly. So when one sees newspaper headlines warning of the euro overtaking the dollar, it is important to realize that this is unlikely to happen for a long time. But if it does happen, the consequences are likely to be large. It would mean the end of America’s “exorbitant privilege.” It would be not just a ten-year “system” coming to an end, but the end of a century of U.S. economic hegemony. General discussion: Paul Krugman saw the crucial question as not whether China will stop accumulating foreign exchange reserves, but whether it will diversify its reserve portfolio, switching to the euro or possibly the yen. He also noted that loss of reserve currency leadership could arise not only from diversification of existing reserves, but also from diversification at the margin. He speculated that this would have huge exchange rate implications. Kenneth Rogoff was skeptical of the authors’ claim that demand for U.S. assets by Asia’s official sector can explain today’s low 14. See, for example, Bergsten (1975), Dooley, Lizondo, and Mathieson (1989), and Eichengreen and Mathieson (2000). Chinn and Frankel (2005) provide a comprehensive review.
Michael Dooley and Peter Garber
205
dollar interest rates, and he drew a parallel with recent work by Glenn Hubbard and Eric Engen, who concluded that the recent huge U.S. budget deficits do not matter for interest rates. Rogoff also doubted that the huge pool of surplus Chinese labor entering the global market is driving interest rates down, noting that standard models would predict the opposite and that both the capital share of income in the advanced industrial economies and corporate profits are surging. Michael Dooley replied that the pressure on interest rates came not directly from labor supply but from China’s very high saving rate: those savings, effectively, are exported to the rest of the world. Rogoff emphasized the role of Japan, whose current account surplus in recent years is several times China’s and whose reserves are larger. He noted that arguments based on labor surplus and the collateral hypothesis do not apply to Japan. He added that other Asian economies also run larger surpluses than China and together accumulate more reserves than China, but they are much more open and their capital markets are more integrated than Europe’s were in the 1960s. Gian Maria Milesi-Ferretti was also skeptical about the authors’ hypothesis that emerging economies accumulate U.S. reserves as a form of collateral. He reminded the panel that many of the countries that have grown rapidly, including Korea, Malaysia, Thailand, and Indonesia, have typically been borrowers on international markets. The main exception has been Taiwan, which is a special case for various reasons. Singapore, which is now a large creditor, borrowed heavily in its early stages of development. Milesi-Ferretti also noted that a country’s external balance can depend importantly on the terms of trade. Asia’s emerging economies are mostly commodity importers and are today running large current account surpluses despite unfavorable terms of trade. If there is a temporary component in the current high prices of oil and other commodities that these countries import, the structural surpluses may be even larger than they appear. Richard Cooper questioned the presumption of many observers that the Chinese renminbi is undervalued. He noted that China has substantially effective controls on resident capital outflows, an investment rate of 40 percent of GDP but an even higher saving rate, and a modest current account surplus relative to the size of the economy. Given these facts and the fact that exchange rates in the rest of the world are generally floating, Cooper saw little evidence that the renminbi is undervalued. Furthermore, he expected a reform of the Chinese banking system and, eventually, full currency con-
206
Brookings Papers on Economic Activity, 1:2005
vertibility. In that event the renminbi might well depreciate, because capital outflows would increase as wealthy Chinese households begin to invest abroad to diversify their portfolios. He added that one already observes increasingly aggressive foreign direct investment by China, with Chinese oil firms just one example.
Michael Dooley and Peter Garber
207
References Aizenman, Joshua, Brian Pinto, and Artur Radziwill. 2004. “Sources for Financing Domestic Capital—Is Foreign Saving a Viable Option for Developing Countries?” Working Paper 10624. Cambridge, Mass.: National Bureau of Economic Research (July). Bergsten, C. Fred. 1975. The Dilemmas of the Dollar. New York University Press. Caballero, Ricardo J. 2004. “The Wrong Call: The Euro Is No Match for the Dollar.” Massachusetts Institute of Technology (December). Caballero, Ricardo J., and Arvind Krishnamurthy. 2001. “International and Domestic Collateral Constraints in a Model of Emerging Market Crises.” Journal of Monetary Economics 48: 513–48. _________. 2003. “Excessive Dollar Debt: Financial Development and Underinsurance.” Journal of Finance 58, no. 2: 867–93. Chinn, Menzie, and Jeffrey Frankel. 2005. “Will the Euro Eventually Surpass the Dollar as Leading International Reserve Currency?” Working Paper 11508. Cambridge, Mass.: National Bureau of Economic Research. Despres, Emile, Charles P. Kindleberger, and Walter S. Salant. 1966. “The Dollar and World Liquidity: A Minority View.” The Economist (February 5). Dickson, Paul. 1978. The Official Rules. New York: Delacorte Press. Dooley, Michael P., David Folkerts-Landau, and Peter Garber. 2003. “An Essay on the Revived Bretton Woods System.” Working Paper 9971. Cambridge, Mass.: National Bureau of Economic Research (September). _______. 2004a. “The Revived Bretton Woods System: The Effects of Periphery Intervention and Reserve Management on Interest Rates and Exchange Rates in Center Countries.” Working Paper 10332. Cambridge, Mass.: National Bureau of Economic Research (March). _______. 2004b. “Direct Investment, Rising Real Wages and the Absorption of Excess Labor in the Periphery.” Working Paper 10626. Cambridge, Mass.: National Bureau of Economic Research (July). _______. 2004c. “The US Current Account Deficit and Economic Development: Collateral for a Total Return Swap.” Working Paper 10727. Cambridge, Mass.: National Bureau of Economic Research (September). Dooley, Michael P., J. Saul Lizondo, and Donald J. Mathieson. 1989. “The Currency Composition of Foreign Exchange Reserves.” International Monetary Fund Staff Papers 36, no. 2: 385–434. Dornbusch, Rudiger. 1973. “Devaluation, Money and Nontraded Goods.” American Economic Review 63, no. 5: 871–80. Eichengreen, Barry. 2004. “Global Imbalances and the Lessons of Bretton Woods.” Working Paper 10497. Cambridge, Mass.: National Bureau of Economic Research (May).
208
Brookings Papers on Economic Activity, 1:2005
_______. Forthcoming. “Chinese Currency Controversies.” Asian Economic Papers. Eichengreen, Barry J., and Donald J. Mathieson. 2000. “The Currency Composition of Foreign Exchange Reserves: Retrospect and Prospect.” Working Paper wp/00/131. Washington: International Monetary Fund. Eichengreen, Barry, and Charles Wyplosz. 1993. “The Unstable EMS.” BPEA, no. 1: 51–124. Eichengreen, Barry, Paul Masson, and others. 1998. “Exit Strategies: Policy Options for Countries Seeking Greater Exchange Rate Flexibility.” Occasional Paper 168. Washington: International Monetary Fund (April). Frankel, Jeffrey. 1995. “Still the Lingua Franca: The Exaggerated Death of the Dollar.” Foreign Affairs 74, no. 4: 9–16. _______. 2005a. “The Twin Deficits.” Notes presented at a meeting of the Bellagio Group, Amsterdam, January 20–21. _______. 2005b. “On the Renminbi: The Choice between Adjustment under a Fixed Exchange Rate and Adjustment under a Flexible Rate.” Working Paper 11274. Cambridge, Mass.: National Bureau of Economic Research (April). Frankel, Jeffrey A., and Andrew K. Rose. 1996. “Currency Crashes in Emerging Markets: An Empirical Treatment.” Journal of International Economics 41, no. 3–4: 351–66. Goldstein, Morris. 2004. “Adjusting China’s Exchange Rate Policies.” Washington: Institute for International Economics (May). Goldstein, Morris, and Nicholas R. Lardy. 2004. “What Kind of Landing for the Chinese Economy?” Policy Briefs in International Economics PB04-7. Washington: Institute for International Economics (November). _______. 2005. “China’s Role in the Revived Bretton Woods System: A Case of Mistaken Identity.” Working Paper 05-2. Washington: Institute for International Economics. Grilli, Vittorio U. 1986. “Buying and Selling Attacks on Fixed Exchange Rate Systems.” Journal of International Economics 20: 143–56. Hsieh, Chang-Tai, and Christina D. Romer. 2001. “Was the Federal Reserve Fettered? Devaluation Expectations in the 1932 Monetary Expansion.” Working Paper 8113. Cambridge, Mass.: National Bureau of Economic Research (February). Kindleberger, Charles P. 1965. “Balance of Payments Deficits and the International Market for Liquidity.” Essays in International Economics 46. International Economics Section, Department of Economics, Princeton University. ________. 1995. “Dollar Darkness.” The International Economy (May/June). McKinnon, Ronald. 1996. The Rules of the Game: International Money and Exchange Rates. MIT Press. Mundell, Robert A. 1971. Monetary Theory. Pacific Palisades, Calif.: Goodyear.
Michael Dooley and Peter Garber
209
Nurkse, Ragnar. 1945. “Conditions of International Monetary Equilibrium.” Essays in International Economics 4. International Economics Section, Department of Economics, Princeton University. Obstfeld, Maurice. 1993. “The Adjustment Mechanism.” In A Retrospective on the Bretton Woods System, edited by Michael D. Bordo and Barry Eichengreen. University of Chicago Press. Prasad, Eswar, and Shang-Jin Wei. 2005. “The Chinese Approach to Capital Inflows: Patterns and Possible Explanations.” Working Paper 11306. Cambridge, Mass.: National Bureau of Economic Research (May). Razin, Assaf, and Gian Maria Milesi-Ferretti. 1998. “Current Account Reversals and Currency Crises: Empirical Regularities.” Working Paper 6620. Cambridge, Mass.: National Bureau of Economic Research (June). Roubini, Nouriel, and Brad Setser. 2004. “The US as a Net Debtor: The Sustainability of the US External Imbalances.” Stern School of Business, New York University, and University College, Oxford (November). Salant, Stephen W. 1976. “Exhaustible Resources and Industrial Structure: A Nash-Cournot Approach to the World Oil Market.” Journal of Political Economy 84, no. 5: 1079–93 . Solomon, Robert. 1977. The International Monetary System: An Insider’s View, 1945-1976. Harper and Row. Summers, Lawrence H. 2004a. “The United States and the Global Adjustment Process.” Speech at the Institute for International Economics, Washington, March. _________. 2004b. “The U.S. Current Account Deficit and the Global Economy.” Per Jacobsson Lecture, Washington, October 3. Triffin, Robert. 1960. Gold and the Dollar Crisis. Yale University Press.
SEBASTIAN EDWARDS University of California, Los Angeles
Is the U.S. Current Account Deficit Sustainable? If Not, How Costly Is Adjustment Likely to Be? MANY ANALYSTS IN academia, the private sector, and applied research institutions express increasing concern about the growing U.S. current account deficit. There is a general sense that current global imbalances are unsustainable and that adjustment must come sooner rather than later. The unprecedented magnitude of the U.S. current account deficit and the United States’ growing net foreign indebtedness have fueled these worries, with many analysts arguing that, unless something is done, the world will move toward a major financial crisis.1 Some have gone as far as to suggest an imminent collapse of the dollar and a global financial meltdown.2 Underlying this view is the fact that, if the deficit continues at its current level, U.S. net international liabilities will eventually reach 100 percent of GDP, a figure widely considered to be excessively large.3 The source of financing of the U.S. current account deficit has also become a matter of concern. A number of authors have argued that, by relying on foreign and particularly Asian central banks’ purchases of 1. Although most of these alarmist discussions have taken the form of newspaper opinion pieces, there have also been a few policy papers on the subject. See, for example, Roubini and Setser (2004). 2. See, for example, Roubini and Setser (2004). For an excellent set of papers on the subject, see Bergsten and Williamson (2004). 3. For example, in a very clear discussion of this issue, Michael Mussa said, “[T]here probably is a practical upper limit for US net external liabilities at something less than 100 percent of US GDP and, accordingly . . . current account deficits of 5 percent or more of US GDP are not indefinitely sustainable” (Mussa, 2004, p. 114).
211
212
Brookings Papers on Economic Activity, 1:2005
Treasury securities, the United States has become extremely vulnerable to sudden changes in expectations and economic sentiments.4 Robert Skidelsky recently argued in the New York Times that the value of the dollar is one of the most important sources of political tension between the United States and Europe. Arguing that “[U]nilateralism is not more acceptable in currency matters than in foreign policy,” Skidelsky points out that, The United States is the only major country proclaiming itself indifferent to its currency’s value. In countries running persistent current account deficits, governments normally—indeed must—reduce domestic consumption. But so far, the United States has relied on other countries to adjust their economies to profligate American spending. . . .5
There is, however, an alternative view. Some authors have argued that, in an era of increasing financial globalization and rapid U.S. productivity gains, it is possible—indeed, even logical and desirable—for the United States to run very large current account deficits for a very long period (say, a quarter of a century). In this view, growing international portfolio diversification implies that the rest of the world will be willing to accumulate large U.S. liabilities during the next few years, maybe even in excess of 100 percent of U.S. GDP. From this perspective, since the U.S. current account deficit poses no threat, there are no fundamental reasons to justify a significant fall in the value of the dollar. 6 This paper analyzes the relationship between the dollar and the U.S. current account, with particular attention to the issue of sustainability and the mechanics of current account adjustment. I develop a portfolio model of the current account and show that, even under a very positive scenario where foreigners’ (net) demand for U.S. assets doubles from its current level, the U.S. current account will have to go through a significant adjustment in the not-too-distant future. Indeed, one cannot rule out a scenario where the U.S. current account deficit shrinks abruptly by 3 to 6 percent of GDP. To get an idea of the possible consequences of such an adjustment, I also analyze the international historical evidence 4. See, for example, Martin Wolf’s article in the Financial Times, “Funding America’s Recovery is a Very Dangerous Game,” October 1, 2003, p. 15. 5. “Winning Back Europe’s Heart: Rogue Dollar,” New York Times, February 20, 2005, p. 9. 6. Dooley, Folkerts-Landau, and Garber (2004a, 2004b); see also Cooper (2004) and Caballero, Farhi, and Hammour (2004).
Sebastian Edwards
213
on current account reversals. The results of this empirical investigation indicate that significant current account reversals have tended to result in large declines in GDP growth. The U.S. Dollar and the Current Account: A Thirty-Year Perspective This section analyzes the behavior of the U.S. real exchange rate (RER) and current account since the adoption of floating exchange rates in the early 1970s.7 I begin by discussing the course of the U.S. RER and current account during that period and the changing nature of the U.S. trade-weighted RER index. I argue that the last thirty years of U.S. RER behavior can be divided into six distinct phases. Second, I discuss the most recent data on the U.S. current account, including its sources of financing. And third, I provide some international evidence on current account imbalances during the last three decades. This comparative analysis helps in placing the recent U.S. experience in historical context. Six Phases of Real Exchange Rate Behavior Figure 1 presents quarterly data for the U.S. current account balance as a percentage of GDP and for the Federal Reserve’s trade-weighted index of the U.S. RER, both for the period 1973–2004; in this figure, as in the rest of the paper, the RER index is defined such that an increase represents a real appreciation. The figure shows that current account deficits have become increasingly large since 1992. It also shows that, over the first decade of floating exchange rates (1973–82), the United States ran small current account surpluses or deficits, which averaged to a small surplus of 0.04 percent of GDP. In contrast, over 1983–2004 the current account balance was in deficit, on average, by 2.4 percent of GDP. Figure 1 also 7. Because of space considerations, I do not discuss in detail some important issues such as the stationarity of the RER and its (changing) volatility through time. Most recent analyses based on panel data have found that the RER is stationary and that its half-life cycle is less than the three to five years traditionally considered to be the consensus view. See Choi, Mark, and Sul (2004). An analysis of U.S. RER volatility indicates that for the complete period 1975–2004 the U.S. RER index exhibited one of the highest volatilities in the sample. Only the British pound, the Japanese yen, and the euro had greater volatility. Within this period, the RER volatility of the U.S. dollar was at its highest in 1985–89, which roughly corresponds to phase III (a period of rapid depreciation) in figure 2.
214
Brookings Papers on Economic Activity, 1:2005
Figure 1. Real Exchange Rate and Current Account Balance, 1973–2004 Index (March 1973 = 100)a
Percent of GDP Real exchange rate (right scale)
6
120
4
100
2
80
0 –2 Current account (left scale)
–4 –6
I 1975
II 1979
1983
III 1987
IV 1991
V 1995
1999
VI 2003
Source: Bureau of Economic Analysis, National Income and Product Accounts, International Transactions Accounts; Federal Reserve data. a. Price-adjusted Major Currencies index.
shows that during the whole period under consideration the RER index fluctuated within a wide range: from 91.2 at its lowest point, it rose to 136.3 at its highest, with a mean for the period of 105.3. Finally, the figure demonstrates an apparent negative correlation between the tradeweighted real value of the dollar and the current account balance: periods of a strong dollar have tended to coincide with periods of larger current account deficits. Although the relationship is not exact, the synchronicity between the two variables is quite high: the contemporaneous coefficient of correlation between the logarithm of the RER index and the current account balance is −0.53; the highest value for the correlation coefficient (−0.60) is obtained when the log of the RER is lagged three quarters. Recent policy debates about the value of the dollar illustrate the massive changes that have occurred in U.S. trade relations during the last three decades. Discussions of the dollar in the early 1970s dealt almost exclusively with bilateral exchange rates—both nominal and real—between the dollar and the currencies of other industrial countries; more recently the debate has increasingly focused on the dollar’s value in terms of the currencies of emerging economies, including the Chinese renminbi, the Korean won, and the Malaysian ringgit. During the last few years the Mexican peso has also become an important determinant of the trade-
Sebastian Edwards
215
weighted value of the dollar; this was not the case when the Smithsonian Agreement was abandoned and the regime of floating exchange rates began in 1973. Between 1995 and 2005 the renminbi’s weight in the Federal Reserve trade-weighted RER index of the dollar rose from 5.7 percent to 11.4 percent, and the Mexican peso’s from 7.0 percent to 10.0 percent. Meanwhile the yen’s weight declined from 16.5 percent to 10.6 percent. In fact, the trade-weighted RER of the dollar is dominated today by the Asian nations, which (excluding India) have a combined weight in the index of 38.8 percent. The currencies of commodity-exporting countries, as a group, are also very important, with a weight of 24.6 percent. Finally, the launching of the euro in 1999 has marginalized the British pound: although its weight of 5.2 percent is still quite respectable, the pound is no longer among the top five currencies in the index. The situation was quite different in 1998, when the pound had a higher weight than the currencies of all but one of the countries that would eventually adopt the euro: in that year the deutsche mark had a weight of 6.4 percent, and the pound a weight of 5.9 percent.8 One can distinguish in figure 1 six distinct phases in U.S. RER behavior for the thirty-two-year period 1973–2004. A summary of these six phases is also in large measure a summary of the history of the international financial system since the inception of floating:9 —Phase I, 1973:1–1978:4. The early years of floating were characterized by a depreciating trend for the dollar in real terms, with the decline in value cumulating to 18.1 percent over the twenty-four quarters. During this period the standard deviation of monthly log differences of the RER index was 0.0205. During the early part of this phase (1973–76), the current account was in surplus, but this turned into a small deficit in 1977 and 1978. —Phase II, 1979:1–1985:1. During the next twenty-five quarters the dollar experienced a 49.3 percent appreciation in real terms. Meanwhile the current account went into deficit, which reached 2.9 percent of U.S. GDP in 1984:4. The standard deviation of monthly log differences of the RER index was 0.022, slightly higher than in phase I. In view of the dollar’s strengthening and the related increase in the U.S. current account deficit, on 8. In 2005 the euro has a weight of 18.8 percent in the Federal Reserve index; in 1995 the currencies that the euro later replaced had a combined weight of 17.3 percent. 9. Figure 1 presents the Federal Reserve’s broad RER index. The same six phases are also apparent when alternative indexes are used.
216
Brookings Papers on Economic Activity, 1:2005
September 22, 1985, the members of the Group of Five major industrial countries (the G-5: the United States, Japan, the United Kingdom, France, and Germany) agreed to implement concerted and coordinated interventions in the foreign exchange market. As part of this agreement, which came to be known as the Plaza Agreement, the G-5 countries also committed themselves to put in place coordinated macroeconomic policies that would reduce the costs of the global adjustment process. —Phase III, 1985:2–1988:4. During the period following the Plaza Agreement, the dollar experienced a rapidly depreciating trend, with a peakto-trough change in the index of −28.7 percent. RER volatility increased substantially during these fifteen quarters: the standard deviation of monthly log differences of the RER index was 0.0268. The current account deficit continued to grow, however, until in mid-1987 it stabilized at around 3.6 percent of U.S. GDP. From that point onward the current account began to improve, and by 1988:4 the deficit had declined to 2.4 percent of GDP. On February 22, 1987, the ministers of finance and central bank governors of the G-6 (the former G-5 plus Canada) released a communiqué, which came to be known as the Louvre Accord, declaring that significant progress had been made in achieving global adjustment, and that “Further substantial exchange rate shifts among their currencies could damage growth and adjustment prospects in [the G-6] countries.” The communiqué went on to say that the G-6 “agreed to cooperate closely to foster stability of exchange rates around current levels.”10 —Phase IV, 1989:1–1995:2. During the next phase the dollar continued to depreciate in real terms, but at a much lower rate than in the preceding phase: during these twenty-six quarters the total real depreciation was 10 percent. The standard deviation of monthly log differences of the RER index over this period was 0.0232, and the current account balance continued to improve, until in 1991:1 the United States posted its first current account surplus in many years. The current account balance averaged −1.15 percent of U.S. GDP during this phase. —Phase V, 1995:3–2002:1. This phase was characterized by a troughto-peak real dollar appreciation of 33.4 percent (although, as figure 1 shows, between 1998:1 and 1999:4 there was a short-lived period of real depreciation). Interestingly, during this phase RER volatility declined significantly: the standard deviation of monthly log differences of the RER 10. Quoted from the text of the Louvre communiqué, available at www.g8.utoronto. ca/finance/fm870222.htm.
Sebastian Edwards
217
index was 0.0196. This phase was also characterized by an increasing current account deficit: whereas in late 1995 and early 1996 the deficit was on the order of 1.5 percent of U.S. GDP, by early 2002 it was hovering just below 4 percent of GDP. In 1999, for the first time in many years, the U.S. federal government posted a budget surplus. —Phase VI, 2002:2–2004:4. In the most recent (and continuing) phase, the real value of the dollar has accumulated a 14 percent real depreciation. The current account deficit has continued to widen, exceeding 5 percent of U.S. GDP toward the end of the sample. RER volatility has increased slightly: the standard deviation of log differences of the RER index was 0.0212. Other important macroeconomic developments during this phase included a worsening of the U.S. fiscal position and stiff increases in the prices of oil and other commodities. Figure 2 breaks down the U.S. current account balance for 1973 through 2004 into its main components: the balance of trade in goods and services, the balance of trade in nonfinancial services, the income account, and transfers, all as percentages of GDP on a yearly basis. As the top left panel shows, large and persistent trade surpluses preceded the era of large current account deficits: already in the late 1970s the trade account was negative, and since mid-1976 it has had only one surplus quarter (1992:2).11 The top right panel shows that since 1996 the trade surplus in nonfinancial services has declined steadily, so that by 2004 it was only 0.3 percent of GDP. The income account remains positive, as the bottom left panel of the figure shows. The surplus has declined sharply since 1980, but given that for many years now the U.S. international investment position has been negative—that is, the United States has been a net debtor—the fact that the income account is still positive may seem surprising. The reason is that the return on U.S. assets held by foreigners has been systematically lower than the return on foreign assets in the hands of U.S. nationals. Finally, the bottom right panel shows that the transfers account has been negative in every year except one since 1946. Recently the transfers deficit has been stable at approximately 0.7 percent of GDP. Recent Current Account Imbalances Table 1 presents data on the U.S. current account deficit and its financing for the period 1990–2004. The nature of external financing of the 11. Mann (2004) shows that most of the U.S. trade deficit is explained by deficits in automobiles and consumer goods.
1979
1985
1991
1997
Source: Bureau of Economic Analysis, International Transactions Accounts.
1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0
Income
2003
–1.0
–0.5
0.0
0.5
1.0
0.0
0.2
0.4
0.6
0.8
1979
1.2
1 0 –1 –2 –3 –4 –5 –6 1.0
Percent of GDP
Percent of GDP
Goods and services
Figure 2. Components of the Current Account Balance, 1973–2004
1985
1991
Transfers
Services
1997
2003
−75.1
−44.9
8.6
Claims reported by banks
−79.0
Current account balance
97.9
81.8
55.7 127.4
87.3
138.7
221.3
7.9
−5.2
0.8
44.6 n.a. n.a.
24.8
130.4
18.0
1997
76.2
4.2
−15.1
36.4
32.1 n.a. n.a.
16.6
28.6
−26.7
1998
233.8
−22.0
−21.5
64.5
182.6 n.a. n.a.
22.4
-44.5
52.3
1999
478.0
−31.7
31.9
162.1
338.0 267.7 93.0
5.3
-70.0
42.5
2000
416.6
−7.5
57.6
24.7
309.2 300.3 12.6
23.8
-14.4
23.1
2001
178.6 241.8 −63.2
16.6
113.4
250.1
2003
323.2 360.1 −36.8
14.8
108.1
358.1
2004
569.9
66.1
32.6
542.7
65.2
55.1
614.0
−15.6
−41.5
−62.4 −133.9 −133.0
301.4 269.8 37.5
21.5
100.4
110.3
2002
3.7 −48.0 −82.0 −118.0 −109.5 −120.2 −136.0 −209.6 −296.8 −413.4 −385.7 −473.9 −530.7 −665.9
43.5
37.4
Source: Bureau of Economic Analysis, U.S. International Transactions and International Investment Position. a. Each financing item is reported net (inflows minus outflows). b. Debt and equity do not add up to total securities because of rounding.
58.0
Net financing
3.4
100.1
−32.6
11.3
14.4
13.2
−35.0
8.0
17.3
Claims reported by nonbanks
−5.4
−46.0 n.a. n.a.
17.4
147.0
−41.0
−45.1 n.a. n.a.
12.3
91.5
133.4
1996
−34.0
34.3
100.1
1995
11.3 −14.7 −28.4 −32.6
13.4
24.4
44.9
1994
Foreign direct investment
Securitiesb Debt Equity
15.4
37.1
70.4
1993
23.4
18.8
Currency
18.8
44.4
1992
−6.2 n.a. n.a.
-2.5
Foreign private purchases of U.S. Treasuries
23.2
1991
18.9
31.8
Change in reserves
−27.2 −10.5 −19.1 −66.2 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.
1990
Item
Table 1. Net Financial Flows and Current Account Balance, 1990–2004a Billions of dollars
220
Brookings Papers on Economic Activity, 1:2005
deficit has changed significantly in the last few years. Whereas from 1997 to 2000 inflows of foreign direct investment (FDI) contributed in an increasingly important way to financing the deficit, net FDI inflows fell sharply in 2001 and have been negative since then. Also, after four years (1999–2002) in which net equity flows were positive, these became negative in 2003–04. Table 1 also shows that during 2003 and 2004 the U.S. current account deficit was fully financed through net fixed-income flows (the sum of the first, second, and fifth rows in the table). Official foreign purchases of government securities—“Change in reserves” in table 1— played a particularly important role in financing the 2003 and 2004 current account deficits. A number of analysts have argued that reliance on foreign central bank purchases of Treasury securities has made the United States particularly vulnerable to sudden changes in expectations and economic sentiments.12 Current account imbalances are reflected in changes in a country’s net international investment position (NIIP): deficits result in a deterioration of the NIIP, and surpluses in an improvement. Figure 3 shows that the U.S. NIIP as a percentage of GDP has become increasingly negative since the mid-1980s: in 2004 U.S. net international liabilities reached 29 percent of GDP. An important feature of the U.S. NIIP that distinguishes it from those of most other countries is that gross international assets and gross international liabilities are held in different currencies. Whereas more than 70 percent of gross foreign assets held by U.S. nationals are denominated in foreign currency, approximately 95 percent of gross U.S. liabilities in the hands of foreigners are denominated in dollars. This means that U.S. net liabilities as a percentage of GDP are subject to “valuation effects” stemming from changes in the value of the dollar: a dollar depreciation reduces the value of U.S. net liabilities; as a result, the deterioration of the U.S. NIIP during 2002–04 was significantly smaller than the accumulated current account deficit during those two years (table 2). A key question in current account sustainability analyses—discussed in detail below—is, What is the “reasonable” long-run equilibrium ratio of U.S. net international liabilities to GDP? The higher this ratio, the greater will be the sustainable current account deficit. According to some, the current ratio of almost 30 percent of GDP is already excessive; 12. See, for example, the article by Martin Wolf cited above.
Sebastian Edwards
221
Figure 3. Net International Investment Position, 1976–2004a Percent of GDP 10 5 0 –5 –10 –15 –20 –25 –30 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 Source: Bureau of Economic Analysis, National Income and Product Accounts, International Investment Position. a. Data for 2004 are author's projection. Direct investment positions are valued at current cost.
others believe that a NIIP-to-GDP ratio of up to 50 percent would be reasonable.13 One of the first things that undergraduate students of open-economy macroeconomics learn is that a country’s current account is equal to the difference between its national saving and its domestic investment. Over the years a number of authors have argued that a worsening of a current account balance that stems from an increase in investment is very different from one that results from a decline in national saving. Some have gone as far as to argue that very large deficits in the current account “don’t matter,” as long as they are the result of increased (private sector) investment.14 As figure 4 shows, the recent deterioration of the U.S. current account has largely corresponded to a decline in national saving, and in particular of public and household saving, rather than a rise in investment. A simple implication of this trend—and one that is emphasized by most authors—is that an improvement in the U.S. current account will require not only an RER adjustment, but also an increase in the national saving rate to GDP. Symmetrically, a correction 13. See Obstfeld and Rogoff (2004) and Mussa (2004). 14. Corden (1994).
222
Brookings Papers on Economic Activity, 1:2005
Table 2. Net International Investment Position and Current Account Balance, 1998–2004 Billions of dollars Item
1998
1999
2000
2001
2002
2003
2004
Net international liabilities position Change from previous year Current account deficit Valuation changesa
900.0
775.5
1,388.7
1,889.7
2,233.0
2,430.7
n.a.
79.3
−124.5
613.3
500.9
343.3
197.7
n.a.
209.6
296.8
413.4
385.7
473.9
530.7
665.9
−130.2
−421.3
199.8
115.2
−130.6
−333.0
n.a.
Source: Bureau of Economic Analysis, U.S. International Transactions and International Investment Position. a. Current account deficit plus valuation changes equals change in net international investment position from previous year.
of current global imbalances will also require a decline in Europe’s and Japan’s saving rates, or an increase in their investment rates, or some combination of the two. The U.S. Current Account Deficit in International Perspective How large are the recent U.S. current account deficits from a comparative point of view? And how large is the U.S. net international liabilities position compared with those of other developed countries through history? The top panel of table 3 presents data on the distribution of current account balances (as a percentage of GDP) in the world economy, as well as in six country groups—the industrial countries, Latin America, Asia, the Middle East, Africa, and Eastern Europe—for 1971 through 2001 and 1984 through 2001. (The data are unweighted averages for each country.) At almost 6 percent of GDP, today’s U.S. current account deficit is very large indeed from a historical and international perspective: it is in the top decile of all the deficits recorded by all industrial countries in the first thirty years of floating exchange rates. Since 1971 the United States has been the only large industrial country that has run current account deficits in excess of 5 percent of GDP. This reflects the unique position of the United States in the international financial system: U.S. assets have been in high demand, allowing the United States to run large and persistent deficits. On the other hand, this fact also suggests that the United States is moving into uncharted waters. As Mau-
Sebastian Edwards
223
Figure 4. Investment and Saving, 1970–2003a Percent of GDP Household saving Corporate saving Public saving Foreign saving Investment
15
10
5
0
–5
1972
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
Source: Bureau of Economic Analysis, National Income and Product Accounts, International Investment Position. a. All series are on a net basis.
rice Obstfeld and Kenneth Rogoff,15 among others, have shown, if the U.S. deficit continues at its current level, in twenty-five years U.S. net international liabilities will exceed those recorded by any other country, as a percentage of GDP, in modern times. During the last thirty years the only industrial countries to have had current account deficits in excess of 5 percent of GDP have been small ones: Australia, Austria, Denmark, Finland, Greece, Iceland, Ireland, Malta, New Zealand, Norway, and Portugal. What is even more striking is that very few countries, industrial or developing, have had large current account deficits that lasted for more than five years. Table 4 lists those countries that have had “persistently large” current account deficits at some time during the period 1970–2001. For purposes of this table, I define a country as having a “large deficit” if, in any year, its current
15. Obstfeld and Rogoff (2004).
224
Brookings Papers on Economic Activity, 1:2005
Table 3. Distribution of Current Account Deficits by World Region, 1970–2001 Percent of GDP Region 1970–2001 Industrial countries Latin America and Caribbean Asia Africa Middle East Eastern Europe All countries 1984–2001 Industrial countries Latin America and Caribbean Asia Africa Middle East Eastern Europe All countries
No. of countries
Mean
Median
1st decile
1st quartile
3rd quartile
9th decile
22 34
0.6 5.4
0.7 4.1
−3.8 −2.5
−1.6 1.1
3.0 8.0
4.8 16.9
22 51 12 25 169
3.0 6.3 0.0 3.9 3.9
2.7 5.3 1.4 3.0 3.3
−7.1 −3.4 −18.8 −2.4 −5.0
−0.6 1.2 −5.0 0.3 −0.1
6.3 9.9 6.4 6.1 7.1
11.3 16.9 13.6 10.7 13.1
25 34
0.2 5.1
0.3 3.7
−4.7 −2.5
−2.3 1.1
2.7 7.0
4.8 17.0
22 51 12 25 169
2.2 5.9 2.3 4.0 3.8
2.4 4.6 1.5 3.1 3.0
−8.0 −3.5 −12.4 −2.5 −4.8
−1.3 0.9 −4.0 0.3 −0.4
5.9 9.1 6.3 6.6 6.7
10.2 16.2 14.9 10.9 12.9
Source: Author’s calculations using data from World Bank, World Development Indicators.
account deficit exceeded its region’s ninth decile.16 I then define a “persistently large deficit” country as one that has at some time had a large deficit, as defined above, for at least five consecutive years.17 The resulting list in table 4 is extremely short, and none of the countries listed is large. This illustrates the fact that, historically, periods of large current account imbalances have tended to be short lived and have been followed by periods of current account adjustment. Table 5 presents data on net international liabilities, as a percentage of GDP, for a group of industrial countries that have historically had a large negative NIIP position.18 The picture that emerges is different from that in 16. Notice that the thresholds for defining “large” deficits are year- and region-specific, with a different threshold for each region each year. 17. For an econometric analysis of the persistence of current account deficits see Edwards (2004). See also Taylor (2002). 18. Data for the United States are from the Bureau of Economic Analysis. For the other countries the data are from the Lane and Milesi-Ferretti data set up to 1997. I have updated them using national current account balance data. The updated figures should be interpreted with a grain of salt, because I have not corrected them for valuation effects.
Sebastian Edwards
225
Table 4. Countries with Persistently Large Current Account Deficits by World Region, 1970–2001a Region and country Industrial countries Ireland New Zealand Latin America and Caribbean Guyana Nicaragua Asia Bhutan Africa Guinea-Bissau Lesotho Eastern Europe Azerbaijan
Period 1978–84 1984–88 1979–85 1984–90, 1992–2000 1982–89 1982–93 1995–2000 1995–99
Source: World Bank, World Development Indicators, various years. a. A persistently large deficit is defined as one that exceeded the ninth decile for the country’s region for at least five consecutive years.
table 4: a number of industrial nations have had—and continue to have— a significantly larger net international liabilities position relative to GDP than does the United States. This suggests that, at least in principle, the U.S. NIIP could continue to deteriorate for some time into the future. But, even if this should happen, at some point the process must come to an end, and the U.S. position as a percentage of GDP will have to stabilize. It makes a big difference, however, at what level it does stabilize. For example, if, in the steady state, foreigners are willing to hold at most the equivalent of 35 percent of U.S. GDP in the form of net U.S. assets, the United States could sustain a current account deficit of only 2.1 percent of U.S. GDP.19 If, on the other hand, foreigners’ net demand for U.S. assets were to grow to 60 percent of GDP—which, as table 5 shows, is approximately the ratio of (net) foreign holdings of Australian assets to that country’s GDP—the sustainable U.S. current account deficit would be 3.6 percent of GDP. And if foreigners are willing to hold (net) U.S. assets equivalent to 100 percent of U.S. GDP—the figure that Mussa considered implausible in the statement footnoted above—the sustainable U.S. current 19. This calculation assumes a 6 percent rate of growth of nominal GDP going forward. See below for an analytical discussion and the relevant equations.
226
Brookings Papers on Economic Activity, 1:2005
Table 5. Net Stock of Liabilities in Selected Industrial Countries, Selected Years Percent of GDP Country
1980
1985
1990
1995
2000
2003
Australia Canada Denmark Finland Iceland New Zealand Sweden United States
n.a. 34.7 n.a. 14.6 n.a. n.a. n.a. −12.9
n.a. 36.3 n.a. 19.0 n.a. n.a. 20.9 −1.3
47.4 38.0 n.a. 29.2 48.2 88.7 26.6 4.2
55.1 42.4 26.5 42.3 49.8 76.6 41.9 6.2
65.2 30.6 21.5 58.2 55.5 120.8 36.7 14.1
59.1 20.6 13.0 35.9 66.0 131.0 26.5 22.1
Source: Bureau of Economic Analysis, U.S. International Transactions and International Investment Position, various years; Lane and Milesi-Ferretti (2001).
account deficit could be as large as 6 percent of GDP, or approximately its current level.20 Since there are no historical precedents for a large, advanced nation running persistently large current account deficits, it is extremely difficult to get a clear idea of how foreigners’ demand for U.S. assets will behave in the future. Given this lack of historical precedent, a reasonable strategy is to model the RER dynamics to be expected if, as posited by Michael Dooley and his coauthors,21 among others, foreigners’ demand for U.S. assets continued to increase. This is the approach followed in the next section.
The Analytics of Current Account and Real Exchange Rate Adjustment The current account and the RER are endogenous variables jointly determined in a general equilibrium context. The key question is how these two variables will move in response to a given exogenous shock—a decline in capital inflows, say—if the other main variables, including economic growth and the rate of unemployment, do not deviate significantly from their long-term equilibrium paths. A number of authors have recently addressed this issue using a variety of simulation and econometric models.
20. Mussa (2004). 21. Dooley, Folkerts-Landau, and Garber (2004a).
Sebastian Edwards
227
Most have asked what amount of RER adjustment would be “required” to achieve a certain current account balance. Among these, some have considered the case where the deficit is completely eliminated,22 whereas others have investigated the reduction of the deficit to a smaller but still positive level. Appendix table A-1, which summarizes a selection of these studies, shows that they have used different methodologies and reached different conclusions.23 All, however, find quite large required adjustments in the trade-weighted value of the dollar.24 The estimates from the studies summarized in table A-1 are much larger than those discussed in most investment bank newsletters and in the media.25 A Portfolio Model of the Current Account and the Real Exchange Rate From an analytical perspective the process of current account adjustment may be broken down into two components: the dynamics of changes in net foreign assets, and the “transfer” associated with changes in a country’s net foreign assets position. Changes in international investors’ willingness to hold U.S. assets will affect total absorption of saving and relative prices, including the RER. An increase in foreigners’ rate of accumulation of U.S. assets will allow the United States to increase its absorption, generating a real appreciation and a current account deficit. In a similar way, a reduction in the rate at which foreigners accumulate the country’s assets—or, worse, a reduction in their holdings of domestic assets—will result in a slowdown or a drop in absorption and a decline in the relative price of nontradables, that is, a real depreciation. These changes in absorption and the concomitant adjustment in relative prices are reminiscent of discussions of the “transfer problem,” which go back at least to the debates between John Maynard Keynes and Bertil Ohlin during the 1920s. In large countries such as the United States, however, the 22. For example, Obstfeld and Rogoff (2000, 2004); Blanchard, Giavazzi, and Sa (this volume). 23. See also the studies by Mann (2003, 2004), which extend her pioneering 1999 model. 24. See, for example, Blanchard, Giavazzi, and Sa (this volume). 25. Although many financial market practitioners do believe that the dollar will weaken, they tend to expect more moderate adjustments. See, for example, the forex publications of some major investment banks.
228
Brookings Papers on Economic Activity, 1:2005
story is more complex. First, as mentioned above, changes in relative prices have valuation effects on net foreign asset holdings, which will feed back into the dynamics of net foreign asset accumulation or decumulation.26 Second, in a large country, changes in aggregate expenditure are likely to affect the international terms of trade, and thus the general equilibrium outcome of the original shock. the basic model. Consider the following barebones portfolio model of the current account.27 Equation 1, which is the basic equation for the external sector (expressed in domestic currency), states that the current account deficit (CAD) is equal to the trade deficit (TD) plus the income account (net income payments to the rest of the world, IA) plus net transfers to the rest of the world (NT):28 (1)
CAD t = TD t + IA t + NTt .
The income account, in turn, is equal to IA t = iD tf − i *F dt , where i is the interest rate paid on (gross) domestic assets in the hands of foreigners Dtf, and i* is the interest rate on (gross) foreign assets held by domestic residents Ftd. Since equation 1 is expressed in domestic currency, F td = Et F td ∗ , where E is the nominal exchange rate, defined as units of domestic currency (in this case dollars) per unit of foreign currency, and F td * denotes (gross) foreign assets held by domestic residents, expressed in foreign currency. Equation 1 can then be rewritten as 26. This effect has been emphasized by Lane and Milesi-Ferretti (2001, 2004a, 2004b), Tille (2003), and Gourinchas and Rey (2005), among others. For a discussion of valuation effects in the context of current account sustainability in emerging economies, see Edwards (2003). 27. To concentrate on the problem at hand and to keep the analysis tractable, I have made a number of simplifications; I have made no attempt to construct a full general equilibrium model. Recent papers that have constructed portfolio models of the current account include Blanchard, Giavazzi, and Sa (this volume), Edwards (1999, 2002), Gourinchas and Rey (2005), and Kraay and Ventura (2002). 28. I have defined current account balances such that a deficit is a positive number. In equation 1, then, negative numbers refer to a surplus.
Sebastian Edwards
(1′)
229
CAD t = TD t + iδ t + ( i − i*) Ftd + NTt ,
where δ is net domestic assets in the hands of foreigners, δt = D tf − F td. The terms iδt and (i − i*)F td capture valuation effects on the current account, recently emphasized by a number of authors.29 Equation 2 is a portfolio balance equation that summarizes net international demand for the domestic country’s assets δt. Domestic and foreign assets are assumed to be imperfect substitutes. The variable α is the share of foreigners’ wealth that they are willing to hold in the form of the domestic country’s assets; W is world wealth, and Wc is the domestic country’s wealth. The variable αjj is the domestic country’s asset allocation on its own assets, that is, the share of their wealth that domestic residents choose to hold in domestic assets. I assume that there is home bias in portfolio decisions; this is reflected in the fact that α and (1 − αjj) are below international market shares of domestic and foreign wealth. There is no need, however, to assume that foreign and domestic investors have the same degree of home bias. Hence the portfolio balance equation can be written as (2)
δ t = α (Wt − W ct ) − (1 − α jj ) W ct .
An important question is how the asset allocation shares α and αjj are determined. Under standard portfolio theory, α and αjj will depend on expected real returns (i, i*), perceived risk (µ, µ*), and the degree of segmentation of international financial markets. Here, however, I make the simplifying assumption that α and αjj are exogenously determined. This assumption allows me to focus on the effects of exogenous changes in portfolio allocations—that is, exogenous changes in α and αjj—on net asset dynamics and the current account. More specifically, I consider the case where changes in α and αjj reduce the initial extent of home bias. Later I discuss how the results are altered if some degree of substitutability between domestic and foreign assets is allowed. World wealth in foreign currency W* and in domestic currency W are related by W *t = Wt /Et. Domestic and foreign interest rates are related by i = i* + (dEe/E) + (µ − µ*) + k, where (dEe/E) is the expected rate of depreciation of the domestic currency, and k is a term that captures the effect of capital controls; in a world of perfect capital mobility, k = 0. 29. Including Lane and Milesi-Ferretti (2004a, 2004b) and Gourinchas and Rey (2005), among others.
230
Brookings Papers on Economic Activity, 1:2005
Note that in this model “investors” refers both to private and public investors, including foreign central banks. Indeed, as already pointed out, recent discussions have emphasized the key role played by foreign (and especially Asian) central banks in helping finance the U.S. current account deficit. The counterpart of a current account deficit is the change in the country’s (net) assets in the hands of foreigners: (3)
CADt = ∆δ t .
Equation 4 defines the trade deficit: (4)
TDt =
∑p
m i
mi − ∑ pix xi ,
where pim and pix are prices of importables and exportables in domestic currency; mi is demand for importables, which is assumed to depend on the real exchange rate e, the international price of importable goods, the country’s real income y, and other factors, including the degree of trade protection v. Demand for exports xi, on the other hand, depends on the RER, the international price of exportables, rest-of-world real income y*, and other factors u: (5)
mi = mi ( e, y, v ); xi = xi ( e, y*, u ) .
Variables mi and xi, in turn, may be interpreted as excess demand for importables and supply of exportables, respectively, in the domestic country. The basic version of the model assumes that the law of one price holds for both importables and exportables: pim = Epim*; pix = Epix*. However, in the simulation exercises below, alternative assumptions can be made, including that exporters and importers price to market. Equation 6 is the equilibrium condition for the nontradable goods market in the domestic country, where S tN is the supply of nontradables in period t, assumed to depend on the RER and other factors z, and D tN is demand for nontradables: (6)
S tN ( et , zt ) = DtN ( et , yt ).
The domestic price level P is assumed to be a geometric average of the nominal prices of tradable goods (importables and exportables) and nontradable goods:
Sebastian Edwards
231 (1− a − b )
Pt = ( ptm ) ( p tx ) ( p Nt ) a
b
.
Equation 7 defines the real exchange rate: (7)
et = Pt Et P*, t
where P*t is the foreign price level. As before, an increase in e represents a real appreciation. The interpretation of this model is simple. The domestic country can run a current account deficit only to the extent that foreign investors are willing to increase their net holdings of domestic assets—that is, to the extent that ∆δt > 0. Once ∆δt is known, and for given values of other key variables, it is possible to derive the real exchange rate e consistent with the prevailing current account deficit or surplus. A particularly interesting exercise, given the current U.S. situation, is to analyze how exogenous changes in portfolio preferences—that is, changes in α, αjj, or both—will affect the current account and the RER. Closing the model would require specifying a number of market clearing conditions, including the saving and investment equations for the world economy; and the world market clearing conditions for each importable and exportable good. These equilibrium conditions determine endogenously both interest rates and all relevant tradable goods prices. Doing this, however, would make the model significantly more complex than is required for dealing with the problem at hand. For this reason I work with a partial equilibrium version of the model, under alternative assumptions regarding these variables’ behavior.30 It is important to emphasize that current account adjustment not only implies changes in the RER, but also requires changes in saving and investment in the domestic country (here the United States) and the rest of the world. From a policy perspective the adjustment in domestic saving would be greatly facilitated by an increase in public sector saving. PORTFOLIO EQUILIBRIUM, DYNAMICS, AND CURRENT ACCOUNT SUSTAINABILITY. External sustainability requires that a country’s net external liabilities stabilize at a level compatible with foreigners’ net demand for these claims, as specified by equation 2. On the assumption that the domestic country’s wealth is a multiple λ of its (potential or full30. Most recent models on global imbalances and the U.S. current account have used a partial equilibrium framework in the simulation phase.
232
Brookings Papers on Economic Activity, 1:2005
employment) GDP, and that its wealth is a fraction β of world wealth W, it is possible to rewrite (international) net demand for the country’s assets as δ = [αθ − (1 − αjj)]λY, where Y is potential GDP, and θ = (1 − β)/β = EW f*/Wc, where W f* is rest-of-world wealth expressed in foreign currency. Denoting γ* = [αθ − (1 − αjj)]λ, then δ = γ*Y. This means that in long-run equilibrium net international demand for the domestic country’s assets can be expressed as a proportion γ* of its potential or sustainable GDP. The determinants of this factor of proportionality γ* depend on relative returns and the perceived risk of the domestic country and of the rest of the world, as well as on the degree of integration of international financial markets. If g is the country’s sustainable rate of growth and π the country’s long-term rate of inflation, the “sustainable” ratio of the current account deficit to GDP is given by (8)
CAD/Y = ( g + π ) αθ − (1 − α jj ) λ = γ * ( g + π ) .
Notice that, if [αθ − (1 − αjj)] < 0, domestic residents’ demand for foreign assets exceeds foreigners’ demand for domestic country assets. Under these circumstances the country will have to run a current account surplus in order to maintain a stable net external assets-to-GDP ratio. Most studies of the sustainability of the U.S. current account have used equations of this type. Mussa,31 for example, argues that in long-term equilibrium γ * is likely to be around 0.50.32 In long-run equilibrium the sustainable trade balance will be given by TD/Y = (g − r)γ*, where r is the real interest rate. In this model, as in earlier models developed by myself and by Aart Kraay and Jaume Ventura,33 additional saving will be allocated in a way that maintains domestic and foreign assets in the same proportion as in the original portfolio. Kraay and Ventura have shown that models that combine this assumption with the assumption of transactions costs in investment go a long way toward explaining international current account behavior in a large number of countries. If the degree of perceived riskiness of the domestic country (exogenously) declines, α, and thus γ *, will increase. As a result, the sustainable 31. Mussa (2004). 32. See also Edwards (1995), Ades and Kaune (1997), and O’Neill and Hatzius (2004) for current account sustainability analyses of this type. 33. Edwards (1999); Kraay and Ventura (2002).
Sebastian Edwards
233
current account balance will deteriorate (see equation 9). Equally important, changes in portfolio allocation, generated by changes in α or αjj, will generate a dynamic adjustment process, during which the current account will differ from its long-run sustainable level. This transitional dynamic can be incorporated into the model through the following equation: (9)
( CAD/Y )
t
= ( g + π ) γ *t + ψ ( γ *t − γ t − 1 ) − κ ( CAD/Y )t − 1 − ( g + π ) γ *t .
According to equation 9, short-term deviations of the current account from its long-run level can result from two forces. The first is a traditional stock adjustment term (γ *t − γt−1) that captures deviations between the demanded and the actual stock of the country’s assets in the hands of foreign investors. The coefficient ψ is the speed of adjustment, which will depend on a number of factors, including the degree of capital mobility in the country in question. The second force affecting this dynamic process, captured by −κ[(CAD/Y)t−1 − (g + π)γ *t ] in equation 9, is a self-correcting term, included to ensure that some form of consumption smoothing is present. The importance of this self-correcting term will depend on the value of κ.34 Whether the dynamic representation in equation 9 is appropriate is, in the final analysis, an empirical matter. As I show below, under certain parameterizations this model does a very good job at tracking the behavior of the U.S. current account during the last few years. The dynamic behavior for the net stock of the domestic country’s assets in the hands of foreigners, as a percentage of GDP, will be given by γt = [γt−1 + (CAD/Y)t−1](1 + g + π)−1. Consider the case where for some exogenous reason the home bias in the rest of the world is reduced—that is, α in equation 2 increases. This will result in an increase in the sustainable current account deficit (equation 8). It will also unleash a dynamic adjustment process, captured by equation 9. During this transitional period the current account deficit will exceed its new long-run (higher) sustainable equilibrium; that is, the current account deficit will overshoot its new sustainable level. During the transition the trade account will move according to the following equation: ∆(TD/Y)t = ∆(CAD/Y)t − ∆(iγ*t )− ∆[(i − i*)(Fd/Y)]t − ∆(NT/Y)t. From 34. If ψ = κ = 0, the current account will jump from one sustainable level to the next. There are many reasons to assume that both ψ and κ are different from zero, including the existence of adjustment costs in consumption.
234
Brookings Papers on Economic Activity, 1:2005
equations 4 through 6, and after making some assumptions regarding the behavior of other key variables, such as the international terms of trade and interest rates, the following equation for the current account may be derived (to simplify the notation, the mi and xi have been aggregated into broad import and export categories): (10)
∆ ( CAD/Y )t − ∆ ( iγ ∗t ) + ∆ ( i − i∗ ) ( F d /Y )t + ∆ ( NT /Y )t + [ σ x (1 + ε e ) − σ m (1 + ηe )] eˆ + ( σ m − σ x ) ( π − π∗ ) + σ m η y g − σ m ε y ∗g∗ + σ m pˆ ∗m − σ x pˆ ∗x − ( σ m − σ x ) ( g + π ) ,
where σm and σx are ratios of imports and exports to GDP; ηe < 0 and εe > 0 are the price elasticities of imports and exports, respectively; ηy and ε*y are the elasticities of imports and exports with respect to domestic and foreign income, respectively; g and g* represent rates of real GDP growth at home and in the rest of the world, respectively; π and π* are domestic and world inflation, respectively; pˆ*m and pˆ *x are the rates of change in international prices of imports and exports, respectively; and eˆ is the rate of change of the real exchange rate. From this equation it follows that, in order for a real depreciation to improve the trade balance (and, other things equal, the current account), it is required that [σx(1 + εe) − σm(1 + ηe)] > 0.35 Although equation 10 is not a reduced-form equation, it is useful for undertaking a number of simulation exercises.36 For example, with equations 2, 3, 9, and 10, and under assumed values for growth, inflation, and interest rates and changes in the international terms of trade, it is possible to analyze how changes in portfolio preferences will affect the trajectories of the current account and the RER. Simulation Results The barebones model developed above may be used to compute the current account and RER adjustments consistent with shifts in portfolio 35. Under balanced initial trade, this expression becomes the traditional MarshallLerner condition. 36. In equation 10 I have assumed that di = di* = 0. Since α and αjj are exogenous, this assumption does not affect the behavior of the RER. Later in the paper I discuss the way in which changes in interest rates and other variables such as the international terms of trade affect the results.
Sebastian Edwards
235
preferences by foreign and domestic investors, including a reduction in the home bias in portfolio investment decisions.37 A first step in this analysis is to calibrate the model. Table A-2 in the appendix presents the parameter values used in the base-case simulation; most of these values are taken from existing studies of the U.S. and world economies. I have selected the values of ψ and κ that best track the actual dynamics of the U.S. current account between 1996 and 2004; the best results are obtained for ψ = 0.30 and κ = 0.20. I also assume that foreigners’ demand α for U.S. assets increased gradually from 0.205 to 0.300 between 1996 and 2004 (see the values for αHistorical, and αInitial in table A-2). As may be seen from the top left and bottom right panels of figure 5, for these parameter values the model tracks actual current account and RER behavior for 1996–2004 quite closely.38 One limitation of this type of simulation exercise is that it is difficult to forecast how foreign investors’ net demand for U.S. assets will behave in the future. It is precisely for this reason that a number of authors have avoided the issue and have instead computed the RER adjustment “required” to eliminate the current account deficit completely.39 Here I take a different approach: instead of assuming that the current account deficit has to be reduced to zero or some other arbitrary number, I analyze the dynamic of the current account under alternative assumptions regarding foreigners’ net demand for U.S. assets. I am particularly interested in understanding what is likely to happen under an optimistic scenario where foreigners’ demand for U.S. assets continues to grow. What makes this approach particularly interesting is the finding that, even under such a scenario, it is highly likely that, in the not-too-distant future, the U.S. current account will undergo a significant reversal. As table A-2 shows, in these simulation exercises I assume a gradual portfolio adjustment over the next five years. I assume that α increases from its current value of 0.30 to 0.40 by 2010 and that αjj falls from 0.73 37. In fact, there are indications that the process of international capital market integration will continue in the future, as some of the largest emerging economies, including China, are increasingly allowing their nationals to invest abroad. See, for example, “China to Seek Full Currency Conversion,” Financial Times, February 28, 2005, p. 6. 38. To obtain the best possible historical fit for the model, I incorporated into the historical simulation changes in the terms of trade that track what was observed in 1996–2004. 39. Obstfeld and Rogoff (2000, 2004). For similar approaches see Mussa (2004) and Blanchard, Giavazzi, and Sa (this volume).
Actual
2004
2010
2016
2022
Net U.S. international liabilities
Simulated
Current account deficit
Source: Author’s calculations using model described in the text.
20
30
40
50
60
Percent of GDP
2
4
6
Percent of GDP
Figure 5. Results of Base-Case Simulations
100
110
120
Index
1
2
3
4
5
Actual
2004
Percent of GDP
2010
Simulated
2016
Real exchange rate
Trade deficit
Trade account deficit
2022
Sebastian Edwards
237
to 0.71 during the same period. This adjustment implies a reduction in home bias both in the rest of the world and in the United States. In the base-case scenario the assumed portfolio adjustment is equivalent to foreigners doubling their net demand for U.S. assets to an amount equal to 60 percent of U.S. GDP. This is a very large number. Indeed, it implies that, under the assumptions that g = 0.03 and π = 0.023, during the five years from 2005 to 2010 the U.S. NIIP will deteriorate by a further $5.7 trillion. Before proceeding, the following assumptions made in the base-case scenario deserve some comment (see table A-2 for details). First, I assume that the United States and the rest of the world grow at the same rate (g = g*). This is consistent with the idea that, while the United States will grow faster than Europe and Japan, the rest of the world—including China and India—will continue to grow very rapidly. In a number of alternative simulations I consider different values for growth. A second assumption concerns the values of the key elasticities, which have been taken from existing studies on the U.S. and global economies.40 Two important characteristics of these elasticities are that the income elasticity for U.S. imports exceeds that for imports by the rest of the world (the socalled Houthakker-Magee phenomenon), and that the RER elasticity of U.S. imports exceeds (in absolute terms) that of U.S. exports. Finally, in the base-case scenario I assume that the adjustment has no effect on the international terms of trade ( pˆm* = pˆx* = 0); in alternative simulations I consider the case where the terms of trade change. BASE-CASE SIMULATIONS. Figure 5 presents the results of this basecase exercise. In these simulations, 2005 should be interpreted as the “initial” period; the previous eight years (the shaded area) represent recent history. The figure presents simulation results for the current account, the trade account, net U.S. assets held by foreigners, and the trade-weighted RER index (and, for the eight historical years, the actual RER index). The most salient features of the base-case simulation are the following: —Under the deliberately optimistic assumption of a significant increase in foreign net demand for U.S. assets, the current account deficit continues to increase until it peaks at 7.3 percent of U.S. GDP. From that point onward the deficit declines toward its new steady state of 3.2 percent of GDP. 40. See Hooper, Johnson, and Marquez (2000).
238
Brookings Papers on Economic Activity, 1:2005
—Once the current account deficit reaches its peak, the reversal is quite sharp. In the base-case scenario, during the first three years of adjustment the deficit is reduced by 3.1 percent of GDP. The reversal of the trade deficit is even sharper. The reason is that, with a larger net U.S. debtor position, net payments (interest and dividends) to foreign investors increase significantly relative to GDP. —As the bottom right panel shows, once the process of current account reversal begins, the trade-weighted RER index falls rapidly: during the first three years of adjustment, the accumulated real depreciation is 13.3 percent. By the time the new, sustainable current account deficit is reached, the accumulated depreciation amounts to 22.5 percent. This result is roughly in line with other studies (table A-1). In alternative simulations in which the valuation effect of dollar depreciation on the U.S. net foreign asset position is ignored, the resulting real depreciation is larger: for example, in the first three years of adjustment the accumulated depreciation is 16.8 percent. —This simulation also indicates that the new steady state is associated with a sharp depreciation: the RER falls to 19.1 percent below its initial (2005) level. Alternative assumptions regarding growth, inflation, interest rates, the terms of trade, elasticities, and other key parameters will, of course, affect the results. Except when the changes in the assumptions are extreme, however, the main qualitative result holds: even under very optimistic assumptions regarding foreigners’ net demand for U.S. assets, the current account deficit is likely to go through a large reversal in the not-too-distant future. ALTERNATIVE PORTFOLIO CHOICES. An important question is how sensitive these results are to portfolio choices. To explore this issue, I report in figure 6 results from a second simulation exercise, which assumes that, after increasing their net holdings of U.S. assets to 60 percent of U.S. GDP by 2010, foreign investors make a new portfolio adjustment and gradually reduce their desired holdings of U.S. assets to 50 percent of U.S. GDP. (The bottom left panel of the figure depicts the trajectory of net foreign holdings of U.S. assets in this simulation.) As the figure shows, in this case the current account reversal is significantly more abrupt, as is the depreciation in terms of the trade-weighted RER index. In the first three years of the adjustment, the current account deficit declines by 5.3 percent of GDP, and the accumulated depreciation
Source: Author’s calculations using model described in the text.
2022
90
20
2016
100
30
2010
110
40
2004
120
50
Net U.S. international liabilities
0
2
4
Actual
2004
Percent of GDP
Index
Actual
Simulated
Current account deficit
Percent of GDP
2
4
6
Percent of GDP
Figure 6. Results of Simulations Using Alternative Assumptions
2010
2016
Simulated
Real exchange rate
Trade account deficit
2022
240
Brookings Papers on Economic Activity, 1:2005
is 23.7 percent. Moreover, as the top right panel of the figure shows, by the third year of the adjustment (2011 in the simulation), the trade deficit has turned into a surplus. It is important to keep in mind that this simulation still assumes that the long-run net demand by foreigners for U.S. assets is still significantly higher—20 percent of GDP higher, to be precise—than today. Because of space considerations, I have not presented the results of more pessimistic scenarios in which foreigners reduce their net demand for U.S. assets below the current level. Suffice it to say that in those scenarios the current account reversal is even more pronounced, as is the concomitant real depreciation. DOES ADJUSTMENT NEED TO BE ABRUPT? The results presented in figures 5 and 6, and in particular the abrupt current account reversal that takes place after the deficit peak is reached, depend on the assumptions made about parameters ψ and κ; different values of these parameters would result in different dynamics. For instance, if in the future the dynamic of the adjustment process changes, such that ψ declines while κ increases, this would result in a more gradual convergence of the current account deficit to its new, sustainable level. To take a concrete example, values of ψ = 0.20 and κ = 0.35 would result in an accumulated compression of the current account of 1.9 percent of GDP during the first four years of the adjustment process. This is a significantly less drastic adjustment than the 6 percent of GDP obtained in figure 6, and it shows that an abrupt collapse in the deficit is not unavoidable. Furthermore, in this simulation the current account deficit would peak at 6.2 percent of GDP (results not shown) rather than at 7.3 percent as in figure 6. The process of net accumulation of U.S. assets by foreigners may also differ from what I have assumed in both simulations. For instance, if they slow their accumulation of U.S. assets, or if they stretch the process over a longer period, the eventual adjustment would be less abrupt than is depicted in figures 5 and 6. The real depreciation of the trade-weighted dollar might also be less pronounced. This would be the case, for example, if U.S. saving were to increase in the next few years, moving closer to its historical average. In that case expenditure reduction would play a more significant role in the adjustment, and expenditure switching (through dollar depreciation) would be less important. The simulations discussed above assumed an exogenously given rate of growth of GDP. This, of course, need not be the case in reality. It is likely, in fact, that current account reversals of the type and magnitude
Sebastian Edwards
241
suggested by the simulation results will have an effect on real economic activity.41 In the next section I use a new, comparative cross-country data set to investigate the real consequences of current account reversals in the world economy since 1971. This comparative analysis will help give some idea of the possible effects of a U.S. current account reversal like that in the simulations in figures 5 and 6.
How Costly Are Current Account Reversals? An International Comparative Analysis The main message of the simulation exercises just presented is that, even under optimistic scenarios where foreigners’ demand for U.S. assets increases significantly, the U.S. current account experiences a significant reversal in the not-too-distant future. But what will be the nature of the adjustment process? I address this issue here by analyzing the international experience with current account reversals in the period 1971–2001. Although the U.S. case is unique, both because of the size of its economy and because the dollar is the world’s main vehicle currency, an analysis of the international experience will shed some light on the likely nature of the adjustment. A particularly important question is whether this adjustment will entail real costs in the form of slower (or negative) growth and higher unemployment. Previous studies have generated conflicting results: after analyzing the evidence from a large number of countries, Gian Maria Milesi-Ferretti and Assaf Razin conclude that major current account reversals have not been costly: “reversals,” they claim, “are not systematically associated with a growth slowdown.42 Jeffrey Frankel and Eduardo Cavallo, on the other hand, conclude that sudden stops of capital inflows (a phenomenon closely related to reversals) have resulted in growth slowdowns.43
41. See the pioneering study on current account reversals by Milesi-Ferretti and Razin (2000). See also Edwards (2004). 42. Milesi-Ferretti and Razin (2000, p. 303). 43. Frankel and Cavallo (2004).
242
Brookings Papers on Economic Activity, 1:2005
In what follows I analyze several aspects of current account reversals, including44 —their incidence —the relationship between reversals and sudden stops of capital inflows —the relationship between reversals and depreciation —the factors determining the probability of a country experiencing a reversal, and —the costs, in terms of slower growth, of reversals. In analyzing these issues I rely on two complementary statistical approaches: First, I use nonparametric tests to analyze the incidence and main characteristics of current account reversals. Second, I use panel regression-based analyses to estimate the probability of a country experiencing a reversal, and the cost of such a reversal in terms of a short-term decline in output growth. Although the data set covers all regions of the world, in an effort to shed light on the U.S. case, I emphasize the experience of large countries.45 Current Account Reversals during 1971–2001: The International Evidence I use two definitions of current account reversals: I define a type I reversal as a reduction in the current account deficit of at least 6 percent of GDP within a three-year period, and a type II reversal as a reduction in the current account deficit of at least 4 percent of GDP in one year. (In both cases the reversal is recorded as occurring in the year when the episode ends. For example, if a country’s current account deficit declined by 7 percent of GDP between 1980 and 1982, the episode is recorded as having taken place in 1982. Also, for an episode to count as a current 44. In Edwards (2004) I used a smaller data set to investigate reversals in emerging economies. In that paper, however, I did not consider the experience of large or industrial countries with reversals. I also used a very simple framework for analyzing growth. In contrast, in this section I use a two-step dynamic of growth approach. 45. Croke, Kamin, and Leduc (2005) recently analyzed the nature of current account adjustments in industrial economies. Their analysis differs from mine in several respects. First, they concentrate on milder current account adjustments; second, their sample includes only industrial countries; and, third, they are interested in analyzing whether there is likely to be a “disorderly” adjustment, defined as a situation of financial disruption.
Sebastian Edwards
243
account deficit reversal, the initial balance has to be indeed a deficit.46) Thus, in a type I reversal, the magnitude of the adjustment is more pronounced than in a type II reversal but is distributed over a longer period.47 Table 6 presents data on the incidence of both types of reversal for the complete sample of countries as well as for each of the six groups of countries considered in the previous section. For the overall sample the incidence of type I reversals is 9.2 percent, and that of a type II reversal is 11.8 percent. The incidence of reversals among the industrial countries is much smaller, however, at 2.7 percent and 2.0 percent, respectively. Indeed, the Pearson χ2 and F tests reported in table 6 indicate that the hypothesis of equal incidence of reversals across regions is rejected strongly. The industrial countries that experienced type I current account reversals during the period are Finland (in 1978 and 1994), Greece (1988), Ireland (1984), New Zealand (1977–78 and 1988–89), Norway (1979–80, 1989, and 2000), and Portugal (1979 and 1984–85). Those that experienced type II reversals are Austria (1982), Canada (1982), Greece (1986), Iceland (1983 and 1986), Ireland (1975), Italy (1975), Malta (1997), New Zealand (1978), Norway (1989), and Portugal (1982–83 and 1985). With the exception of Italy and Canada, all of these countries are economically very small, underlining the point that there are no historical precedents of large countries undergoing profound current account adjustments. As pointed out above, this implies that the results reported here on current account reversals should be interpreted with a grain of salt and should not be mechanically extended to the case of the United States. The analysis presented above distinguished countries by their stage of development and world region. An alternative way of dividing the sample, and one that is particularly relevant for deriving lessons for the United States, is by economic size. I define “large” countries as those whose GDP placed them in the top 25 percent of the sample distribution in 1995 (by this criterion there are forty-one “large” countries in the sample). For the period 1971–2001 the incidence of type I reversals 46. These definitions differ somewhat from those used in other studies, including Freund (2000), Milesi-Ferretti and Razin (2000), Edwards (2002), and Guidotti, Villar, and Sturzenegger (2003). 47. Notice that it is possible for a country to have experienced both a type I and a type II reversal during the same historical episode.
244
Brookings Papers on Economic Activity, 1:2005
Table 6. Incidence of Current Account Reversals by World Region, 1970–2001 Percenta Type I reversal Region or country group Industrial countries Latin America and Caribbean Asia Africa Middle East Eastern Europe All countries Summary statistics Uncorrected Pearson χ2 (5) Design-based F test (5, 12,500) P-value
Type II reversal
No reversal
Reversal
No reversal
Reversal
97.3 92.0 88.3 88.3 86.6 90.7 90.8
2.7 8.0 11.7 11.7 13.4 9.3 9.2
98.0 87.7 87.7 83.4 85.0 88.9 88.2
2.0 12.3 12.3 16.6 15.0 11.1 11.8
37.31 7.46 0.00
67.42 13.08 0.00
Source: Author’s calculations using data from World Bank, World Development Indicators, various years. a. Number of reversal episodes divided by the product of all countries in the group and all years, times 100.
among large countries is 5.3 percent, and that of type II reversals is 6.8 percent. Current Account Reversals and Sudden Stops of Capital Inflows In the last few years a number of authors have analyzed episodes of sudden stops of capital inflows into a country.48 From an analytical perspective, sudden stops and current account reversals should be closely related phenomena, but there is no reason for them always to occur together. Indeed, because of changes in international reserves, it is perfectly possible for a country to suffer a sudden stop without simultaneously experiencing a current account reversal. However, in countries with floating exchange rates, changes in international reserves tend to be relatively small, and, at least in principle, the relationship between sudden stops and reversals should be stronger. To investigate formally the relationship between these two phenomena, I define a “sudden stop” episode as an abrupt and major reduction in capital inflows to a country that until that time had been receiving large volumes of foreign capital. Specifically, I impose the following criteria: capital inflows into the country in question during the two years preceding 48. See Calvo, Izquierdo, and Mejia (2004); Edwards (2004).
Sebastian Edwards
245
the episode must have been larger (relative to GDP) than those of three quarters of the countries in its region; and net capital inflows must have declined by at least 5 percent of GDP in the year of the episode itself.49 Table 7 presents summary statistics, for three country samples, on the coincidence of sudden stops and current account deficit reversals (under both definitions of the latter). The first sample consists of large countries, defined, as stated above, as those whose GDP is in the top quartile of the sample distribution; the second consists of industrial countries only; and the third is the complete sample. The bottom panel of the table shows, in the first column, that 21.1 percent of all countries experiencing a sudden stop also faced a type I current account reversal, and 15.0 percent of those with type I reversals also experienced (in the same year) a sudden stop. The bottom panel also shows, in the second column, that 51 percent of all countries subjected to a sudden stop faced a type II current account reversal, and that 26.7 percent of those experiencing a type II reversal also suffered (in the same year) a sudden stop. The χ2 tests indicate that in both cases the hypothesis of independence between reversals and sudden stops is rejected. The data for the industrial countries show that the joint incidence of type I reversals and sudden stops is rather low for this group. In fact, according to the χ2 test, the null hypothesis of independence between the two phenomena cannot be rejected. The relationship between sudden stops and type II reversals, however, is somewhat stronger than for type I reversals among this group: the hypothesis of independence is rejected (χ2 = 23.6; p = 0.00). The results for large countries are similar to those for industrial countries. An analysis of the lead-lag structure of reversals and sudden stops suggest that sudden stops tend to occur either before or at the same time (during the same year) as current account reversals. Indeed, a series of nonparametric χ2 tests rejects the hypothesis that current account reversals precede sudden stops (results not shown). Current Account Reversals and the Exchange Rate An important policy question—and one that is particularly relevant for the current policy debate in the United States—is whether current account 49. To check the robustness of the results, I also used two alternative definitions of sudden stops, which considered a reduction in inflows of 3 and 7 percent of GDP in one year. Detailed results using these definitions are not reported here.
246
Brookings Papers on Economic Activity, 1:2005
Table 7. Conditional Incidence of Current Account Reversals and Sudden Stops of Capital Inflows, 1970–2001 Percent Country sample and eventa Large countries Reversal Sudden stop Sudden stop Reversal χ2(1) P-value Industrial countries Reversal Sudden stop Sudden stop Reversal χ2(1) P-value All countries Reversal Sudden stop Sudden stop Reversal χ2(1) P-value
Type I reversal
Type II reversal
9.3 7.0 1.3 0.26
25.5 15.6 27.5 0.00
5.0 7.1 0.4 0.51
18.2 28.6 23.6 0.00
21.1 15.0 26.6 0.00
51.0 26.7 262.5 0.00
Source: Author’s calculations from data in World Bank, World Development Indicators, various years. a. x|y denotes the probability of x occurring given the occurrence of y in the same year.
reversals have historically been associated with unusually large depreciations. The starting point for my analysis of this issue is the construction of an index of “external pressures” along the lines suggested by Barry Eichengreen and others:50 (11)
I t = ∆E E − ( σ E σ R )( ∆R R ) ,
where ∆E/E is the rate of change of the nominal exchange rate, ∆R/R is the rate of change of international reserves, and σE and σR are the standard deviations of changes in the RER and in international reserves, respectively. Traditional analyses define a crisis to have occurred when It exceeds the mean of the index plus k standard deviations. My crisis indicator Ct thus takes a value of 1 (crisis) or zero (no crisis) according to the following rule:51 50. Eichengreen, Rose, and Wyplosz (1996). 51. The pioneering work here is that by Eichengreen, Rose, and Wyplosz (1996), who suggested that the index (equation 11) also include changes in domestic interest rates. The original index, however, has limited use in broad comparative analyses, because most emerging and transition economies do not have long time series on interest rates. For this reason, most empirical analyses are based on a restricted version of the index such as equation 11.
Sebastian Edwards
(12)
247
Ct =
{
1, if I t ≥ mean ( I t ) + kσ 1 0, otherwise.
Using equation 12, I define two currency crisis indicators: The first (crisis type A) uses the traditional crisis index and assigns Ct a value of 1 when k ≥ 3. The second (crisis type B) looks to the nominal exchange rate to determine the value of Ct. In this case It = ∆E/E, and Ct = 1 if It ≥ mean(It) + kσE and 0 otherwise. In this case the country experiences a large depreciation without a major loss in international reserves. This indicator is more relevant for the case of floating exchange rate countries, where changes in international reserves are usually minimal. I computed a number of two-way frequency tables similar to table 7 using both crisis definitions and both definitions of current account reversals. I also performed χ2 tests for independence of occurrence of these phenomena. Table 8 presents data on the shares of current account reversals of both types that are accompanied by crises. Results are presented for the same three samples as above: large countries, industrial countries, and all countries, under three different lag structures (no lag between reversal and crisis, crisis lagged one period, and crisis lagged two periods into the reversal).52 The results suggest that, historically, current account reversals and currency crises have occurred jointly in a large proportion of cases. Consider, for example, the case of crisis type A and reversal type I for the sample of large countries: 26.7 percent of countries with a type I reversal experienced a contemporaneous type A crisis; 43.1 percent experienced such a crisis in the second year of the reversal episode; and 34.5 percent of the reversals were accompanied by a crisis in the third (and final) year of the reversal episode. Table 8 also shows that industrial countries with reversals tended to experience currency crises during the initial year of the reversal episode. The table also reports p-values for χ2 tests of the independence of reversals and currency crises; in most cases the null hypothesis that the two are independent is rejected at conventional levels.
52. Data on the percentage of crises that also correspond to reversals are available from the author on request. The results of the χ2 tests confirm those discussed above. I also used GDP distributions for other years to define large countries and obtained similar results.
248
Brookings Papers on Economic Activity, 1:2005
Table 8. Incidence of Current Account Reversal Associated with Currency Crisisa Percent of countries with reversal episodes Reversal and crisis are contemporaneous Type of reversal and country sample Type I reversal Large countries Industrial countries All countries Type II reversal Large countries Industrial countries All countries
Crisis follows reversal by one year
Crisis follows reversal by two years
Type A crisis
Type B crisis
Type A crisis
Type B crisis
Type A crisis
Type B crisis
26.7 (0.09) 6.7 (0.49) 21.2 (0.10)
16.1 (0.01) 0.0 (0.43) 9.1 (0.38)
43.1 (0.00) 25.0 (0.16) 25.6 (0.00)
17.2 (0.00) 12.5 (0.10) 10.3 (0.08)
34.5 (0.00) 50.0 (0.00) 22.2 (0.01)
13.8 (0.05) 12.5 (0.11) 9.8 (0.09)
31.2 (0.00) 28.6 (0.09) 20.2 (0.05)
18.2 (0.00) 14.3 (0.07) 10.0 (0.03)
42.9 (0.00) 35.7 (0.01) 23.8 (0.00)
15.6 (0.00) 0.0 (0.43) 11.5 (0.00)
29.5 (0.01) 26.7 (0.11) 16.7 (0.86)
12.8 (0.04) 6.7 (0.67) 8.2 (0.47)
Source: Author’s calculations using data from World Bank, World Development Indicators, various years. a. Numbers in parentheses are p-values of the χ2 test.
Table 9 presents data on the distribution of exchange rate changes for countries with type I current account reversals.53 The top panel reports results for the nominal exchange rate (relative to the dollar; here a positive number indicates a depreciation), and the bottom panel for the tradeweighted RER index. These changes are calculated as the cumulative exchange rate change for the period from three years before the reversal to the year of the reversal. For comparison I have also included the distribution of three-year nominal exchange rate changes for a control group of countries that did not experience a current account reversal during 1970–2001. The results in the top panel indicate that countries experiencing reversals have tended to have significantly larger nominal depreciations than the control group. Consider, for example, the case of large countries: the average depreciation associated with a reversal episode in 53. Data on countries experiencing type II reversals are not reported here, but the results are similar and are available from the author on request.
Sebastian Edwards
249
Table 9. Mean Cumulative Changes in Exchange Rates Following Type I Current Account Reversalsa Percent Countries experiencing type I reversal
Countries not experiencing reversal
Kruskal-Wallis test (p-value)b
Nominal exchange rate Large countries Industrial countries All countries
33.1 18.9 27.5
9.2 3.2 9.5
0.00 0.19 0.00
Real exchange ratec Large countries Industrial countries All countries
−1.4 9.3 −4.0
0.04 1.6 3.6
0.12 0.55 0.00
Country sample
Source: Author’s calculations. a. Data are cumulative changes over the three years beginning with the year of the current account reversal. b. The null hypothesis is that the data from the two samples have been drawn from the same population. c. A positive number indicates a real appreciation.
those countries that suffered reversals is 33 percent versus only 9.2 percent for the control group. To test formally whether the nominal exchange rate behaved differently in reversal and control countries, I performed a series of nonparametric Kruskal-Wallis χ2 tests on the equality of the distribution of the cumulative depreciation. The null hypothesis is that the data for the reversal countries and those for the control group have been drawn from the same population. As table 9 shows, in the majority of cases (two out of three) the null hypothesis is rejected at conventional levels. The bottom panel of table 9 presents data for the cumulative change in the RER for the reversal and the control groups. Large countries experienced a rather small real depreciation on average (1.4 percent) in the period surrounding a current account adjustment, a result that is not statistically different from that for the control group (p = .12). For the complete sample the χ2 test indicates that the treatment and the control groups are drawn from different populations. Perhaps surprisingly, for the industrial countries the cumulative average change in the RER is an appreciation, not a depreciation. The average accumulated depreciations (both nominal and real) in the reversal countries reported in table 9 are very small compared with the “required” depreciations calculated in a number of studies, includ-
250
Brookings Papers on Economic Activity, 1:2005
ing the simulations reported earlier in this paper. Obstfeld and Rogoff,54 for example, estimate that eliminating the U.S. current account deficit would require a real depreciation of between 16 and 36 percent. Blanchard, Giavazzi, and Sa have done estimates that indicate a required depreciation of the trade-weighted dollar of 40 percent or more.55 One of many possible reasons for these differences is that the United States is a very large country, whereas the countries that have experienced reversals are much smaller. Also, the elasticities may be different for the United States than for the average reversal country. Yet another possibility has to do with the level of economic activity and aggregate demand. Most recent models of the U.S. current account assume that the economy stays on a full-employment path. It is possible, however, that countries that have experienced reversals have also gone through economic slowdowns, and that a reduction in aggregate demand contributed to the adjustment effort. The Probability of Experiencing a Current Account Reversal To better understand the forces behind current account reversals, I estimated a number of equations on the probability of experiencing a reversal, using panel data and the following empirical model: (13)
ρ jt = 1, if ρ*jt > 0 0, otherwise.
(14)
ρ*jt = α jt + ε jt .
Variable ρjt takes a value of 1 if country j experienced a current account reversal in period t, and zero if it did not. Whether the country experiences a current account reversal is assumed to be the result of an unobserved latent variable ρ*, jt which in turn is assumed to depend linearly on vector jt. The error term εjt is given by a variance component model: εjt = vj + µjt, where vj is independent and identically distributed (i.i.d.) with zero mean and variance σ v2 and µjt is normally distributed with zero mean and variance σµ2 = 1. The data set used covers eighty-seven countries for the 1970–2000 period; data are not available for every country for every
54. Obstfeld and Rogoff (2004). 55. Blanchard, Giavazzi, and Sa (this volume).
Sebastian Edwards
251
year, however. See appendix table A-3 for exact definitions and data sources. In determining the specification of this probit model, I followed the literature on external crises and included the following covariates:56 the oneyear-lagged ratio of the current account deficit to GDP; a “sudden stop” dummy that takes the value of 1 if the country experienced a sudden stop in capital inflows in the previous year; an index of the occurrence of sudden stops in the same region in the same year (to capture the effect of regional contagion); the one-year-lagged ratio of gross external debt to GDP;57 the one-year-lagged rate of growth of domestic credit; the oneyear-lagged ratio of the country’s fiscal deficit to GDP; and the logarithm of the country’s initial GDP per capita. The regressions were performed with and without the fiscal deficit variable for both measures of current account reversal. Table 10 presents the results of estimating this variance-component probit model for a sample of large countries, defined as before. The vast majority of the coefficients have the expected sign, and most are significant at conventional levels. The results may be summarized as follows: A larger current account deficit increases the probability of a reversal in the following year, as does a sudden stop of capital inflows. Countries with higher GDP per capita have a lower probability of a reversal. The results do not provide strong support for the contagion hypothesis: the variable that measures the incidence of sudden stops in the county’s region is significant in only one of the four equations (although its sign is always positive). There is evidence that an increase in a country’s gross external debt increases the likelihood of a reversal, and that larger public sector deficits increase the probability of a type II reversal. Countries with looser monetary policy, as measured by growth in domestic credit, also have a higher probability of experiencing a reversal. Although the United States is a very special case, the results reported in table 10 provide some support for the idea that, during the last few years, the probability of the United States experiencing a current account reversal has increased. Indeed, the United States has experienced a steady increase in some important determinants
56. See, for example, Frankel and Rose (1996), Milesi-Ferretti and Razin (2000), and Edwards (2002). 57. Ideally, one would want to have data for net debt; however, data on net liabilities are unavailable for most countries.
252
Brookings Papers on Economic Activity, 1:2005
Table 10. Probability of Current Account Reversals in Large Countries: Random-Effects Probit Regressionsa Type I reversal Independent variable
b
Ratio of current account deficit to GDP Occurrence of sudden stop in country Index of sudden stops in region Ratio of external debt to GDP Domestic credit growth Ratio of fiscal deficit to GDP Initial GDP per capita No. of observations No. of countries
10-1 0.05 (1.65)* 0.82 (2.06)** 0.78 (0.66) 0.01 (2.81)*** 0.001 (2.50)** −0.004 (0.12) −0.28 (2.19)** 545 36
10-2 0.05 (1.63)* 0.83 (2.08)** 0.80 (0.68) 0.01 (2.88)*** 0.001 (2.52)** −0.29 (2.23)** 582 37
Type II reversal 10-3 0.19 (5.46)*** 0.93 (2.46)** 1.42 (1.54) 0.001 (0.29) 0.0002 (1.65)* 0.05 (1.85)* −0.15 (1.57) 557 36
10-4 0.19 (5.53)*** 0.83 (2.24)** 1.64 (1.84)* 0.001 (0.32) 0.0003 (1.71)* −0.16 (1.66)* 597 37
Source: Author’s regressions. a. Results obtained from estimating the model in equations 13 and 14 in the text on unbalanced panel data for the sample of large countries. Numbers in parentheses are z statistics (in absolute value); all equations include country dummy variables, results for which are not reported. Asterisks indicate statistical significance at the ***1 percent, **5 percent, and *10 percent levels. b. All independent variables are lagged one period.
of reversals, such as its gross international debt, its fiscal deficit, and the current account deficit itself. Current Account Reversals and Growth I investigate next the relationship between current account reversals and real economic performance, with particular attention to the following issues: whether, historically, abrupt current account adjustments have had an effect on GDP growth; whether sudden stops and current account reversals have had similar impacts on growth; and whether the effects of reversals depend on the structural characteristics of the country in question, including its economic size, its openness to trade, and the extent to which it restricts capital mobility. In addressing these issues, I emphasize the case of large countries; as a comparison, however, I also provide results for the complete sample of large and small countries. Previous analyses of the real effects of current account reversals have reached different conclusions. Milesi-Ferretti and Razin, for example,
Sebastian Edwards
253
used before-and-after analyses as well as cross-country regressions to address this issue, concluding that “reversal events seem to entail substantial changes in macroeconomic performance between the period before and the period after the crisis but are not systematically associated with a growth slowdown.”58 On the other hand, in a previous paper I used dynamic panel regression analysis and concluded that major current account reversals had a negative effect on investment, and that they had “a negative effect on GDP per capita growth, even after controlling for investment.”59 GROWTH EFFECTS OF CURRENT ACCOUNT REVERSALS AND SUDDEN STOPS: AN ECONOMETRIC MODEL.
The point of departure for the empirical analysis is a two-equation formulation for the dynamics of real growth of GDP per capita in country j in period t. Equation 15 is the long-run GDP growth equation, and equation 16 captures the growth dynamics: (15)
g j = α + x jβ + rj θ + j .
(16)
∆g jt = λ g j − g jt −1 + ϕv jt + γu jt + ε jt .
I use the following notation: g˜ j is the long-run rate of real growth in GDP per capita in country j; xj is a vector of structural, institutional, and policy variables (identified below) that determine long-run growth; rj is a vector of regional dummies; α, β, and θ are parameters to be estimated; and j is an error term assumed to be heteroskedastic. In equation 16, gjt is the rate of growth of GDP per capita in country j in period t. The terms vjt and ujt are shocks, assumed to have zero mean and finite variance and to be mutually uncorrelated. Specifically, vjt is assumed to be an external terms-of-trade shock, whereas ujt captures other shocks, including current account reversals and sudden stops of capital inflows. εjt is an error term, which is assumed to have a variance component form, and λ, ϕ, and γ are parameters that determine the particular characteristics of the growth process. Equation 16 has the form of an equilibrium correction model and states that the actual rate of growth in period t will deviate from the long-run rate of growth because of the existence of three types of shocks: vjt, ujt, and εjt. Over time, however, the actual rate of growth will tend to 58. Milesi-Ferretti and Razin (2000, p. 303, emphasis added). 59. Edwards (2002, p. 52). In a recent paper, Guidotti, Villar, and Sturzenegger (2003) consider the role of openness in an analysis of import and export behavior in the aftermath of a reversal. See also Frankel and Cavallo (2004).
254
Brookings Papers on Economic Activity, 1:2005
converge toward its long-run value, with the rate of convergence given by λ. Parameter ϕ in equation 16 is expected to be positive, indicating that an improvement in the terms of trade will result in a (temporary) acceleration in the rate of growth, and that negative terms-of-trade shocks are expected to have a negative effect on gjt.60 From the perspective of the present analysis, a key issue is whether current account reversals and sudden stops reduce growth; that is, whether coefficient γ is significantly negative. In estimating equation 16 I used dummy variables for sudden stops and reversals. An important question, addressed in detail below, is whether the effects of different shocks on growth are different for countries with different structural characteristics, such as the degree of trade and capital account openness. Equations 15 and 16 are estimated using a two-step procedure. In the first step I estimate the long-run growth equation 15 using a cross-country data set. These data are averages for 1974–2001, and the estimation corrects for heteroskedasticity. These first-stage estimates are then used to generate long-run predicted growth rates to replace g~j in the equilibrium error correction model (equation 16). In the second step I estimate equation 16 using the generalized least squares (GLS) method for unbalanced panels; I use both random effects and fixed effects estimation procedures (only the former are reported here). The data are annual data for 157 countries for 1970–2000; data are not available for every country for every year, however. (See appendix table A-3 for exact data definitions and sources.) In estimating equation 15, I followed the standard literature on growth, as summarized by Robert Barro and Xavier Sala-i-Martin, Jeffrey Sachs and Andrew Warner, and David Dollar, among others.61 I assume that the long-run rate of growth of GDP g˜ j depends on a number of structural, policy, and social variables: the equation includes the logarithm of initial GDP per capita, the investment ratio, the secondary education coverage rate (as a proxy for human capital), an index of the degree of openness of the economy to trade and capital flows, the ratio of government consumption to GDP, and regional dummies. The results obtained from these first-stage estimates are not reported but are available upon request. 60. See Edwards and Levy-Yeyati (forthcoming) for details. 61. Barro and Sala-i-Martin (1995); Sachs and Warner (1995); Dollar (1992).
Sebastian Edwards
255
Table 11 presents the results of the second-stage estimation of the growth dynamics equation (equation 16), using random effects. The top panel reports results for the sample of large countries, and the bottom panel for the complete sample. The equations whose results are reported in columns 11-1 and 11-2 include the type I and type II reversal dummies, respectively. Column 11-3 includes the sudden stops indicator and neither reversal dummy. Columns 11-4 and 11-5 include both the sudden stops indicator and the type I or the type II reversal variable, respectively, as regressors.62 The results in table 11 may be summarized as follows: The estimated coefficient on the growth gap is, as expected, positive, significant, and smaller than 1. The estimates are on the high side (between 0.66 and 0.72), suggesting that, on average, deviations between long-run and actual growth get eliminated rather quickly. For instance, according to the results in column 11-1, approximately 85 percent of a shock to real growth in GDP per capita will be eliminated within three years. Also, as expected, the estimated coefficients on the terms-of-trade shock are always positive and statistically significant, indicating that an improvement in the terms of trade results in an acceleration, and a deterioration in a deceleration, in the rate of growth of real GDP per capita. As may be seen from columns 11-1 and 11-2, the coefficients on both the current account reversal variables are significantly negative, indicating that reversals result in a deceleration of growth. For large countries these results suggest that, on average, a type I reversal is associated with a reduction of GDP growth by 2.1 percentage points. This effect is eliminated gradually as g converges toward g˜ j. In the case of type II reversals, the estimated negative effect on GDP growth is even larger, at −4.1 percentage points. The results in column 11-3 show that countries that have experienced a sudden stop of capital inflows have also tended to experience a reduction in GDP growth: for large countries the point estimate is −2.4 percentage points. This is the case whether or not the country in question has also suffered a current account reversal. The equations reported in the last two columns in table 11 include both the current account reversal (type I or type II) and sudden stop indicators. The results in column 11-5 suggest that higher costs of adjustment have been 62. In the analysis that follows, and in order to focus the discussion, I concentrate on the effects of current account reversals.
256
Brookings Papers on Economic Activity, 1:2005
Table 11. Impact of Current Account Reversals and Sudden Stops on Economic Growth: Random-Effects GLS Regressionsa Sample and independent variable Large countries Growth gapb Change in terms of trade Type I current account reversal Type II current account reversal Sudden stop of capital inflows Constant No. of observations No. of countries Adjusted R2 All countries Growth gapb Change in terms of trade Type I current account reversal Type II current account reversal Sudden stop of capital inflows Constant No. of observations No. of countries Adjusted R2
11-1 0.67 (21.20)*** 0.09 (7.88)*** −2.12 (3.94)***
11-2 0.72 (25.33)*** 0.10 (10.30)***
11-3 0.68 (22.82)*** 0.08 (7.99)***
11-4 0.66 (20.54)*** 0.08 (7.34)*** −2.11 (3.89)***
−4.13 (9.34)*** −0.28 (2.10)** 799 41 0.41
−0.21 (1.70)* 846 41 0.50
−2.36 (3.99)*** −0.31 (2.36)** 811 41 0.45
0.82 (40.26)*** 0.07 (11.77)*** −1.04 (3.00)***
0.82 (42.10)*** 0.08 (12.65)***
0.81 (40.18)*** 0.07 (11.31)***
−2.39 (3.99)*** −0.18 (1.36) 764 41 0.42
0.82 (38.93)*** 0.07 (11.10)*** −0.73 (2.03)**
−2.01 (6.64)*** −0.30 (2.26)** 1,723 90 0.48
−0.15 (1.16) 1,821 90 0.49
−1.23 −1.02 (2.82)*** (2.28)** −0.27 −0.26 (2.62)*** (2.33)** 1,641 1,546 81 81 0.51 0.52
11-5 0.71 (24.60)*** 0.10 (9.52)*** −3.74 (7.94)*** −1.37 (2.36)** −0.18 (1.39) 810 41 0.50
0.82 (40.76)*** 0.08 (12.18)*** −1.80 (5.50)*** −0.53 (1.19) −0.14 (1.32) 1,635 81 0.51
Source: Author’s regressions. a. Results obtained from estimating the model in equations 15 and 16 in the text on unbalanced panel data. The dependent variable is the change in the growth rate of GDP per capita (in percentage points). Numbers in parentheses are t statistics (in absolute value); all regressions include country dummy variables, results for which are not reported. Asterisks indicate statistical significance at the *** 1 percent, **5 percent, and *10 percent levels. b. Difference between estimated long-run and actual annual growth rates of real GDP, in percentage points.
associated with type II reversals: the coefficient on the type II dummy variable is more than twice as large (in absolute terms) as that on the sudden stop variable in the same equation. According to this equation, countries that have experienced both a reversal and a sudden stop experienced,
Sebastian Edwards
257
on average, a decline in growth in GDP per capita of 5.1 percentage points. To summarize, the results presented in table 11 are revealing and cast some light on the likely costs of a future current account reversal in the United States. Historically, large countries that have suffered such reversals have experienced deep reductions in GDP growth. These estimates indicate that, on average, and with other factors unchanged, the decline in growth in GDP per capita has been in the range of 2.1 to 4.1 percentage points in the first year of the adjustment. Three years after the initial adjustment, GDP growth will still be below its long-run trend. EXTENSIONS, ENDOGENEITY, AND ROBUSTNESS. Here I discuss some extensions of the model and examine the robustness of the estimates, including possible endogeneity bias. Specifically, I address the role of countries’ structural characteristics in determining the costs of adjustment, present results from instrumental variables GLS regressions with random effects, and consider the effects of changes in the terms of trade. Openness and the costs of adjustment. Recent studies on the economics of external adjustment have emphasized the role of openness to trade. Guillermo Calvo, Alejandro Izquierdo, and Luís-Fernando Mejia, Frankel and Cavallo, and I, among others, have found that countries that are more open to international trade tend to incur a lower cost of adjustment to a current account reversal.63 These studies, however, do not distinguish between large and small countries or between openness in the trade account and openness in the capital account. To investigate whether openness has historically affected the cost of external adjustment in large countries, I added two interactive regressors to equation 16: the first interacts the reversal indicator with trade openness, and the second with an index of the country’s degree of international capital mobility. Trade openness is proxied by the fitted value of the ratio of imports plus exports to GDP obtained from a gravity model of bilateral trade.64 The index on international capital mobility is one that I developed in a previous paper;65 the index ranges from 0 to 100, with higher numbers denoting greater capital mobility. The results, presented in table 12, show that the coefficients
63. Calvo, Izquierdo, and Mejia (2004); Frankel and Cavallo (2004); Edwards (2004). 64. The use of gravity trade equations to generate instruments in panel estimation was pioneered by Jeffrey Frankel. See, for example, Frankel and Cavallo (2004). 65. Edwards (forthcoming).
258
Brookings Papers on Economic Activity, 1:2005
Table 12. Impact of Trade Openness and Capital Mobility on Growth in Large Countries: Random-Effects GLS Regressionsa Independent variable Growth gapb Change in terms of trade Type I current account reversal Type I current account reversal × trade openness indicator Type I current account reversal × capital mobility indicator Type II current account reversal Type II current account reversal × trade openness indicator Type II current account reversal × capital mobility indicator Constant No. of observations No. of countries Adjusted R2
12-1
12-2
12-3
12-4
0.67 (21.17)*** 0.09 (7.78)*** −3.48 (1.98)** 0.27 (2.47)** −0.007 (0.24)
0.67 (21.12)*** 0.09 (7.83)*** −3.84 (4.42)*** 0.27 (2.55)**
0.68 (22.35)*** 0.09 (8.77)***
0.68 (22.40)*** 0.09 (8.79)***
−1.92 (1.83)* −0.02 (0.58) −0.05 (1.70)* −0.16 (1.26) 793 41 0.43
−4.12 (7.94)*** −0.04 (1.27)
−0.28 (2.14)** 794 41 0.38
−0.29 (2.19)** 793 41 0.38
−0.16 (1.27) 793 41 0.43
Source: Author’s regressions. a. Results obtained from estimating the model in equations 15 and 16 in the text on unbalanced panel data, with the addition of the variables interacting current account reversals with trade openness and capital mobility indicators. The dependent variable is the change in the growth rate of GDP per capita (in percentage points). Numbers in parentheses are t statistics (in absolute value); all regressions include country dummy variables, results for which are not reported. Asterisks indicate statistical significance at the ***1 percent, **5 percent, and *10 percent levels. b. Difference between estimated long-run and actual annual growth rates of real GDP, in percentage points.
on the reversal indicators continue to be significantly negative, as they were in table 11. The coefficient that interacts trade openness with the presence of a type I reversal is significantly positive in columns 12-1 and 12-2 in table 12. The point estimate in both is 0.27, indicating that external adjustment is less costly in countries with higher trade ratios. However, the coefficient that interacts trade openness and the dummy for reversals is not significant when the type II reversal indicator is used. The coefficient that interacts capital account openness and reversal is not significant in any of the regressions. Endogeneity and instrumental variables estimates. The results discussed above were obtained using a random-effects GLS procedure for unbalanced panels, and under the assumption that the reversal variable is exogenous. However, whether a reversal takes place may be affected by
Sebastian Edwards
259
the country’s growth performance, and thus endogenously determined. To deal with this issue I reestimated equation 16 using an instrumental variables GLS panel procedure. The following instruments were used: the one- and two-period-lagged ratio of the current account deficit to GDP; a lagged sudden stop dummy, which takes the value of 1 if the country experienced a sudden stop in the previous year; the same regional contagion variable used in the previous analysis; the one-year-lagged ratio of external gross debt to GDP; the one-year-lagged ratio of net international reserves to GDP; the one-year-lagged rate of growth of domestic credit; and the logarithm of initial GDP per capita. As the results in table 13 show, the coefficients on the reversal indicators are significantly negative, confirming that, historically, current account reversals have had an adverse effect on growth. The absolute values of the estimated coefficients, however, are larger than those obtained with the random-effects GLS procedure (top panel of table 11). Terms-of-trade effects. The estimation that yielded the results in table 11 controlled for terms-of-trade changes. That is, the coefficients on the type I and II reversal variables capture the effect of a current account reversal with the terms of trade held constant. As discussed above, however, external adjustment in large countries is very likely to affect the terms of trade. The exact nature of that effect will depend on a number of factors, including the relevant elasticities and the extent of home bias in consumption. To get an idea of the effect of current account reversals when international prices are allowed to adjust, I reestimated equation 16 excluding the terms-of-trade variable for the sample of large countries. The full results are not reported here, but the estimated coefficients on the reversal variables were smaller in absolute terms than those in table 11: −2.43 versus −2.12 in table 11 for type I reversals, and −3.63 versus −4.13 for type II reversals. These results suggest that, for this sample, external adjustment has been associated, on average, with an improvement in the international terms of trade. Robustness tests and other extensions. To test the robustness of the results, I also estimated several alternative versions of equation 16 for the sample of large countries. In one of these exercises I introduced lagged values of the reversal indicators as additional regressors. The results (not reported here) indicated that lagged values of these indexes were not significant at conventional levels. I also varied the definition of “large countries,” but this likewise did not dramatically affect the results.
260
Brookings Papers on Economic Activity, 1:2005
Table 13. Impact of Current Account Reversals on Growth in Large Countries: Instrumental Variables Regressionsa Independent variable Growth gapb Change in terms of trade Type I current account reversal
13-1
13-2
0.86 (18.50)*** 0.06 (3.87)*** −9.40 (4.55)***
0.89 (20.50)*** 0.11 (6.86)***
Type II current account reversal Constant No. of observations No. of countries Adjusted R2
0.24 (1.27) 514 34 0.41
−12.24 (7.40)*** 0.38 (1.95)* 538 34 0.40
Source: Author’s regressions. a. Results obtained from estimating the model in equations 15 and 16 in the text on unbalanced panel data using an instrumental variables GLS procedure. The instruments used were the one- and two-period-lagged ratio of the current account deficit to GDP; a lagged sudden stop dummy equal to 1 if the country experienced a sudden stop in the previous year; the regional sudden stop index used in table 10; the one-year-lagged ratio of external gross debt to GDP; the one-year-lagged ratio of net international reserves to GDP; the one-year-lagged rate of growth of domestic credit; and the logarithm of initial GDP per capita. Numbers in parentheses are t statistics (in absolute value); all regressions include country dummy variables, results for which are not reported. Asterisks indicate statistical significance at the ***1 percent, **5 percent, and *10 percent levels. b. Difference between estimated long-run and actual annual growth rates of real GDP, in percentage points.
Concluding Remarks The results reported in this paper illustrate the uniqueness of the current U.S. external situation. Never before in modern economic history has a large industrial country run persistent current account deficits of the magnitude posted by the United States since 2000. This development can be explained in the context of a portfolio model of the current account, where, for a number of reasons—the end of the Cold War, the Internet revolution, and the liberalization of international capital movements in most countries—foreign investors have increased their (net) demand for U.S. assets. Indeed, by increasing their holdings of U.S. assets to 30 percent of their wealth, foreigners have provided American residents with sufficient funds to run the large current account deficits of the last few years. The future of the U.S. current account—and thus of the dollar— depends on whether foreign investors will continue to add U.S. assets to their investment portfolios. As a way of sharpening the discussion, I have deliberately made a very optimistic assumption, namely, that during the
Sebastian Edwards
261
next five years foreigners’ net holdings of U.S. assets, as a proportion of U.S. GDP, will double from their current level. The simulation model indicates that, even under this optimistic assumption, in the not-toodistant future the United States is likely to go through a significant external adjustment. Indeed, one cannot rule out a scenario in which the U.S. current account deficit shrinks abruptly by 3 to 6 percent of GDP. According to the simulations, this type of adjustment would imply a cumulative real depreciation of the trade-weighted dollar in the range of 13 to 23 percent during the first three years of the adjustment. To obtain an idea of the possible consequences of this type of adjustment, I analyzed the international evidence on current account reversals. The results of this empirical investigation indicate that major current account reversals have been associated with large declines in GDP growth: In large countries, with other factors unchanged, the decline in growth in GDP per capita has averaged in the range of 2.1 to 4.1 percentage points in the first year of the adjustment. Three years after the initial adjustment, GDP growth is still below its long-run trend. The results presented in this paper are revealing and suggest that the United States is likely to experience a major adjustment in the not-toodistant future. However, many questions are still unresolved and will require additional research. These include the following: —How does the behavior of foreign central banks, including their future demand for U.S. assets, affect the likelihood and magnitude of a U.S. current account reversal? A particularly important question involves the appropriate international reserves policy for central banks in a world where most exchange rates have at least some flexibility. A number of analysts are concerned that the Asian central banks will reduce their demand for U.S. assets, unleashing an abrupt collapse in the value of the dollar. —How exactly does the adjustment process work in large countries? Although I have concentrated on a group of countries that I defined as “large,” in fact all of the countries in my sample that have experienced current account reversals have much smaller economies than the United States. In particular, more analysis is needed of the consequences for global interest rates of a major U.S. current account adjustment. —How do nominal exchange rates behave in a current account adjustment episode? Most models of the U.S. current account imbalance, including the portfolio model presented here, have focused on the RER.
262
Brookings Papers on Economic Activity, 1:2005
However, estimating the way in which the adjustment process will affect nominal exchange rates is not a trivial matter. The actual adjustment in nominal exchange rates will depend on pass-through coefficients, as well as on the exchange rate policies followed by some important U.S. trade partners, including China, Japan, and other Asian countries. —Finally, how does private sector saving, and in particular household saving, behave in the United States? To the extent that household saving increases—as a result of a decline in home prices, for instance—the current account deficit will decline without putting pressure on the value of the dollar. Likewise, if saving in foreign countries declines, the current account surplus in the rest of the world will also tend to decline, helping to achieve global balance. Of course, what matters for current account behavior is aggregate national saving. Therefore the behavior of public sector saving is a fundamental variable for future current account and real exchange rate behavior in the United States and the rest of the world.
Main assumptions Income elasticity of imports (1.7) exceeds that of exports (1.0). Base-case scenario assumes no RER adjustment for dollar. Scenario with dollar adjustment assumes real depreciation of 25 percent. Structural adjustment scenario assumes that exports’ elasticity increases to 1.3.
Elasticity of substitution between tradables and nontradables is assumed equal to 1. Assumes 6 percent annual nominal interest rate and NIIP equal to 20 percent of GDP. Tradables output is assumed to be 25 percent of GDP. Assumes that full employment is maintained.
Analyzes rates of return obtained by foreign owners of U.S. assets.
Methodology
Model tracks U.S. NIIP through time. Analyzes trajectory of NIIP under three scenarios and asks whether trajectories are sustainable. Elasticities-based adjustment mechanism. Considers two scenarios for global growth.
Develops and calibrates optimizing model of small open economy with two goods: tradable and nontradable. Output is exogenous; prices are assumed to be flexible; monetary policy stabilizes price level. Analyzes effect on RER of exogenous shock that results in reduction of current account deficit by 4.4 percent of GDP.
Analyzes trajectory of NIIP as percentage of GDP.
Mann (1999)
Obstfeld and Rogoff (2000)
O’Neill and Hatzius (2002)
Author(s)
Main results
(continued)
Analysis finds it unlikely that U.S. will continue to attract foreign buyers for its assets at observed
In base case, elimination of current account deficit implies 16 percent real depreciation, and 12 percent nominal depreciation of the dollar. Scenario assuming tradables share of GDP is 15 percent results in real depreciation of 20 percent. Effect on nominal value of dollar could be even higher if reduction in current account deficit is very rapid.
In base-case scenario NIIP becomes increasingly negative and current account is unsustainable in medium run. In real depreciation scenario current account deficit is within sustainable range even in a 10-year horizon. In structural adjustment scenario current account deficit is 3 percent of GDP in10-year horizon, if global economy performs well.
Table A-1. Comparison of Selected Studies of U.S. Current Account Adjustment and the Dollar, 1999–2005
APPENDIX A
Wren-Lewis (2004)
Author(s)
Current account deficit of 2 percent of GDP is consistent with nominal exchange rates of 88 yen to dollar and 1.18 dollars to euro. Under positive technological shock, “sustainable” current account deficit may be larger and consistent with nominal exchange rates of 89–100 yen to dollar rate and 1.11–1.19 dollars to euro. Estimates that if China has current account surplus of 1 percent of GDP, nominal exchange rate would be 6.71 renminbi to dollar.
To determine initial conditions, estimates “underlying” (or cyclically adjusted) current account balances. Considers three possible longterm scenarios corresponding to current account deficits of 1, 2, and 3 percent of GDP. Uses partial equilibrium model of small economy with three goods (including nontraded good). Elasticities and other parameter values taken from regression analysis and from OECD data set.
Calibrates partial equilibrium model to obtain set of bilateral RERs consistent with attaining specified (exogenous) current account deficits. No attempt is made to determine sustainable level of U.S. current account. Considers effect of U.S. fiscal shock and of U.S. technological shock.
low rates of return, and thus its current account deficit is clearly unsustainable. Return to sustainable deficit (2 percent of GDP) will imply real depreciation of as much as 43 percent.
Argues that, with exception of FDI, these rates of return have been modest. Shows that FDI has declined significantly as source of current account deficit financing.
Argues that at the observed levels of current account deficits, the NIIP is moving toward the levels of Canada, Australia, and New Zealand. It is difficult to believe that this is possible for a large country such as the U.S. Estimates “required” RER depreciation in order to bring current account deficit to 2 percent and NIIP not to surpass 40 percent.
Main results
Main assumptions
Methodology
Table A-1. Comparison of Selected Studies of U.S. Current Account Adjustment and the Dollar, 1999–2005 (continued)
Estimates RER path consistent with equilibrium in nontradable goods market. RER is assumed to depend on country’s net foreign asset position and on relative productivity.
Analyzes trajectory of NIIP and argues that it is unlikely to continue to grow at current pace. If it did, it would reach 100 percent of GDP. Argues that challenge is for RER adjustment to be gradual and not disrupt growth. Argues that U.S. fiscal adjustment is necessary for smooth correction of imbalances. No attempt is made to calculate “outer limit” of U.S. NIIP. Analyzes RER adjustment compatible with a gradual reduction of current account deficit to 2 percent of GDP and NIIP between 40 and 50 percent.
Benassy-Quere and others (2004)
Mussa (2004) On basis of results from large econometric models, assumes that 1-percentage-point reduction of U.S. current account deficit is associated with 10 percent real depreciation.
Model is estimated simultaneously for fifteen currencies. Data on net foreign assets obtained from Lane and Milesi-Ferretti (2004) and relative productivities obtained as ratio of consumer to producer price index. No attempt made to impose external equilibrium condition. Results provided for two cases: using dollar as numeraire and using euro as numeraire.
(continued)
Calculates that further real depreciation of 20 percent relative to mid-2004 values is needed to achieve long-term current account deficit of 2 percent of GDP. Discusses policies that would assist adjustment process: fiscal consolidation in the U.S. to help keep U.S. demand growth below pace of output growth, and more expansive monetary policy in Europe and Japan. Concludes that “some” international policy cooperation is likely to help the adjustment process.
Extent of currency misalignment depends on how broad is adjustment. Using dollar as numeraire, estimates that euro was undervalued by between 1.2 and 7.6 percent in 2003, and yen by between 14.3 and 22.1 percent in 2001.
Main assumptions Estimates trade balance equation and uses resulting coefficients to compute real depreciation “required” to achieve different current account adjustment targets. Trade equation also includes foreign and U.S. demand growth.
Ratio of current account deficit to tradables is 25 percent; current account deficit is 5 percent of GDP. Output is exogenously given in both countries. NIIP is 20 percent of GDP. Domestic country produces 22 percent of world tradables. Simulation is done for alternative values of elasticities and under different assumptions regarding changes in tradables output and military spending.
Methodology
Update of O’Neill and Hatzius (2002) model. Analyzes trajectory of NIIP as percentage of GDP and finds that path is not sustainable. Introduces role of productivity gains into original framework. Analyzes composition of capital flows into U.S. Incorporates role for valuation effects.
Extends Obstfeld-Rogoff (2002) model to two-country world. Terms of trade are now endogenous. Incorporates valuation effects of exchange rate changes on NIIP. Assumes elimination of current account deficit, that is, reduction equal to 5 percent of GDP.
O’Neill and Hatzius (2004)
Obstfeld and Rogoff (2004)
Author(s)
Assuming constant output, elimination of current account deficit implies real depreciation between 14.7 and 33.6 percent. If tradables output increases by 20 percent, required real depreciation ranges from 9.8 to 22.5 percent. If permanent increase in military expenditure occurs, required real depreciation ranges from 16.0 to 36.1 percent.
Reduction of current account deficit to 3 percent of GDP would imply real depreciation on order of 21.6 to 23.6 percent. Reduction of current account deficit to 2 percent would imply real depreciation on order of 32.1 to 34.1 percent. Elimination of current account deficit would imply real depreciation on order of 53 to 55 percent (significantly greater than estimated by Obstfeld and Rogoff, 2004).
Main results
Table A-1. Comparison of Selected Studies of U.S. Current Account Adjustment and the Dollar, 1999–2005 (continued)
Considers dynamics of adjustment. Considers valuation effects of changes in dollar. Simulates model under certain assumptions for values of key parameters (elasticities, portfolio shares, and others). Asks what real depreciation of dollar is required to eliminate current account deficit?
Uses portfolio model to analyze U.S. current account behavior. Assumes changes in portfolio preferences in world economy.
Blanchard, Giavazzi, Sa (this volume)
Sources: Literature cited.
First scenario considers a constant RER for dollar. Second scenario considers constant trade deficit at 5 percent of GDP and real depreciation of approximately 7 percent. Third scenario considers faster growth rate of exports and substantial (50 percent) depreciation and assumes gradual elimination (by 2012) of fiscal deficit.
Uses macro aggregate model to project U.S. current account. Imposes exogenous assumptions on RER and analyzes current account path.
Roubini and Setser (2004)
Estimates range for required real depreciation (today). After incorporating valuation effects, this range is estimated at between 40 and 90 percent.
In first scenario current account deficit is 13 percent of GDP in 2012. In second scenario current account deficit is 9 percent of GDP in 2012. In third scenario NIIP stabilizes at approximately 55 percent of GDP and current account deficit declines gradually, reaching 4.3 percent of GDP in 2012.
268
Brookings Papers on Economic Activity, 1:2005
Table A-2. Parameter Values for Variables Used in the Simulations Variable
Parameter value
Portfolio adjustment model WWorld Initial WUS Initial αinitial
$80 trillion $36 trillion 0.300
αjj, initial
0.730
αfinal
0.400
αjj, final
0.710
αhistorical
0.205
αjj, historical
0.800
λ γ*initial
3.0 0.290
γ*final γ*historical
0.600 0.150
Transfer problem g
0.03
g*
0.03
π π*
0.023 0.023
i
0.043
i*
0.053
ηe
−1.10
Definition and comments World wealth in 2005 U.S. wealth in 2005 Foreigners’ demand for U.S. assets in (early) 2005a U.S. residents’ demand for U.S. assets in (early) 2005 Foreigners’ portfolio allocation to U.S. assets in 2010. An alternative simulation assumes that, after reaching 0.40, α declines gradually to 0.365 in 2014. U.S. residents’ demand for U.S. assets in (early) 2010. An alternative simulation assumes that, after reaching 0.71, αjj rises to 0.72 in 2014. Foreigners’ demand for U.S. assets in (early) 1996. The move to the “initial” current value of 0.30 is assumed to have been gradual. U.S. residents’ demand for U.S. assets in (early) 1996 Wealth-to-GDP ratio Value of [αθ − (1 − αjj)]λ in (early) 2005 (see text) Value of [αθ − (1 − αjj)]λ in 2010 Value of [αθ − (1 − αjj)]λ in 1996 Assumed long-term sustainable annual rate of growth of U.S. GDP Annual growth rate of rest-of-world GDP (including emerging economies as well as Europe and Japan) Long-term annual rate of U.S. inflation Long-term annual rate of foreign inflation; some simulations used a value of 0.03. Long-term real U.S. interest rate; some simulations used a value in the range 0.05 to 0.065. Long-term real rest-of-world interest rate; some simulations used values in the range 0.06 to 0.075. Price elasticity of U.S. imports; this is slightly below the consensus value; a range of values was used in other simulations. (continued)
Sebastian Edwards
269
Table A-2. Parameter Values for Variables Used in the Simulations (continued) Variable
Parameter value
εe 0.50 ηy 1.50 εy 1.20 σm 0.14 σx pˆm*
0.09 0
pˆx* 0 ψ 0.30 κ 0.20
Definition and comments Real exchange rate elasticity of U.S. exports (approximately the consensus value); sensitivity analyses used a range of 0.2 to 0.6. Consensus value for income elasticity of U.S. imports Consensus value for income elasticity of U.S. exports Share of imports as a fraction of U.S. GDP in 2004 Share of exports in U.S. GDP in 2004 Rate of change in world price of imports; in alternative simulations a range of −0.05 to −0.10 was used. Rate of change in world price of exports; in alternative simulations a range of 0.05 to 0.07 was used. Coefficient measuring rate of adjustment from actual to demanded asset stock; value chosen to obtain best possible fit for 1996–2004. Coefficient measuring rate of adjustment of absorption to change in income; value chosen to obtain best possible fit for 1996–2004 period.
Source: Author’s model described in the text. a. The adjustment period for α and αjj is assumed to be five years.
270
Brookings Papers on Economic Activity, 1:2005
Table A-3. Variable Definitions and Data Sources Variable
Definition
Source
Type I current account reversal
Reduction in an existing current account deficit by at least 6 percent of GDP over three years.
Author’s determination based on data from World Bank, World Development Indicators, various years
Type II current account reversal
Reduction in an existing current account deficit by at least 4 percent of GDP in one year
Author’s determination based on data from World Bank, World Development Indicators, various years
Sudden stop
Reduction in net capital inflows by at least 5 percent of GDP in one year. The country must have received an inflow of capital larger than its region’s third quartile during the preceding two years.
Author’s determination based on data from World Bank, World Development Indicators, various years
Type A currency crisis
Dummy variable equal to 1 when an index of external pressures exceeds its mean by 3 standard deviations
Author’s determination based on international reserves and nominal exchange rate data from International Monetary Fund, International Financial Statistics, various years
Type B currency crisis
Dummy variable equal to 1 when an index of external pressures exceeds its mean by 3 standard deviations exclusively through changes in the nominal exchange rate
Author’s determination based on nominal exchange rate data from International Monetary Fund, International Financial Statistics, various years
Nominal exchange rate
Dollars per local currency unit
International Monetary Fund, International Financial Statistics, various years
Real exchange rate
Bilateral real exchange rate calculated using consumer price indexes in both countries
Author’s calculations using nominal exchange rate and consumer price index data from International Monetary Fund, International Financial Statistics (continued )
Sebastian Edwards
271
Table A-3. Variable Definitions and Data Sources (continued) Variable
Definition
Source
Change in terms of trade
Change in capacity to import for a given amount of exports, in terms of constant local currency
World Bank, World Development Indicators
Ratio of reserves to GDP
Net international reserves divided by GDP
World Bank, World Development Indicators
Domestic credit growth
Growth rate of domestic credit in percent a year
World Bank, World Development Indicators
Ratio of external debt to GDP
Total external debt divided by GDP
World Bank, World Development Indicators
Ratio of fiscal deficit to GDP
Overall government budget deficit divided by GDP
World Bank, World Development Indicators
GDP per capita
Real GDP per capita in 1995 dollars
World Bank, World Development Indicators
Index of capital mobility
Index from 0 to 100, with higher values indicating greater capital mobility
Edwards (forthcoming)
Trade openness indicator
Exports plus imports divided by GDP
World Bank, World Development Indicators
Comments and Discussion Kathryn M. E. Dominguez: The U.S. current account deficit at the end of 2004 reached 5 percent of GDP, a remarkably high number and far outside the experience of any other large developed country. This paper by Sebastian Edwards examines the factors that have led to such a large imbalance, attempts to forecast how long deficits of this magnitude can be sustained, and analyzes the likely near-term consequences for the U.S. economy of a reversal of the current account balance. Current account deficits have been the norm for the United States for some twenty-five years, just as surpluses have been the norm for many developing countries and many of the rest of the world’s developed countries. In theory a deficit, even if persistent, is not necessarily cause for concern. A country can finance deficits only if the world perceives it to be a good credit risk. Indeed, the reason many developing countries are forced to run surpluses is that they lack access to deficit financing. So, if the world has been willing to finance U.S. current account deficits for over a quarter of a century, why the concern? Edwards makes the case that the reason for concern is that the U.S. current account deficit is not sustainable: the world will not be willing to continue to finance U.S. deficits on the current scale into the future. He argues further that, even if net demand for U.S. assets continues to increase, a current account reversal is inevitable, which in turn will result in a significant reduction in U.S. growth. If Edwards’s predictions are accurate, the implications for the U.S. economy are quite bleak. I will begin by discussing some of the underlying factors that have led to the recent large U.S. current account deficits, and which are likely to influence the future path of global imbalances. Next I will raise some disagreements with some of the assumptions made in the portfolio balance 272
Sebastian Edwards
273
model used by Edwards to simulate U.S. current account dynamics into the future. I will then discuss whether evidence from smaller countries regarding the links between current account reversals and economic growth is relevant for the U.S. situation. Finally, I will return to Edwards’s predictions about the likely downturn for the U.S. economy in light of the evidence from the data and the model. Edwards’s figure 1, which graphs the U.S. current account balance and the dollar real exchange rate since 1973, provides some historical context and shows the close association of dollar appreciations and U.S. current account deficits. His table 1 shows that the sources of financing for these deficits have changed significantly over the past few years. In particular, foreign direct investment and other equity flows, which were important in the 1990s, have been replaced with net fixed-income flows, consisting largely of purchases of U.S. Treasury securities by foreign central banks. This shift in financing has had the favorable consequence (from the point of view of the U.S. income account) of U.S. investors receiving higher returns on the foreign assets they hold than foreigners have received on their U.S. assets. In addition, because the bulk of U.S. liabilities held by foreigners are denominated in dollars, whereas the bulk of foreign assets held by U.S. investors are denominated in foreign currency, the recent dollar depreciation has led to positive valuation effects, which, in turn, have improved the U.S. net international investment position. An examination of the factors driving the movements in these data is warranted, especially if these factors are expected to persist. In this context it is interesting to contrast the role of exchange rate policies in the 1980s with that in the more recent period. In the mid-1980s several foreign countries joined with the United States in coordinated interventions to bring down the value of the dollar and correct global imbalances. In the more recent period there has been no such coordinated attempt on the part of the United States or the rest of the world to intervene against the dollar. Quite the opposite: central banks in Asian countries have been intervening to support the value of the dollar relative to their own currencies, by building dollar reserves to the tune of about $2 trillion. China alone holds around $610 billion in dollar reserves and Japan $840 billion. Their purchases of low-return U.S. Treasury securities have sustained and indeed amplified global imbalances by serving to both finance the U.S. deficit and maintain a high value of the dollar. Edwards discusses this role of foreign central banks in sustaining the U.S. deficit in the context of the United States’ increasing vulnerability to
274
Brookings Papers on Economic Activity, 1:2005
a change in sentiment toward the dollar on the part of these few key players. An alternative view is that, if the Asian central banks come to believe that it is in their best interests to help correct global imbalances, they can look to the policies followed by central banks in the 1980s to help bring about an orderly change in the value of the dollar. Another issue that is little discussed in the paper, but is clearly a driving force behind the current global imbalances, is differences in economic growth rates. The U.S. economy in the 1990s sustained the longest expansion in its recorded history. At the same time, Europe and Japan largely experienced at best lackluster growth. These growth differentials affected the U.S. current account balance in two ways. First, faster relative U.S. growth led to higher aggregate demand in the United States for both domestic products and imports, relative to foreign demand for U.S. products. Second, because the U.S. economy was booming in both absolute and comparative terms, global investors were attracted to U.S. assets, which in turn helped finance the deficit and maintain a strong dollar. Any future changes in growth differentials are similarly likely to affect the size and sustainability of the U.S. current account deficit. Edwards’s figure 4 shows U.S. investment and saving rates since 1970. What is striking is how different today’s saving rates look relative to historical norms. In the past decade and especially in the last few years, the net household saving rate has been unusually low. This, combined with the dramatic fall in net public saving in recent years, has reduced total U.S. saving to its lowest level in a quarter of a century. Investment in recent years has also been relatively low, although well within the range of historical norms during periods of slow growth. A country’s current account deficit, of course, equals the difference between saving and investment, and if we believe that the U.S. saving rate will eventually revert to historical norms, this provides additional reason for optimism that the current account deficit will improve. Edwards introduces a partial equilibrium version of a simple portfolio balance model of the current account to simulate how potential changes in the world’s appetite for U.S. assets will influence current account and real exchange rate adjustment. A number of the simplifying assumptions implicit in the model are likely to have important implications for the results. In particular, relative asset allocation shares (the shares of their wealth that foreign and U.S. investors allocate to foreign and domestic assets) are assumed to be exogenously determined and subject to home bias. The model
Sebastian Edwards
275
is then used to analyze how an exogenous change in these portfolio allocation preferences will influence the current account. In his base-case simulation, Edwards assumes that foreign net holdings of U.S. assets rise to 60 percent of U.S. GDP by 2010 (starting from 30 percent at the end of 2004), that the U.S. economy and the rest of the world economy grow at the same rate, that the U.S. saving rate remains at its current low level, that the income elasticity for U.S. imports is higher than for rest-of-world imports, and that the international terms of trade remain unchanged. The key result is that, even given the assumed doubling in foreign demand for U.S. assets, the U.S. current account deficit will eventually decline to a steady state of 3.2 percent of GDP, with a relatively sharp reversal occurring in 2007, bringing about a 3 percent of GDP reduction in the deficit in three years. The accumulated real dollar depreciation over this period is 22.5 percent. The virtue of Edwards’s model is its simplicity. It allows the reader to easily follow the mechanics of how a change in portfolio preferences influences all the other key variables (net assets held by foreigners, the trade balance, the real exchange rate, and the international wealth transfer associated with these changes). The problem with the model is that its assumptions more or less guarantee the main result, namely, that even with an increase in foreign holdings of U.S. assets, the U.S. current account will experience a sharp reversal. Edwards briefly discusses how a relaxation of some of his assumptions would likely influence (and in most cases soften) his most dire predictions, but the main message remains that a reversal is inevitable. The final section of the paper uses cross-country evidence to estimate the likely costs of a U.S. current account reversal in terms of growth and employment. The analysis involves examining the incidence of current account reversals since 1970 for various categories of countries, the correlation of reversals and sudden stops in capital inflows, the correlation of reversals and exchange rate changes, and, finally, the correlation of reversals and GDP growth rates. Edwards is candid about a serious problem with this analysis: since the United States is unique both because of its size and because of the role of the dollar in the global economy, it is not clear that the cross-country evidence marshaled here has any relevance for the United States. Indeed, Edwards finds no historical examples of large countries that have experienced a large current account reversal along the lines predicted by his portfolio balance model.
276
Brookings Papers on Economic Activity, 1:2005
Yet the statistical information provided is interesting in its own right, even if its relevance to the U.S. experience is uncertain. The international evidence suggests that current account reversals, especially when combined with sudden stops in capital inflows, result in significant reductions in rates of GDP growth. It is certainly hard to argue that a dramatic fall in the current account deficit could occur without major economic dislocation in any country, including the United States. What is less clear is whether the United States is likely to experience a dramatic current account reversal or a sudden stop in capital inflows. Indeed, the fact that many of the factors driving the large current account deficit are unusual relative to historical (or international) norms suggests that predicting future dynamics based on recent experience may not be appropriate. Does an exit strategy exist for the United States that would allow it to forestall a sharp current account reversal? This question evokes a whole series of related questions that seem worth considering before coming to any conclusions about the future course of the U.S. economy. How might a rise in U.S. interest rates (which would likely both attract more foreign investment and decrease the U.S. income account) muddy the waters? Might higher interest rates, in turn, influence household saving rates? Will the U.S. government continue to run large budget deficits, and, if so, might U.S. households turn Ricardian and save more? Might an internationally coordinated intervention strategy to gradually lower the value of the dollar work to improve global imbalances? How might an increase in rest-of-world growth rates (or a fall in U.S. growth rates) influence trade and investment patterns? In summary, Edwards has provided a stimulating paper that argues plausibly that the U.S. current account deficit is both unsustainable and likely to lead to a fall in U.S. economic growth. This conclusion relies on the assumption that many of the key U.S. and foreign macroeconomic variables will continue along their course of the past few years. Time will tell whether or not this assumption is a valid one. Given the rather bleak implications of Edwards’s analysis, my hope is that the data prove him wrong. Pierre-Olivier Gourinchas: Sebastian Edwards has written an ambitious paper on an important topic. The paper starts with a thirty-year perspective on U.S. current account developments. It introduces and calibrates a portfolio balance model designed to help in understanding the developments of recent years. It then uses the model to project the adjustment path for the
Sebastian Edwards
277
dollar’s real exchange rate and the U.S. current account. Finally, it discusses the growth implications of current account adjustment in light of the experience of other countries. I cannot stress enough the importance of the topic: few macroeconomic questions are as pressing today as that of the sources and the implications of growing world external imbalances,1 and few issues are receiving as much attention from academics and policymakers alike. In that respect the paper stands resolutely on the alarmist side of the current debate, claiming that the U.S. external position is not sustainable. Even under optimistic assumptions about foreigners’ appetite for U.S. assets, the paper finds, “the [U.S.] current account will have to go through a significant adjustment in the not-too-distant future.” This adjustment would not be immediate, but it would be dramatic when it arrives: the benchmark projection finds that the current account deficit first increases to 7.3 percent by 2009, then experiences an abrupt reversal that brings it down to 3.2 percent by about 2018. Such a reversal in the current account would be accompanied by a sharp real depreciation of the dollar of 22.5 percent once the correction begins. Finally, the paper argues, a current account reversal of this magnitude is also typically associated with a significant slowdown in economic activity. READING THE TEA LEAVES. The paper starts with a thorough analysis of the buildup in U.S. external imbalances. Like other papers before it,2 it emphasizes two important elements of the current situation. First, the source of financing of the current account deficit has shifted away from foreign private investors to foreign central banks, and away from equity and direct investment to fixed-income vehicles, especially U.S. Treasury securities. Second, the deficit has been associated in recent years with a decline in U.S. national saving, especially household and public saving. Since the financing of the deficit by foreign central banks does not reflect market forces, it follows that U.S. households and government are artificially living beyond their means. This interpretation, shared by many commentators, puts the blame squarely on U.S. domestic factors. The willingness of Asian central banks, concerned about the value of their currency against the dollar, to finance the deficits only serves to maintain the gravity-defying properties of the deficits and of the dollar exchange rate and makes the U.S. position more vulnerable. 1. See Rajan (2005). 2. For example, Roubini and Setser (2004).
278
Brookings Papers on Economic Activity, 1:2005
Although this is a compelling story, it is possible to construct a different reading of the situation. First, the decline in national saving can be an equilibrium outcome, without any shift in the U.S. saving schedule. In fact, if the main impetus behind the current imbalances were a decline in U.S. saving, those imbalances should be associated with high world real interest rates. This was the case in the early 1980s in the United States, the period of so-called twin deficits. Yet what do we observe? According to the Federal Reserve, the ten-year yield on Treasury inflation-indexed securities has declined from 2.29 percent in January 2003 to 1.63 percent in February 2005. An appealing solution to what has become known as Greenspan’s conundrum was put forth by Federal Reserve Governor Ben Bernanke.3 He observes that the combination of large deficits and low real interest rates could be the result of a shift in the net saving schedule of the rest of the world, through either reduced foreign investment or increased foreign saving (the saving glut hypothesis). This shift could be a consequence perhaps of the recent current account reversals in Asia as well as the growth slowdown in Europe. It is also instructive to analyze the sources of financing of the U.S. deficits. As mentioned above, a common interpretation, shared by this paper, is that the current account is increasingly financed through reserve accumulations. I want to argue that the picture is quite different when one considers gross flows together with net flows (table 1). Edwards emphasizes that net foreign direct investment and equity flows for the United States turned negative in 2003–04. Hence the net financing had to have come from net reserves and net debt accumulation. Yet consider gross equity and direct investment liability flows in the table. After a decline between 2000 and 2003, these flows increased again in 2004, from $37.3 billion to $56.2 billion for equities and from $39.9 billion to $115.5 billion for direct investment. What this means is that the negative net flows come from the larger U.S. gross purchases of equity ($93 billion in 2004) and direct investment ($248.5 billion). The increase in gross U.S. foreign liabilities totaled $1.4 trillion in 2004. This total capital inflow helped finance a $666 billion current account deficit and $818 billion in foreign asset acquisition by U.S. investors (with a $52 billion statistical discrepancy). There is really no meaningful way in which the $358 billion in net reserve accumulation has to have served to finance the current account deficit. One could just as accurately 3. Bernanke (2005).
Sebastian Edwards
279
Table 1. Net and Gross Financial Flows of the United States Billions of dollarsa Item
1998
1999
2000
2001
2002
2003
2004
Reserves, net −26.7 Foreign private purchases of 28.6 U.S. Treasuries Currency 16.6 32.1 Securities, netb Debt, gross assets 22.8 Debt, gross liabilities 110.7 Net debt 87.8 Equity, gross assets 101.4 Equity, gross liabilities 45.6 Net equity −55.7 Foreign direct 36.4 investment, net Assets 142.6 Liabilities 179.0 Claims reported by −15.1 nonbanks, net Claims reported 4.2 by banks, net Other −1.2 Net financing 75.0
52.3 44.5
42.5 −70.0
23.1 −14.4
110.3 100.4
250.1 113.4
358.1 108.1
22.4 182.6 1.9 185.9 184.0 114.3 112.9 −1.4 64.5
5.3 338.0 15.2 267.4 252.2 106.7 192.5 85.8 162.1
23.8 309.2 −24.5 274.4 298.9 109.1 119.5 10.4 24.7
21.5 16.6 301.4 178.6 −33.5 −28.1 229.3 213.7 262.8 241.8 17.6 100.4 56.2 37.3 38.6 −63.2 −62.4 −133.9
14.8 323.2 −2.2 357.9 360.0 93.0 56.2 −36.8 −133.0
224.9 289.4 −21.5
159.2 321.3 31.9
142.3 167.0 57.6
134.8 72.4 32.6
173.8 39.9 55.1
248.5 115.5 −41.5
−22.0
−31.7
−7.5
66.1
65.2
−15.6
−2.1 231.7
−1.8 476.3
−1.6 415.0
−0.9 569.0
−2.5 542.7
−0.2 614.0
296.8 65.1
413.4 −62.8
385.7 −29.3
473.9 −95.0
530.7 −12.0
665.9 51.9
Memoranda: Current account balance Statistical discrepancy
209.6 134.6
Source: Bureau of Economic Analysis, U.S. International Transactions, table 1. a. Items may not sum to totals because of rounding. b. Excluding U.S. Treasury securities.
say that it served to finance the $248.5 billion in foreign direct investment acquisitions by U.S. firms. In fact, as my figure 1 shows, although the share of official flows into the United States has recently increased dramatically, from about zero to 25 percent of foreign-owned assets in the United States, this follows just as dramatic a collapse, from 25 percent to –5 percent between 1996 and 1998. Figure 1 highlights that official purchases represent—except for the few years of the equity bubble—a relatively stable share of gross liability flows. As a rough estimate, foreign official assets provide at most only a quarter of the total capital inflows into the United States. CURRENT ACCOUNT REVERSALS AND ASSET SUBSTITUTION. The portfolio balance model developed in the paper predicts large but delayed current
280
Brookings Papers on Economic Activity, 1:2005
Figure 1. Share of Official Assets in Total Foreign Purchases of U.S. Assetsa Percent
40 Excluding bank and nonbank claims
30 20 All claims
10 0
1992
1994
1996
1998
2000
2002
Source: Bureau of Economic Analysis, International Transactions Accounts. a. Figure reports official purchases as a share of all foreign claims in the United States and as a share of all foreign claims excluding bank and non bank claims. Bank and nonbank asset and liability claims roughly offset each other.
account and exchange rate adjustments. I must admit that I was initially puzzled by the difference between the estimates of this paper and those by Olivier Blanchard, Francesco Giavazzi, and Filipa Sa in this volume. After all, both papers use the same model (in Edwards’s paper, equations 1 through 4) and a similar calibration of the model’s parameters. Both models emphasize the sort of valuation effects that have received considerable attention recently.4 Yet Blanchard, Giavazzi, and Sa predict a gradual adjustment of the current account (and a gradual depreciation of the dollar, cumulating to 54 percent), whereas, as already noted, Edwards predicts that the current account will keep worsening for the next four years before reversing sharply between 2009 and 2012. The answer lies in the specification of the asset demand side of both models. Whereas Blanchard and his coauthors assume that investors’ demand for assets depends upon the expected excess return (the expected rate of dollar depreciation when local currency returns are constant and equal), Edwards assumes that the portfolio shares are exogenous. This is an extreme 4. See Gourinchas and Rey (2005) and Lane and Milesi-Ferretti (2004b).
Sebastian Edwards
281
assumption: it implies that there is no substitution between domestic and foreign assets, regardless of their relative returns. Edwards assumes that the degree of home bias in investment decreases exogenously between 2004 and 2010. Crucially, it decreases more for foreign investors (the share of U.S. assets in foreign portfolios increases from 0.3 to 0.4) than for U.S. investors (the share of foreign assets in U.S. portfolio increases from 0.27 to 0.29). Abstracting from the short-term dynamics represented by equation 10, this decline in relative home bias is what sustains the increase in current account deficits for the next four years (top left panel of Edwards’s figure 5) and the continued appreciation of the dollar over the same period (bottom right panel of figure 5). But consider what happens when foreigners stop increasing the share of U.S. assets in their portfolios. At that point the drying up in foreign financing requires a drastic current account reversal, which can only be triggered by a real dollar depreciation. Hence the predictions of the current account reversal in terms of timing and magnitude come mostly from the assumptions about the path of relative asset shares. The sharp dollar depreciation that occurs when the adjustment begins delivers crushingly low returns to foreigners on their U.S. investment. They lose 22 percent in three years if local currency returns are equal and constant. One would expect that, faced with such negative excess returns, investors would demand less U.S. assets today. But a lower demand for U.S. assets today would precipitate the adjustment in the exchange rate and the current account. This indicates that the assumption of exogenous portfolio shares, although useful to calibrate and simplify the model, is eventually too extreme. Besides the assumptions on portfolio shares, the assumption of a common growth rate in the United States and the rest of the world matters also.5 Consider, for instance, what would happen if the rest of the world is expected to grow faster than the United States. Given stable portfolio shares, foreigners will want to purchase more U.S. assets as they grow richer. Hence the supply of foreign capital will not collapse abruptly. 5. This is especially important when looking at the steady state of the model. In the benchmark calibration, while the trade deficit stabilizes, the dollar keeps depreciating in real terms. What is happening is a consequence of the Houthakker-Magee paradox. If the income elasticity of imports exceeds that of exports, the real depreciation must occur along a balanced growth path with common growth rates at home and abroad and a stable ratio of the trade balance to GDP. It is possible to have a stable real exchange rate if the rest of the world grows faster than the United States (Krugman’s 45-degree rule; Krugman, 1989).
282
Brookings Papers on Economic Activity, 1:2005
CURRENT ACCOUNT REVERSALS AND GROWTH. The last part of the paper emphasizes that a current account reversal is likely to trigger a growth slowdown in the United States. This section presents a rich set of empirical results. This is an area where Edwards has made numerous important contributions, and I will limit myself to two brief observations. The first is that his projected real depreciation of the dollar does not allow for the possibility of a U.S. growth slowdown. Yet it is possible that a decline in output growth would improve the current account, and as imports drop with domestic income, the residual decline in the real exchange rate should be comparatively smaller than in the absence of a slowdown. More important, although I agree with the general conclusion of this section—indeed, it is hard to see how a current account reversal of the size that the paper projects could occur without major disruptions in the U.S. economy—it is important to keep in mind that the U.S. situation is unprecedented in many ways besides the size of the external deficit. In particular, never before in peacetime has the center country of the international monetary system accumulated net liabilities on such a scale. It is reasonable to analyze current account reversals for other economies—most of which are small relative to gross world product—using the small-country theoretical apparatus. It is more questionable to do the same for the United States, as Edwards acknowledges. General equilibrium considerations— already essential in understanding the source of the current account deficit in the first place—are crucial in understanding the possible rebalancing that needs to occur. World asset and good prices cannot be taken as given. To conclude, it is the great merit of this paper to offer an ambitious contribution to the debate on the sustainability and ultimate adjustment in U.S. external imbalances. The paper’s emphasis on the portfolio allocation problem is also most welcome.
General discussion: Panel members discussed Edwards’s model’s prediction of a sharp depreciation of the dollar leading to current account reversal and possible crisis. Olivier Blanchard observed that these predictions result from the absence of valuation effects and zero substitutability between domestic and foreign assets in the model. With valuation effects, a smaller depreciation of the dollar would reduce the share of dollar assets in foreign portfolios to the level consistent with foreign investors’ preferences. And, with some substitutability between domestic and foreign assets, the depreciation would likely be much more gradual, starting as soon as investors
Sebastian Edwards
283
anticipated the possibility of a change in portfolio preferences. Indeed, it would already have started. Richard Cooper believed the model’s assumption that foreign investors would target a certain share of U.S. assets in their portfolios was too rigid. Although the assumption of a preference shift to more dollar assets can explain the recent accumulation, in the simulations, when the new, preferred portfolio is achieved, the demand for U.S. assets falls abruptly, leading to a crisis. Cooper suggested that a more natural baseline might call for some fraction of rest-of-world saving to be invested in the United States for an extended period, in which case the model would predict a gradual slowdown in demand for U.S. assets and a much softer landing. Peter Garber observed that Edwards’s baseline model does not explain how the rest of the world would absorb $500 billion of added saving that would arise from a reversal in the U.S. current account equal to 4 percent of U.S. GDP. He conjectured that the added saving would feed back into foreign investors’ behavior, possibly changing the model’s predictions in a meaningful way. Gian Maria Milesi-Ferretti said he found it difficult to judge how far the present, decade-long trend toward greater global diversification of portfolios would go or how it might end. The world may be transitioning to a new steady state, but we have little guidance about where that new steady state will be. This implies that the history of current account reversals is not a good benchmark for the present U.S. situation, but one cannot rule out the possibility that some shock will lead to a crisis and return investors’ behavior to what it was in the 1990s. Barry Eichengreen questioned the robustness of the paper’s empirical analysis, comparing it to a similar recent analysis by Hilary Croke, Steven Kamin, and Sylvain Leduc. Their sample includes more industrialized countries and a less restrictive definition of reversal, which leads them to predict sizable current account reversals in some large countries, including France, the United Kingdom, and the United States. Unlike Edwards, however, they do not find large negative output effects associated with reversals, possibly indicating that large countries are able to adjust more smoothly. Sebastian Edwards replied that the differences between his results and those of Croke, Kamin, and Leduc arise from the very mild episodes of exchange rate adjustment that qualify as reversals in their analysis. He found it unsurprising that such gradual adjustments are not disruptive. Hélène Rey professed puzzlement at Edwards’s finding that a sudden stop or reversal of capital inflows has more pronounced effects the higher the
284
Brookings Papers on Economic Activity, 1:2005
degree of trade openness. Theory, as embodied in the model by Obstfeld and Rogoff in this volume, holds that a more closed economy requires a larger relative price adjustment, which suggests that the disruption to the economy will be more substantial. Believing there might be an omitted-variable bias in Edwards’s results, she suggested including the degree of liability dollarization, which has been a robust explanatory variable in the sudden stop literature, as an independent variable. Edwards agreed that his results were surprising, but he added that, in the sudden stop literature, liability dollarization plays no role in the impact of a sudden stop or reversal, but only on the probability of their happening. Milesi-Ferretti conjectured that the reason the empirical analysis found current account reversals not to be associated with large exchange rate swings is that changes in the terms of trade are often associated with current account reversals. As examples he noted that some countries in the sample, such as Norway and New Zealand, have highly volatile terms of trade. If the price of their exports doubles, their current account balances improve without a weakening of their currencies. Several panelists stressed the importance of valuation effects in preventing the U.S. net debt position from increasing substantially in recent years. Cooper pointed out that, in 2003, the U.S. current account deficit reached $530 billion, yet the U.S. net asset position declined by only $98 billion. The valuation effects that account for this difference are due both to exchange rate effects and to differing returns on foreign and domestic gross portfolios. Richard Portes observed that the depreciation that contributed to these valuation effects cannot go on forever. Milesi-Ferretti added that, to the extent valuation effects from differences in asset returns hold up the dollar, the trade deficit will grow faster, requiring a larger depreciation eventually.
Sebastian Edwards
285
References Ades, Alberto, and Federico Kaune. 1997. “A New Measure of Current Account Sustainability for Developing Countries.” New York: Goldman Sachs Emerging Markets Economic Research. Barro, Robert J., and Xavier Sala-i-Martin. 1995. Economic Growth. New York: McGraw-Hill. Benassy-Quere, Agnes, Pascale Duran-Vigneron, Amina Lahreche-Revil, and Valerie Mignon. 2004. “Burden Sharing and Exchange Rate Misalignments within the Group of Twenty.” In Dollar Adjustment: How Far? Against What? edited by C. Fred Bergsten and John Williamson. Washington: Institute for International Economics. Bergsten, C. Fred, and John Williamson, eds. 2003. Dollar Overvaluation and the World Economy. Special Report 16. Washington: Institute for International Economics (February). ________. 2004. Dollar Adjustment: How Far? Against What? Washington: Institute for International Economics. Bernanke, Ben S. 2005. “The Global Saving Glut and the U.S. Current Account Deficit.” The Sandridge Lecture, Virginia Association of Economics, Richmond, March 10. Available at www.federalreserve.gov/boarddocs/speeches/ 2005/200503102. Blanchard, Olivier, Francesco Giavazzi, and Filipa Sa. 2005. “The U.S. Current Account and the Dollar.” Working Paper 11137. Cambridge, Mass.: National Bureau of Economic Research (February). Caballero, Ricardo, Emmanuel Farhi, and Mohamad L. Hammour. 2004. “Speculative Growth: Hints from the U.S. Economy.” Working Paper 10518. Cambridge, Mass.: National Bureau of Economic Research (May). Calvo, Guillermo A., Alejandro Izquierdo, and Luis-Fernando Mejia. 2004. “On the Empirics of Sudden Stops: The Relevance of Balance-Sheet Effects.” Working Paper 10520. Cambridge, Mass.: National Bureau of Economic Research (May). Choi, Chi-Young, Nelson Mark, and Donggyu Sul. 2004. “Unbiased Estimation of the Half-Life to PPP Convergence in Panel Data.” Working Paper 10614. Cambridge, Mass.: National Bureau of Economic Research (July). Cooper, Richard. 2004. “America’s Current Account Deficit Is Not Only Sustainable, It Is Perfectly Logical Given the World’s Hunger for Investment Returns and Dollar Reserves.” Financial Times, November 1. Corden, W. Max. 1994. Economic Policy, Exchange Rates, and the International System. University of Chicago Press. Croke, Hilary, Steven B. Kamin, and Sylvain Leduc. 2005. “Financial Market Developments and Economic Activity during Current Account Adjustments in Industrial Economies.” International Finance Discussion Paper 827. Washington: Board of Governors of the Federal Reserve System (February).
286
Brookings Papers on Economic Activity, 1:2005
Dollar, David. 1992. “Outward-Oriented Developing Economies Really Do Grow More Rapidly: Evidence from 95 LDCs, 1976–1985.” Economic Development and Cultural Change 40, no. 3: 523– 44. Dooley, Michael P., David Folkerts-Landau, and Peter Garber. 2004a. “The Revived Bretton Woods System: The Effects of Periphery Intervention and Reserve Management on Interest Rates and Exchange Rates in Center Countries.” Working Paper 10332. Cambridge, Mass.: National Bureau of Economic Research (March). ________. 2004b. “Direct Investment, Rising Real Wages and the Absorption of Excess Labor in the Periphery.” Working Paper 10626. Cambridge, Mass.: National Bureau of Economic Research (July). Edwards, Sebastian. 1995. Crisis and Reform in Latin America: From Despair to Hope. Oxford University Press for the World Bank. ________. 1999. “Crisis Prevention: Lessons from Mexico and East Asia.” Working Paper 7233. Cambridge, Mass.: National Bureau of Economic Research (July). ________. 2002. “Does the Current Account Matter?” In Preventing Currency Crises in Emerging Markets, edited by Sebastian Edwards and Jeffrey A. Frankel. University of Chicago Press. ________. 2003. “Debt Relief and the Current Account: An Analysis of the HIPC Initiative.” World Economy 26, no. 4: 513–31. ________. 2004. “Thirty Years of Current Account Imbalances, Current Account Reversals and Sudden Stops.” International Monetary Fund Staff Papers 51 (special issue): 1–49 (January). ________. Forthcoming. “Capital Controls, Sudden Stops and Current Account Reversals.” In International Capital Flows, edited by Sebastian Edwards. University of Chicago Press. Edwards, Sebastian, and Eduardo Levy-Yeyati. Forthcoming. “Flexible Exchange Rates as Shock Absorbers.” European Economic Review. Eichengreen, Barry, Andrew K. Rose, and Charles Wyplosz. 1996. “Contagious Currency Crises.” Working Paper 5681. Cambridge, Mass.: National Bureau of Economic Research (July). Frankel, Jeffrey A., and Eduardo A. Cavallo. 2004. “Does Openness to Trade Make Countries More Vulnerable to Sudden Stops, or Less? Using Gravity to Establish Causality.” Working Paper 10957. Cambridge, Mass.: National Bureau of Economic Research (December). Frankel, Jeffrey A., and Andrew K. Rose. 1996. “Currency Crashes in Emerging Markets: An Empirical Treatment.” Journal of International Economics 41, no. 3–4: 351–66. Freund, Caroline L. 2000. “Current Account Adjustment in Industrialized Countries.” International Finance Discussion Paper 692. Washington: Board of Governors of the Federal Reserve System (December).
Sebastian Edwards
287
Gourinchas, Pierre-Olivier, and Hélène Rey. 2005. “International Financial Adjustment.” Working Paper 11155. Cambridge, Mass.: National Bureau of Economic Research (February). Guidotti, Pablo E., Agustin Villar, and Federico Sturzenegger. 2003. “Aftermaths of Current Account Reversals: Export Growth or Import Compression.” Paper presented at the Eighth LACEA Meeting, Puebla, Mexico, October. Hooper, Peter, Karen Johnson, and Jaime Marquez. 2000. “Trade Elasticities for G-7 Countries.” Princeton Studies in International Economics 87. Princeton University. Kraay, Aart, and Jaume Ventura. 2002. “Current Accounts in the Long and Short Run.” Working Paper 9030. Cambridge, Mass.: National Bureau of Economic Research (June). Krugman, Paul R. 1989. “Differences in Income Elasticities and Trends in Real Exchange Rates.” Working Paper 2761. Cambridge, Mass.: National Bureau of Economic Research. Lane, Philip R., and Gian Maria Milesi-Ferretti. 2001. “The External Wealth of Nations: Measures of Foreign Assets and Liabilities for Industrial and Developing Countries.” Journal of International Economics 55, no. 2: 263–94. ________. 2004a. “International Investment Patterns.” Discussion Paper 4499. London: Centre for Economic Policy Research (July). ________. 2004b. “Financial Globalization and Exchange Rates.” CEPR Discussion Paper 4745. London: Centre for Economic Policy Research (November). Mann, Catherine L. 1999. Is the U.S. Trade Deficit Sustainable? Washington: Institute for International Economics. ________. 2003. “How Long the Strong Dollar?” In Dollar Overvaluation and the World Economy, edited by C. Fred Bergsten and John Williamson. Special Report 16. Washington: Institute for International Economics (February). ________. 2004. “The U.S. Current Account, New Economy Services, and Implications for Sustainability.” Review of International Economics 12, no. 2 (special issue): 262–76 (May). Milesi-Ferretti, Gian Maria, and Assaf Razin. 2000. “Current Account Reversals and Currency Crises: Empirical Regularities.” In Currency Crises, edited by Paul Krugman. University of Chicago Press. Mussa, Michael. 2004. “Exchange Rate Adjustments Needed to Reduce Global Payments Imbalance.” In Dollar Adjustment: How Far? Against What? edited by C. Fred Bergsten and John Williamson. Washington: Institute for International Economics. Obstfeld, Maurice, and Kenneth Rogoff. 2000. “Perspectives on OECD Economic Integration: Implications for U.S. Current Account Adjustment.” In Global Economic Integration: Opportunities and Challenges. Kansas City, Mo.: Federal Reserve Bank of Kansas City (August).
288
Brookings Papers on Economic Activity, 1:2005
________. 2004. “The Unsustainable U.S. Current Account Position Revisited.” Working Paper 10869. Cambridge, Mass.: National Bureau of Economic Research (November). O’Neill, Jim. 2003. “Features of a Dollar Decline.” In Dollar Overvaluation and the World Economy, edited by C. Fred Bergsten and John Williamson. Special Report 16. Washington: Institute for International Economics (February). O’Neill, Jim, and Jan Hatzius. 2002. “US Balance of Payments: Still Unsustainable.” Global Economics Paper 70. New York: Goldman Sachs (March). ________. 2004. “US Balance of Payments: Unsustainable, But. . . .” Global Economics Paper 104. New York: Goldman Sachs (March). Rajan, Raghuram G. 2005. “Global Current Account Imbalances: Hard Landing or Soft Landing.” Speech at the Credit Suisse First Boston Conference, March 15. Available at www.imf.org/external/np/speeches/2005/031505.htm. Roubini, Nouriel, and Brad Setser. 2004. “The U.S. as a Net Debtor: The Sustainability of the U.S. External Imbalances.” Stern School of Business, New York University, and University College, Oxford (August). Sachs, Jeffrey D., and Andrew Warner. 1995. “Economic Reform and the Process of Global Integration.” BPEA, no. 1: 1–95. Taylor, Alan M. 2002. “A Century of Current Account Dynamics.” Working Paper 8927. Cambridge, Mass.: National Bureau of Economic Research (May). Tille, Cedric. 2003. “The Impact of Exchange Rate Movements on US Foreign Debt.” Current Issues in Economics and Finance 9, no. 1: 1–7 (January). Wren-Lewis, Simon. 2004. “The Needed Changes in Bilateral Exchange Rates.” In Dollar Adjustment: How Far? Against What? edited by C. Fred Bergsten and John Williamson. Washington: Institute for International Economics.
DEAN BAKER Center for Economic and Policy Research J. BRADFORD DELONG University of California, Berkeley PAUL R. KRUGMAN Princeton University
Asset Returns and Economic Growth IT is difficult to see how real U.S. GDP growth can be as rapid in the next half-century as it has been in the last. The baby boom is long past, and no similar explosion of fertility to boost the rate of labor force growth from natural increase has occurred since or is on the horizon. The modern feminist revolution is two generations old: no reservoir of potential female labor remains to be added to the paid labor force. Immigration will doubtless continue—the United States is likely still to have only one-twentieth of the world’s population late in this century and to remain vastly richer than the world on average—but can immigration proceed rapidly enough to make the labor force grow as fast in the next fifty years as it did in the past fifty? Productivity growth, the other possible source of faster GDP growth, is a wild card: although we find very attractive the arguments of Robert Gordon for rapid future productivity growth,1 his is not the consensus view; this is shown most strikingly by the pessimistic projection of the Social Security trustees that very long run labor productivity growth will average 1.6 percent a year.2 A slowing of the rate of real economic growth raises challenges for the financing of pay-as-you-go social insurance systems that rely on a rapidly expanding economy to provide generous benefits for the elderly at relatively low tax rates on the young. An alternative way of financing such systems
1. Gordon (2003). Oliner and Sichel (2003) and Kremer (1993) provide additional reasons to be very optimistic about future productivity growth. 2. Board of Trustees of the Federal Old Age and Survivors Insurance and Disability Insurance Trust Funds (2005; all citations from this report are for the intermediate projection). Contrast this with the 2.0 percent average annual rate of economy-wide labor productivity growth from the fourth quarter of 1989 through the first quarter of 2005.
289
290
Brookings Papers on Economic Activity, 1:2005
is to prefund them, and for that reason projections of future rates of return on capital play an important role in today’s economic policy debates. The solutions to many policy issues depend heavily on whether historical real rates of return—especially the 6.5 percent or so annual average realized rate of return on equities—are likely to persist: the higher are likely future rates of return, the more attractive become policies that, at the margin, shift some additional portion of the burden of financing social insurance onto the present and the near future, thus giving workers’ contributions the power to compound over time. We believe that the argument for prefunding—that slowing economic growth creates a presumption that the burden of financing social insurance should be shifted back in time toward the present—is much shakier than many economists recognize.3 It is our belief that if forecasts of slower real GDP growth come to pass, then it is highly likely that future real returns to capital will likewise be significantly below past historical averages. In our view the links between asset returns and economic growth are strong: the algebra of capital accumulation and the production function and the standard macrobehavioral analytical models that economists use as their finger exercises suggest this; arithmetic suggests this as well, for we cannot see any easy way to reconcile current real bond, stock dividend, and stock earnings yields with the twin assumptions that asset markets are making rational forecasts and that rationally expected real rates of return will be as high in the future as they have been in the past half-century. Our basic argument is very simple. Consider a simple chart of the supply and demand for capital in generational perspective (figure 1). The supply of capital—the amount of investable assets accumulated by savers— presumably follows a standard (if probably steeply sloped) supply curve,4 with relative quantities of total saving and thus of capital plotted on the horizontal axis, and the price of capital—that is, its rate of return—on the vertical axis. The demand for capital by businesses will, of course, depend 3. An argument challenged, for reasons similar to but not exactly aligned with those we discuss here, in Cutler and others (1990). 4. Supply is likely to be steeply sloped because of opposing income and substitution effects. An increase in the rate of return increases the total lifetime wealth of savers, which presumably increases their consumption when young and so diminishes their saving. An increase in the rate of return also increases the incentive to save, which presumably increases saving. The net effect—which we believe to be positive—is likely to be relatively small.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
291
Figure 1. The Supply and Demand of Capital and the Rate of Return Rate of return
Supply of saving
Demand under rapid economic growth
Demand under slow economic growth
Capital
on the rate of return demanded by the savers who commit their capital to businesses: the higher this required rate of return, the lower will be business demand for capital—and the more eager will businesses be to substitute labor for capital in production. The demand for capital by businesses depends on many other factors as well, from which we single out two: —The rate of growth of the labor force. Labor and capital are complements. A larger labor force for firms to hire from will raise the marginal product of capital for any given level of the capital stock, making businesses more willing to pay higher returns in order to get hold of capital. —The rate of improvement in the economy’s level of technology. Better technology—also a complement to capital—will boost business demand. What is the effect of a slowdown in economic growth—through either a fall in the rate at which the labor force grows, or a fall in the rate at which technology and thus equilibrium labor productivity increase—on this equilibrium? Assume that these changes do not affect the saving behavior of
292
Brookings Papers on Economic Activity, 1:2005
the accumulating generation:5 then they affect only the demand curve and not the supply curve. Each of these shocks moves the demand curve leftward: having fewer workers reduces the marginal product of capital and hence firm demand for capital; slower productivity growth does the same. The equilibrium capital stock falls, and the rate of return that savers can demand, while still finding businesses willing to invest what they have saved, falls as well. Slower economic growth brings with it lower real rates of return. We make our case as follows. After first laying out what we see as the major issues to be resolved, we discuss how the algebra of the production function and capital accumulation suggests that rates of return and rates of growth are strongly linked. We then analyze the standard, very simple, macrobehavioral models that economists use to address these issues and find that they, too, lead us to not be surprised by a strong positive relationship between economic growth and asset returns. We then turn to the arithmetic: starting from current bond, stock dividend, and stock earnings yields, we find it arithmetically very difficult to construct scenarios in which asset returns remain at their historic average values when real GDP growth is markedly slowed. Next we turn to what we regard as the most interesting possibility for escape from this bind. In the late nineteenth century, slower growth in the British economy was accompanied by no reduction in returns on British assets, as Britain exported capital on a scale relative to the size of its economy never seen before or since. Could the United States follow the same trajectory? Yes. Is it likely to? Not without a huge boost to national saving. Before concluding, we turn to a brief analysis of the equity premium. Much argument and some analysis of the dilemmas of the U.S. social insurance system point to the large historical value of the equity premium in America as a potential source of excess returns. We argue, however, that once one has conditioned on the level of the capital-output ratio, returns on 5. As Gregory Mankiw points out in his comment on this paper, and as we discuss below, in the standard Ramsey model a reduction in the rate of natural increase does affect the saving of the accumulating generation—and shifts the saving supply curve inward exactly as much as investment demand shifts inward, keeping the real rate of return unchanged. This is due to the powerful bequest motive behind the assumption of an infinitely lived representative household whose utility for a given level of consumption per capita is linear in the size of the household.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
293
balanced portfolios in the long run depend only on the physical return to capital and the margins charged by financial intermediaries. (However, attitudes toward risk do affect the long-run capital-output ratio.) They do not depend on the equity premium or the price of risk. We conclude that if economic growth over the next century falls as far as envisioned by forecasts like those in the 2005 Social Security trustees’ report, then it is not very likely that asset returns will match historical experience. If the stock market today is significantly overvalued and about to come back to earth, if the distribution of income undergoes a significant shift away from labor and toward capital, or if the United States massively boosts its national saving rate and runs surpluses on the relative scale of pre–World War I Britain, for more than twice as long as Britain did— then a real GDP growth slowdown need not entail a significant reduction in asset returns. But these seem to us to be possible, not probable, scenarios, and not the central tendency of the distribution of possible futures that is a real economic forecast.
Issues The United States is in all likelihood undergoing a minor demographic transition: from a twentieth century in which the population’s rate of natural increase was high, to a twenty-first century in which, many suspect, fertility will be at or below levels consistent with zero population growth. This will translate into a slowdown in growth in labor input. From 1958 to 2004, total hours worked in the economy grew at 1.5 percent a year as the entrance of the baby-boomers—male and female—and their successors into the labor force vastly outweighed a decline in average hours worked. The Social Security Administration’s 2005 trustees’ report projects that hours worked will grow at only 0.3 percent a year from 2015 through 2045.6 Meanwhile some economists—although far from all—are projecting a slowdown in productivity growth.7 The Social Security Administration foresees economy-wide labor productivity growing at only 1.6 percent a year in 2011 and thereafter. In contrast, between 1995 and 2004 economywide labor productivity grew at 2.5 percent a year, between 1990 and 2004 6. Board of Trustees (2005). 7. See Oliner and Sichel (2003); Gordon (2003); Nordhaus (2005).
294
Brookings Papers on Economic Activity, 1:2005
it grew at 2.0 percent a year, and between 1958 and 2004 at 1.9 percent a year.8 Thus, less than a decade from now, the Social Security forecasters at least see a significant change in both key factors in economic growth: a fall of 1.2 percentage points a year in the rate of growth of labor input, and a fall of between 0.3 and 0.9 percentage point, depending on whether one takes the long 1958–2004 or the short 1995–2004 baseline, in labor productivity growth. The total growth slowdown forecast to hit in a decade or less is thus in the range of 1.6 to 2.2 annual percentage points of real GDP. What implications will this growth slowdown—if it comes to pass— have for asset values and returns? One position, taken implicitly by the Social Security Administration and explicitly by others,9 is that there is no reason to expect asset returns to be lower in the future. Whereas U.S. economic growth is determined by productivity growth and labor force growth in the United States, U.S. asset returns are determined by time preference, the intertemporal elasticity of substitution in consumption, and attitudes toward risk, all in a global economy. Why should they be connected? Thus, we hear, past asset performance is still the best guide to future returns. We take a contrary position. Yes, safe asset returns are equal to the marginal utility of saving, stock market returns equal safe asset returns plus the cost of bearing equity risk, and the United States is part of a world economy. Yes, economic growth is equal to productivity growth plus labor force growth. But only in the case of a small open economy with fixed exchange rates are asset returns determined independently of the rate of economic growth. In a large open economy, they are jointly determined and will be linked.10
8. An alternative breakdown would distinguish 1958–73, during which economy-wide labor productivity growth averaged 2.6 percent a year; the productivity slowdown period of 1973–95, when it averaged 1.2 percent a year; and the post-1995 “new economy” period, when it averaged 2.6 percent a year. Much depends on whether one interprets the 1973–95 productivity slowdown period as an anomalous freak disturbance to the economy’s normal structure, or as just one of those things one can expect to see every half-century or so. 9. Council of Economic Advisers (2005). 10. Even in a small open economy, real returns on assets and rates of economic growth will be linked unless the real exchange rate is fixed. Even perfect arbitrage by mammoth amounts of risk-neutral foreign capital only equalizes expected rates of return at home and abroad calculated in foreign currency. With a flexibly changing real exchange rate, the rate of return in foreign currency is not the same as the rate of return in domestic currency.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
295
Perhaps an analogy will be helpful. In international trade, the trade balance is the difference between what exporters are able to sell abroad and home demand for imports. In international finance, the trade balance is the difference between national saving and national investment. How can this be? Why should a change in exporters’ success at marketing abroad change either national saving or national investment? Great confusion has been caused throughout international economics over how, exactly, to think of the connection. We believe that claims that national economic growth is unconnected with asset returns reflect a similar failure to grasp the whole problem. This is an important issue to get straight now, because the relative attractiveness of pay-as-you-go versus prefunded social insurance systems depends to some degree on the gap between the return on capital r and the rate of real economic growth n + g, the sum of the rate of growth of employment n and the rate of growth of labor productivity g. If we are willing to be simple Benthamites, with a social welfare function that shamelessly makes interpersonal comparisons of utility, the argument is straightforward. The higher is the rate of economic growth n + g relative to the return on capital r, the more attractive do pay-as-you-go social insurance systems become. When n + g approaches r, pay-as-you-go systems appear to be very cheap and effective ways of increasing social welfare by passing resources down from the rich and numerous future to the poorer and less numerous present. By contrast, the larger is r relative to n + g, the greater are the benefits of prefunding social insurance systems. Prefunded systems can use high rates of return and compound interest to reduce the wedge between productivity and after-contribution real wages. They thus sacrifice the possibility of raising social welfare by moving wealth from the richer distant future to the near future and the present, but in return they gain by reducing the social insurance tax rate and thus its deadweight loss. And whenever we make utilitarian arguments other than those of pure Pareto-preference for why one set of policies is superior to another set, we are all, in our hearts, secret Benthamites. Thus, to the extent that the political debate over the future of social insurance in America is conducted in the language of rational policy analysis, getting the gap between r on the one hand and n + g on the other hand right is important. Policies predicated on a false belief that r is much larger relative to n + g than it is will unduly burden today’s and tomorrow’s young people and will leave many disappointed when returns on
296
Brookings Papers on Economic Activity, 1:2005
assets turn out to be less than anticipated and prefunding leaves large unexpected holes in retirement financing. Policies predicated on the belief that n + g is higher relative to r than it is pass up opportunities to lighten the overall tax burden and still provide near-equivalent income security benefits in the long run.
Algebra Let us begin by distinguishing a number of different rates of return. In this paper we use r to stand for a physical gross marginal product of capital, and we assume that it is the product of a Cobb-Douglas production function: (1)
r =α
Y . K
We distinguish this physical capital rate of gross profit r from the net rate of return on a balanced financial portfolio rf and from the net rate of return on equities re. Only under the assumptions of constant depreciation rates δ, constant financial markups, and a constant price and amount of risk is the mapping among these three straightforward. Toward the end of this paper we briefly consider the equity premium, but otherwise we assume that depreciation rates, financial markups, and other factors that could vary the wedges between r, rf , and re are unimportant. Thus we will move back and forth between these three different rates of return: things that raise or lower the return on stocks will also raise or lower the return on bonds and (after the capital stock has adjusted) the physical marginal product of capital as well. Robert Solow studied a constant-returns Cobb-Douglas production function with α as the returns-to-capital parameter, and with Y, K, L, and E as aggregate output, the capital stock, the supply of labor, and the level of labor-augmenting technology, respectively:11 (2)
11. Solow (1956).
Y = K α ( EL )1− α .
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
297
Assume constant rates of labor force growth n, of labor-augmenting technical change g, of depreciation δ, and of gross saving s. In the closedeconomy case, in which all of domestic capital K is owned by domestic residents and in which all of national saving goes into increasing the domestic capital stock, we know that, along a steady-state growth path of the economy, (3)
K s = . Y n+g+δ
This tells us that, along such a growth path, (4)
n + g + δ r = α . s
If permanent shocks that reduce n + g cause the economy to transit from one steady-state growth path to another, the rate of return on capital falls, with the change in r being (5)
∆r = ( α s ) ( ∆n + ∆g ).
As long as α is greater than or equal to s—that is, as long as the economy is not dynamically inefficient12—the reduction in r will be greater than one for one. From this algebra we would expect the roughly 1.5-percentagepoint reduction in the rate of real GDP growth forecast by the Social Security Administration to carry with it a greater than 1.5-percentage-point reduction in r. These are steady-state results. How relevant are they for, say, the seventyfive-year standard forecast horizon used in analyses of the Social Security system? In the Solow model the capital-output ratio approaches its steadystate value at an exponential rate of −(1 − α)(n + g + δ), which, at historical values, is roughly 3.6 percent a year. That closes half the gap to the steady-state capital-output ratio in twenty years. After seventy-five years the capital-output ratio has closed 93 percent of the gap between its initial and its steady-state value. In this simple Solow setup, only three things can operate to prevent a permanent downward shock to n + g from reducing r. Perhaps the depreciation 12. We have every reason to believe that the economy is dynamically efficient, in that capital in the steady state exceeds the “golden rule” level. See Abel and others (1989).
298
Brookings Papers on Economic Activity, 1:2005
rate δ could fall. We have been unable to think of a coherent reason why a reduction in labor force growth n or labor productivity growth g should independently carry with it a reduction in δ. (However, the reduction in r could plausibly carry with it an extension of the economic lives of equipment and buildings, and so bring about a partly offsetting fall in δ that would moderate the decline in r.) Or perhaps the production function could shift to increase the capital share of income α. Last, perhaps a permanent downward shock to n + g could also bring about a reduction in the saving rate s. If it were the case that (6)
s ds = − ( dn + dg ) , n + g + δ
then the rate of return r would be constant. There is a reason to think that a fall in n would carry with it a reduction in s: an economy with slower labor force growth is an aging economy with relatively fewer young people and, presumably, if the young do the bulk of the saving, a lower saving rate. (A decline in g, however, would tend to work the other way: the income effect would tend to raise s.)
Analysis Are the effects just discussed plausibly large enough to keep the rate of return on capital constant at the rate of economic growth? To assess that, we need to model saving decisions, which requires moving from algebra to model-based analysis. The Ramsey Model We move now from Solow to Ramsey-Cass-Koopmans.13 Consider a version of this Ramsey model in which the representative household has the following utility function:
∑ (1 + β ) ∞
(7)
t=0
13. See Romer (2000).
−t
(U (C ))N t
1− λ t
,
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
299
where β is the pure rate of time preference, Ct is consumption per household member, and Nt is the number of members of the representative household, growing according to N t +1 = (1 + n ) N t ,
(8)
where n now measures growth in the size of the household. In the standard Ramsey model setup as presented by David Romer,14 the parameter λ equals zero, so that the household utility function becomes
∑ (1 + β ) ∞
(9)
t=0
−t
(U (C ))N . t
t
This choice drives the result that changes in labor force growth do not have long-run effects on steady-state capital-output ratios or rates of return. But, to us at least, this assumption seems artificial. If it is indeed the case that the utility function is that specified in equation 9, then the more members of the household, the merrier: household utility is linear in the number of people in the household but suffers diminishing returns in consumption per capita. A household with this utility function, provided it has control over its own fertility, would choose to grow as rapidly as possible; that would be the way to make individual units of consumption contribute as much as possible to total household utility. It seems reasonable to allow λ to be greater than zero and so have a utility function with diminishing returns both with respect to household consumption per capita and with respect to household size. There is yet another reason to be uncomfortable with the assumption that λ = 0. If the term “golden rule” were not already taken in the growth theory literature, we would use it here, for λ = 0 requires that those household members making decisions in period t love others (the new household members joining in period t + 1) as they love themselves. They assemble the household utility function by treating the personal utility that others receive in the future from their consumption per capita as the equivalent of their own personal utility. Since we cannot call this the “golden rule,” we instead call it perfect familial altruism. If 1 > λ > 0, there is imperfect familial altruism: those making decisions in period t care about the personal utility of extra family members in period t + 1, but not as much as they care about their own. 14. Romer (2000).
300
Brookings Papers on Economic Activity, 1:2005
And if λ = 1, decisionmakers in period t act as if they care only about their own personal utility. We are comfortable with altruism; we are uncomfortable with perfect familial altruism. To the extent that changes in population growth are due to changes in rates of international migration, the assumption that λ = 0 is not defensible. The representative agent in period t would then regard the future-period utility of unrelated strangers of different nationality who migrate into the country on an equal footing as her own utility, or the utility of her direct descendants.15 In this version of the Ramsey-Cass-Koopmans model, the first-order condition for the representative household’s consumption-saving decision is (10)
U ´(Ct ) dCt =
(1 + n )
1− λ
(1 + β )
U ´(Ct +1 ) dCt +1 .
If the household faces a net rate of return on financial investments of rf, then (11)
1 + rf dCt = dCt +1 , 1+ n
because resources in period t + 1 must be split among more members of the expanded household. For log utility we then have −λ
(12)
)
(1 + n ) (1 + rf Ct +1 = . Ct (1 + β )
Along the economy’s steady-state growth path, with consumption per worker growing at the rate of labor augmentation g, this becomes (13)
rf = (1 + g ) (1 + n ) (1 + β ) − 1, λ
and in the continuous-time limit, (14)
rf = β + g + λn.
15. Approximately 0.3 percentage point a year of the slowdown in labor force growth projected by the Social Security trustees’ report (Board of Trustees, 2005) is due to a slowdown in immigration.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
301
Looking across steady-state growth paths, one sees that reductions in the rate of output growth per worker g reduce rf one for one in the case of log utility. (They reduce rf by a multiplicative factor γ of the change in g in the case of constant-relative-risk-aversion utility: U(Ct) = [(Ct)1 − γ]/[1 − γ].) Reductions in the rate of labor force growth n also reduce rf except in the case of λ = 0. If 1 > λ > 0, slower rates of labor force growth reduce rf, but less than one for one. And if λ = 1, decisionmakers in period t are not altruistic at all: they act as if they care only about their own personal utility, and reductions in n reduce rf one for one—the same amount as do reductions in g. The Ramsey model converges to a balanced-growth path, and this plus the assumption of a representative agent is sufficient to nail down the relationship between economic growth and asset returns. In the steady state, consumption per capita is growing at rate g, and so the relative marginal utility of consumption per capita one period into the future is
(1 + β ) (1 + g ) −1
(15)
−1
in the case of log utility. And the rate at which consumption per capita can be carried forward in time is
(1 + r )(1 + n )
(16)
f
−1
.
To drive the rate of return on capital rf away from (17)
rf = (1 + g ) (1 + n ) (1 + β ) − 1 λ
requires that the consumption of those agents who are marginal in making the consumption-saving decision in period t grow at a rate different from that of growth in consumption per capita. This requires heterogeneous agents. And the simplest suitable model with heterogeneous agents is the Diamond model. The Diamond Model In the overlapping-generations model of Peter Diamond,16 each agent lives for two periods, works and saves when young, and earns returns on
16. Diamond (1965).
302
Brookings Papers on Economic Activity, 1:2005
capital and spends when old. Thus, for a given generation that is young in period t, their labor income per worker when young wt, their consumption per worker when young cyt, their consumption per worker when old cot+1, the net rate of return on capital rt+1, and the economy’s capital stock per worker in the second period kt+1 are all linked: (18)
wt = cyt + kt +1
(19)
cot +1 = (1 + rt +1 ) kt +1.
With a Cobb-Douglas production function, output per (young) worker when the period-t generation are young—in period t—is α
(20)
yt = E
1− α t
kt , 1 + n
where E is our measure of the efficiency of labor, growing at proportional rate g each period, and where (1 + n) appears in the denominator because n is the rate of population growth per generation. With this production function, labor income is a constant fraction of output per worker, (21)
wt = (1 − α ) yt ,
and the real return on capital will be the residual, capital income, divided by the capital stock: (22)
rt =
αyt αEt1− α ktα −1 = . kt (1 + n )α −1
Once again take time-separable log utility for our utility function, (23)
)
ln ( cyt +
ln ( cot +1 ) 1+β
and look for steady states in capital per effective worker by requiring that (24)
kt = Et k * .
From this we get the following steady-state first-order condition:
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
303
(1 + r ) 1 1 = . cyt (1 + β ) cot +1
(25)
The model can be solved by substituting in the budget constraint, (26)
1 kt − kt + 1 (1 − α ) Et1− α 1 + n a
=
(1 + r ) 1 , ( (1 + β ) 1 + r ) kt +1
to get (27)
1 1 − α k* − k * 1 + g 1 + n α
=
1 , (1 + β ) k *
which leads to 1
(28)
(1 − α ) 1− α k* = . (1 + g ) (1 + n )α ( 2 + β )
Recalling that r = (αk*α−1)/(1 + n)α−1, we have (29)
α(1 + g)(1 + n)(2 + β) r = . (1 − α )
In this equation, the lower the rate of productivity growth g, and the lower the rate of labor force and population growth n, the lower is the rate of return on capital r. Conclusion Thus, in the Diamond overlapping-generations model as well as in the Ramsey model and the Solow model, slower economic growth comes with lower net returns on capital. In the Ramsey model, there is reason to think that reductions in labor productivity growth have a greater effect on rates of return than do reductions in labor force growth: —In the basic Solow algebra, the reduction in gross returns r is proportional to (α/s) times the reduction in growth.
304
Brookings Papers on Economic Activity, 1:2005
—In the Diamond model, the reduction in net returns rf is equal to 2α/(1 − α) times the reduction in labor productivity growth g and, to a first approximation, equal to 2α/(1 − α) times the reduction in labor force growth n. —In the Ramsey model, the reduction in rf is equal (with log utility) to the reduction in labor productivity growth g and, to a first order, to λ times the reduction in labor force growth n (where λ is the degree to which familial altruism is imperfect). At some level, the same thing is going on in all three setups. Reductions in economic growth in these setups are all declines in the rate of growth of effective labor relative to the capital stock provided by previous investment. Effective labor becomes relatively scarcer and capital relatively more abundant. The terms of trade move against capital, and so the return to capital falls. Why, then, does a fall in labor force growth not reduce rates of return in the Ramsey model in the case of perfect familial altruism, λ = 0? Because a reduction in population growth also reduces the utility value of moving consumption forward in time—an important component of the value of saving in the Ramsey model with perfect familial altruism comes from the possibility of dividing the saving among more people in the future and thus escaping the diminishing marginal utility of consumption. Thus the marginal household utility of saving falls in the Ramsey model when population growth falls. This fall reduces the effective supply of capital by as much as the fall in the rate of population growth reduces the effective supply of labor. To the extent that a slowdown in economic growth is driven by a reduction in the rate of immigration, this representative-agent effect in the Ramsey model is not an effect that we want the model to have: perfect familial altruism is not an assumption that anyone would wish to make. These models say that there is some economic reason to believe that a slowdown in economic growth would carry a reduction in asset returns with it. These models are the standard models that economics graduate students and their professors use routinely. They are oversimplified. They are abstract. They are ruthlessly narrow in their conceptions of human motivation and institutional detail. Are they relevant to the real world? Are they telling us something that we should hear when we try to forecast the long-run future?
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
305
Arithmetic Is it possible to imagine scenarios in which asset returns remain close to their historical averages even when real GDP growth slows markedly? Yes. Are any such scenarios plausible forecasts in the sense of being the central tendency of a distribution of possible futures? We believe not. In this section we conduct some simple arithmetic exercises to make our case. Earnings and Returns Jeremy Siegel believes that stocks are “in the middle range of fair market value” and that therefore the current earnings yield of 5.45 percent is a “good long-term estimate of real returns.”17 The sum of dividend payouts, net buybacks, and investment financed by net retained earnings must add up to 5.45 percent of today’s stock values.18 Returns to investors are payouts— dividends and net buybacks—plus the value of investments financed by net retained earnings. Firms, which have traditionally paid out, on average, roughly 60 percent of their accounting profits through dividends and buybacks and rely on retained earnings to finance a substantial share of any increase in their capital stock, have little room to boost risk-adjusted returns by massively expanding payouts, unless they can do so without crippling their earnings growth—that is, unless a good deal of today’s retained earnings are wasted. Firms similarly have little room to boost risk-adjusted rates of return on their equity by cutting back on payouts, unless there are very large wedges between rates of return on retained and reinvested earnings and rates of return in the market—that is, unless firms have been massively underinvesting. Current earnings yields thus suggest that the stock market is in accord with the logic of our algebra and analysis: it is not anticipating the average real return on the stock market of 6.5 percent a year or so realized over the past half century. But reported accounting earnings are not true Haig-Simons earnings (that is, equal to the amount that can be consumed from earnings without 17. Siegel (2005, p. 8). 18. There is a wedge of 0.3 percentage point a year between the GDP deflator and the CPI. Siegel’s estimated real rate of return becomes 5.15 percent a year in the CPI-basis numbers used by the Social Security Administration.
306
Brookings Papers on Economic Activity, 1:2005
changing wealth).19 There is good reason to believe that returns on retained earnings are higher than market returns.20 And it is at least plausible that the wedge between market returns and returns on retained earnings depends on the rate of economic growth: faster growth means higher demand and greater profits if returns to scale are increasing. So the argument that earnings yields do not support high expected equity returns needs to be shored up by an explicit look ahead at how payouts and values might evolve.21 Dividend Yields, Returns, and Growth Begin with the identity that is the Gordon equation for equity prices: (30)
P=
D , re − g
where D are the dividends paid on a stock or an index of stocks, P is the corresponding price, re is the expected real rate of return on equities, and g is the expected permanent real growth rate of dividends. This is a standard way to approach the determinants of equity prices as a whole.22 In this framework the real rate of return on equities is (31)
re =
D + g. P
Returns on an index of stocks differ from the current dividend yield plus the growth rate of economy-wide corporate earnings for two important reasons: —First, g will be less than the growth rate of economy-wide corporate earnings because those earnings are the earnings of newly created companies that were not in the index last period. Corporate earnings are a return to entrepreneurship as well as capital; hence the rate of growth of economy-wide earnings will in general outstrip that of the earnings of the companies represented in a stock index. —Second, dividends are not the only way firms pump cash to shareholders. Stock buybacks decrease the equity base and thus push up the 19. 20. 21. 22.
Haig (1921). See Hubbard (1998). See Baker (1997) for the first argument along these lines of which we are aware. Campbell and Shiller (1988).
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
307
rate of growth of the earnings on the index (as opposed to the earnings of the companies in the index). It is convenient to think of both of these factors as affecting the payout ratio rather than the growth rate, and to replace equation 31 with (32)
re =
D+B + g, P
where B is net share buybacks (buybacks less initial public offerings), and g is now the growth rate of D + B.23 The 2005 report of the Social Security trustees projects a long-run real GDP growth rate of 1.8 percent a year on a GDP deflator basis.24 It projects that labor and capital shares will remain constant in the long run.25 With a long-run gap of 0.3 percentage point between the consumer price index (CPI) and the GDP deflator,26 and with an auxiliary assumption that capital structures are in balance, this is an implicit forecast that the variable g in the Gordon equation will be 1.5 percent a year. Current dividend yields on the Standard and Poor’s (S&P) 500 index are 1.9 percent a year. Current net stock buybacks are 1.0 percent a year. The sum of these is 4.4 percent a year, which is thus the expected real rate of return r in the Gordon equation. That is significantly lower than the 6.5 percent real rate of return that is the historical experience of the American stock market. Possible Ways Out Are there ways to escape from this arithmetic of earnings and payouts? Yes. The U.S. economy is not on a steady-state growth path. Three potential ways out seem most worth exploring: —Perhaps the stock market is currently overvalued and will decline, significantly raising payout yields.
23. Subtracting initial public offerings ensures that the ratio of total economy-wide earnings to the earnings of companies in the index does not grow. Adding gross buybacks takes account of the antidilution effects of narrowing the equity base of companies currently in the index. 24. Board of Trustees (2005, table V.B2). 25. The assumption of a constant income share follows from the derivation of real wage growth from productivity growth, which is discussed on pages 85–88 of Board of Trustees (2005). 26. Board of Trustees (2005, table V.B1).
308
Brookings Papers on Economic Activity, 1:2005
—Perhaps payout growth will be unusually rapid in the near term before slowing to its long-term forecast trend rate of 1.5 percent a year. —Perhaps the distribution of world investment will shift in a way that allows U.S. companies to earn greater and greater shares of their profits abroad. Diamond argues for the first possibility.27 A decline in the stock market, relative to the economy’s growth trend, of 40 percent would carry payout yields up to the 5.0 percent consistent with a long-run real return of 6.5 percent a year and real profit and dividend growth (on a CPI basis) of 1.5 percent a year. Such a scenario is certainly possible: it was the stock market’s experience between the late 1960s and the early 1980s. But we have a hard time seeing it as the central tendency of the distribution of possible futures.28 The second possibility requires payouts—both dividends and net stock buybacks—to grow rapidly in the near term to validate a subsequent real growth rate of 1.5 percent a year and a current expected real return of 6.5 percent a year. If such growth were to be concentrated in the next decade, the real payouts of the companies in the S&P index would have to grow at an average of 8.6 percent a year. Over the past fifty years the earnings on the S&P index have grown at an average rate of 2.1 percent a year. It could happen: perhaps we are in the middle of a permanent shift in the distribution of income away from labor and toward capital. But, once again, we regard these as unlikely scenarios, not as the central tendency of the distribution of possible futures that is a rational forecast. The third way out is the one that we regard as the most interesting possibility. We take it up in the next section.
The Open-Economy Case In any open economy the steady-state Gordon equity valuation equation is as before, except that the rate of growth is not that of the domestic cor-
27. Diamond (2000). 28. Certainly no investment adviser who anticipates that real equity returns will average −0.6 percent a year over the next decade has any business advising clients to shift their portfolio in the direction of equities today. That is true even when the U.S. government is the adviser, and the relatively young future beneficiaries of Social Security are the clients.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
309
porate capital stock g but that of the capital stock owned by American companies, gk: (33)
re =
D+B + gk . P
If foreign companies, on net, invest in America—that is, if the United States on average runs a current account deficit—then the rate of growth of the earnings of American companies in our domestic stock market index will be slower than the rate of growth of economy-wide earnings and of real GDP. The open economy will then deepen rather than resolve the problem of combining slow expected growth with high expected returns. If instead it is American companies that, on net, invest abroad, then the rate of growth of the capital stock, and thus of the earnings of companies in the index, will exceed the rate of growth of the domestic economy g. How much larger? If we look over spans of time long enough for adjustment costs in investment not to be a major factor, the value of the capital stock will be proportional to the size of the capital stock.29 If we assume in addition that companies maintain stable debt-equity ratios, we have (34)
Y gk = g + x , K
where x is that component of the current account surplus (as a share of GDP) that corresponds to American companies’ net investments abroad,30 and Y/K is the ratio of current output to corporate capital.
29. We here dismiss the possibility that investments overseas might provide higher risk-adjusted rates of return in the long run than domestic investments: Tobin’s q = 1 both here and abroad. The Bureau of Economic Analysis reports that as of the end of 2003 the market value of foreign-owned assets in the United States is about $10.5 trillion, compared with foreign assets held by U.S. residents of about $7.9 trillion, yet the associated income flows are about the same. We attribute this difference to a difference in risk. The experience of nineteenth-century British investors with such landmarks of effective corporate governance as the Erie Railroad suggests that, although there are supernormal returns to be earned in the course of rapid economic development, people with offices separated by oceans are unlikely to be the ones who reap them. 30. The phrase “corresponds to American companies’ net investment abroad” is needed to abstract from current account deficits that finance net government consumption or net household consumption.
310
Brookings Papers on Economic Activity, 1:2005
Here, again, we return to arithmetic. Our rate of return on equities is (35)
re =
Y D+B + g + x . K P
From the previous section this is (36)
Y re = 4.4% + x . K
Assuming a capital-output ratio of 3, we then have (37)
x = 3( re − 4.4 percent ).
In words: for any excess of the rate of return on equities over the closedeconomy benchmark case of 4.4 percent a year, three times that figure is the current account surplus associated with net corporate investment overseas needed to produce the higher return. Note that, for a constant rate of return, the needed surplus grows over time. In equations 34 through 36, Y/K is not the physical domestic outputto-capital ratio; it is the ratio of domestic output to total capital owned by American companies—including capital overseas. As overseas assets mount, the needed surplus for constant payout yields mounts as well. Such enormous current account surpluses are possible. Great Britain had them in the quarter-century before World War I, when it ceased to be the workshop of the world and became for a little while its financier.31 Slowing economic growth in the late Victorian and Edwardian eras and reduced investment relative to national saving were cause (or consequence, or possibly both) of the direction of Britons’ saving and of British companies’ investment overseas. We see no signs that the United States will undertake a similar trajectory over the next several generations. And we are impressed by the scales involved: to be consistent with current payout yields, and given a forecast real GDP growth rate of 1.8 percent a year, to achieve 6.5 percent annual returns on equity the current account surplus produced by American net corporate investment abroad would have to begin at 6 percent of GDP and grow thereafter.
31. Edelstein (1982).
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
311
Could such large outward levels of net corporate investment abroad be consistent with relatively balanced overall trade—in other words, could they be offset by large net portfolio investment inside the United States? Not without additional forces at work. The reason is that the open-economy saving-investment relation, (38)
S − NX ≡ I ,
(where NX is net exports) is an identity. Consider the three uses that such large inward portfolio investments could have: —They could be used to purchase securities newly issued by American businesses to finance investment in the United States. The flow of inward portfolio investment would add as much to domestic investment as the outward-directed flow of corporate investment would have subtracted. There would be no slowdown in the rate of growth of the domestic capital stock. Thus the rising domestic capital-output ratio would push down rates of return at home. Since foreigners are making these large portfolio investments in the United States, this fall in domestic rates of return would be associated with a similar fall in foreign rates of return as well. —They could be used to purchase securities newly issued by American businesses to finance investment abroad. In this case, gross foreign direct investment by domestic firms would have to be large enough not only to absorb the difference between domestic investment and domestic saving, and so slow down the rate of growth of the domestic capital stock, but also to neutralize the portfolio capital inflow. We are thus back to square one. —They could be used to purchase already-existing assets from Americans, who then do not reinvest the proceeds either in expanding the domestic capital stock or in further funding American investment abroad, but instead consume the proceeds.32 This means massive dissaving on the part of those who sell their assets to foreigners: a large fall in S. Once again, we see a possible scenario but not the central tendency of the distribution of possible futures that would constitute a forecast.
32. This is the possibility that Mankiw stresses in his comment on this paper: that if domestic saving rates fall sharply, the reduction in the rate of growth of the domestic capital stock required to keep rates of return high can be accomplished without a large current account surplus.
312
Brookings Papers on Economic Activity, 1:2005
The Equity Premium Economists do not have a good explanation of the equity premium. Rajnish Mehra and Edward Prescott titled their well-known paper “The Equity Premium: A Puzzle,” for good reason.33 Stocks have outperformed fixed-income assets by more than 5 percentage points a year for as far back as records go. As Martin Feldstein, former chairman of the Council of Economic Advisers, has often said, it is as if the market’s attitude toward systematic equity risk were that of a rich sixty-five-year-old male with a not-very-healthy lifestyle, whose doctor has told him that he is likely to live less than a decade. Yet we believe that properly structured markets should— and can—mobilize a much deeper set of risk-bearers with a much greater risk tolerance. That they do not appear to have done so is a significant mystery. We find ourselves persuaded by Mehra that the equity premium remains a puzzle, unexplained by rational agents in models that maximize individual utility.34 It is quite possible that a substantial part of the equity premium is a thing of the past, not the future.35 In the distant past the fear of a recurrence of the railroad and other “robber baron” scandals, and in the more recent past the memory of the Great Depression, kept some investors excessively averse to stocks. In addition, the United States has had remarkably good economic luck—a point stressed by Robert Shiller.36 And, over time, as people realized that their predecessors had been excessively fearful of equity risk, rising price-dividend ratios pushed a further wedge between stock and bond returns. But today our arithmetic projects stock returns of 4.4 percent a year, for an equity premium of perhaps 2.5 percentage points, not 5. To the extent to which this past behavioral anomaly was the result of an excessive fear of stocks and an excessive attachment to bonds, it is not clear that its erosion should have an impact on the expected return on a balanced portfolio. The simplest, crudest, and most extremely ad hoc model of the equity premium would embed the stock-versus-bond investment decision in the simplest possible Diamond-like overlapping-generations 33. Mehra and Prescott (1985). 34. Mehra (2003). 35. In conversation, Randall Cohen of the Harvard Business School has been an especially forceful advocate of this point of view. 36. Shiller (2005).
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
313
model, with the capital stock each period being the wealth accumulated when young by the old, retired generation. Assume that each generation, when it saves, invests a share eh of its savings in equities and a share 1 − eh in bonds. Firms, however, are unhappy with such a capital structure. Unwilling to run a significant risk of bankruptcy, they are unwilling to commit less than a share ef, where ef > eh, of their payouts to equity. A smaller cushion—in the sense that a smaller cyclical decline in relative profits would run the risk of missing bond payments and drawing an appointment with a bankruptcy court—is simply unacceptable to entrenched managers. If a physical unit of saving when one is young yields returns to physical capital r when one is old, the rates of return on equity and debt, re and rd, respectively, are then calculated as (39)
e 1 + re = (1 + r ) f eh
(40)
1 − ef 1 + rd = (1 + r ) , 1 − eh
with the equity premium being (41)
e (1 − e f ) 1 + re . = f eh (1 − eh ) 1 + rd
In this excessively simple framework, it does seem highly plausible that (1 + re)/(1 + rd) has fallen with greater household willingness to hold equity, because of institutional changes (such as revisions of the “prudent man” rule, the growth of IRAs and 401(k)s, and lower transactions costs associated with stock trading), the fading memory of 1929, two decades of fabulous bull markets, and increased financial sophistication on the part of households. Thus, even if there were no reasons connected with slowing growth to expect lower returns on capital, one might well anticipate lower returns on equity in the future than in the past. And past decades have seen institutional changes that one would expect, from a behavioral perspective, to boost the share of financial assets channeled to equities.37 37. Barberis and Thaler (2003).
314
Brookings Papers on Economic Activity, 1:2005
A lower rate of return on the assets in a balanced portfolio has powerful implications for economic policy. A lower equity premium seems, to us at least, to have powerful implications for one issue, namely, whether the stock market’s apparent failure to mobilize society’s risk-bearing capacity is a large-scale market failure, and whether a government-run social insurance scheme can and should attempt to profit from (and thus repair) this failure to mobilize society’s risk-bearing resources. The government has the greatest ability of any agent in the economy to manage systematic risk. If other agents are not picking up their share—and if, as a result, there are properly adjusted excess returns to be earned by the government’s taking a direct position itself or assuming an indirect position by reinsuring individuals’ social insurance accounts—why should the government not do so? The difference between, broadly speaking, the economists of the coasts and the economists of the interior is that the first specialize in thinking up clever schemes to repair apparent market failures, whereas the second specialize in thinking up clever reasons why apparent market failures are not really so. Even though we are from the coasts, we find enough reasons to believe that the equity premium will be smaller in the future than in the past to prefer that attempts to exploit it be implemented slowly and gingerly.
Conclusion We see strong reasons to think that, over the long run, rates of return on assets are correlated and causally connected with rates of economic growth. We would expect the reduction in asset returns to be greater for a given reduction in productivity growth than for an equal reduction in labor force growth. We think that reductions in asset returns could be offset and even neutralized by other factors—by capital expropriating some of what has been labor’s share of income, by a failure of today’s stock market values to soberly reflect likely future returns rather than irrational exuberance, or by the United States cutting its consumption beneath its production for generations and following Britain’s pre–World War I trajectory as supplier of capital to the world. But we see these as unlikely (although possible) scenarios. We do not see any of them as the central tendency of the distribution of possible futures that is a proper economic forecast. And although a combination of partial moves in each of the three directions could
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
315
achieve the result, we see no good reason to presume that such a scenario is likely. We see the two strands of our argument—our arithmetic demonstration that equity returns as high in the future as in the past are unlikely, and our analytical arguments that rates of return and rates of growth are likely to move together—as reinforcing each other. Returns must be consistent with the saving decisions of households, the investment decisions of firms, and the technologies of production. But returns must also equal payout yields plus capital gains—only in stock market bubbles can capital gains diverge widely from economic growth, and then only for a little while. Powerful economic forces work to make sure that what the economy’s behavioral relationships produce is consistent with its equilibrium flow-offunds conditions. That is the logic that applies here: if slower economic growth reduces the arena for the profitable deployment of capital, rates of return will fall until less capital is deployed. By how much will they fall? Until—in steady state—payout yields plus retained earnings are equal to profits, and retained earnings are no larger than the sustainable growth of the capital stock permits.
Comments and Discussion N. Gregory Mankiw: This paper by Baker, DeLong, and Krugman is really three papers in one. The first paper is a straightforward review of how population growth affects the return to capital in standard models of economic growth. The second paper is a discussion of what return one should expect for the stock market in the coming decades, given current measures of valuation. The third paper offers some ruminations about the equity premium. What links the three papers is their motivation. President Bush has called for reform of the Social Security system. According to the Social Security actuaries, the system faces large unfunded liabilities. That conclusion, however, is based on a projection that includes much slower labor force growth (and thus economic growth) than the United States has experienced historically. This raises the question of what rate-of-return projections should be assumed as the nation considers possible reforms. When evaluating reform proposals, the Social Security Administration uses a projected real annual return on equities of 6.5 percent (which, given the trustees’ assumption about the risk-free rate, implies an equity premium of 3.5 percent). Paul Krugman has written elsewhere that “a rate of return that high is mathematically impossible unless the economy grows much faster than anyone is now expecting.”1 This three-in-one paper began as an attempt to justify that assertion. I will discuss each of the three papers in turn, before addressing the policy motivation. POPULATION AND GROWTH THEORY. The first paper in this paper reviews several standard neoclassical growth models. The aim is to see what these models predict for the relationship between population growth and the rate of return to capital.
1. “Many Unhappy Returns,” New York Times, February 1, 2005.
316
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
317
The Solow growth model gives a clear answer to this question: slower population growth lowers the rate of return. Because the saving rate is fixed, slower population growth raises the steady-state capital-labor ratio, which in turn means a lower marginal product of capital. The Diamond model gives a similar answer, at least for the functional forms assumed here. The Ramsey model, however, leads to a very different conclusion. In that model the saving rate adjusts so that the rate of return is invariant to the population growth rate. This adjustment of the saving rate is economically intuitive: if there are going to be fewer people in the future, we need to save less for the future. This conclusion is the essence of the analysis presented in a 1990 Brookings Paper called “An Aging Society: Opportunity or Challenge?” written by David Cutler, James Poterba, Louise Sheiner, and Lawrence Summers.2 They used a standard Ramsey model to argue that, “the optimal policy response to recent and anticipated demographic changes is almost certainly a reduction rather than an increase in the national saving rate.” I should note that national saving is currently low by historical standards, but I will not suggest that this is necessarily the “optimal policy response” that Cutler and his coauthors were proposing. Realizing that the Ramsey model does not support the main contention of the paper, Baker, DeLong, and Krugman propose a new but unpersuasive generalization of it. The authors claim that the standard Ramsey model is one of “perfect familial altruism.” That is not how I would describe it. Even the standard Ramsey model includes discounting, so that my utility is weighted more heavily than that of my children and grandchildren. What the proposed generalization does is make the effective discount rate for future utility depend on the population growth rate. When population growth slows, the effective discount rate falls, and this fall in the discount rate blunts the decline in the saving rate that occurs in the standard Ramsey model. Is this generalization appealing? Not to me. As the parent of three children, I can attest that one of the things parents do when child N is born is to assure the N − 1 children that they will be loved just as much. The generalization of the Ramsey model proposed here is, in essence, a denial of this claim. In the end it is clear that the tools of modern growth theory lead to an ambiguous answer about how population growth affects the return to capital. 2. Cutler and others (1990).
318
Brookings Papers on Economic Activity, 1:2005
One can write down textbook models in which the two variables move together (the Solow model), and one can write down models in which they do not (the Ramsey model). The natural response to this theoretical ambiguity is to muster evidence, either from time-series data or from the international cross section, about the actual effect of population growth. This paper, however, presents no evidence one way or the other. Perhaps that is a subject for a future Brookings Paper. STOCK MARKET VALUATION. The second paper in this paper discusses the expected return on the stock market. The authors begin with the observation that the current average earnings yield is 5.23 percent a year, which is about a percentage point lower than the historical average. As a result, they expect future stock returns to be lower than historical averages as well. I give some weight to this piece of evidence. It is possible that the Social Security Administration’s assumption of 6.5 percent a year for equity returns is about a percentage point too high. The risk-free rate assumption of 3 percent a year may also be about a percentage point too high, as judged by current yields on long-term inflation-indexed bonds. The equity premium of 3.5 percent, however, seems about right. After observing the earnings yield, the authors consider stock market valuation from the perspective of the famous Gordon formula, according to which the expected return on a share of stock equals the current dividend yield plus the projected growth rate of dividends per share. Although the Gordon formula has a long and venerable tradition, I don’t think it provides a particularly edifying approach here. For a neoclassical economist, the starting point for thinking about the role of dividends in stock valuation is the classic Modigliani-Miller theorems, which tell us that the dividend payout is irrelevant to the value of the firm. It seems unnatural for purposes of stock valuation to focus on the level and growth of a variable that, to a first approximation, does not matter. If the dividend yield is approximately irrelevant, as Modigliani and Miller tell us, then it is easy to imagine that it could undergo a major change in the years to come. Looking ahead, it seems plausible to me that dividend payouts broadly construed could rise significantly. If we are about to experience a period of slower economic growth because of demographic change, then firms might well have fewer profitable investment opportunities and, as a result, might decide to pay out a larger percentage of their earnings. There are several ways this could occur. One possibility is by increasing normal dividends or share repurchases. Another is through corporate reorganizations.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
319
Corporate managers might find cash takeovers and acquisitions more profitable than internal expansion. Cash purchases of other businesses take money out the corporate sector and are, in essence, a form of share repurchases. They are another way to increase dividends, broadly construed. OPEN-ECONOMY STOCK VALUATION. The authors then consider a related open-economy issue. Is it possible, they ask, for growth in dividends to significantly exceed growth in the domestic economy because corporations are investing and earning profits abroad? They suggest that this is unlikely, on the apparent ground that it would require something implausible about capital flows. I am not convinced. Suppose that General Electric, seeing fewer profitable investments in the United States, uses some of its earnings to buy a factory in China. That represents a capital outflow from the United States and a current account surplus, which I think is what the authors have in mind. But consider what the Chinese former owners of the factory, who now have dollars from the deal, might do with them. One possibility is that they buy U.S.-produced goods, which would indeed mean a current account surplus for the United States. Another possibility is that they buy U.S. assets. They might even buy stock in General Electric or be given GE stock as part of the transaction. In this case General Electric can diversify abroad while the United States has balanced trade. Here is one scenario that seems plausible to me. With much of the rest of the world, such as China and India, growing so rapidly, U.S. companies will increasingly find profitable opportunities abroad. At the same time, foreigners will increasingly invest in U.S. companies, which will be among the driving forces behind global growth. Under this scenario an increasing share of the earnings of U.S. corporations could come from abroad, without any obvious implications for the U.S. current account. The authors conclude that this is a “possible scenario but not the central tendency,” but they do not cogently explain how they reach this conclusion. At least we can agree it would be a mistake to call this scenario “mathematically impossible.” THE EQUITY PREMIUM. Let me turn now to the last paper in this paper, which concerns the equity premium. Here the authors give us a model that is creative, bizarre, or vacuous, depending on your point of view. Most analysis of the equity premium begins with the premise that it has something to do with the trade-off between risk and return. Not so in this model. Here the household sector decides exogenously what fraction of
320
Brookings Papers on Economic Activity, 1:2005
wealth to put in equities, and the corporate sector decides exogenously what fraction of the capital return to pay out to equity holders. From these two exogenously determined shares, the equity premium emerges. The model reminds me of John Kenneth Galbraith’s worldview. Households are not sufficiently intelligent to make portfolio decisions based on risk and return. Corporate managers are sufficiently immune to market forces that they divide up the economic pie however they see fit. If I took this model seriously, it would do more than inform my view of the equity premium. It would shake my faith in corporate capitalism! But there is a less dramatic way to view this part of the paper. Perhaps the equations presented here should be viewed less as behavioral descriptions and more as accounting identities. If this interpretation is right, then I am at a loss about what purpose these equations serve. They do not seem readily adapted to calibration, to gauge how the equity premium has changed over time. I am comfortable with the authors’ suggestion that the equity premium may be smaller in the future than it has been in the past because institutional changes have made the spreading of risk more efficient. But this model does not shed much new light on this familiar conjecture. IMPLICATIONS FOR SOCIAL SECURITY REFORM. Let me close with a few words about Social Security reform. The authors were drawn to this set of topics because they think it is central to the debate over the president’s reform proposal. I disagree. There are two elements to the president’s proposal. First, the president wants to eliminate the system’s unfunded liabilities by bringing promised benefits into line with the dedicated payroll tax revenue. Various ideas for doing this have been put on the table, such as raising the retirement age and changing indexation rules. Second, the president wants to give workers the option of converting some of their defined-benefit retirement income from Social Security into a defined contribution, which would be placed in a personal account and invested in a broadly diversified portfolio of stocks and bonds. Reasonable people can disagree about the merit of these proposals. I made the case for the president’s proposals as I see it in a recent article in the New Republic.3 The case for reforming benefits is that the government should not promise more than it has the wherewithal to pay. The case for moving
3. “Personal Dispute,” The New Republic, March 21, 2005.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
321
Social Security from a defined-benefit to a defined-contribution structure is that it gives individuals more choice and control over their retirement income and the government greater transparency in its finances. These arguments are not based on any particular estimate of the average return to capital or of the equity premium. I don’t think the key issue in the debate over Social Security is whether, over the next century, the riskfree annual return will be 1 or 3 percent, or whether the equity premium will be 2 or 4 percent. So even if I agreed with the arguments raised in this paper and lowered my estimates of rates of return, it would not change my mind about the need to reform Social Security or the kinds of reforms that are desirable. I would guess that, in their hearts, the authors agree with me about this. To see if I am right, I would like them to ponder the following question: Suppose that, next week, the stock market falls by 50 percent, so that dividend and earnings yields double. Would Baker, DeLong, and Krugman suddenly change their minds and favor President Bush’s proposal for Social Security reform? I suspect they would not. If I am right, this suggests that although the paper raises some interesting questions about the future of asset returns, as far as the debate over Social Security goes, it is largely a non sequitur. William D. Nordhaus: What are the prospective returns on stocks? This is a question with multi-trillion-dollar stakes, and so it is always of much interest to many people. It has recently become a political as well as an economic question with the proposal by the Bush administration to introduce private (or personal) accounts as a component of Social Security pensions. This paper by Baker, DeLong, and Krugman argues that the returns assumed in the Bush administration’s calculations are too high, that historical returns do not provide a reliable benchmark for future returns, and that structural changes to the U.S. economy are likely to lower returns in the future. The last act of the political-economic drama is missing from this play, however, for the authors do not discuss in any detail the consequences of their findings for Social Security. My comments are primarily directed to the question of estimating the prospective long-run returns to equity. There are several approaches to this question. One is to examine historical returns. Estimates here are highly dependent on the time period. A second approach, which the paper uses, is based on the Gordon equation. This approach is difficult to apply because
322
Brookings Papers on Economic Activity, 1:2005
it relies on price-per-share data, ignoring the fact that the number of shares can change over time. To provide some perspective, I will look at the “fundamental” return on stocks by examining the rate of return on corporate capital, using different approximations. Before going down this road, two caveats should be mentioned. First, the returns data customarily examine the returns to nonfinancial domestic corporations. This component of profits is, unfortunately, a declining share of corporate profits as measured in the national accounts. After making up around 85 percent of corporate profits in the decade after World War II, the share of the domestic nonfinancial sector has declined to slightly over 50 percent in the last three years. The second shortcoming in using national accounts data is that there are major accounting differences between reported book profits and national accounts profits.1 Book profits contain several inappropriate items (such as gains on pension plans), whereas national accounts profits are more comparable over time but are limited to income earned on domestic production. Moreover, neither concept is conceptually equivalent to a measure of true income. My table 1 shows several measures of the returns to capital based on national accounting data on nonfinancial corporations. The first column presents a first approximation, the real rate of return after tax, measured as total property return divided by the replacement or market value of real assets. Total property return includes both interest and profits after corporation taxes in the numerator, with the current value of fixed capital, software, inventories, and land in the denominator. According to this first approximation, the return to capital averaged 6.1 percent a year over the 1960–2004 period. The return in 2004 was slightly above the long-term annual average, at 6.9 percent. A second approximation would take into account the cyclical nature of profits. I have taken a very simple approach, using as a cyclical variable the difference between the actual unemployment rate and the Congressional Budget Office’s estimate of the non-accelerating-inflation rate of unemployment (NAIRU). The cyclical term in the regression is significant and explains some of the peaks and troughs of the rate-of-return series but does not change the long-term picture. Most important for my purposes is that the latest year (2004) shows a cyclically corrected rate of return of 7.0 percent a year, also slightly above the long-term average (second column in table 1). 1. An excellent discussion of the differences is contained in Petrick (2001).
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
323
Table 1. Alternative Rate-of-Return Measures and Q Ratios, Various Periods, 1960–2004 Percent a year except where stated otherwise Domestic nonfinancial corporations Prospective tenyear return with Average earnings Not cyclically Cyclically Average reversion of Q yield on corrected corrected b Q ratioc toward 1a,d S&P 500e Average rate of profita
Period 1960–69 1970–79 1980–89 1990–99 2000–04
7.1 5.3 5.6 6.5 5.9
6.8 5.4 6.1 6.6 5.9
0.85 0.62 0.55 1.11 1.13
7.8 8.7 10.4 5.9 5.2
5.7 9.0 8.9 4.8 3.6
1960–2004
6.1
6.2
0.82
7.5
6.7
2000 2004
6.0 6.9
5.5 7.0
1.65 0.97
2.5 7.2
3.6 4.9
Sources: Bureau of Economic Analysis, “Note on the Profitability of Domestic Nonfinancial Corporations, 1960–2001,” Survey of Current Business, September 2002, pp. 17–20, updated by the author with data on profits and capital from www.bea.gov; Federal Reserve data from www.federalreserve.gov/releases/z1/current/default.htm; Data Resources, Inc. (DRI); Standard and Poor’s data from www2.standardandpoors.com/spf/xls/index/SP500EPSEST.XLS a. Property income after tax (national accounts concept) divided by the current or replacement cost of real assets (capital, land, inventories). b. Profits estimated using a cyclical correction derived from regressing the rate of profit in the first column on the difference between the unemployment rate and the Congressional Budget Office’s estimate of the non-accelerating-inflation rate of unemployment (NAIRU) using annual data. c. Ratio of the market value of equities plus net financial assets to the replacement cost of real assets, all for U.S. nonfinancial corporations (constructed from Federal Reserve data). d. Estimates constructed by adjusting the cyclically corrected return on the assumption that Q reverts to 1 at an exponential rate of 10 percent a year. e. Historical data from DRI, using reported earnings, updated by the author using data from Standard and Poor’s.
A third approximation would take into account that Q (the ratio of the market value of capital to its replacement cost) may differ from 1. This is controversial in financial economics, with some economists holding that a Q ratio that differs from 1 is not possible because of the fine arbitrage of markets. Readers who believe that can simply skip the discussion of this third approximation. Baker, DeLong, and Krugman assume that Q (by which I think they mean average Q) equals 1, but it is worth thinking about what difference it would make to the results. Incorporating a nonunitary Q into the prospective returns analysis is complicated, but I discussed the essence of the matter in an earlier Brookings Paper.2 That paper examined several cases, but the most interesting is the one in which Q fluctuates, because of animal spirits or irrational exuberance, and 2. Nordhaus (2002).
324
Brookings Papers on Economic Activity, 1:2005
then reverts to 1 at an exponential rate of 10 percent a year (a rate consistent with historical data). The third column of table 1 shows an estimate of Q for nonfinancial corporations, and the fourth column shows the prospective return under the reversion model of the behavior of Q. The Q-adjusted and cyclically adjusted prospective return in 2004 was 7.2 percent a year. As it turns out, this makes very little difference to the estimate using 2004 data, because Q was very close to 1 for 2004. It would have made a substantial difference at the peak of the stock market bubble in 2000 or at the trough of irrational malaise in the 1970s and early 1980s. A final calculation would look at the Standard and Poor’s earnings-price ratio (last column of table 1). Ignoring accounting and coverage differences, the earnings-price ratio should equal the after-tax rate of return on capital divided by average Q. This relationship is reasonably close for the entire 1960–2004 period. However, the two approaches provide quite different answers for 2004. The ratio for 2004 was 4.9 percent. The implicit Q in the Standard and Poor’s returns is seriously inconsistent with the national accounts estimates just presented: to get the same rate of return would require a Q ratio of 1.44. I suspect most of the difference is due to differences in sectoral coverage. Financial corporations have a much higher rate of return to capital than nonfinancial corporations, and the share of financial corporations has risen in recent years. As a result the implicit Q will be higher for the Standard and Poor’s calculations, which includes financial firms. Overall, this look at the “fundamentals” suggests a prospective return of around 7 percent a year starting from a base year of 2004. The above analysis of prospective returns assumes more or less unchanged underlying conditions. The major thrust of the paper, however, is to examine the impact of potential structural shifts on the rate of profit and hence on the rate of return on equities. The authors note correctly that most closedeconomy models would predict a decline in the rate of profit with a decline in the growth of labor inputs or of labor-augmenting technological change, all else unchanged. The authors’ discussion of rates of return in an open economy raises more questions than I can answer. However, at least two developments could well raise the return on equities. The first is the impact of correcting the U.S. current account imbalance. Virtually every study of this issue (including those in the present volume) suggests that at some point the dollar is in for a big real depreciation. Such a depreciation would be expected to raise domestic profits as well as provide capital gains on foreign assets.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
325
A second and more complicated issue involves the potential opening of foreign economies to U.S. capital. That process has begun but is by no means complete. It may turn out that opening foreign markets would raise the potential return on foreign investment, which would be part of U.S. stock returns. The final point I would emphasize concerns uncertainty. The fact is that we cannot precisely estimate the prospective return on stocks. The authors’ estimates are serious ones. Their analysis concludes that the appropriate estimate for future real returns is 41⁄2 percent a year, less 1 or 2 percentage points should a slowdown occur. My estimates suggest something more like 7 percent a year, with an uncertain adjustment for future changes. These are central tendencies, even before taking into account inherent variabilities. It may be possible to reduce this factor-of-two uncertainty, but I doubt it. Given the inherent uncertainty, policies should be robust to a significant range of possible outcomes. Policies should also be intrinsically robust— that is, able to withstand a financial meltdown without causing a political meltdown. I suspect that a pension system where (as is increasingly the case) most private pensions and a substantial part of public pensions are defined-contribution plans is not politically robust to two or three decades of plausibly low returns. In the end, I may not agree with all of the authors’ numbers and analyses. But I do agree with their central conclusion, slightly restated: Given the uncertainty about the prospective equity premium, policymakers should be very hesitant to base major public policy programs such as Social Security on the continued existence of a large equity premium and high stock returns in the future. Those who design policies should be able to answer the following questions: What would be the economic results if returns were at the low end of the plausible range? Who will compensate those who realize low or negative returns? And how will that compensation occur? General discussion: Several panelists commented on the substantial uncertainty surrounding forecasts of population, output, and rates of return for the long horizons relevant to Social Security. Robert Gordon declared himself highly skeptical of the prediction by the Social Security trustees, adopted by the authors, of a substantial long-term decline in U.S. economic growth caused by a decline in both population and productivity growth. In particular, he doubted the official assumption that immigration will remain constant, so that immigrants are projected to make up a declining share of
326
Brookings Papers on Economic Activity, 1:2005
the population. According to historical experience, even without illegal immigration, immigration as a fraction of the population has increased steadily since about 1955, and the number of immigrants arriving annually has been growing at a rate between 3 and 4 percent a year. Moreover, Gordon expected substantial increases in U.S. income and therefore a steady increase in the demand for immigrants, as well as no shortage of supply. He noted that the conservative assumption in his Fall 2003 Brookings Paper made a large difference in the outcome. In particular, that paper shows that, if annual immigration does no more than rise gradually over the next twenty years from its current 0.4 percent of the population to 0.5 percent, the total U.S. population seventy-five years from now will be 610 million, instead of 425 million as currently projected. The implied population growth rate is 1 percent rather than 0.3 percent as Baker, DeLong, and Krugman assume. Gordon also believed the trustees’ projected decline in productivity puts too much weight on the dismal years of growth from 1973 to 1995. Dean Baker agreed, noting that the trustees use average productivity growth in the last four completed business cycles to make their projections, thus ignoring the recent strong performance. Gordon observed that productivity growth in the past twenty, fifty-five, or eighty years has been substantially higher than the official projections. He also noted that, as long as benefits are indexed to wages, a reduction in productivity growth automatically reduces Social Security obligations. Discussion turned to the authors’ predictions of lower rates of return on equity investment, conditional on slower growth of population and productivity. Henry Aaron suggested that, with slightly higher assumptions about immigration and total factor productivity growth and a modestly lower saving rate, the rate of return would not fall much below its historical average of about 6.5 percent a year. He did not think this an implausible combination of events. Olivier Blanchard agreed that there is substantial uncertainty about future rates of return and that Aaron’s scenario is possible. But he thought it wise to worry about the adverse tail of the distribution of outcomes. He viewed the paper as arguing simply that there is a significant risk that growth rates will be low and that it is precisely in those cases that investment in equities, whether in the Social Security trust fund or in private accounts, will be disappointing. William Brainard supported the life-cycle model as the framework for analyzing the effect of demographic changes on saving, much preferring
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
327
it to the assumption of infinitely lived families in the Ramsey model. But, although he found the authors’ use of the two-period life-cycle model clarifying, he wished they had drawn on the extensive literature that examines more realistic versions of the model and estimates empirically the effects of changes in birth rates, life expectancy, and immigration on saving. Martin Baily thought it important to distinguish between the rate of return on domestic assets and the return on the U.S. stock market. Foreign earnings are important to U.S. corporations and the U.S. stock market: in the past ten years or so, U.S. companies have been quite successful at earning high rates of return overseas, even though the rate of return on foreign capital in general is lower than the return on capital in the United States. They are able to do this because of proprietary technology and better management methods, advantages that, Baily warned, may not persist. He noted that the other large, mature industrial regions, Japan and Europe, have even lower birth rates than the United States, as does China. These dramatic demographic trends overseas are going to play an important role in the return on capital within a global context. William Brainard likewise emphasized the distinction between domestic and national capital stocks and the importance of taking into account likely changes in global saving. In the authors’ analysis, growth in the domestic labor force affects the rate of return on the domestic capital stock, including capital owned by foreigners as well as capital owned by U.S. investors. U.S. investors, on the other hand, hold claims on the returns on capital abroad, both through their ownership of U.S. multinational firms and through their ownership of foreign-headquartered firms. Brainard amplified Baily’s comment about global trends, noting that the reduction in saving required to maintain the rate of return on capital in the face of reduced U.S. labor force growth was similar in magnitude to net foreign investment in the United States today. Gradual elimination of that capital inflow, in the absence of increases in national saving, would largely avoid the capital deepening that, in the authors’ analysis, forces down the return to capital. Aaron agreed with Gregory Mankiw that Social Security solvency and Social Security privatization are two distinct issues. Aaron argued, however, that privatization as proposed by the administration would significantly aggravate the solvency problem, increasing the projected seventy-fiveyear deficit by slightly more than a third from the level projected by the actuaries. Aaron also believed the plan posed significant risks to individuals: In the administration’s proposal, what matters far more than the average
328
Brookings Papers on Economic Activity, 1:2005
long-range rate of return over an individual’s lifetime is the return over the final years of the individual’s working life. A significant market decline just before a worker retires or becomes disabled could be devastating to that worker. Aaron believed acceptance of so great a risk was inconsistent with the fundamental purpose of social insurance. Benjamin Friedman commented that the pros and cons of privatization would make for an interesting debate regardless of whether Social Security has unfunded liabilities. Although he agreed with Mankiw that the two issues would best be discussed separately, he observed that it was President Bush, not the present authors, who had chosen to confound them, and he regarded Mankiw’s complaint about the authors linking them as an implicit criticism of the president.
Dean Baker, J. Bradford DeLong, and Paul R. Krugman
329
References Abel, Andrew B., N. Gregory Mankiw, Lawrence H. Summers, and Richard J. Zeckhauser. 1989. “Assessing Dynamic Efficiency: Theory and Evidence.” Review of Economic Studies 56, no. 1: 1–20. Baker, Dean. 1997. Saving Social Security with Stocks: The Promises Don’t Add Up. New York: The Century Foundation. Barberis, Nicholas C., and Richard H. Thaler. 2003. “A Survey of Behavioral Finance.” In Handbook of the Economics of Finance, edited by George Constantinides, Milton Harris, and Rene Stulz. Amsterdam: Elsevier. Board of Trustees of the Federal Old Age and Survivors Insurance and Disability Insurance Trust Funds. 2005. The 2005 Annual Report of the Board of Trustees of the Federal Old Age and Survivors Insurance and Disability Insurance Trust Funds. Government Printing Office. Campbell, John Y., and Robert J. Shiller. 1988. “Stock Prices, Earnings, and Expected Dividends.” Journal of Finance 43, no. 3: 661–76. Council of Economic Advisers. 2005. “Three Questions About Social Security” (www.whitehouse.gov/cea/three-quest-soc-sec.pdf [February 4]). Cutler, David M., James M. Poterba, Louise M. Sheiner, and Lawrence H. Summers. 1990. “An Aging Society: Opportunity or Challenge?” BPEA, no. 1: 1–56. Diamond, Peter A. 1965. “National Debt in a Neoclassical Growth Model.” American Economic Review 55, no. 5: 1126–50. _________. 2000. “What Stock Market Returns to Expect for the Future?” Social Security Bulletin 63, no. 2: 38–52. Edelstein, Michael. 1982. Overseas Investment in the Age of High Imperialism. Columbia University Press. Geanakoplos, John, Olivia S. Mitchell, and Stephen P. Zeldes. 1998. “Would a Privatized Social Security System Really Pay a Higher Rate of Return?” Working Paper 6713. Cambridge, Mass.: National Bureau of Economic Research (August). Gordon, Robert. 2003. “Exploding Productivity Growth: Context, Causes, and Implications.” BPEA, no. 2: 207–79. Haig, Robert M. 1921. “The Concept of Income—Economic and Legal Aspects.” In The Federal Income Tax, edited by Robert M. Haig. Columbia University Press. Hubbard, R. Glenn. 1998. “Capital-Market Imperfections and Investment.” Journal of Economic Literature 36, no. 1: 193–225. Kremer, Michael. 1993. “Population Growth and Technological Change: One Million BC to 1990.” Quarterly Journal of Economics 108, no. 3: 681–716. Mehra, Rajnish. 2003. “The Equity Premium: Why Is It a Puzzle?” Working Paper 9512. Cambridge, Mass.: National Bureau of Economic Research (February). Mehra, Rajnish, and Edward C. Prescott. 1985. “The Equity Premium: A Puzzle.” Journal of Monetary Economics 15 (March): 145–61.
330
Brookings Papers on Economic Activity, 1:2005
Nordhaus, William D. 2002. “The Recent Recession, the Current Recovery, and Stock Prices.” BPEA, no. 1: 199–220. _________. 2005. “The Sources of the Productivity Rebound and the Manufacturing Employment Puzzle.” Working Paper 11354. Cambridge, Mass.: National Bureau of Economic Research (May). Oliner, Steven D., and Daniel E. Sichel. 2003. “The Resurgence of Growth in the Late 1990s: Is Information Technology the Story?” Updated version. Washington: Federal Reserve Board. Petrick, Kenneth A. 2001. “Comparing NIPA Profits with S&P 500 Profits.” Survey of Current Business (April): 16–20. Romer, David. 2000. Advanced Macroeconomics, 2nd ed. McGraw-Hill. Shiller, Robert J. 2005. “The Life-Cycle Personal Accounts Proposal for Social Security: A Review.” Working Paper 11300. Cambridge, Mass.: National Bureau of Economic Research (May). Siegel, Jeremy. 2005. “Long Term Returns and the Demographic Crisis.” Washington: American Enterprise Institute (www.aei.org/doclib/20050310_Siegel.pdf [March 11]). Solow, Robert. 1956. “A Contribution to the Theory of Economic Growth.” Quarterly Journal of Economics 70, no. 1: 65–94.
