TRANSITION, TAXATION AND THE STATE
Transition and Development Series Editor: Professor Ken Morita Faculty of Economic...
23 downloads
550 Views
657KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
TRANSITION, TAXATION AND THE STATE
Transition and Development Series Editor: Professor Ken Morita Faculty of Economics, Hiroshima University, Japan The Transition and Development series aims to provide high quality research books that examine transitional and developing societies in a broad sense – including countries that have made a decisive break with central planning as well as those in which governments are introducing elements of a market approach to promote development. Books examining countries moving in the opposite direction will also be included. Titles in the series will encompass a range of social science disciplines. As a whole the series will add up to a truly global academic endeavour to grapple with the questions transitional and developing economies pose. Also in the series: Estonia, the New EU Economy Building a Baltic Miracle? Edited by Helena Hannula, Slavo Radoševic and Nick von Tunzelmann ISBN 0 7546 4561 4 The Periphery of the Euro Monetary and Exchange Rate Policy in CIS Countries Edited by Lúcio Vinhas de Souza and Philippe De Lombaerde ISBN 0 7546 4517 7 The Institutional Economics of Russia’s Transformation Edited by Anton N. Oleinik ISBN 0 7546 4402 2 The Polish Miracle Lessons for the Emerging Markets Edited by Grzegorz W. Kolodko ISBN 0 7546 4535 5 Organizational Change in Transition Societies Josef Langer, Niksa Alfirevic and Jurica Pavicic ISBN 0 7546 4464 2 Beyond Transition Development Perspectives and Dilemmas Edited by Marek Dabrowski, Ben Slay and Jaroslaw Neneman ISBN 0 7546 3970 3
Transition, Taxation and the State
GERARD TURLEY National University of Ireland, Galway
© Gerard Turley 2006 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the publisher. Gerard Turley has asserted his moral right under the Copyright, Designs and Patents Act, 1988, to be identified as the author of this work. Published by Ashgate Publishing Limited Gower House Croft Road Aldershot Hampshire GU11 3HR England
Ashgate Publishing Company Suite 420 101 Cherry Street Burlington, VT 05401-4405 USA
Ashgate website: http://www.ashgate.com British Library Cataloguing in Publication Data Turley, Gerard Transition, taxation and the state. - (Transition and development) 1. Taxation - Russia (Federation) 2. Taxation - Europe, Eastern 3. Tax administration and procedure - Russia (Federation) 4. Tax administration and procedure - Europe, Eastern 5. Russia (Federation) - Economic conditions - 19916. Europe, Eastern - Economic conditions - 1989 I. Title 336.2'00947 Library of Congress Cataloging-in-Publication Data Turley, Gerard. Transition, taxation and the State / by Gerard Turley. p. cm. -- (Transition and development) Includes bibliographical references and index. ISBN 0-7546-4368-9 1. Tax collection--Europe, Eastern. 2. Tax collection--Former Soviet republics. 3. Tax collection--Russia (Federation) I. Title. II. Series. HJ320.E852T87 2005 336.2'00947--dc22 2005021078 ISBN 0 7546 4368 9 Printed and bound by Athenaeum Press Ltd, Gateshead, Tyne & Wear.
Contents List of Tables List of Figures List of Abbreviations Foreword Acknowledgements
vii ix xi xiii xv
1
Introduction
2
Tax, Transition and the State: The Case of Russia
11
3
Budget Softness and the Russian Enterprise Sector
41
4
Tax Arrears and the Romanian Enterprise Sector
63
5
Effective Tax Administration in Transition Countries
97
6
Tax Capacity and Effort in Transition Countries
125
7
Conclusion
147
Bibliography Index
1
155 167
To Monica
‘The revenue of the state is the state. In effect all depends upon it, whether for support or for reformation ...’. Edmund Burke Reflections on the Revolution in France
This publication was grant-aided by the Publications Fund of National University of Ireland, Galway
List of Tables Table 1.1 Table 2.1 Table 2.2 Table A.1 Table B.1 Table C.1 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 3.7 Table 3.8 Table 3.9 Table D.1 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7 Table 4.8 Table 4.9 Table 4.10
Tax Revenue/GDP Ratios, 1989-1998 Index of Governance and Enterprise Restructuring/Reform, 1994-1999 Revenue Share of GDP, 1992-1999 Development of Market-Supporting Institutions, 2000 Governance, State Capture and Intervention Small Business Sector in Russia, 1998-2000 (000s) Distribution of Firms (437 firms) Difficulty of Obtaining Bank Credit on Commercial Terms (Percentage of Firms) Ordered Logit Results for Difficulty of Obtaining Credit in 1999 Difficulty of Obtaining Government Assistance (Percentage of Firms) Financing a Shortfall in Working Capital OLS Regression Results for Financing a Shortfall in Working Capital Levels of/Changes in Debts and Overdue Debts (Percentage of Assets) Regression Results for Levels of Debts and Overdue Debts Regression Results for Real Changes in Debts and Overdue Debts Extracts from Q uestions on Enterprise Finance EBRD Transition Indicators for Romania (and Visegrad Four), 1994-1998 Tax/GDP Ratios, 1994-1998 Stock/Flow Analysis of Romel’s Tax Liabilities in 1997, at Current Prices (in Thousand Lei) Stock/Flow Analysis of Romel’s Tax Liabilities in 1997, at Constant (End-Year) Prices (in Thousand Lei) Flow Analysis of Romel’s Tax Liabilities in 1997, at Constant (End-Year) Prices (in Thousand Lei) Stock/Flow Analysis of 9,000 odd Firms in the Romanian Enterprise Sector in 1997 (in Billion Lei) Stock of Tax Debt and Overdue Tax Debt at end-1997 Flows of Tax Debt January-December 1997 Groups of Firms with Tax Arrears in 1997 Compared Tax Arrears in Transition Economies
7 19 24 34 35 36 44 47 48 49 51 53 55 56 58 59 65 66 69 70 72 73 74 75 77 79
viii
Table E.1 Table F.1 Table F.2 Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table G.1 Table H.1 Table H.2 Table H.3 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table I.1 Table J.1
Transition, Taxation and the State
Consumer Price Index, January-December 1997 Selected Indicators for Tractorul UTB S.A.,1990-1998 Selected Indicators for Brasov (County) and Romania, 1997 Benchmark Tax Rates, Gross and Net Equivalents Statutory Tax Rates and Tax/GDP Ratios for 25 TEs, 1997 Statutory and Effective Taxation, 1997 Bribe Tax and Corruption for TEs Data Sources for the 25 TEs VAT Effective/Statutory Ratios, 1992-1998 SST Effective/Statutory Ratios, 1992-1998 CIT Effective/Statutory Ratios, 1992-1998 Determinants of Tax Capacity Tax/GDP ratios GDP per capita, Agriculture and Export Shares of GDP for TEs Tax Capacity Regression Results Tax Capacity Estimates Tax Ratio and Tax Effort Rankings Country Data Tax Effort Estimates
84 87 93 103 105 107 111 116 117 118 119 130 133 134 136 137 139 143 145
List of Figures Figure 5.1a Figure 5.1b Figure 5.1c Figure 5.2a Figure 5.2b Figure 5.2c Figure 5.3a Figure 5.3b Figure 5.4a Figure 5.4b Figure 6.1
VAT Effective/Statutory Ratio 1997 vs Progress in Transition SST Effective/Statutory Ratio 1997 vs Progress in Transition CIT Effective/Statutory Ratio 1997 vs Progress in Transition VAT Normalised Tax Yield (NTY) 1997 vs Progress in Transition SST Normalised Tax Yield (NTY) 1997 vs Progress in Transition CIT Normalised Tax Yield (NTY) 1997 vs Progress in Transition VAT Effective/Statutory Ratio and the Average Bribe Tax SST Effective/Statutory Ratio and the Average Bribe Tax VAT Normalised Tax Yield (NTY) and Average Bribe Tax SST Normalised Tax Yield (NTY) and Average Bribe Tax Tax Ratio and Est. Tax Capacity
120 120 120 121 121 121 122 122 123 123 142
This page intentionally left blank
List of Abbreviations ACIR BEEPS CBR CEE CIS CIT CMEA CPI CPI EBRD E/S EU FIGs FSU GDP GKO IBFD IMF MPS NBF NNT NTE NTY OECD OFZ POF PSAL RA RTS SAL SBC SME SNA SOEs SOF SST
Advisory Commission on Intergovernmental Relations Business Environment and Enterprise Performance Survey Central Bank of Russia Central and Eastern European Commonwealth of Independent States Corporate Income Tax Council for Mutual Economic Assistance Consumer Price Index Corruption Perceptions Index European Bank for Reconstruction and Development Effective/Statutory European Union Financial–Industrial Groups Former Soviet Union Gross Domestic Product Short-term Government Bonds (Treasury Bills) International Bureau of Fiscal Documentation International Monetary Fund Material Product System Net Bank Financing Net New Taxes Net Tanzi Effect Normalised Tax Yield Organisation for Economic Co-operation and Development Treasury Bonds Private Ownership Fund Private Sector Adjustment Loan Régies Autonomes Representative Tax System Structural Adjustment Loan Soft Budget Constraint Small and Medium-Sized Enterprises System of National Accounts State-Owned Enterprises State Ownership Fund Social Security Tax
xii
STS TEs UN VAT WB
Transition, Taxation and the State
State Tax Service Transition Economies United Nations Value-Added Tax World Bank
Foreword Revenue erosion and lax payments discipline, manifesting itself in low revenue mobilisation, ineffective tax collection and widespread non-compliance, have been a problem in many ex-socialist countries since the start of transition. This book examines the problems of tax payments discipline and collection in the transition countries of Central and Eastern Europe and the former Soviet Union in the context of economic transition from a centrally planned economy to a market economy, the transformation of a socialist state to a capitalist state, the nexus between government and business, and the persistence of the soft budget constraint. Much of the literature on the revenue decline in transition countries has focused on either economywide transitional phenomena or on taxpayers’ non-compliance. In contrast, this monograph examines the problem from the position of the tax collector, that is, the state, and its ability or willingness to collect taxes. Using the concept of János Kornai’s soft budget constraint, the book examines the problem of budget softness and tax payments discipline in the postsocialist transition economies during the first decade of transition. Appropriate methodologies are applied to new data for the transition economies with the purpose of revealing incidences of budget softness (or hardness) and, more generally, measuring the degree to which the enterprise sector in postsocialist countries is not tax compliant. Tax collection problems, arising from economic, administrative or political factors, are investigated in the context of transition. The results indicate tax collection problems arise for a number of reasons, including budget softness but also because of a general poor payments discipline, corruption and bribery, ineffective tax administration and low tax capacity and tax effort arising from both economic and political factors. Furthermore, our evidence indicates that cross country differences are not small, state control matters and, for many transition countries, tax administration and political constraints, as opposed to tax design and economic constraints, are more pressing problems. Transition, Taxation and the State outlines the tax collection and discipline problems (particularly in the context of the soft budget constraint and the state-enterprise relationship legacy of the socialist era) that the postsocialist state in transition countries experienced in the first decade of transition. As for the state (or the ‘tax state’ to use Joseph Schumpeter’s expression) and tax revenue performance in the second decade…let us wait and see the evidence. Gerard Turley National University of Ireland, Galway
This page intentionally left blank
Acknowledgements This book is based on research entitled Soft Budget Constraints, Tax Payments Discipline and Tax Collection in Transition Economies, undertaken at the Centre for Economic Reform and Transformation (CERT), Heriot-Watt University, Edinburgh under the direction of Mark E. Schaffer. His guidance, encouragement and thoughtful insights were invaluable and I hereby wish to acknowledge his significant contribution to this monograph. I also wish to express my gratitude to Paul Hare of Heriot-Watt University and the late George Blazyca of Paisley University who gave me the idea for this book. I wish to pay gratitude to colleagues and friends at Heriot-Watt University, Edinburgh and the National University of Ireland, Galway. A special word of thanks must go to Michael Cuddy of the National University of Ireland, Galway, Alan Bevan formerly of the European Bank for Reconstruction and Development (EBRD) and Giovanni Mangiarotti formerly of the Aston Business School, Birmingham for their direction and support. Also, my thanks to everybody who assisted me during my time in Romania and Russia. With respect to Chapter Two, I wish to thank Alan Bevan, Roger Clarke, Michael Cuddy, Paul Hare, Marina Pavlushevich, Mark Schaffer and seminar audiences at the National University of Ireland, Galway and BASEES, Fitzwilliam College, Cambridge for useful comments. The analysis in Chapter Three arose from an enterprise restructuring project in Russia involving CERT (Heriot-Watt University, Edinburgh), LBS (London), NERA (London) and BEA (Moscow). Thanks must go to Alan Bevan, Boris Kuznetsov, Giovanni Mangiarotti and Mark Schaffer for their assistance and helpful suggestions. Useful comments were also received at a BEA conference in Moscow. The work in Chapter Four was part of a PhareACE project carried out for the Government of Romania. I wish to thank Lucian Croituru, Terry Green, Constantin Munteanu, Alina Potter and Mark Schaffer. I am also grateful to the Romanian Ministry of Finance for supplying me with a dataset. My thanks to Alan Bevan and Paul Hare and to seminar audiences at Heriot-Watt University, Edinburgh and at CREEB, Buckinghamshire for helpful comments. The research in Chapter Five builds on collaborative work conducted for the EBRD, to whom I am most grateful. I am also grateful to Michael Alexeev, Alan Bevan, Wendy Carlin, Mark Schaffer and participants at the July 2000 National University of Ireland, Galway workshop on ‘Institutions and their Change in Transition Economies’ for helpful comments. Useful suggestions were also received at a Foundation for Fiscal Studies seminar in Dublin and a Scottish West Coast seminar in Glasgow. The analysis in Chapter Six was part of a research project entitled Public Finance in Transition Countries: Problems of Compliance, Opportunism and
xvi
Transition, Taxation and the State
Soft Budget Constraints that involved a number of partners including myself and Mark Schaffer, and was financed by PhareACE 1998 Programme to which I want to express my appreciation. I am also indebted to Alan Bevan, formerly of the EBRD and Matthias Reister of the National Accounts Section of the UN Statistics Division for providing data. My thanks to National University of Ireland, Galway, Heriot-Watt University, Edinburgh, the British Council, the European Commission, Galway Development Services International, the Romanian Centre for Economic Policies, the EBRD, the UK Department for International Development and the Adam Smith Institute for their financial assistance and support. I wish to thank Ashgate Publishing Company and its staff for the work undertaken in the publication of this book. In particular, my thanks to Brendan George, Senior Commissioning Editor of the Transition and Development series, Ken Morita, Academic Series Editor of the Transition and Development series, and last, but by no means least, Carolyn Court and Pam Bertram for their assistance in the preparation of the manuscript for publication. As always, I wish to thank Claire Noone and Imelda Howley of National University of Ireland, Galway for their time and support in helping me to understand the finer points of formatting. I bear full responsibility for all errors and any omissions. Finally, and on a personal note, I wish to thank my family for the patience and support that they have shown during the time that it has taken to research, write and edit this book – almost, but not quite, as long as transition itself.
Chapter One
Introduction One of the stylised facts of economic transition from a centrally planned economy to a market economy was the decline in tax revenue. Given the nature of transition where the role and size of government is reduced, a fall in tax revenue was predicted. The revenue erosion, particularly evident in the early years of transition, is often explained in the context of a change from the repressive and distortionary tax system of Soviet times where taxes (and subsidies) were the key mechanisms for fiscal redistribution from profitable to unprofitable enterprises and where tax collection from large state-owned enterprises (SOEs) was a simple task, to a more Westernstyle tax system where voluntary compliance and self-assessment are the norm and where confrontation between the tax collector and the taxpayer is not uncommon. Furthermore, as transition economies (TEs) witnessed a recession of historical proportions, a decline in the profitability of the SOEs (and other traditional tax bases), an expansion of the unofficial economy and a rise in tax evasion, corruption and bribery, the tax share of GDP was to fall even further. After a decade of transition, and despite a recovery in tax revenues in a number of leading transition countries by the late 1990s, the tax ratio in some TEs had fallen to levels below what is considered normal in market economies. Many regard this fall as excessive and view the decline in tax revenue as a serious obstacle in the attempts to finance public expenditure, redistribute income and, at the same time, embrace effective fiscal policy. Much of the literature on the revenue decline in TEs has focused on economywide transitional phenomena (the transformational recession or the rise in the unofficial economy are two examples) or on taxpayers’ non-compliance (related to, for example, rising tax evasion and a primitive tax culture) as explanations for the fall in tax revenues. In contrast, this book examines the problem of revenue erosion from the position of the tax collector, that is, the state, and, in particular, the weakness of the state as creditor, i.e. its (in)ability or (un)willingness to collect taxes. An appropriate theoretical framework for analysing the role of the state as creditor and the volatile state-enterprise relationship that exists in postsocialist times, particularly as it relates to tax payments and collection, is János Kornai’s soft budget constraint (hereinafter SBC). In this context, the book examines the problem of budget softness and tax payments discipline in transition countries and, particularly in Russia, a country where tax collection is considered a problem (despite the observation that tax revenue relative to national income is not unusually low given Russia’s level of economic development). In Russia and in other transition countries, the falling tax share of national income needs to be seen in the context of the fluid state-enterprise relationship that is common in the transition from plan to market, the capacity of
2
Transition, Taxation and the State
the reconstituted state and its willingness to tax, the nexus between government and business and the persistence of the SBC in the transition setting. In respect of tax collection, the research undertaken and reported in this monograph indicates that problems arise for a number of reasons, including budget softness but also because of a general poor payments discipline, corruption and bribery, ineffective tax administration, poorly designed intergovernmental fiscal relations and low tax capacity and tax effort arising from both economic and political factors. 1.1 Paradigms of Economic Transition The standard paradigm of economic transition that dominated in the early years of the 1990s and had the support of the international financial organisations (more notably the International Monetary Fund and, less so, the World Bank) is often referred to as the Washington Consensus.1 This orthodox or mainstream approach views the transition from a centrally planned to a market economy as a reform process emphasising the universality of the laws of the market and the undoubted economy-wide efficiency gains accruing from the standard policy prescriptions of the trinity of liberalisation (‘getting prices right’), stabilisation and privatisation. This blueprint for transition, based on the spontaneity of markets, traditional neoclassical price theory and general equilibrium theory, promotes the primacy of policy reforms and economic fundamentals and the replication or transplantation of international best-practice institutions (with the emphasis on laws and the legal and regulatory framework) of the West to the ex-socialist countries of the former Soviet bloc (a kind of utopia based on ‘societal engineering’). Although the Washington Consensus emerged from a different set of conditions, it argues that these one-size-fits-all market-oriented reforms are appropriate to any setting, including the postsocialist Central and Eastern European (CEE) and former Soviet Union (FSU) countries. A knowledge or experience of the state socialist system and the centrally planned economy is not required. Policy reform strategies are implemented along a scorched-earth approach, with textbook reforms being designed by technocrats and introduced as rapidly and comprehensively as possible in view of the reform complementarities that exist. Not surprisingly, in terms of the speed of the radical reforms, this approach is often referred to as the ‘big bang’ or ‘shock therapy’ view of transition. This also applies to the economic role of the state 1 Strictly speaking, the Washington Consensus refers to a set of policy guidelines for most Latin American countries in the late 1980s for which, it was argued, a consensus was reached among Washington-based international agencies, the US government and mainstream economists. John Williamson of the Institute for International Economics, the person who coined the phrase, viewed these reforms as the lowest common denominator of policy advice by ‘Washington’ to Latin American countries as of 1989. The ten economic reforms focused primarily on structural adjustment policies of price and trade liberalisation, macroeconomic stabilisation and fiscal discipline, deregulation of entry barriers and privatisation of stateowned enterprises.
Introduction
3
where what is required is a depoliticisation of the economy, a break of the nexus between government and business and a dismantling of the state, or, according to critics of the Washington Consensus approach, in the extreme case of neoliberal market fundamentalism, state desertion. The institutional-evolutionary paradigm, more popular within academic circles, views transition as a large scale institutional transformation where the focus is on the institutional underpinnings of capitalism appropriate to the specific conditions of each country and in accordance with the initial conditions at the outset of transition. This approach is critical of the revolutionary vision of transition and, instead, views transition as a process involving systemic change in the face of great uncertainty and complexity, unlike the competitive neoclassical model and its notion of equilibrium, which is inherently static. As opposed to equilibrium processes, the emphasis of the institutional-evolutionary approach is on the dynamics of institutional change within an evolutionary perspective, based on contracting and noncooperative games in modern microeconomic theory. Here, the focus is on the gradual development of the institutional supports or arrangements for a market economy, accepting the dangers of institutional voids and that not all existing or inherited institutions, organisational forms or social capital are redundant. Transitional second-best institutions and the preservation of social capital can be both worthwhile and necessary in order to prevent further economic disruption. Whereas the Washington Consensus view of transition is a top-down approach, the institutional-evolutionary perspective is a bottom-up view focusing on the institutional design of market economies, the importance of social norms and the organic development of the private sector. Markets and economic agents do not exist in a vacuum but in an institutional framework – ‘the rules of the game’ – that facilitates exchange and interaction. Institution building and the provision of a framework for well-functioning market structures and organisations is the focus of this approach and it argues for the gradual or incremental implementation of sequenced reforms (often through experimentation and learning by doing) in order to ensure growing support for policies. In the institutional-evolutionary approach, although there is recognition for the need to reduce the role of the state, the emphasis is on a reconstituted state and improving state capacity (so as to, among other things, enhance the market environment) as opposed to a weakened state. It also stresses the path dependency of system development and is mindful of the historical continuity and the communist legacy unlike the ahistorical, tabula rasa Washington Consensus approach (Clague and Rausser 1992; Roland 2000; Bönker et al. 2002). Although an outline of the two major paradigms of transition is useful in the context of this book, many observers feel that the debate between the two approaches and, in particular, the controversy between ‘shock therapy’ and gradualism and the tendency to label countries as either one or the other, has not been very helpful and has unintentionally diverted attention away from some of the more important aspects of economic transition. One such feature of transition is the SBC.
4
Transition, Taxation and the State
1.2 Definition and Interpretation of the SBC The incentive problem inherited from the socialist system known as the SBC takes its name from the budget constraint faced by households in standard microeconomic theory. The budget constraint was first extended to organisations and firms by Kornai (1979, 1980) and applied to the socialist economies of Central and Eastern Europe. The budget constraint is softened when a firm is not held to a fixed budget, but finds its budget constraint non-binding. The enterprise sector is said to exhibit a SBC when there is a recurring or persistent expectation of a refinancing or bailout of loss-making firms; firms receive financial assistance because they are loss making and the expectation of aid is close to certainty as the external support is more than just a once-off event. The channels or mechanisms by which the ex post rescue of unprofitable firms takes place vary, from budgetary subsidies and tax arrears (by government) to inter-enterprise arrears (by trade suppliers and utility companies) to overdue loans (by banks). Since transition began over a decade and a half ago, the more traditional forms of the SBC, namely, arbitrary pricing and direct subsidies, have given way to new and, often, more implicit instruments, such as tax arrears and overdue payables to banks, trade suppliers and utilities. Another mechanism that is evident in predominately FSU countries, where banking intermediation is generally underdeveloped, is non-cash payments (barter, promissory notes, offsets) by firms to its creditors. Either way, this expectation of a bailout influences and undermines the ex ante behaviour and incentives of firms. The objective of the organisation that is bailed out is straightforward, namely, survival. The motive of the rescuer varies, depending on the interpretation of the SBC (as there is no consensus on a precise definition of the concept). For the purposes of this book Kornai’s bureaucratic hierarchical model (1980, 1992a) is used, as opposed to Dewatripont and Maskin’s dynamic commitment model (1995) and Stiglitz’s gambling banks model (1994), of a paternalistic and benevolent state willing to support unviable firms in order to avoid politically and socially costly layoffs. A fourth interpretation is Shleifer and Vishny’s politicians and firms model (1994), a theory that is employed in Chapter Two. Using the Kornai model, the SBC syndrome can be viewed as a theory of exit and, thus, complements Schumpeter’s theory of creative destruction (Schumpeter 1911). Whereas Schumpeter focused on the birth or creation of organisations, the SBC phenomenon explains the survival (or demise) of organisations. By preventing certain firms from bankruptcy, the SBC alters the natural selection process inherent in a competitive environment (whether it be market socialism, transition or market economy). Since the term first appeared in 1979, there have been a number of different explanations of the SBC. According to Kornai (1979, 1980), the source of the budget softness, in the context of the socialist system, is the paternalism of the state. Firms are not responsible for losses, or for profits, hence, the levelling effect. This explanation is system-specific, focuses on political considerations and is based on the vertical relationship between superior and subordinate. In contrast, the explanation advanced by Dewatripont and Maskin (1995) focuses on economic causes, namely the inability
Introduction
5
to commit to no bailout ex post and the centralised financial system. Using a gametheoretic model, the SBC is viewed as a time-consistency problem in the presence of irreversible investment.2 With asymmetric information and adverse selection, bad projects get refinanced – the phenomenon of ‘throwing good money after bad’. A different explanation, espoused by Stiglitz (1994), argues that, in the context of the financial system, soft budgets arise when financial institutions have an incentive to make large gambles when appraising projects. In this explanation, an insolvent bank may be willing to invest in a risky project because the bank will become solvent if the gamble pays off and, in the case of the project turning bad, will be no worse off than it was before the loan was made i.e. still insolvent. A fourth model is presented by Shleifer and Vishny (1993, 1994), where subsidies are paid to enterprises to retain excess employment. These transfers result from bargaining that takes place between managers of firms and government politicians, the latter driven by non-economic objectives, namely their own narrow self-interest and self-preservation. 3 In the context of transition and economic theory, there are two broad perspectives on the SBC. One approach is the Washington Consensus view, which treats the SBC as an exogenous variable and a matter of direct policy choice. The implication here is that political will is all that is required in order for the budget constraint of firms to be hardened. The alternative perspective is the institutional-evolutionary approach that views the SBC as endogenous to the institutional set-up. Here, the hardening of budget constraints is possible only as an outcome of institutional change. Credible commitment to hardening budget constraints is a matter of devising suitable institutional mechanisms and arrangements (Roland 2000). 1.3 Evidence of SBC in Transition Countries Despite the problems with the short time horizon (the start date for transition is generally accepted to be 1990 for CEE countries and 1992 for FSU countries) and the usual data problems, the empirical evidence on the SBC is now quite considerable (Schaffer 1998; Roland 2000; Kornai 2001). The evidence appears to indicate that in the early years of transition, trade credit arrears and overdue payables to banks were often the main instruments of budget softness as opposed to the budgetary subsidies and arbitrary pricing mechanisms of earlier pre-transition years. Among other changes, price liberalisation and reform of the public finances ensured that these old channels of the SBC were to disappear (in 2 Schaffer (1989) also presents a game-theoretic model where the centre is unable to make credible commitments to the enterprise. Unlike Kornai’s SBC where the paternalism of the state is known, it is possible, under imperfect information, for the state to build a reputation for toughness and, in doing so, is able to impose hard budget constraints on the enterprise. 3 A useful taxonomy for classifying SBC models, in the context of ex ante and ex post (in)efficiencies, is outlined in Mitchell (2000). This includes the SBC model based on bank or creditor passivity (Mitchell 1998).
6
Transition, Taxation and the State
the case of arbitrary pricing) or to decline in importance (in the case of budgetary subsidies). As transition progressed, banks and trade suppliers imposed hard budget constraints on firms and customers by insisting on payment, and, in the case of late or non-payment, refusing to extend new flows of finance to delinquent firms (Schaffer 1998). These firms, often loss-making or unviable, managed to stay afloat by extracting new forms of assistance, either from their workforce (in the form of wage arrears), from utilities (in the form of payment delays or non-cash payments), or, more commonly, from the state (in the form of tax and social security contributions arrears). In the case where the SBC manifests itself in the form of inter-enterprise arrears or overdue payables to banks, the source of the budget softness is frequently the state. In the case of the former, firms accumulate trade credit arrears in anticipation of a general government bailout (for example, by means of a clearing scheme). In the case of the latter, banks with large non-performing loans (to non-viable and/or favoured firms) are often state-owned or subject to political interference or operate under a SBC regime i.e. insolvent banks expect a bailout by government. This perception of government is particularly true in some of the FSU republics where the state is often viewed as the softest creditor; a perception that is tolerated and sometimes even instigated by the state itself. These forms of budget softness were also evident in CEE countries but have diminished as transition has progressed. As the state, through its tax system, appears to be the main source, and instrument, of budget softness, this monograph addresses the issue of state governance with respect to taxation and, in particular, the problems of tax discipline and collection. The motivation for focusing on taxation and tax collection stems from observing the general decline in tax revenues witnessed by the majority of ex-socialist countries since the start of transition with a view to providing, from the perspective of the tax collector (i.e. the state) as opposed to the taxpayer (i.e. for our purposes, the enterprise) or the general economic environment (i.e. the transition setting), an explanation for revenue erosion. In the wider context of the SBC, it is the government-firm relation that is of interest here, with government acting as the supporting organisation or funding source and the firm acting as the budget constraint organisation. Other relations where the SBC is evident include government-government (Wildasin 1997), government-bank (Mitchell 1998) and bank-firm (Dewatripont and Maskin 1995). In trying to identify evidence of budget softness, researchers encounter endogeneity and causality problems when examining the relationship between the financial position of a firm (usually gauged by some measure of profitability) and its rescue (usually measured by some form of financial aid). Although we encounter the same measurement problems when investigating the various channels of budget softness, we then proceed to focus on the more general matter of tax discipline, compliance and collection. Tax collection problems can arise for a number of reasons, including budget softness (Chapter Three) but also because of a general poor payments discipline (Chapter Four), ineffective tax administration (Chapter Five) or because of low tax capacity and effort arising from economic and noneconomic factors (Chapter Six).
Introduction
7
1.4 Soft Taxation in Transition Countries An important feature of the evolving state-enterprise relationship in TEs is the tax system. The transition to a market economy involved replacing the old system (dominated by the turnover tax and transfers from state enterprises) with a tax system more in line with market economies. On a practical level, this involved the introduction of Western-type corporate and personal income taxes, a value-added tax (VAT) and improving tax administration. On a higher level, it involved detaching the enterprise sector from the state. This has proven more difficult than expected (EBRD 1999). As a result, tax arrears, poor tax collection and revenue erosion were common in many of the transition countries during the first decade. Table 1.1, where the tax/GDP ratios are reported for the transition countries of Central and Eastern Europe and the former Soviet Union, shows the general decline in tax revenues. This fall in tax revenues is particularly true of the laggard reformers (Bulgaria, Romania and Russia for example) – as opposed to the rapid reformers (Czech Republic, Estonia and Poland for example) or those
Table 1.1 Tax Revenue/GDP Ratios, 1989-1998 Albania Bulgaria Croatia Czech Republic Hungary FYR Macedonia Poland Romania Slovak Republic Slovenia Average (CEE)a Armenia Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Moldova Russia Tajikistanb Turkmenistan Ukraine Uzbekistan Average (FSU)a
1989 .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. 38.5c
1990 42.2 43.0 .. .. 46.6 .. .. 35.8 .. .. 41.9 .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
1991 26.7 37.7 .. .. 47.9 .. 35.4 36.2 .. .. 36.8 .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ..
1992 16.4 33.1 .. 38.8 45.6 .. 36.5 34.2 39.4 41.7 35.7 20.5 31.1 .. 30.8 8.2 21.5 14.6 .. 30.3 20.8 35.9 .. .. 41.6 26.4 25.6
1993 18.1 28.9 .. 41.2 45.2 .. 39.1 31.5 36.5 42.9 35.4 13.1 33.2 .. 36.5 2.0 16.7 14.8 36.1 27.8 21.1 31.7 .. .. 41.1 28.4 25.2
1994 19.1 31.8 40.1 40.3 42.9 41.9 39.6 28.3 38.8 42.6 36.5 13.1 16.9 45.3 38.8 4.6 12.3 13.6 36.1 31.4 26.4 30.9 40.6 16.4 39.1 23.3 25.9
1995 17.7 29.3 41.4 39.7 41.9 38.8 38.7 28.9 42.0 42.3 36.1 12.7 10.4 39.0 37.8 5.4 11.0 15.0 35.1 31.6 28.8 28.3 12.8 19.2 34.8 27.7 23.3
1996 15.3 26.5 41.5 38.8 43.5 37.9 38.3 27.1 41.1 41.3 35.1 12.9 14.2 38.6 37.1 10.9 11.3 13.2 33.7 28.8 27.4 28.3 11.7 14.4 34.7 32.3 23.3
1997 13.5 26.6 40.3 38.0 42.9 36.1 37.5 28.0 38.4 40.4 34.2 16.3 17.0 43.2 37.1 13.0 12.2 12.5 34.8 32.0 29.9 29.3 13.3 20.8 35.6 27.7 25
1998 15.9 30.6 43.5 36.9 41.2 34.4 35 28.2 37.2 41.2 34.4 16.9 15.0 43.7 37.1 13.4 16.2 14.4 34.3 32.9 29.0 29.2 11.7 16.2 31.8 29.4 24.7
Notes: a. Unweighted average. b. The data for 1995 are from 11 May–31 Dec. c. Estimate of tax revenue/GDP ratio for the USSR. Source: OECD 1991; Tanzi and Tsibouris 2000; Author’s calculations.
8
Transition, Taxation and the State
countries that have not started or have delayed reforms (Belarus and Uzbekistan for example) – where the state has proven either too weak or too corrupt to collect taxes, to enforce the laws governing taxation and to resist the demands of vested interests and well-connected firms.4 Mitra and Stern (2003) observe a ‘U-shaped’ (a decline followed by a recovery) trend in tax revenues over the transition period. This is true when we examine transition beyond the first half dozen years and it is particularly evident in the advanced transition countries, of which many managed to halt the decline in the tax ratio and, in some cases (Estonia and Slovenia for example), even reverse the trend. This pickup in tax revenues should not be surprising given the recovery in national output and the improvement in the administrative capacity of the tax systems. This book purports to address the earlier period in transition when a decline in tax revenue was predicted but the extent, and the cross-country variation, of the actual decline was both dramatic and surprising (Cheasty 1996). We use the SBC framework to conceptualise the tax collection problems. The empirical evidence on the SBC indicates that the tax system is a mechanism commonly used by which the budget constraint of firms can be softened. In particular, late, non-cash, or non-payment of tax liabilities by loss-making firms (although not exclusively as tax arrears of high profile profitable firms in Russia and elsewhere are not uncommon) suggests that the hardening of budget constraints, expected in transition countries in the 1990s, did not fully materialise (Kornai 2001). We begin by looking at taxation in the context of the transition from a planned to a market economy. Under the socialist system, high tax shares of GDP were common; tax ratios in excess of 50 per cent of GDP were not unheard of. As the transition process accelerated, many of these countries witnessed a dramatic decline in the tax/GDP ratio. Russia, with late, non- and in-kind payments of tax evident during the 1990s, is an appropriate setting within which to begin our analysis.5 We then examine the role of taxation in the context of the SBC. In particular, we apply appropriate (in some cases, modified) methodologies to new data for the TEs with the purpose of revealing incidences of budget softness (or hardness) and, more generally, measuring the degree to which the enterprise sector in TEs is not tax compliant. Tax collection problems, either arising from economic, administrative or political factors, are investigated in the context of transition. Our evidence indicates that cross country differences are not small, state control matters and, for many transition countries, tax administration and political constraints, as opposed to tax design and economic constraints, are more pressing problems. In Chapter Two we set out to assess the problems of taxation in the context of the Russian State, its relationship with the enterprise sector and the transition from plan to market. It is argued that Russia’s well-known tax collection problem is a manifestation of an ineffective and poorly governed state, supported by weak institutions. Central to this problem is the symbiosis of politics and economics that 4 The other set of TEs where tax revenues plummeted were the countries involved in war and internal strife (Azerbaijan, Georgia and Tajikistan for example). 5 In this book, the term ‘Russia’ is used for the ‘Russian Federation’.
Introduction
9
was common in Soviet times and is still prevalent in postsocialist Russia. In the early years of transition, Russia witnessed an ineffective state and a systemic crisis of governance. This crisis of a weak state and poor governance, manifesting itself in a severe fiscal crisis, culminated in the August 1998 crash. The chapter begins with a short analysis of the models of government. Following this, we focus on the weak state and the symbiotic relationship or nexus that exists between government and business. The symptoms of an ineffective state that were evident in Russia’s fiscal system are described later. The chapter concludes with an account of the August 1998 fiscal crisis and some tentative policy prescriptions. In Chapter Three, a survey of over 400 Russian manufacturing firms is used to look for evidence of SBC. We distinguish between three categories of creditors – banks, suppliers, and government – and use several different measures of the SBC – ease/difficulty of obtaining credit, how a shortfall in working capital would be financed, and observed levels of and changes in debts and overdue debts. The evidence suggests that on average suppliers and banks are unwilling to extend credit to unprofitable enterprises and hence are imposing hard budget constraints on firms. Subsidies from government of various forms are not uncommon, and we find evidence, albeit weak, that these are more likely to be allocated to loss-making firms. Since transition began, overdue tax liabilities have emerged as one route by which firms have their budget constraints softened. We examine this issue in Chapter Four. Using a stock/flow analysis on comprehensive firm-level data, for Romania, we establish that many firms are obtaining cheap working capital from the government by means of tax arrears; stocks and flows of overdue tax debts indicate a sizeable tax arrears problem in Romania. There is also some evidence to indicate that the problem is not confined to loss-making firms. Profitable firms with flows of tax arrears suggest a general poor payments discipline problem. Chapter Five looks at the issue of tax administration and compliance in transition countries. Wide differences between effective or realised average tax rates and tax yields that would result if statutory tax rates were strictly applied indicate tax compliance and collection problems. Due to the greater politicisation of tax systems in TEs, we would expect the shortfalls in effective tax yields for TEs to be larger than a benchmark for the mature market economies where tax systems are well established, the administrative capacity is stronger and tax arrears are tolerated less frequently. We find a positive correlation between progress in economic transition and effective tax administration. For slow reformers, the effectiveness of tax collection appears to vary with the extent of state control. Those TEs that have maintained the apparatus of the state have done well in tax collection compared to those countries where there is evidence of state decay. In Chapter Six we examine the issue of tax capacity (potential) and effort (performance). The decline in the tax/GDP ratio for many ex-socialist countries has raised fears of further revenue erosion and low tax potential for these countries in the post-transition era. The purpose of this chapter is to apply a tax capacity methodology, favoured by the International Monetary Fund (IMF) in the 1970s to study international
10
Transition, Taxation and the State
differences in tax ratios among developing countries, and extend it to the TEs using data for the 1990s. The results indicate that many Commonwealth of Independent States (CIS) transition countries resemble more closely the low- and middle-income countries of Asia and Latin America than their fellow transition countries of Central and Eastern Europe and the Baltic States, and that TEs, in general, are no worse at tax collection than other countries of similar economic structure and development. Although the tax capacity and tax effort estimates indicate that binding constraints for some of the TEs may be low incomes and a high agriculture share of GDP, political and administrative constraints cannot be dismissed as possible reasons for low tax revenues and tax collection problems. The main findings of the research concerning tax collection problems in transition countries are summarised in Chapter Seven. Since the start of transition, most ex-socialist countries have witnessed declines in tax/GDP ratios. Tax collection problems have emerged for a number of different reasons, including the SBC and the nature of the postsocialist state-enterprise relationship. As with other studies of the TEs, our research indicates that cross-country differences are significant, institutions (in our case, an effective tax administration system) matter, and that, as with nontransition countries, taxation is a vexed topic with economic, political and social dimensions.
Chapter Two
Tax, Transition and the State: The Case of Russia 2.1 Introduction The power to impose and collect tax is a defining characteristic of a modern capitalist state. It was Joseph Schumpeter who, prior to his appointment as Finance Minister in the Austrian Government after World War I, described the capitalist state as the ‘taxcollecting state’. In his 1918 paper entitled ‘The Crisis of the Tax State’, he wrote that ‘…“tax” has so much to do with “state” that the expression “tax state” might almost be considered a pleonasm.’ (Schumpeter 1918). For Schumpeter, the origins of the capitalist state (at least in Germany and Austria) arose, not from political needs, but from fiscal needs. It is the fiscal demands and the tax-collecting powers, born out of a ‘common exigency’ that distinguishes the modern capitalist system from its predecessor, the feudal system, he argued. In contrast, the origins of the socialist state (and all its manifestations including the SBC) are the political one-party system (Kornai 1992a). One of the characteristics of this system, as observed by Mancur Olson, was its ability to collect an extraordinary amount of ‘taxes’ (Olson 2000). It is within this context that we examine the transition from the socialist state to the capitalist state that took place at the end of the 20th century. As our primary interest lies in a sovereign state’s power to tax, it is appropriate to assess Russia’s (the most important of the successor states to the Soviet Union) transition to a ‘tax-collecting state’. Since transition began, many ex-socialist countries have experienced a dramatic fall in tax revenue. Russia is no exception. In 1992, total government revenue as a share of GDP was 39.3 per cent; by 1998 the revenue/GDP ratio had fallen to 33.4 per cent. Moreover, federal revenues fell from 16.6 per cent of GDP in 1992 to only 11 per cent in 1998, of which two percentage points was non-cash (IMF 2000a). This inability to collect ‘sufficient’ taxes is a manifestation of an ineffective and poorly governed state, supported by weak institutions whose authority is not legitimate in the eyes of its citizens.1 Many Russians view the efforts of the central government as the private affairs of the elite and certainly not aimed at enhancing the common interests of its citizens. This crisis of a weak state and poor public governance, manifesting itself in a severe fiscal débâcle, culminated in the August 1998 crash. It was Sergei Kiriyenko, the Russian 1 According to one commentator, ‘The revenue extraction process in Russia reveals an inherent mistrust between state and society’ (Easter 2003).
12
Transition, Taxation and the State
PM at the time of the August crisis, who said, ‘If the state does not learn to collect taxes, it will cease to exist.’2 This is an outline of the chapter. We begin in Section 2.2 with a short theoretical analysis of taxation and the models of government. In Section 2.3 we focus on the weak state and the symbiotic relationship that exists between government and business. The symptoms of an ineffective state that are evident in Russia’s fiscal system are described in Section 2.4. In Section 2.5 we outline the events of the August 1998 crash – a fiscal crisis that had its origins in a weak and ineffective Russian state. The policy implications arising from the analysis of poor state governance are examined in the final section of the chapter. 2.2 The Theory Our primary interest is in taxation and different models of government. In a wellfunctioning market economy, governments oversee the provision of public goods and the orderly payment of taxes, where taxes are defined as compulsory, unrequited (in the sense that benefits provided by government to taxpayers are not normally in proportion to their payments), nonrepayable payments exacted by government for public purposes. In postsocialist Russia, the payment of taxes has been problematic. What follows is a stylised account of tax payments (and its sister activity, bribery) in the context of transition from plan to market. When an enterprise produces goods/services or engages in trade or employs workers, it incurs tax liabilities of all sorts (for example, VAT or export tax or social security tax). Once a tax liability arises and is known to the tax authorities, the enterprise can do one of three things. One, the tax is paid, either in cash or non-cash form. Non-monetary tax payments were not uncommon in Russia during the 1990s. Explanations vary from liquidity shortages and the demonetisation of the economy that resulted from tight monetary policy to poor financial intermediation leading to lower transaction costs for barter than for non-barter to the facilitation of tax avoidance or evasion whereby the barter transaction circumvents the use of bank accounts and the chances of confiscation of revenue by the tax authorities (Gaddy and Ickes 1998; Commander and Mumssen 1999). 2 Ironically, Jeffrey Sachs, much maligned for his advice on Russia’s transformation, did recognise the dangerous prospect of state collapse or insolvency in Russia. In 1994, in warning about the dangers of a ‘bad’ equilibrium for the Russian State, he wrote, ‘Several simple examples of immediate relevance to Russia illustrate the risks of state collapse.’ He went on to list six, namely tax evasion, criminality, regional separatism, flight from the rouble, foreign debt overhang and panic by government creditors (Sachs 1995). Although Sachs was quite right in his analysis, namely of a state in near collapse, it was his policy recommendations, including his claim, at that time, that radical reform by itself revives the collapsed state, that were to attract most public attention. For lessons on government collapse in transition countries, including Russia, see Roland (2000).
Tax, Transition and the State: The Case of Russia
13
Two, the tax is not paid (or paid with such a delay that the tax payment is, given interest charges and inflation rates in the early years of transition, in effect, a subsidy.) The delinquent taxpayer (for our purposes, the enterprise) can be either a profitmaker or a loss-maker. If it is a case of profit-making firms not paying taxes, the problem is generally one of a poor payments discipline (aside from the possibility of individual and legitimate tax disputes with the tax authorities). Alternately, if it is a case of loss-making firms not paying taxes, the problem is one of a SBC (Kornai 1980, 1986).3 Three, payments are made arising from the tax liability but are paid in the form of a bribe to government officials in order to, among other things, reduce or eliminate the known tax liability. This behaviour, both on the part of the recalcitrant firm and the politician or bureaucrat, can be described by reference to Shleifer and Vishny’s predatory government and corruption model (Shleifer and Vishny 1993, 1998). We begin, however, with János Kornai’s analysis of how the socialist system comprised a paternalistic, all-powerful, helping-hand state that gave rise to the SBC and financial indiscipline on the part of state-owned enterprises (Kornai 1980, 1992a). Unlike private firms in a market economy where the state is, in general, neither paternalistic nor pervasive, SOEs in a socialist economy may face ‘soft’ budget constraints.4 The budget or financial constraint is softened when the state bails out or refinances loss-making firms. The recurring expectation of a bailout, in whatever form it takes, influences ex ante the behaviour of enterprises. This helping-hand model of government depended on a strong and benevolent state whose primary aim was to provide for the enterprise which meant maintaining employment, preventing firm closures and job losses and, ultimately, ensuring security. As socialist economies are transformed into market economies, the SBC syndrome was expected to diminish. It is true that many of the features of Kornai’s socialist system have indeed disappeared. There are no longer shortages of consumer goods, nor do people spend time and energy queuing for goods. Many firms operate in the private sector offering a variety and quality of goods that were not available under the Soviet system; many firms that remain in the state sector are forced by market conditions to restructure. These economies are now more integrated with the market economies of Western Europe. Of course, it was expected that some aspects of Kornai’s paternalism might linger longer than others, or might manifest themselves in new ways. For example, some loss-making firms (not to mention healthy firms) are still extracting subsidies from the state, but by means of tax arrears and/or utility arrears rather than by explicit on-the-budget subsidies. The incidence of tax arrears and other non-market economic phenomena suggest that, in some respects, the state is still ‘soft’ in its relationship with certain loss-making 3 A loss-making firm will not be liable for profit tax but it will have other tax liabilities (VAT and social security taxes, for example). 4 Of course, the SBC is usually defined in the context of the market socialist system where a price mechanism existed rather than to the classical socialist system where money was passive and plans were expressed in terms of physical quantities and output targets.
14
Transition, Taxation and the State
enterprises. Despite these incidences, for countries like Poland and Hungary, the ‘helping hand’ has been replaced by Adam Smith’s ‘invisible hand’.5 For Russia and other FSU states, neither the helping-hand model nor the invisible-hand model can adequately explain the behaviour of the state with respect to some firms in the stateowned (and, indeed, privatised) enterprise sector.6 In addition, Kornai’s model of a paternalistic state and the SBC has the state redistributing profits from ‘profitable firms’ to ‘loss-making’ firms, the so-called levelling effect. Given that price controls were set with social objectives in mind, enterprises were not responsible for losses. Profits were redistributed through various subsidy and tax channels. Bargaining took place along the vertical chain between managers, local bureaucrats and central planners. In postsocialist Russia, some politically-connected, profitable firms manage to engage in socially-costly rent-seeking behaviour, namely, lobbying for subsidies and tax concessions from the government while, at the same time, tolerate delayed trade payments, or payments in kind, from their customers. In turn, politicians (often with short time horizons) extract bribes in return for these lucrative favours, whether they are subsidies, cheap credits, licenses, quotas or permits. It is in this context that we draw upon the second of the theories, that of Andrei Shleifer and Robert Vishny’s ‘Politicians and Firms’ model (Shleifer and Vishny 1993, 1994).7 Shleifer and Vishny’s model is predicated on a predatory, activist, 5 The symbiosis of politics and economics that manifested itself in different ways was far more ingrained in the Soviet Union that it was in Central and Eastern Europe and this observation may go some way in explaining the cross-country performance differences since 1990. On the subject of performance, although Russia’s record since 1992 is considerably worse than most of Central and Eastern Europe, it compares favourably to many other CIS countries. This is often ignored in the transition literature on Russia and Eastern Europe. 6 It is important to acknowledge that the enterprise sector that exists in Russia today is not monolithic. It is made up of state, privatised and de novo private firms, profitable and unprofitable firms, viable and nonviable firms, domestic and foreign firms and so on. In this chapter, we are primarily concerned with the nonviable firms and the well-connected firms (sometimes one and the same) that are, at times, kept alive by political interference. 7 With respect to the interaction between bureaucrats (government officials) and entrepreneurs (private sector actors), there are different models of government, as explained by Frye and Shleifer (1997) and, again, by Shleifer and Vishny (1998). There is the marketfriendly ‘invisible hand’, the more interventionist-type ‘helping hand’ and the predatory ‘grabbing hand’. As the transition from Soviet-style socialism (a kind of ‘clenched hand’) to a market economy based on private property and enterprise (Adam Smith’s ‘invisible hand’) continues, Russia seems to be caught halfway, in a state of limbo. Although Frye and Shleifer (1997) refer to the state-enterprise relationship in China as an example of the helping-hand model, we extend its applicability (without changing the basic features of the model i.e. a wellorganised, interventionist, welfare-maximising government) and use it to explain the stateenterprise relationship that existed in other socialist countries pre-transition. As indicated by Frye and Shleifer, these models of government are ‘ideal types’; in reality, governments may be a mixture of all or a combination of some. A variant of these types of government models is presented in Treisman (1995).
Tax, Transition and the State: The Case of Russia
15
grabbing-hand state whose aim is self-preservation and wealth accumulation, leading to rent seeking, corruption and bribes.8 Using standard economic theory of industrial organisation, Shleifer and Vishny (1993) show that the type of organised corruption that was prevalent in the former Soviet Union (where the person paying the bribe was assured of getting the government good) was less damaging than the unorganised crime and corruption that became prevalent in Russia in the early years of the reform period. According to Shleifer and Vishny (1994), bureaucrats are self-interested economic agents often acting independently of each other who use their power and position to pursue their own agendas. They did so under the socialist system by restricting output and keeping prices low, resulting in endemic shortages (Shleifer and Vishny 1992). As bureaucrats did not have any legitimate cash flow rights to goods produced in the state sector, corrupt officials turned their attention to creating, or taking advantage of, shortages in order to enhance their opportunities for collecting bribes. The underground barter economy was used to produce goods that the state failed to supply. In postsocialist Russia, the well-connected, state-owned or privatised firms, often run by the old elite (the nomenklatura) frequently engage in rent or subsidyseeking behaviour rather than profit-seeking behaviour.9 By lobbying, they manage to extract soft credits, subsidies and tax concessions from the state, often in return for kickbacks. Effort is exerted and resources are wasted in the pursuit of these rents. The effect of this is that compliant firms, often de novo private firms, are penalised. The behaviour of these compliant firms is adversely affected as they, in turn, bargain with government officials for tax concessions and soft loans. The cosy networks and personal exchange that exist between officials of government and managers of these enterprises reduce the risk of exposure and, in doing so, reduce the relative transaction costs of bribery. In this type of environment, firms are more likely to seek a maximisation of political connections – relational capital - rather than of profits (Gaddy and Ickes 1998; Åslund 1999). In developed market economies, the job of market-friendly, representative institutions like the political parties, the media and social organisations is to ensure that the type of distortive and corrupt bureaucratic behaviour described above does 8 Others, including Jan Winiecki (1992) and Anders Åslund (1994, 1999) have highlighted corruption and the rent-seeking activities of the old elite in explaining Russia’s problems in its transition to a market economy. Of course, a predatory view of the state is not new and has followers in both the neoclassical and Marxian traditions. 9 In the early years of reform in Russia when inflation was high, certain sectors (agriculture and coal, for example), enterprises and bankers gained large rents, either in the form of subsidised credits, loans at negative real interest rates or arbitrage opportunities. As Treisman (1998) and Hellman (1998) have argued, these winners gained at the expense of the population at large. According to Treisman, the Russian government managed to coopt the winners and, in the process, achieved lower inflation. Hellman argues, contrary to conventional wisdom, that reform governments must restrain these ‘winners’ (by expanding political participation and competition) as they are more of a threat to successful reform than the ‘losers’.
16
Transition, Taxation and the State
not go unpunished. In Russia, these ‘checks and balances’ were either open to abuse or were simply too weak to prevent the rent-seeking activities of politicians, government officials and nomenklatura-appointed enterprise managers. The political and institutional vacuum that existed in Russia allowed the rent-seeking elite to pursue their political objectives (securing voters’ support and maintaining power) and, at the same time, to enrich themselves by accumulating vast sums of personal wealth (Åslund 1999). The more efficient allocation of resources that the transition to capitalism was promised to bring was hindered by the activities of these rent-seeking interest groups (Olson 1965, 2000; McFaul 1995; EBRD 1999). These special interest groups competed for the resources of the state (sometimes colluding with government officials) rather than seeking greater market share. Due to a number of factors, including the large rent-seeking opportunities, the weak central authority, the lack of representative institutions and the failure to build a sufficient constituency for market-oriented reforms, these entrenched groups have been able to advance their own narrow interests at the expense of the reform process and the welfare of the population at large. In summary, Russia’s ‘peculiar’ (a word often used to describe the postsocialist Russian economic system) transition from a planned to a market economy has been a story of winners and losers, and powerful vested groups with conflicting and narrow interests, made stronger by the weakness of political parties, of democratic institutions and of the rule of law. The early years of transition left the Russian state in a position of limbo, some place between the ‘helping hand’ and the ‘invisible hand’, what Shleifer and Vishny call the ‘grabbing hand’. This view was espoused by Boris Yeltsin in his State of the Federation address in March 1999 when he talked about Russia getting stuck halfway, and left with ‘…a hybrid of the two systems…’ (i.e. plan and market). This vacuum has being partly filled by oligarchs, rent seekers, corrupt government officials and the mafia. This is what George Soros calls ‘robber’ capitalism, Grigory Yavlinsky calls ‘phony’ capitalism, Marshall Goldman calls ‘bastard’ capitalism and John Gray calls ‘mafia-dominated anarcho-capitalism’. It manifests itself in a number of different ways; the problem of tax collection is one of these manifestations. But this is only one side of the story. There is another side to the story. Russia, in many respects, resembles a market economy. The majority of its output, estimated at 70 per cent in 1999, is produced in the private sector (EBRD 1999). For most goods, prices are not controlled and are close to world levels. Certainly, some Russians are better off today than they were over a decade ago, despite the setbacks associated with transition. As for taxation,the tax/GDP ratio in Russia is higher than the ratio for some of the Organisation for Economic Co-operation and Development (OECD) countries and some of the exsocialist transition countries. Given Russia’s level of national income, this level of government capacity is quite surprising. Here is one of postsocialist Russia’s many paradoxes. This is the other side of the story, a side that is often neglected and seldom recognised. Leaving this aside, the rest of this chapter explores the relationship between the ‘tax-collecting state’ and the ‘tax-paying enterprises’ in a particular transition economy, namely Russia (in the first decade of transition). In effect, it is
Tax, Transition and the State: The Case of Russia
17
the story of an economy that has experienced a transformation (albeit traumatic and uneven) from an administrative-command economy to a market economy but whose state is in crisis; an incapacitated state that has failed to change in line with the economy and whose actions, and in some cases inactions, have adversely affected the behaviour of enterprises, not only state and privatised, but also the new private sector (often with new firms going underground or failing to grow, or in some cases, even start up, due to the predatory nature of government and its officials, particularly at regional and local level).10 2.3 The State and the Enterprise Sector in Russia In this section, we provide an explanation for Russia’s tax problems in the context of the state-enterprise relationship. We argue that the real cause of the problem is the ineffective state, the behaviour of an enterprise sector that is often dominated by powerful vested interests and the non-market relationship that exists between the state and parts of the enterprise sector in postsocialist Russia. We will deal with each of these in turn. 2.3.1 The State As we see it, the problem of the state as exists in Russia is twofold. Firstly, although the state has succeeded in building institutions required in a democracy (a constitution, a parliament, elections and political parties, an independent judiciary, a media with freedom constitutionally guaranteed), its progress in building institutions that are supportive of a market economy has been less impressive (Appendix A). Among others, these institutions are legal (property rights, courts, bankruptcy procedures), regulatory (monopolies, financial sector and securities markets), administrative (civil service and public administration), financial (adequate financial intermediation, supervision and prudential regulation), fiscal (tax code, tax administration) and social (obligations to workers, welfare provision). Without these institutions in place and working well, Russia cannot properly function as a capitalist democracy. Just as Kornai (1992a) emphasised the importance of ‘norms’ under the socialist system, we need to emphasise the importance of institutions in the new market system. Ironically, much of this requires a strong state. Secondly, the actions and inactions of the state in postsocialist Russia are at odds with its responsibilities in a market economy. On the one hand, it is not doing what 10 This view of a dysfunctional Russian state is shared by others, including Sapir (1999), Brown (1999), Gustafson (1999) and Nagy (2000). For example, in their separate accounts of the August 1998 crisis, they explicitly refer to ‘a large-scale failure of state power’ (Sapir 1999); ‘a bloated state that …was incapable of performing satisfactorily such basic functions as collecting taxes, paying public service workers, and maintaining law and order’ (Brown 1999); ‘...a failure of the central state’ (Gustafson 1999) and ‘…the on-going meltdown of the state’ (Nagy 2000).
18
Transition, Taxation and the State
it should be doing, whether that is enforcing the rule of law, building an effective public administration, ensuring property rights and contract enforcement, providing essential public services, legitimising the informal sector, ensuring a social contract or encouraging competition. The national state has, in effect, abdicated on these responsibilities. Weakened, it became particularly vulnerable to the small groups of lobbyists and vested interests (agricultural lobby, bankers, energy sector) that sought privileges and favours. In particular, it is the politically-connected firms that have managed to ‘capture’ the state in pursuit of their own narrow interests (EBRD 1999; Hellman and Schankerman 2000). This leads us to the other dimension of governance in Russia today (Appendix B). In sharp contrast to the above, the state is frequently doing what it should not be doing whether it is softening budget constraints of nonviable firms, indiscriminately bailing out insolvent ‘banks’, exchanging government goods in return for kickbacks, stifling enterprise with punitive taxes, incessant inspections, costly regulations and draconian penalties, keeping power at the regional level at the expense of local government, and maintaining, or in some cases, fostering features of the old system (the privileges and favours, the ‘patronclient’ networks, the regulations, the bureaucracy, the extortion and so on). This view of a state that is irresponsible in its inactions (weak and ineffective) and actions (predatory and interventionist) is shared by others, including Åslund (1994), Mau (2000) and Nagy (2000). Peter Evans (1992) remarks that the ‘…conjunction of leviathan and the invisible hand…’ is not as contradictory or as uncommon as it might appear.11 In that context, the problem in Russia, as distinct from other countries where a symbiotic relationship between government and business is common, (Japan and Korea, for example) is that the Russian state lacks the ‘embedded autonomy’ prevalent in so-called developmental states (Evans 1992). 2.3.2 The Enterprise Sector The changes in the enterprise sector in Russia since transition began have been mixed, with evidence of significant sectoral and regional differences. Whereas the resources sector generates substantial value added and much needed hard currency earnings, the value added and foreign currency earnings generated by the manufacturing sector are less substantial. This is a legacy of Soviet times when negative value added was quite common and enterprises did not trade directly with foreign firms, other than through the trading bloc, the Council for Mutual Economic Assistance (CMEA). Although many of these enterprises have been privatised, they (pre-1998 crisis) have 11 The publication’s editors, Stephan Haggard and Robert R. Kaufman write that, ‘For governments to reduce their role in the economy and expand the play of market forces, the state itself must be strengthened.’ Evans supports this thesis of a complementarity between state and market by quoting from such figures as Karl Polanyi, who in The Great Transformation wrote ‘The road to the free market was opened and kept open by an enormous increase in continuous, centrally organised and controlled interventionism’ and Max Weber, writing in Economy and Society that ‘Capitalism and bureaucracy have found each other and belong intimately together.’
Tax, Transition and the State: The Case of Russia
19
been slow to restructure (Frydman et al. 1996; Blasi et al. 1997). Table 2.1 reports the EBRD’s indicator of governance and enterprise restructuring for a number of transition economies, including Russia, for the period 1994–1999. It is evident from the table that the improvement in enterprise restructuring in Russia, in the judgement of the EBRD, has been slow and erratic, since 1994.
Table 2.1
Estonia Georgia Hungary Kazakhstan Lithuania Russia Ukraine
Index of Governance and Enterprise Restructuring/Reform, 1994-1999 1994 3.0 1.0 3.0 1.0 2.0 1.7 1.0
1995 3.0 2.0 3.0 1.0 2.0 2.0 2.0
1996 3.0 2.0 3.0 2.0 3.0 2.0 2.0
1997 3.0 2.0 3.0 2.0 2.7 2.0 2.0
1998 3.0 2.0 3.3 2.0 2.7 2.0 2.0
1999 3.0 2.0 3.3 2.0 2.7 1.7 2.0
Note: The classification system is based on the judgement of the EBRD’s Office of the Chief Economist. The scores represent the following: 1 = SBC (lax credit and subsidy policies weakening financial discipline at the enterprise level); few other reforms to promote corporate governance. 2 = moderately tight credit and subsidy policy but weak enforcement of bankruptcy legislation and little action taken to strengthen competition and corporate governance. 3 = significant and sustained actions to harden budget constraint and to promote corporate governance effectively (e.g. through privatisation combined with tight credit and subsidy policies and/or enforcement of bankruptcy legislation). 4 = substantial improvement in corporate governance, for example, an account of an active corporate control market; significant new investment at the enterprise level. 4+ = standards and performance typical of advanced industrial economies: effective corporate control exercised through domestic financial institutions and markets, fostering market-driven restructuring. Source: EBRD 1999.
In post-privatisation Russia, corporate governance mechanisms are generally weak with insiders retaining control. This resulted from the mass privatisation programme where the stakeholders (incumbent managers and workers) were co-opted in order to secure their support for the privatisation scheme. Much of the capital stock in the Russian state-owned enterprise sector is old and obsolete. Modes of organisation are often antiquated. Labour hoarding by some state-owned and private enterprises is evident, with some social provisions still provided by the enterprise. In rural Russia, many of the mono-enterprise towns remain, isolated from the marketplace and even more isolated from the international capitalist world. Labour mobility is often restricted by enterprise provision of social services, derisory unemployment benefits, wage arrears and, sometimes even, the registration (propiska) system (Broadman and Recanatini 2001). Despite the reforms (and the difficulty in acquiring verifiable evidence), many enterprises in Russia, over a decade into transition, are still likely to
20
Transition, Taxation and the State
operate with two sets of books, a shell or daughter company, an offshore subsidiary and have unregistered activities. Regular payments to the mafia, government officials and local politicians are not uncommon. Enterprises are as likely to engage in late payment, informal, personalised, private methods of contract enforcement and other non-market activities as they are to engage in prompt payment, formal, anonymous, legal methods of contract enforcement and other market activities. As with other transition economies, differences in enterprise performance and restructuring in Russia may be accounted for by a number of factors, including ownership, market structure and competition, corporate governance, finance and budget softness (Djankov and Murrell 2002). Since privatisation began, improvement in enterprise performance has been delayed, and in some cases hampered, by lack of investment finance, domestic and foreign competition, regional administrative barriers and inadequate corporate governance mechanisms (Angelucci et al. 2002). Of course, the enterprise sector in Russia is not monolithic. We can distinguish between state, privatised and new private, with de novo private firms showing strong performance relative to both state-owned and privatised firms (Appendix C). Each of these three categories is likely to have a different relationship with the state. Often, the incumbent (state or privatised) is protected by the (subnational) government whereas de novo private firms are often harassed by the state (McKinsey 1999; Broadman 2002). We now examine the state-enterprise relationship in more detail. 2.3.3 The State-Enterprise Relationship Central to the behaviour of enterprises and the slow pace of restructuring is the political economy nexus that exists between the state and the enterprise sector or, more particularly, between subnational government and incumbent state or privatised firms. In a well-functioning market economy, the certain provision of public goods in exchange for the orderly payment of taxes is a central feature of the state-enterprise relationship. This is not the case in Russia (or in many other FSU countries). In the case of a federalist state with weak central authority, the state-enterprise relationship is more likely to be dominated by subsidies, tax breaks, payment arrears, assetstripping and bribes than by public goods and compulsory, unrequited payments to government i.e. taxes. Although the state has been weakened in many respects since the dismantling of state socialism, the state still retains ‘interventionist’ features of both the helpinghand and the grabbing-hand models with respect to its relationship with the enterprise sector. Some politically-connected firms manage to extract rent from the state in return for bribes to government officials and bureaucrats. This is the case of the activist, predatory state ensuring its survival, while at the same time enriching public officials (Shleifer and Vishny 1994; EBRD 1999). Likewise, it is not uncommon for loss-making firms to be bailed out by the state (particularly at the regional level of government), by means of tax and utility arrears. This is the case of the SBC phenomenon (Pinto et al. 2000; Kornai 2001). In this case, it would appear that its behaviour is influenced by a desire to maintain employment and prevent social
Tax, Transition and the State: The Case of Russia
21
unrest. McKinsey (1999) found that ‘… policies are often put in place to achieve social objectives, namely protecting existing jobs, but in many cases, the suspicion is that they also serve the personal financial interests of government officials in collusion with businessmen…’. What has changed is the way that the state channels funds to these loss-making enterprises. Arbitrary pricing and budgetary subsidies have been replaced by new, and often implicit, ways of extracting subsidies from the state, including tax exemptions, arrears and offsets, utilities arrears, non-payment and, in some cases, barter. This intervention by the state (and in increasing amounts by regional government) prevents a reallocation of resources to the sectors of the economy where there is potential for further value added. Although Russia may be a ‘market’ economy in the sense that most of the output, officially or unofficially, is produced in the private sector and resources are allocated through a decentralised price mechanism, it still retains some features of the old Soviet socialist system. One of these features is the state-enterprise relationship that still lingers, preventing the state from carrying out its ‘market’ duties and preventing enterprises from restructuring. By condoning informal activities, reneging on its obligations to its suppliers and its workforce, taking bribes and by allowing rent-seeking behaviour, accepting tax offsets, negotiating tax liabilities with well-connected enterprises, allowing mutual indebtedness and barter chains to flourish, tolerating arrears, allowing the resources sector to subsidise the manufacturing sector, and by maintaining ‘relational capital’ with the enterprise sector, the state is implicated in the economic and political crisis of the 1990s in Russia, culminating in the rouble devaluation and the debt default of August 1998. Since 1992, governments in Russia have been described at various different stages as soft, weak, corrupt or interventionist. With respect to its relationship with the enterprise sector, we can say that Russian governments have proved to be all of the above at some stage or other. In a market economy the state has responsibilities to the enterprise sector. Among others, it is sometimes an owner (although not often), a creditor (tax collection agency), a legal guardian (courts and contract enforcement, bankruptcy) and a regulatory body (to promote competition and prevent monopoly practices). With respect to the Russian enterprise sector, we can say that as an owner the state is often corrupt, sometimes paternalistic; as a creditor it is sometimes ‘soft’; as a legal guardian it is weak and open to abuse; and, finally, as a regulatory body, it is frequently bureaucratic and interventionist (while, at other times, ineffective and even absent). The state-enterprise relationship suffers as a result, failing to develop into a market-type relationship as exists in mature market economies and in some of the leading transition economies (Hungary and Poland, for example). To summarise, Russia has experienced a ‘soft’ transition in terms of the symbiosis between politics and economics, or what ex-acting PM Yegor Gaidar refers to as the ‘…intertwining of property and power…’ and of ‘…business and
22
Transition, Taxation and the State
bureaucracy…’ (Winiecki 1992; Gaidar 1999a).12 This is a legacy of the socialist political-economic system, described as ‘…the unique symbiosis of the state with society and the economy.’ (Kaminski 1996).13 This is not a case for state desertion, but for the relationship between the state and the enterprise sector to be made more formal within an appropriate institutional setting. Although this symbiotic relationship is best exemplified by the existing stateenterprise relationship, the implications reach beyond the enterprise sector. For example, the results for many ‘ordinary’ Russians have been catastrophic, despite the increase in individual liberty and choice. In the early years, this was reflected in rising mortality rates, greater income inequality and even larger numbers of people living in ‘poverty’. In later years, it has been reflected in rising crime and in a general mistrust of state institutions and in the entire political process. Recall Thomas Hobbes treatise Leviathan wherein he stated that life without an effective state is ‘…solitary, poor, nasty, brutish and short.’ (Hobbes 1651). As the state continued to weaken, many Russians found themselves disengaging from the national state. It is not surprising that the general public do not have trust in the government or its institutions when arrests and charges against state employees of the Central Bank of Russia (CBR), Goskomstat, the Federal Bankruptcy Service, the State Tax Service (STS) and other state agencies for fraud, evasion and other criminal activities were not uncommon throughout the 1990s. What are the symptoms of an ineffective state? How does the symbiotic relationship between government and business manifest itself? How can the umbilical cord of the state be detached without damaging the fiscal capacity or the tax powers of the state? In the next section, we focus on Russia’s weakened fiscal system and the problems of fiscal discipline and tax collection.
12 Of course, this ‘…intertwining of property and power…’ can be traced back to preSoviet and Tsarist times, as far back as the 14th century and remained a characteristic of the Russian state throughout most of its history (Pipes 1974; Hedlund 1999). 13 This symbiotic relationship is portrayed in Ericson’s depiction of the Russian economic system as characteristic of ‘industrial feudalism’ with regional fiefdoms (Ericson 1999). He describes the state-enterprise relationship as one where there ‘…is a tendency at all levels to look to governments, with their ability to command resource flows, for direction, support and the solution of economic problems’ and where there ‘…is a tendency for governments to look to business organisations for access to the resources needed to maintain their power and control.’ Easter (2003) makes a similar claim when writing, ‘While the state may be dependent on large corporations for tax revenue, Russia’s corporate elites remain dependent on state patronage for wealth and status.’ In support of this, one of the key findings of the EBRD/WB BEEPS survey was that ‘States and firms continue to be tied together in a web of interactions in which the state provides a wide range of direct and indirect subsidies to firms, while firms provide public officials with some combination of control over company decisions and bribes.’ (EBRD 1999).
Tax, Transition and the State: The Case of Russia
23
2.4 Taxation and Russia’s State-Enterprise Relationship A central feature of the state-enterprise relationship in any economy is the fiscal system and taxation. Moreover, since transition began a decade ago, nothing portrays the fusion between government and business in Russia better than the taxation issue. In a market system, the tax obligation of an enterprise is determined by tax law. In the socialist system, the ‘tax’ obligation of an enterprise was determined by the bureaucrats and was subject to change, depending on the circumstances. In the early years of postsocialist Russia, the tax obligation of many enterprises was determined, in theory by statutory tax legislation, but in practice, by personal relations and open to negotiation and subject to lobbying and bargaining. Ex-acting PM Yegor Gaidar, in relating the tax collection problem to the SBC and the interface between the state and the enterprise in postsocialist Russia, writes ‘…the problem of tax collection was not a problem of tax administration in the usual sense. It was more a political struggle about what constituted the essence of the emerging economic system, whether it was to be a system in which the relationship between the state and the enterprises was to be regulated by law or whether it would be business as usual, based on political influence and personal contacts’ (Gaidar 1999b). Taxation in the traditional (pre-reform) socialist system was very different to the Western-style tax system, both in terms of objective and composition. Taxes were mostly transfers within the public sector, from profitable SOEs, that is, stateowned enterprises with surpluses, to loss-making SOEs. For the enterprise, taxes were implicit and negotiable. As for the state, taxes were easy to collect as taxpayers, that is, enterprises, tended (relative to the capitalist system) to be fewer in number and larger in size. In addition, the State Bank, in processing enterprise payments, acted as a tax collection agency. The main forms of ‘tax’ revenue were enterprise profits tax (= profit remittances), turnover tax and payroll tax. The collapse of the socialist system meant that fiscal institutions, as exist in market economies, had to be built from scratch. Statutory tax systems were required in order to raise enough revenue to pay for the provision of public goods. Income tax systems (both corporate and personal), value-added taxes and tax administration systems were all introduced in the early years of transition. Russia was confronted with a Western-style tax system where voluntary compliance and confrontation between the tax collector and the taxpayer are the norm. For these and other reasons, it was inevitable that the transition to a market economy would bring a fall in the tax share of GDP. By 1998, the total government revenue/GDP ratio was 33.4 per cent. Moreover, federal revenues were only 11 per cent of GDP, of which two percentage points were in non-cash form (see Table 2.2).14 We use total government revenue as a share of 14 This problem of insufficient federal government revenues was even recognised by Anders Åslund (a critic of government intervention) who, in the immediate aftermath of the crisis, (in 1999), wrote ‘...Russia’s federal revenues are small, and the central government can hardly manage without additional resources…’ Admittedly, there were improvements in (federal) tax revenue in 1999, due largely to discretionary changes in policy, namely, the reintroduction of export taxes and the centralisation of tax receipts.
Transition, Taxation and the State
24
GDP as this is a good measure of government capacity. Table 2.2 shows that since transition began, Russia’s government capacity has weakened. This is synonymous with recent empirical evidence that poorly performing governments (in this case Russia), in contrast with better performing governments (many of the ex-socialist Central European countries), collect fewer taxes (La Porta et al. 1999).15
Table 2.2 Revenue Share of GDP, 1992-1999 1992
1993
1994
1995
1996
1997
1998
1999
Enlarged Government Revenue*
39.3
36.2
34.6
34.1
33.5
36.5
33.4
35.6
Federal Government Revenue
16.6
13.7
11.8
12.9
12.5
12.3
11.0
13.4
- Cash
n.a.
n.a.
11.4
11.0
9.1
10.0
9.0
13.4
- Non-Cash
n.a.
n.a.
0.4
1.9
3.4
2.3
2.0
0.0
of which
Note: * Consolidated revenues (incorporating extrabudgetary funds), including both cash and non-cash items. n.a. = not available. Source: IMF 2000a.
The problem of poor tax collection is more than an administrative problem. The problem is more deep-rooted, with numerous possible causes. For a start, there are a large number of enterprises that are either ‘unprofitable’ or operate in the informal economy. Two, the tax base is narrow and as a consequence there is an excessive tax burden on those liable for tax. This encourages those liable for tax to find ways of reducing their tax bill, legally or illegally. For example, competition for enterprise tax revenues between the different levels of government in conjunction with the practice of tax shifting between regions provides an opportunity for enterprises to lower their effective tax rate (see below). Three, the tax system that operated throughout the 1990s included a plethora of tax concessions and tax breaks, not to mention tax amnesties. Four, the government is often found negotiating the tax obligations of well-connected enterprises. Five, tax liabilities are often overdue, and, in some cases, either written off or reduced. Six, the use of barter and non15 Likewise, Shleifer and Treisman (2000) note that ‘The most market-friendly governments in the late twentieth century tended to be those that collected and spent the most revenues.’
Tax, Transition and the State: The Case of Russia
25
monetary exchange makes tax detection more difficult. Seven, tax administration is undoubtedly poor. Tax collection was made more difficult by the de facto dual subordination of tax administrators to the central authorities and the subnational governments (see Chapter Five). It is also true that Russia has made less progress than other transition countries in moving away from taxing corporations toward taxing individuals. This is a legacy of the socialist system when the state expropriated ‘profits’ from the enterprises. In postsocialist Russia (as is common in many other CIS countries), many of these same enterprises, some privatised, are not short of ways of evading detection from the tax authorities. Among others, it includes the use of money surrogates, affiliates or subsidiaries, multiple bank accounts or barter transactions. For the Russian federation, tax collection is further complicated by the chaotic evolution, since reforms began, of the system of intergovernmental fiscal relations.16 Russia, with its highly centralised formal system combined with subnational informal autonomy (as opposed to the days of the Soviet Union when subnational governments were mere administrative branches of central government), does not have a well-designed system of fiscal federalism (Lavrov et al. 2000). Among other things, this results in an intense struggle between the federal government and regional governments over revenue-sharing. The system of intergovernmental fiscal relations and fiscal federalism in Russia is deficient in a number of respects. For one, during the mid-1990s the share of revenues accruing to the subnational government increased in an ad hoc and arbitrary fashion, from 31 per cent in 1992 to 38 per cent in 1995, all at the expense of a weakened federal government (Freinkman et al. 1999). Two, federal government has assigned much of its social spending responsibilities to the regions; equal transfers of revenues have not been forthcoming. These unfunded mandates often resulted in unclear responsibilities and in the case of subnational governments, expenditure arrears. Three, with income differentials between the regions growing, intergovernmental transfers have often been discretionary, non-transparent and politically motivated. Four, bilateral power-sharing agreements between the federal government and individual regions (in particular, some of Russia’s 21 republics) have undermined the fiscal system. The centre is often found negotiating from a position of weakness with centrifugal, resource-rich regions whose demands range from devolution of spending responsibilities or conferring of special tax powers to political autonomy. It would seem that the relationship between federal government and the republics is determined more by economic resources, political influence, personalities and bargaining skills than by de jure rights. 16 An excellent account of fiscal decentralisation in the context of subnational SBC, that is, the (in)ability of central government to credibly commit to a no-bailout policy when dealing with subnational jurisdictions and the danger of these lower level governments passing up liabilities to the centre, is given in Jonathan Rodden et al. (2002) Fiscal Decentralisation and the Challenge of Hard Budget Constraints. In the case of Russia, see Freinkman et al. (1999) Subnational Budgeting in Russia, Preempting a Potential Crisis.
26
Transition, Taxation and the State
Five, there is a lack of adequate tax autonomy at subnational level. For taxes that are permanently assigned to subnational governments, limitations, in respect of bases and rates, are often set by federal government. For taxes shared, the revenuesharing arrangements are often customised and subject to yearly shared rates. Six, the incentives for government officials at subnational levels are often skewed toward low tax (particularly in cash) collection, excessive spending and greater concessions to local enterprises. In particular, the revenue sharing schemes between regional and local governments (with subsidies constituting the largest share of local government spending) provide little or no fiscal incentive for the local government to mobilise revenues or to increase their tax base as any increase in local government own revenue is ‘taxed away’ by regional government (Martinez-Vazquez and Boex 1999; Zhuravskaya 2000). More generally, ‘…Tax sharing inherently contains a strong element of political bargaining that is, in itself, a source of “softness”’ (Lavrov et al. 2000).17 Another tax rights problem in Russia is the overlapping tax base and ‘tragedy of the commons’ problem, leading to ‘overgrazing’ (Berkowitz and Li 1999). During the mid-1990s, subnational governments in Russia had the authority to introduce new taxes within their jurisdictions. Often, the result was high aggregate tax rates, depressed economic activity and firms driven underground.18 Despite recent improvements, intergovernmental fiscal arrangements at the end of the 1990s remain non-transparent, complicated and uncertain, making for a federal system that undermines the fiscal responsibility of subnational governments. As one public finance expert remarked ‘…the Russian Federation’s current fiscal structure with its “shared” taxes, obscure transfers, and duplicative expenditure roles seems designed to frustrate rather than facilitate effective subnational government.’ (Bird 2000). In the context of financial discipline, the system of intergovernmental fiscal relations described above, that is of unclear expenditure assignments, weak revenue autonomy, transfers that are subject to frequent changes and negotiation, and few restrictions on subnational borrowing (all contrary to the principles underlying the idea of ‘market-preserving’ federalism), lends itself to soft subnational budget constraints (Rodden et al. 2002). One implication of intergovernmental relations and fiscal federalism in Russia is the weak position that the federal government often finds itself in vis-à-vis the regional governments and the enterprise sector. For example, if the federal government makes a move against the utilities in an attempt to collect taxes, supplies are threatened. Alternately, if the federal government moves on the regional governors, taxes are withheld. Likewise, if the federal government moves on the large enterprises, the 17 Actual revenue sharing rates have differed from the sharing rates as stated in budget law on account of differences in the extent of tax arrears between federal and subnational government, various bilateral treaties signed with individual regions (republics) and, in some cases, refusal to remit tax collection to the federal government (Martinez-Vazquez and Boex 1999). 18 In the context of fiscal federalism, subnational government incentives and tax rights, there is the obvious contrast, since reforms began, between the Russian and Chinese experience. For more on this, see Berkowitz and Li (1999) and Roland (2000).
Tax, Transition and the State: The Case of Russia
27
managers and regional governors often collude. It is not uncommon for managers of large enterprises to play one level of government off against the other; more often than not it is the federal government that loses out (Treisman 1999; Shleifer and Treisman 2000).19 For some, the fiscal problem is as much on the expenditure side (where some spending is quite wasteful, particularly at the regional level, and some is used for private benefit) as on the revenue side. Russia, with a GDP per capita less than $2,000 in 1998, does not have the economic capacity to be a big spender.20 Whereas federal non-interest spending has declined as a share of GDP, regional spending and expenditure arrears have both increased. Those favouring spending cuts advocate further reductions in budgetary subsidies where they are above ‘normal’ levels, cuts in government bureaucracy and a reform of entitlements so that social provisions are targeted to where they are most needed. Given the limited means of the national state, much of the welfare spending is misdirected, dispersed without regard to the income levels of the recipients (Åslund 2002). Yet, an equally valid argument can be made for more expenditure in the areas of public sector pay, health services, education and the environment (Shleifer 1997). Tax arrears, or overdue tax liabilities in the enterprise sector, have been of particular concern to the Russian government. On the issue of tax arrears, all enterprises are not the same. What follows is a stylised account of tax arrears in the enterprise sector. For the purpose of explaining tax arrears, the enterprise sector can be divided into four categories. Some enterprises pay their taxes on time; more pay their taxes but with a delay, some in cash and some in-kind; others who could pay, do not pay; and, finally, there are the firms that cannot pay their taxes and do not pay taxes.21 The first category forms a disproportionately large part of the corporate tax actually paid by the enterprise sector and is not politically strong. The second category uses tax liabilities as a source of cheap working capital in times of financial difficulty; possibly because the government is considered to be the softest creditor and is paid only after employees, banks and trade suppliers have been paid. The third category is either politically sensitive or politically connected. The final category is continually in financial and economic difficulty but does not (for example, because of weak bankruptcy practices) or is not allowed to (for paternalistic reasons) die. By 1998, much of the tax payments made to the government were not in cash form but in non-monetary forms. According to the OECD, 40 per cent of all taxes paid 19 The different games played and the ingenious schemes used by the respective stakeholders (federal and subnational governments, enterprises, tax authorities) in Russia’s federal tax system are aptly described in Shleifer and Treisman (2000). 20 EBRD, 2001. In purchasing power parity terms, the figure is closer to $4,000. 21 Pre-1998, one of these delinquent firms is Gazprom, the gas monopoly and Russia’s richest company. Gazprom often acts like a quasi-fiscal institution and has been described as ‘the state within the state’. It has political connections, secures favourable tax breaks, has had large tax arrears, often conducts barter-trade and continued to supply non-paying customers. The mutual debts that existed between the government, Gazprom and other enterprises made the budgetary process for the federal government all the more difficult.
28
Transition, Taxation and the State
to the federal government in 1997 were non-monetary (OECD 1997). In sum, we feel that the state must shoulder most of the blame for its inability to collect taxes. Many of the reasons for the low compliance result either directly or indirectly from discretionary government policy. Ultimately, like so many of the other problems that exist in Russia today, the problems are one of economic incentives, institutional weakness, arbitrary government action and an ineffective state. The problem of tax payments (and the related problem of non-monetary forms of payment) can be summarised as follows. The incidence of tax and inter-enterprise (trade credit) arrears, offsets and other forms of non-monetary payments paints a picture of mutual indebtedness between certain enterprises, between certain enterprises and government and between the federal government and certain regional governments. Some enterprises, whether profitable or unprofitable, do not pay their bills and stay in business. Loss-making enterprises are subsidised either by i. profitable enterprises (many of which are in the energy sector) through nonpayment, ‘discounted’ payment, in-kind payment or late payment of bills;22 ii. government through the non-payment or late payment of taxes;23 iii. workers through wage arrears.24 In the socialist system, manufacturing enterprises were subsidised by underpriced inputs and low interest charges on the use of capital. In postsocialist Russia, the industrial sector is similarly subsidised, with only a change in the form of this subsidisation. Funds are channeled to the enterprises via quasi-fiscal and fiscal 22 Gaddy and Ickes ‘virtual economy’ argument is that enterprises are engaged in exaggerated-price barter transactions whereby the value of a good or service is exaggerated in order to hide the negative value added generated by manufacturing enterprises (Gaddy and Ickes 1998). Value-creating enterprises are also faced with constraints. For example, the energy companies in Russia are prohibited from disconnecting non-paying customers. In return, the government tolerates tax arrears and arrears to the extra-budgetary funds by the utilities. In response to overdue receivables, some utilities have responded by rescheduling or writing off debts in return for an ownership stake in these recalcitrant firms. In this scenario, the utilities can been viewed as conduits, whose purpose it is to channel funds to favoured and/or troubled firms. 23 The SBC syndrome. One difference between the SBC in Kornai’s paternalistic model and the budget softness as currently evident is the absence of any fiscal redistribution from profitable to unprofitable firms (the levelling effect). Whereas this was a feature under the socialist system, in postsocialist Russia some profitable firms are also benefiting (by means of tax arrears, export licenses, etc...). This phenomenon is better explained by the Shleifer-Vishny model (1994) where these firms are using their political connections and bribing government officials to take advantage of a weak but self-preserving state and corrupt officials. 24 There is a popular perception that it is the Russian household sector that ultimately carries the burden through non- or late payment of wages, and of social and welfare benefits. It is not uncommon for some enterprises to use wages arrears to strengthen their case for concessions from the government. This may also be evidence of a subnational SBC when subnational wage arrears get rescued by federal bailouts (OECD 2000a; Lavrov et al. 2000).
Tax, Transition and the State: The Case of Russia
29
institutions. Some loss-making manufacturing enterprises stay in business with the aid of tax arrears, non- or late payment of bills to their suppliers and utility companies or non- or late payment of wages to their workforce. These economy-wide, nonmarket activities are allowed to persist and proliferate because they are ‘tolerated’, ‘facilitated’ and, in some cases, even ‘instigated’ by the state. Since the collapse of the Soviet Union direct subsidies have fallen. Since the tightening of monetary policy, growth in new bank credit has decelerated. As enterprises became more credit constrained, they sought out new ways of surviving. Some combination of non-payment of bills (to utilities, suppliers, government, workers), and late payment to suppliers (inter-enterprise arrears) and to government (tax arrears) and use of nonmonetary forms of payment (barter, offsets, veksels25) allowed for an injection of ‘soft’ credit to the enterprise sector. Again, this was sanctioned, in the past by a strong but benevolent state, determined to prevent firm closures and job losses (the ‘helping hand’) and more recently by a weak but predatory state, determined to survive and, in the process, to enrich itself (the ‘grabbing hand’). Commander and Mumssen (1999) make a similar assertion when they argue that ‘…the state increasingly injected net credit into the ailing industrial sector….’ because of ‘…concerns about employment, but also a result of corruption, fraud and tax evasion….’ In its analysis, the EBRD (1999) found that ‘…payments are consistently made to particular interest groups under the guise of employment protection.’ In postsocialist Russia, probably the greatest source of paternalism and budget softness lies with the regional governments. Attempts by the federal government to instigate reforms have been hampered, if not by the Duma, by the regional governments whose influence resides in their control over the use of land, real estate, taxes, utilities and ultimately, local enterprises (state and privatised). In more recent times (a phenomenon that increased as the economic depression deepened), some regions ceased to remit taxes to the ‘centre’ (partly in response to regional transfers committed by the centre but withheld), engaged in unauthorised spending of budget funds, hid additional resources in extrabudgetary funds, extended soft credits to (often, unproductive) enterprises, regained stakes in local enterprises by means of debt-equity swaps and imposed ‘temporary’ price controls on certain commodities.26 This tendency for regions to behave in a quasi-socialist manner has been described by some as evidence of a feudal system where governors treat their regions as fiefdoms (Ericson 1999; Fairbanks 1999). This crisis of a weak state and poor governance manifested itself in a severe fiscal crash in August 1998, the topic of the next section. 25 A veksel is a promissory note issued usually by an enterprise, it is passed through the chain and usually presented for payment by the end-holder who usually gets paid in kind with the products of the issuer. In Russia, where there is a secondary market in veksels with the value of the veksel depending on the credibility of the issuer, veksels are viewed as close substitutes for money. 26 Given the way fiscal federalism works in Russia, taxes get passed up to the central authorities (hence, the ability to stop payments to the centre) and enterprise subsidies are expenditure items assigned to subnational governments (hence, subsidies paid by regional and local governments to ‘favoured’ enterprises).
30
Transition, Taxation and the State
2.5 The 1998 Russian Crisis27 As stated previously, the long-term problem with the Russian economy is a failure on the part of the state to meet its obligations with respect to the functioning of a market economy. As we have seen, this includes the failure to enforce laws, to collect taxes, to pay wages and pensions, to repay investors and international donors, to encourage competition in the enterprise and banking sectors, to foster market-type institutions and to provide an adequate social security net.28 Its relationship with the enterprise sector is problematic. It either condones non-market behaviour by some enterprises (subsidies, cheap credits and other rent-seeking activities) or bails out loss-making enterprises (by way of tax arrears, offsets and concessions). In either case, we have evidence of the grabbing-hand or the helping-hand state. This crisis of the state culminated in the August crash of 1998, described by many commentators as the worst economic and political crisis in Russia since the dissolution of the USSR (Illarionov 1999; Malleret et al. 1999; Nagy 2000). A brief account of the August 1998 crisis follows. On July 13, one month before the August 17th débâcle, agreement was reached between the international financial organisations and Russia on a loan package worth US$22.6bn in total, of which US$11.2bn was additional monies agreed by the IMF. For many, this was viewed as both helpful and necessary if a financial crisis was to be averted in Russia. For others, it was seen as yet another bailout, with the inevitable consequences to follow. Critics of the package believed that the terms agreed (fiscal and structural reforms, particularly relating to tax) were unlikely to be met by the Russian Government and Parliament. Of the total, US$4.8bn was disbursed at the end of July, down from the anticipated US$5.6bn. A further $300m was disbursed by the World Bank, under SAL III. Another measure aimed at boosting market confidence and providing some relief to the Treasury was the swapping of short-term rouble-denominated debt for dollar-denominated debt (Eurobonds). The initial reaction to the package was positive, with a fall in interest rates and a rise in share prices. As we know, this was short-lived. In its short period in charge, Kiriyenko’s Government concentrated on the fiscal system, and in particular on reducing the budget deficit. By 1998, the federal budget deficit had been reduced, with current revenue covering non-interest current expenses. However, many of the new Government’s tax collection initiatives were of an administrative nature. The Government did succeed in introducing a much needed new Tax Code but there were delays in enacting Parts II and III. Other measures
27 The author acknowledges the substantial Russian-language literature that is available on this vexed topic. The literature used for this research is in the English language and taken from Western journals. A small number of the authors are Russian. 28 Given the theory of the SBC and the premise made in this chapter, that is, of a weak state (particularly as creditor), it is interesting to note that Gunnar Myrdal, when writing about the developing countries of South Asia, describes those states that fail to enforce basic obligations as ‘soft states’ (Myrdal 1968).
Tax, Transition and the State: The Case of Russia
31
were inevitably rejected by the populist communist-dominated Duma. Some of the remaining proposals had to be passed by presidential decree. By late summer as market sentiment deteriorated further, net financing from treasury bills issues was negative i.e. the Government was making net repayments of treasury bills. The Government could no longer roll over treasury bills as holders demanded payment on redemption dates. Matters were made worse by the fact that external reserves continued to fall, forcing the CBR to draw down the IMF monies in order to support the much-maligned rouble. Eventually, the Government could no longer meet its financial obligations. On August 17th, the Government conceded. In essence, there were three separate announcements. First, a change in the exchange rate bands was revealed. There was to be a new ‘currency corridor’. The existing currency bands (+/-15 per cent) with upper and lower limits of 5.27 roubles and 7.13 roubles respectively around a central parity value of 6.2 roubles per dollar were to be replaced with wider limits of 6.0 roubles and 9.5 roubles respectively. This was a de facto devaluation of the lower band of 33 per cent. Second, a 90-day moratorium on private debt repayments to foreign creditors was announced. Although described by the IMF as a ‘temporary restriction on capital payments abroad’, it amounted to a unilateral default on private foreign debt. Third, a mandatory debt conversion of GKOs and OFZs maturing on or before 31 December 1999 into longer-term maturity debt instruments and a suspension of trading in the domestic Treasury bill market was announced. This amounted to a unilateral rescheduling of rouble-denominated government debt and its aim was to give the Government some breathing space in its attempt to meet its obligations. The Government also announced that certain restrictions on foreign exchange operations were to be imposed. The combined decision to devalue and default, prompted by an unsustainable fiscal position and excessive public sector debt, brought an abrupt end to Russia’s commitment to low inflation, to be achieved by a mix of (possibly prolonged) tight monetary policy and an (seemingly inflexible) exchange rate regime. In this author’s view, the August 1998 crisis was a fiscal crisis that had its origins in a weak and ineffective state. As in other countries where regime changes have arisen from major economic and political crises, it was hoped that the events of August 1998 and the subsequent changes in government might provide Russia with the opportunity to rebuild an effective and well-functioning state. 2.6 What is to be Done? Despite the doom and gloom preached by others, economic reforms in Russia have made some progress, albeit uneven. The same assessment cannot be made in relation to political and institutional reforms. These are the areas that the government most needs to focus on. Further reform of the political and electoral process and of government institutions is required, recognising the need for greater competition, democracy and transparency. Despite local, regional and national elections in the past decade
32
Transition, Taxation and the State
or so, the political reforms that took place in the last days of the Soviet Union and in the early days of the market reform period were never fully completed. The party system, as exists in democracies worldwide, has not been fully developed in Russia. This is particularly true at subnational level where local membership is low. Moreover, there are various aspects of the government and its apparatus that need to change. Russia’s postsocialist political governance and institutions are, in one observer’s opinion, not so different from the central authorities that existed under the Soviet system (Shevtsova 1999). The structures, organisations and the personnel of government, the relationship between the federal, regional and local governments, and the functions of government need to be redefined and reformed in order for corruption and pilfering to decline, economic reform to continue and public support to be restored. Although this redefinition will require a reconstruction of government in many respects, it is imperative that the government fulfils its expenditure obligations and contracts. Institution building must continue, recognising that institutions take time to develop. Above all, many elements of the modern state – the judiciary and the legislature, for example – need to be strengthened. Among other things, this will require a meritocratic bureaucracy insulated from political pressures and interest groups. A well-functioning, well-financed public administration would serve Russia well. As we know from the experience of other countries, the fight against corruption requires a mixture of policies. More legal reform, greater competition and transparency, an improvement in accounting standards, a more simple, less burdensome and more transparent tax system, less discretionary power for state officials and more deregulation (of government controls, licenses and permits) are required. The building of public institutions alongside the promotion of civic groups can help in the fight against corruption. On the taxation front, there are several issues that need to be addressed. With respect to tax reform, tax rules need to be transparent and not subject to arbitrary changes or political interference. There needs to be a reduction in the number of taxes. Loopholes and exemptions need to be tightened. Tax administration structures need to be improved and strengthened (see Chapter Five). As regards taxation and fiscal federalism, the bargaining and open negotiation between the various different levels of government must be avoided. Tax (and expenditure) assignments, tax sharing and intergovernmental transfers need to be better designed, largely in accordance with general public finance principles, to improve accountability, autonomy and, in the context of subnational SBC, fiscal discipline of regional and local governments. More generally, in order for tax compliance to improve there needs to be a change in the economic and fiscal incentives faced by taxpayers, subnational governments and the tax officials of the regional and local branches of the Ministry of Taxation and Fees (formerly the STS). Rather than aggressively pursuing the small and mediumsized businesses as is commonly practised by tax officials, the tax authorities need to use their resources in the pursuit of the large, politically-connected enterprises that are behind in their tax payments. Many of these enterprises either belong to one of the powerful financial-industrial groups (FIGs) or are part of an oligarch empire.
Tax, Transition and the State: The Case of Russia
33
These reforms will help to change the perception that the Russian tax system is too complicated, highly distortionary, rife with delinquency and grossly unfair.29 In the short term, the federal government in Russia can restore political support, legitimacy and credibility by committing itself to these political, tax and institutional reforms. In the medium to long term, attention must focus on the provision of adequate public services, policies to tackle issues of unemployment, inequality and poverty and a commitment to a stable, strong but limited state with a well-defined capacity and set of responsibilities.30 We conclude this chapter by returning to Schumpeter’s writings and, in particular, his paper on the crisis of the tax state. He wrote that the ‘…collapse of Russia is a very special case which does not belong here…The fiscal collapse in Russia, too, was only a consequence of the anti-capitalistic will.’ Although this refers to a different time in Russian history, it is equally appropriate now. The fiscal collapse in Russia at the end of the 20th century may not have arisen from an ‘anti-capitalistic will’, but rather from a failure to realise that capitalist-oriented economies require capitalist-oriented (tax-collecting) states. Remember it was Schumpeter, a supporter of capitalism and capitalists, who (inaccurately) predicted the demise of capitalism and the rise of socialism.31 Despite the setbacks of the past 15 years, Russia and other ex-socialist countries have indeed experienced, in a reasonably short period, a similar transition. But in this case, the transition has been from socialism to capitalism. One feature of this transition has been the emergence of new channels of support through which the state, in Russia and elsewhere, bails out loss-making enterprises. We return to the concept of the SBC and its various instruments in the next chapter. Appendix A: Institutional Reforms The development of institutions that support a democratic, market economy is central to a successful transition in the long run (North 1990). In this context, transition can be viewed as the process of institutional change and adaptation. These institutions, whether they be legal, fiscal, financial, administrative, commercial, social or political are necessary in order for a market economy to flourish. Since the early years of transition, the EBRD has reported annual cross-country progress in reform using transition indicators. These indicators measure advances in institution building, with respect to markets, enterprises and financial institutions. 29 Many of the tax reforms outlined above have been implemented since 1999. For a recent account of these reforms, read IMF (2002). For an account of Russia’s tax collection problems in the wider context of political constraints and postsocialist state building, read Easter (2003). 30 The recent improvement in Russia’s economic performance since 1999 does not, in the author’s view, change either the basic premise or the lessons outlined in this chapter. 31 Schumpeter’s prediction for Russia (notwithstanding the 70 odd years of the administrative economy) has proven more accurate. He wrote ‘…the return to the tax state is certain particularly in the Russian case’ (Schumpeter 1918).
Transition, Taxation and the State
34
Using these transition indicators, we can attempt to measure the extent of institutional building. In the Transition Report 2000, progress in institutional development is measured as the unweighted average of the transition indicators for the following: competition policy; enterprise reform and corporate governance; banking sector and non-banking financial institutions. We report the year 2000 scores for a selection of transition countries, including Russia, in Table A.1.
Table A.1 Development of Market-Supporting Institutions, 2000 Transition Country Albania Croatia Czech Republic Estonia Georgia Kazakhstan Moldova Poland Russia Ukraine Turkmenistan
Institution Index 1.9 2.7 3.2 3.1 2.0 2.2 2.1 3.3 1.9 2.1 1.0
Source: EBRD 2000.
We can see from the table that Russia’s progress in building institutions that support a well-functioning market economy is poor, compared with institutional reform achieved by the leading transition economies, namely Czech Republic, Estonia and Poland. Moreover, Russia has achieved no more, and in some cases slightly less, in terms of institutional development than some of the so-called laggard reformers, namely Georgia, Moldova and Ukraine. Although the demand for supporting institutions is generated by markets, the supply of institutions depends on a state’s capacity. In Russia, a more effective state is needed in order, among other things, to support and develop these market institutions. Appendix B: Governance, State Capture and Intervention It is often said that Russia’s transition to a market economy has been a story of poor public governance, entrenched interests and bureaucratic intervention. In the Transition Report 1999, the EBRD published cross-country ratings for all of these three aspects of transition, based on an EBRD/WB Business Environment and Enterprise Performance Survey (BEEPS) conducted on over 3,000 enterprises in 20 transition countries, including Russia.
Tax, Transition and the State: The Case of Russia
35
The quality of governance is based on the extent to which the state, as viewed by businesses, provides a number of basic services in an efficient and non-discriminatory manner. These services are macroeconomic stability, infrastructure, law and order and a fair tax and regulatory environment. Of the 20 countries surveyed, Russia’s governance index, with a score of 1.16, was the fourth worst, ahead of only Romania, Kyrgyzstan and Moldova. The EBRD’s indicator of state capture – a form of grand corruption - measures the extent of undue influence on government policy-making by powerful vested interests. In the survey, firms were asked to assess the impact on their business of the sale of parliamentary legislation and presidential decrees to private interests. In Russia, over 40 per cent of firms surveyed reported significant impact from state capture. When measuring the extent of firms’ influence on government policy-making, the results for the regional government level in Russia (nearly 10 per cent of firms compared with the national figure of 5.5 per cent) are even more indicative of the enterprise sector’s capacity to influence government decision–making. Interestingly, it appears that for the transition countries the quality of governance and the extent of state capture are highly correlated. Higher levels of state capture are strongly associated with poorly governed states. Russia, classified as a high-capture economy, is a good example of a poorly governed transition country where the state has been ‘captured’ by powerful vested interests.
Table B.1 Governance, State Capture and Intervention Transition Country Azerbaijan Belarus Bulgaria Hungary Kyrgyzstan Lithuania Russia Slovak Republic Slovenia Uzbekistan Average (20), unweighted
Governance indexa
State captureb
Intervention indexc
1.53 1.57 1.38 1.98 0.85 1.54 1.16 1.65 1.95 1.83 1.47
59.21 12.5 38.24 13.79 26.67 17.14 40.87 20 8.26 6.85 25.86
17.7 42.5 16.1 41.8 24.1 21.2 18.2 43.4 26.5 29.6 23.4
Notes: a. An average of the ratings given by firms across nine key functions, with the index ranging from a low of 0.82 for Moldova to a high of 1.98 for Hungary. b. The percentage of firms reporting that the sale of parliamentary votes or presidential decrees had a significant or very significant impact on their business. c. The average across seven dimensions of intervention of the percentage of firms reporting intervention ‘sometimes or more frequently’, with the index ranging from a high of 43.4 for the Slovak Republic to a low of 8.6 for Armenia. Source: EBRD 1999; Hellman and Schankerman 2000; EBRD’s Office of the Chief Economist.
36
Transition, Taxation and the State
As for the level of state intervention, firms in the survey were asked whether their business operations were subject to intervention. Enterprise decisions relating to pricing, investment, employment, sales, wages, mergers/acquisitions and dividends were considered. The index was calculated as the average across all the aforementioned dimensions of intervention of the percentage of firms reporting intervention ‘sometimes or more frequently’. The results indicate a wide diversity across the transition countries both in terms of the level of intervention by the state and in terms of the types of decision-making in which the state intervenes. The level of reported state intervention in enterprise decisions in Russia, at 18.2, is low compared to many other transition countries including, surprisingly, the leading transition countries. The suggestion here is that it may be the nature and quality rather than the level of intervention that is important for transition countries. The ratings for a selection of transition countries, including Russia, are reported in Table B.1. The average for the 20 transition countries in the survey is also included. Appendix C: The Small Business Sector in Russia According to the Federal Law No. 88-ФЗ of 14 June 1995 ‘On State Support for Small Business in the Russian Federation’ the small business definition comprises (i) small enterprises – legal entities (ii) farms and (iii) individual entrepreneurs without legal status. Using the number of persons employed as the criterion, the definition of a small business in Russia is different to the standard definition used by the European Commission. According to the European Commission definition, a small enterprise (as opposed to the broader term of a SME, defined as enterprises which have fewer than 250 employees) is one whose workforce is less than 50 employees. In the Russian definition of a small enterprise, the maximum number of employees is 100 for industry, building and transport, 60 for agriculture, science and technological fields, and 30 for retail trade and consumer services.
Table C.1 Small Business Sector in Russia, 1998-2000 (000s) Small Enterprises Farms Individual Entrepreneurs*
1998 868 270 3,141
1999 891 261 3,593
2000 879 262 3,865
Note: * From the Ministry of Taxation and Fees. Source: Малое Предпринимательство в России 2001, Госкомстат России.
For a number of well-documented reasons (see below), the number of small businesses in Russia is both relatively low and underdeveloped. In the late 1990s, according to official statistics, Russia’s small businesses accounted for 11 per cent of GDP, as compared with levels close to 50 per cent for many market economies. The
Tax, Transition and the State: The Case of Russia
37
percentage of the population employed in these businesses is estimated at 13 per cent, well below the 50–70 per cent levels observed in many developed market economies (Goskomstat, OECD). Even if we allow for the number of small businesses that may be operating in the informal economy and include an estimate for the number of individual entrepreneurs (4.2 million in Jan. 2001, according to Goskomstat) the total figure for small businesses is still below the average for mature market economies.Table C.1 reports data for small businesses in Russia for the period 19982000. Some of the main barriers to small business development in Russia are as follows: Tax system For a vibrant, flexible small business sector to develop, it is important that the tax system within which the sector operates does not act as an obstacle. In market economies, tax systems tend to favour small businesses by offering tax breaks either in the form of low tax rates, rebates, allowances or other forms of assistance. In Russia, the reverse has often been the case, with concessions granted to large enterprises. As for the tax system itself, tax rates tended to be high, there were a plethora of different taxes to be paid (to federal, regional or local government) and, often, it was the case that the normal allowances applicable to small firms in other countries did not apply in the Russian case. Although the more recent reforms, in the guise of the new Tax Code, have improved the situation for the small business sector, there are still problems with the large number of taxes to be paid and the complicated tax administration procedures. Bureaucracy and corruption It is widely known and recognised that Russia suffers from high levels of bureaucracy and corruption.32 With respect to the small business sector, state intervention is often excessive and costly. The legislative and bureaucratic environment hinders small business development. The form of this intervention varies, from the burdensome registration and certification requirements and the excessive number of license regulations to the frequent and arbitrary nature of inspections by tax, fire and safety
32 Transparency International ranked Russia 79th out of 91 countries in the Corruption Perceptions Index (CPI) for 2001. The CPI measures, on the basis of surveys, the perceptions by business people and country analysts of the degree of corruption among public officials and politicians. Although Russia’s relative position improved in 2002 (71 of 102 countries), due mainly to a number of reforms aimed at reducing corruption, Russia has, in the words of the Chairperson of Transparency International, ‘…a long way to go and remains seriously corrupt….’(http://www.transparency.org/pressreleases_archive/2002/2002.08.28.cpi. en.html). The Heritage Foundation’s Index of Economic Freedom for 2001 ranked Russia 127th out of 155 countries. In 2002, Russia received a rank of 131, with no change in its overall score from 2001, of 3.7 (http://www.heritage.org/research/features/index).
38
Transition, Taxation and the State
officials who often use their position to extract bribes from businesses.33 As the state continued to weaken throughout the 1990s in terms of the provision of basic public services, the gap was often filled by the private sector and, within that, the private protection organisations. In return for regular payments, these competing groups would offer certain services to businesses, ranging from debt collection and contract enforcement to security protection. Possible causes of corruption in Russia include the socialist and postsocialist legacy, low levels of economic development and the limited exposure to democracy and the rule of law. This combination of excessive bureaucracy and corruption contributed to a rise in the cost of business for the small business sector. The result has often been the demise of many small businesses or, in many other cases, the decision of many firms to go underground. Banking system In a mature market economy, banks act as financial intermediaries, channeling funds from lenders to borrowers. For start-up businesses and new private firms, the banking sector is a very important source of finance. This is particularly true when alternate sources, such as venture capital and equity shares, are not fully developed. In Russia, there are many problems with the banking system.34 With respect to the small business sector, banks have often tended to ignore small firms. Although there are over 1,000 banks in Russia, many have tended not to engage in normal banking activities (saving and lending) but rather, in many cases, act as clearing houses for large enterprises or engage in non-lending activities (foreign exchange speculation, for example). When banks have engaged in lending activities in the past, it has often been either to the government (prior to the 1998 Russian crisis) or to large state– owned enterprises rather than to new small, private firms. In the past, collateral or guarantee requirements have tended to be excessive, interest rates have been high and volatile and the public’s confidence and trust in banks has been low (for reasons well documented). For small firms and individual entrepreneurs operating in Russia, family and friends, as opposed to the banking sector, are, often, a more realistic source of finance.
33 For a selection of recent sources on the problems of administrative barriers and Russia’s bureaucratic environment as it affects the small business sector, see Moisseev, Alexei and Roland Nash, ‘Small Businesses in Russia: Little Enterprise and Stunted Growth’ Renaissance Capital (Moscow) 2001; The Study of Administrative Barriers to Investment Executive Summary, World Bank 2002 (http://www.worldbank.org.ru/eng/statistics/ereports); Centre for Economic and Financial Research/World Bank Small Business Surveys 2002, which aim to monitor the deregulation process and, in particular, the administrative barriers to small business development in Russia (http://www.cefir.org/news.html). 34 For the problems in the Russian banking system, see the relevant sections in IMF Staff Country Report No. 02/75 Russian Federation, International Monetary Fund, 2002, and OECD Economic Survey of the Russian Federation, Organisation for Economic Co-operation and Development, 2002.
Tax, Transition and the State: The Case of Russia
39
Economic conditions and low levels of domestic demand Favourable economic conditions are considered necessary for self-employment and the small business sector to develop. Unfortunately, less than a favourable market environment has tended to be the case in Russia since market reforms began. It is estimated that, in the year 2000, the level of real GDP in Russia was 63 (1989 = 100).35 It is difficult for firms to operate, let alone take investment decisions, in an environment when real wages are falling, unemployment is rising and domestic demand is stagnant (pre 1999–2004 recovery). For most of the decade, Russia was either in a (transformational) recession or, in the few cases when economic growth was recorded (1997, 1999), it was export-led growth, arising from high oil prices and a rouble devaluation, that was primarily responsible.36 In such circumstances, it is difficult to imagine how a new emerging small business sector can develop, let alone flourish.
35 Annex 3.1 of Transition Report 2001, EBRD, 2001. 36 Various quarterly and monthly updates of Russian Economic Trends, Russian European Centre for Economic Policy (RECEP), Moscow.
This page intentionally left blank
Chapter Three
Budget Softness and the Russian Enterprise Sector 3.1 Introduction In the original analysis of the SBC, paternalism is the key determinant; the preferences of a paternalistic state are to bail out chronic loss-makers (Kornai 1980, 1992a). As the literature on the SBC has grown, so have the explanations for the SBC. Economic, political and institutional reasons have been given to explain the emergence (and persistence) of the SBC (Shleifer and Vishny 1994; Dewatripont and Maskin 1995; Skoog 2000). In this chapter, the interpretation of the SBC syndrome is similar to Kornai’s definition, namely, the financial indiscipline of firms and, in particular, the recurring expectation of a bailout or refinancing of loss-making enterprises or organisations. Loss-making enterprises or organisations get bailed out, ex post, and more often than not, it is the state that is the source of the budget softness. Using a survey of privatised Russian enterprises covering the period 1997–1999 and applying appropriate econometric techniques to the data, we set out in this chapter to report evidence relating to the financial discipline, or indiscipline, of Russian enterprises. From various empirical studies on the SBC phenomenon, for Russia and elsewhere, it appears that there has been a trend, despite considerable cross-country differences, towards a hardening of the budget constraint since transition began over fifteen years ago (Schaffer 1998; Kornai 2001). As for the form of the SBC, arbitrary pricing and budgetary subsidies have been replaced by new, and often implicit, ways of extracting financial aid, including tax arrears and offsets, utilities arrears, nonpayment and, in some cases, barter. In this chapter we examine various instruments of the SBC, as exist in postsocialist Russia. In addition to using recent data that have not been used before for this purpose, the study also covers the period of the 1998 crisis in Russia when there were reports, in the media and elsewhere, of a non-payments crisis and a general deterioration in payments discipline. An alternate hypothesis is that, despite the August 1998 crisis, enterprises and the banking sector still managed to avoid imposing SBC on each other. With respect to financial discipline during this time, the behaviour of the state may have been more problematic. In this chapter we examine three categories of creditors (banks, suppliers and the state) and the different ways (bank loans, trade credit arrears, budgetary subsidies and tax arrears) in which the budget constraint might be softened.1 A fourth possibility 1
A very short review of the literature on arrears is given in De Rosa (2001).
42
Transition, Taxation and the State
that is often alluded to in the transition literature relating to CIS countries is workers, and wage arrears (Alfandari and Schaffer 1996). As before, any evidence or source of budget softness may not lie with the enterprise sector or the banking sector in the form of overdue payables to suppliers and banks respectively, but with the state (and regional government, in particular) in the form of overdue tax debts. The purpose of our analysis is to investigate these possibilities. This is the outline of the chapter. In the next section, we provide a rationale for the sample that was chosen, followed by a description of the survey data. In Section 3.3, the findings of our analysis are presented. The analysis is divided into three parts. The first part examines the ex ante supply of credit from the banking sector and the government. The second part investigates temporary shortfalls in working capital and the sources of finance available to enterprises. An analysis of the actual levels of and real changes in debts and overdue debts forms the third part of our findings. The last section of the chapter contains the main conclusions that arise from our results. 3.2 The Sample and Survey Data2 As part of a project on enterprise restructuring and capacity utilisation in Russia, a survey of 437 enterprises was undertaken in Spring 2000. In view of the overall purpose of the project, which was to analyse the relationships between performance, ownership, competition, corporate governance, restructuring and finance for privatised enterprises, the sample comprises enterprises with a number of specific characteristics. These characteristics, described below, relate to industry, size and region. The enterprises were randomly sampled from a population of firms (stratified by industry, size and region) included in a large database of Russian enterprises.3 The period covered in the questionnaire survey was 1997–1999. As regards industry, only firms from manufacturing industries were included in the sample. This meant that enterprises in extractive industries (which includes Ferrous and non-Ferrous metal industries), energy (electricity, petroleum, gas and coal), construction, trade and services sectors were excluded. More specifically, the sample was confined to enterprises that belong to manufacturing industries according to the Russian Industrial Classification (two-digit OKONKH codes). The industries selected were Chemical and Petro-chemical, Machine Building, Wood & Paper, Stone & Clay (Production of Building Materials), Light industry and Food processing. With respect to size (number of employees), firms with employees of between 100 and 5,000 were included. Firms with less than 100 employees were excluded for a number of reasons. For one, as the focus of the project was on the performance of 2 A more detailed account of the survey methodology is given in Angelucci et al. (2002). 3 The initial list of firms from which the sample was drawn was based on the Goskomstat Enterprise Registry (of about 30,000 medium and large industrial enterprises) data included in the ALBA-Y database.
Budget Softness and the Russian Enterprise Sector
43
privatised enterprises, it meant the exclusion of de novo private firms (which tend to be small) and of small firms that emerged post 1992 as spin-offs of larger enterprises. Two, as small firms in Russia are treated differently in respect of tax and accounting matters, a meaningful comparison with other firms is difficult. Three, such firms account for a very small percentage of industrial output in Russia. Enterprises with over 5,000 employees were excluded from the sample as they tend to operate under conditions that are quite different to the market environment in which the vast majority of firms operate. Three broad size categories were chosen: 100-500 employees, 501-1,000 employees and 1,001-5,000 employees. The average size, in terms of the number of employees, of the surveyed enterprises for 1999 is 891. With respect to the three size categories chosen, the average size is 265, 703 and 1,820 employees respectively. Given the limited number of enterprises in the sample, it was decided that the selected regions would be chosen from four macro-zones, namely, European Russia (Central zone), the Volga, the Urals and Siberia. Moscow city, Moscow oblast, St. Petersburg city and Leningrad (St. Petersburg) oblast were the regions chosen in European Russia. The Volga was represented by Volgograd oblast, Samara oblast and Nizhny Novgorod oblast. Perm oblast, Chelyabinsk oblast and Sverdlovsk (Ekaterinburg) oblast were chosen from the Urals and in Siberia, Omsk oblast, Novosibirsk oblast and Krasnoyarsk krai were chosen. Of the 13 regions, one in each of the four macro-zones has only a small number of observations due to its ‘reserve’ status, i.e. reserved for use if the size and industry quotas could not be met in the other regions belonging to the respective macro-zones. These four regions are marked with an asterisk in Table 3.1. As regards the corporate status of the enterprises at the time of the survey, 306 were joint-stock companies and 113 were closed corporations. Of the remainder, 16 were limited partnerships and 2 were production co-operatives. These specific features of the surveyed enterprises meant that a number of other enterprise-types were excluded from the sample, including fully state-owned enterprises, de novo private firms, enterprises from the military-industrial complex and enterprises that were undergoing bankruptcy proceedings at the time of the survey. We include a short note here on ownership and corporate governance. With respect to ownership, we distinguish between three main categories of owners: insiders, outsiders and the state. Insiders are the workers and managers, whereas Russian enterprises, Russian banks and financial companies (including FIGs), foreign firms and foreign banks constitute the outsiders. The state comprises federal and regional/ local government. Of a total of 365 enterprises for which we have full information on ownership at the start of the year 2000, 59.5 per cent are majority insider-owned and 30.7 per cent are majority outsider-owned, where majority is defined as the group accounting for 50 per cent or more of shares. Less than four per cent are majority state-owned (despite the ex ante restriction, a few such firms crept in) with the remaining 6.3 per cent of firms having no single category with a majority of shares. We also have information on the composition of the board of directors at the time of the survey: of a total of 393 enterprises, about a third are manager-controlled (that is, representatives of managers comprise a majority of seats on the board),
44
Transition, Taxation and the State
workers and outsiders each have a majority in about a quarter of firms; almost a fifth of firms have no single category with a majority of seats. As for the age of the enterprise, the project’s purpose, namely the analysis of privatised enterprises, limits our sample to only those firms that were in existence before 1992 (the year commonly used as the start date for transition in Russia). Of the surveyed enterprises, 175 were privatised in 1992, 117 in 1993 and 46 in 1994. As for the others, they were privatised either before 1992 or after 1994.
Table 3.1 Distribution of Firms (437 firms) Industry Chemicals Machine-building Wood & paper Stone & clay Light industry Food processing Size Observations (1999) Region Central zone Moscow oblast Moscow city St. Petersburg city Leningrad oblast*
Observations 56 108 66 72 72 63 100-500 147
% 33.6
501-1000 139
% 31.8
Percentages 12.8 24.7 15.1 16.5 16.5 14.4 1001-5000 151
% 34.6
No. of firms 122 41 36 30 15
Percentages 27.9 9.4 8.2 6.9 3.4
Volga zone Nizhny Novgorod oblast Samara oblast Volgograd oblast*
115 64 39 12
26.3 14.7 8.9 2.7
Urals zone Sverdlovsk oblast Perm oblast Chelyabinsk oblast*
111 50 43 18
25.4 11.5 9.8 4.1
Siberia zone Novosibirsk oblast Krasnoyarsk krai Omsk oblast*
89 40 37 12
20.4 9.2 8.5 2.7
Note: * = reserve region. Source: Author’s calculations.
Budget Softness and the Russian Enterprise Sector
45
In summary, the sample is reasonably evenly distributed across industries and size classes. In terms of industry, Chemicals and Stone & Clay are over-represented in our sample. Due to the initial selection bias, the average size of the surveyed enterprises is larger than the industrial average. The regional distribution is approximately even across the four macro-zones, with a small number of the thirteen regions therein under-represented (Omsk and Volgograd, in particular). Table 3.1 shows the distribution of firms across industry, region and size. From this description of the surveyed enterprises, it should be obvious that our sample is not, and was not intended to be, representative of Russian industry. The selection of firms was restricted by the project’s objective, namely, the determinants of privatised enterprise performance, and the resultant stratification and sampling criteria applied. This choice of enterprises leads to certain selection biases. For example, our sample is likely to be biased in favour of larger and more profitable enterprises. Nevertheless, we can say that the surveyed enterprises are representative of mid-sized manufacturing firms in Russia’s industrial heartland. In addition, the survey is a useful instrument by which we can investigate the existence and degree of budget softness in the Russian industrial enterprise sector almost a decade after market reforms began. 3.3 Analysis and Findings Managers of privatised industrial enterprises were surveyed using a questionnaire (Appendix D) which dealt with five different aspects of enterprise performance, namely restructuring, competition, ownership, corporate governance and finance.4 In the questionnaire there are a number of questions that relate to the financial discipline of firms and, by implication, budget softness or hardness, whichever may dominate. In particular, managers of firms were asked a number of questions relating to access to bank, equity and state (federal and regional/local) finance. Combining this information with the profit and loss statement/balance sheet data included in the survey questionnaire, we divide our analysis into three parts, reporting the results for the year 1999 only. We begin with the supply of credit from banks and the state, to the enterprise sector. 3.3.1 Supply of Credit Managers of enterprises in the survey were asked about access to short-term and long-term bank credit and to financial assistance from the federal government budget and the regional/local government budget. We use an ordered logit procedure that 4 The questionnaire was developed by the project consortium. The survey data were collected using face-to-face interviews between enterprise managers and staff of the market research company AC Nielsen. Of the 437 enterprises, the respondent was the General Director in 172 cases, the Deputy Director General in 77 cases, the Economics Director in 49 cases and the Finance Director in 40 cases (with others accounting for the remaining cases).
46
Transition, Taxation and the State
has been used before in similar studies involving surveys of enterprises. In earlier studies, Pinto and van Wijnbergen (1994) and Fan et al. (1996) use an ordered logit procedure, with enterprises’ estimates of the ease of obtaining credit as the dependent variable. In the former, managers of Polish state enterprises were interviewed in 1991 and 1992. Profitability, defined as profit over sales, was chosen as the main explanatory variable. With profits positively related to the ease of obtaining credit evident by 1992, it was clear that the bank-enterprise relationship was governed by a hard budget constraint. According to the authors, ‘... commercial banks were letting profitability concerns guide their decision on credit allocation’ (Pinto and van Wijnbergen 1994). The latter study concerns a survey of Russian enterprises that was carried out by the World Bank in 1994. Financial distress, defined as firms that reported ‘usually being loss-making’, was the main explanatory variable. Again, the authors found evidence that loss-makers find it significantly more difficult to obtain short-term bank credit. With creditworthiness being a factor in determining the supply of credit, it appears that ‘…the banking system has some hardness in it’ (Fan et al. 1996). With a polytomous or multiple category dependent variable and where information of an ordinal nature is conveyed, we use an ordered logit procedure. The difficulty of obtaining credit is our dependent variable. We code the different choice categories such that a low value corresponds to credit that is very easily obtained (very easy = 1) and a high value corresponds to credit that is impossible to obtain (impossible = 5). Our main explanatory variable is profitability, measured as after-tax profits over assets, winsorised at 2.5 per cent.5 In our results, a negative coefficient on profitability means that higher profits are associated with credit that is less difficult – easier – to obtain. This would indicate that enterprises are not operating under a SBC. The ‘paternalism’ theory of the SBC implies a non-linear response of creditors to profits – larger losses should imply larger bailouts, but once the firm is profitable, bailouts should cease. An additional possibility is that profitable firms are taxed or are otherwise required to pay more the more profitable they are, a phenomenon Kornai termed ‘levelling’. To allow for the possibility of a non-linear response to profits, we use a quadratic formulation for profit. When estimating the marginal effects we see that the less profitable the firm is, the greater is the impact on the dependent variable of a one-unit change in profitability. For ease of exposition, let us take two values for profit, one at the lower end of the range (say, a -5 per cent profit rate) and one at the upper end of the range (say, a 20 per cent profit rate). In the case of shortterm bank credit, at the negative five per cent profit rate, the marginal impact is -15.3. This compares to a marginal impact of only -4.17 at the profit rate of 20 per cent.
5 In attempting to deal with the problem of extreme values or outliers in our sample, we use a technique referred to as winsorising. By this we mean that observations in the lower and upper tails are identified but instead of removing them, we assign the value of the cutoff (defining the tail) to these observations. For example, if we winsorise at 2.5 per cent, all observations that fall below or above the cut-off point for the lower and upper 2.5 per cent of the distribution are assigned a value equal to the cut-off point. In other words, the five per cent of the sample with the largest and smallest values are affected.
Budget Softness and the Russian Enterprise Sector
47
The impact of an additional unit of profitability is lower at a high rate of profit as compared with a low rate of profit. Aside from profit, other explanatory variables include a size variable (log employment), industry dummies, region dummies and ownership shares. We also include board representation. The results of the ordered logit procedure using the maximum likelihood estimation method are presented in Table 3.3 below. We begin by examining bank credit, both short-term and long-term. Following this, we explore government financial assistance, both federal and regional/local. Short-term and long-term bank credit Table 3.2 reports the survey responses to the question relating to access to bank credit, both short-term and long-term. As expected with the banking sector, longterm credit is more difficult to obtain than short-term credit. Forty four per cent of our firms report that it would be very difficult or impossible to obtain long-term commercial bank lending; the equivalent figure for short-term commercial bank lending is just 18.3 per cent. The various size, region and sectoral dimensions to commercial credit access are discussed in Angelucci et al. (2002).
Table 3.2
Difficulty of Obtaining Bank Credit on Commercial Terms (Percentage of Firms)
Response Very easy Fairly easy Fairly difficult Very difficult Impossible Other
Short-term bank credit 16.7 38.4 23.3 8.9 9.4 3.3
Long-term bank credit 6.1 15.8 28.4 21.1 23.3 5.3
Source: Author’s calculations.
We now use these responses to characterise the supply curve for bank credit. In particular, we are interested in examining the relationship between the financial position of the firm, as reflected in the profitability of the firm, and the supply of credit, reflected in the ease or difficulty of obtaining bank credit. The usual econometric identification problem does not arise here as firms are responding to a question about the availability of credit, and not about whether or not they actually intend to or have borrowed. With these estimates (as opposed to transacted volumes of credit which reflect both the desire of firms to borrow and the willingness of banks to lend), we can be reasonably confident that what is being observed from the data characterises the supply of bank credit. This is important as the phenomenon of the SBC pertains to the supply of credit, or more specifically, the willingness of creditors to supply funds to loss-making enterprises. The regression results are shown in Table 3.3.
Transition, Taxation and the State
48
Table 3.3 Ordered Logit Results for Difficulty of Obtaining Credit in 1999 Dependent Variables Long-term Financial bank credit assistance from federal state budget
Independent Variables
Short-term bank credit
Profit/assets
-13.0*** (2.23) 22.2*** (5.35) -0.198 (0.145)
-10.8*** (2.12) 16.6*** (5.21) 0.157 (0.149)
1.81 (2.35) -2.47 (6.02) -.266 (0.165)
0.456 (2.14) -0.659 (5.39) -0.237 (0.156)
-0.00267 (.00374) 0.00964 (0.0116) χ2 = 8.41 (0.135) χ2 = 24.27** (0.0187) 317
.000880 (0.00378) -0.00476 (0.0114) χ2 = 10.03* (0.0744) χ2 = 24.74** (0.0161) 312
.000986 (.00426) -0.0120 (0.0125) χ2 = 12.89** (0.0244) χ2 = 10.99 (0.53) 318
-.00140 (0.00394) -0.00379 (0.012) χ2 = 12.58** (0.0276) χ2 = 33.68*** (0.0008) 321
not significant
not significant
-15.2 -4.12
-12.5 -4.16
(Profit/assets)2 Size (Log L) Ownership shares: Outsiders State Industry dummies Region dummies Number of observations Marginal impact of profitability evaluated at profit/assets ratio of: -0.05 0.20
Financial assistance from regional/local state budget
Notes: The table reports the estimated coefficients for the main independent variables; the standard errors are shown in parenthesis. For industry and region dummies, the p-values are in parenthesis. The levels of significance are indicated as follows: *** at one per cent, ** at five per cent and * at ten per cent.
In our results for 1999, the coefficients on the two profitability terms for both short-term (column 2) and long-term (column 3) bank credit translate into negative marginal impacts of higher profits on difficulty of obtaining bank credit for virtually the entire range of profitability observed. In other words, larger profits are associated with credit that is less difficult – easier – to obtain. With the evidence indicating that loss-makers are finding it more difficult to obtain bank credit, the implication here is that the average bank is not imposing a SBC on enterprises. This indicates that the supply curve for bank credit is upward sloping, with the supply of credit increasing
Budget Softness and the Russian Enterprise Sector
49
with the profits of a firm. Aside from industry (which matters in the case of longterm credit), the results for all the other explanatory variables are similar for both short-term and long-term bank credit, i.e. size and ownership do not matter; region does matter. When asked about the cost and term structure of a bank loan, the responses of managers were not unexpected. The average expected interest rate, per annum, payable on a short-term bank loan, repayable in less than three months, was 34.7 per cent. As for the length of a bank loan, a large percentage of firms (57.7 per cent) report that they would be unable to obtain a bank loan for more than 12 months duration. In the case of the former, there is a strong regional dimension; in the latter, the raw data indicates a strong industry dimension (Angelucci et al. 2002). Federal and regional/local government assistance This exercise is repeated, with the difficulty of obtaining financial assistance from the federal state budget and the regional/local state budget as the dependent variable. Table 3.4 reports the survey responses to the question relating to access to (any kind of) financial assistance from the federal state budget and the regional/local state budget.6 As expected, it is more difficult to access financial assistance from the federal state budget than from the regional/local state budget. Over sixty per cent of the respondents say that it is impossible to access federal state budget financing, as compared with 45.5 per cent who report that it is impossible to access regional/local state budget financing. Again, Angelucci et al. (2002) report the size, regional and industry dimensions of state financial assistance, both federal and regional/local.
Table 3.4 Difficulty of Obtaining Government Assistance (Percentage of Firms) Response
Federal state budget
Regional/local state budget
Very easy
0.2
0.5
Fairly easy
1.4
1.8
Fairly difficult
11.0
18.8
Very difficult
23.8
31.1
Impossible
60.6
45.5
3.0
2.3
Other Source: Author’s calculations.
6 The different kinds of state financial assistance were not explicitly referred to in this question (q31). They were, however, listed in a previous question (q30). See Appendix D.
50
Transition, Taxation and the State
With respect to the different explanatory variables (with the exception of the region dummies) the results for federal and regional/local state assistance are similar, as shown in Table 3.3. The ordered logit results are reported in Table 3.3. With respect to the impact of profitability, the results for federal and regional/local state assistance are similar: in both cases, the coefficients on profitability are insignificant.7 This indicates that, for our sample, profit has no impact on the ease or difficulty of obtaining government assistance, i.e. profit-makers and loss-makers are equally likely (or unlikely) to obtain government assistance. As the coefficients on ownership are also statistically insignificant, it appears that there is no difference between a privatised enterprise with a small residual state ownership share and one with a large residual state ownership share in terms of the ease or difficulty of obtaining government assistance. Industry matters as our significance tests indicate variation by industry. In contrast, region matters only in the case of financial assistance from the regional/local state budget, as one might expect. In our sample, the regions where it is easiest to obtain assistance from the regional government are Volgograd oblast and Novosibirsk oblast. Moscow oblast and Sverdlovsk oblast are the regions where it is most difficult to obtain assistance from the regional government. The government assistance regressions do not suggest evidence of the classical Kornai state-paternalism SBC, but they are not inconsistent with the other theories explaining government subsidies. The significance of the industry and region dummies suggests regional differences in the softness of subnational governments and industry differences in government ‘tastes’. In a related question, managers of enterprises were asked to specify the individual kinds of financial assistance from the federal and regional/local state budgets that they had received in the preceding three-year period. The kinds of financial assistance listed in the questionnaire are credits, tax holidays, tax exemptions, restructuring of tax debts, direct subsidies and other. Of these types, the two most popular, according to the survey responses, were restructuring of tax debts by federal government (88 firms) and regional/local government (69 firms), and tax holidays from the regional/ local government (82 firms). All other listed kinds of government assistance were less common and, in particular, direct subsidies, with only two (11) firms in receipt of a direct subsidy from the federal (regional/local) government in the previous three years. In sum, the findings from our survey, and contrary to the conclusions by Brana et al. (1999), indicate that the banking system in Russia, on average, is not a source of budget softness.8 The supply curve for bank credit is upward sloping, with respect to profitability, as it normally is in a market economy. As for the state, the evidence is 7 Both separately and (not reported here but included in the regressions) jointly. 8 The inclusion, in the case of Brana et al. (1999), of state-owned enterprises and of data for earlier in the transition era might explain the differences in results between the two studies. In addition, our survey excludes very large enterprises (in terms of employee numbers), often the candidates most likely to be assisted.
Budget Softness and the Russian Enterprise Sector
51
not as clear. Although there is no evidence of budget softness, we cannot infer a hard budget constraint. We find some evidence, however, of the former in the next section. 3.3.2 Financing a Shortfall in Working Capital In our survey, managers of enterprises were asked a hypothetical question relating to temporary shortfalls in working capital and possible methods of finance. The question posed is as follows. Consider the following situation facing your firm. Your incoming cash flow for the quarter is ten per cent lower than you expected. This cash flow is not permanently lost, but merely unexpectedly delayed. However, your working capital needs and level of production remain the same. How would you finance the gap?
In answering the question, managers were asked to indicate their ‘top three or four’ sources and to provide a percentage breakdown of how much each source would contribute to the financing. These sources of finance and their respective percentages (responses of the enterprises’ managers) are listed in Table 3.5.
Table 3.5 Financing a Shortfall in Working Capital Liquidated short-term financial assets Bank borrowing Credit from suppliers Delayed payments to suppliers/utilities Bills of exchange (veksels) Exchange of goods for goods Delayed payments to budget (taxes) and extrabudgetary funds Delayed payments to workers (wages) Subsidy from government Other
Share (%) 0.3 23.8 15.7 13.2 1.4 19.8 12.8 7.9 0.2 5.0
Source: Author’s calculations.
As is evident from the table, bank borrowing, barter (goods for goods) and trade credit from suppliers are the most popular ways of financing a temporary shortfall in working capital, according to the responses given by the enterprises’ managers.9 With respect to bank borrowing and trade credit, this practice is similar to other 9 The popularity of bank credit in financing a temporary shortfall in working capital may come as a surprise. In Russia, it is commonly known that bank lending to the enterprise sector is not very large. However, in our analysis we are dealing with potential rather than actual bank borrowing.
52
Transition, Taxation and the State
transition countries and, indeed, market economies where firms often use trade and bank credit to finance temporary shortfalls in working capital. As for barter (goods for goods), this is not uncommon in Russia (at least as a form of payment in the 1990s) and is well documented in the transition literature (Commander and Mumssen 1999; Seabright 2000). The next most popular sources of finance are delayed payments to suppliers/utilities, tax arrears and delayed payments to workers (wage arrears). These forms of finance, in contrast to those above, are not so common in industrialised countries. The least popular methods of finance, according to the responses given, are bills of exchange (veksels), liquidation of short-term financial assets and government subsidies.10 The five per cent mean value for ‘other sources’ indicates that no important source of finance has been omitted from the analysis. Using ordinary least squares regression analysis, we take profitability, measured as profits over assets, as our main explanatory variable. The other explanatory variables are size (log employment), industry dummies, region dummies and ownership shares. The particular source of finance (in percentage terms), whether it is bank credit, delayed payment of taxes or any of the other sources listed above, is the dependent variable. The regression results are shown in Table 3.6. We summarise the results for a selected number of sources of finance, namely, bank borrowing, trade credit, tax arrears, a budget subsidy and wage arrears. With respect to bank borrowing, the estimated regression coefficients on profitability are significant. Moreover, bank borrowing is positively related to profitability. This indicates that the more profitable the enterprise is, the more it can borrow from banks, all other things equal (in the hypothetical situation where the enterprise is confronted with a temporary shortfall in working capital). As this is what one would expect in a properly-functioning banking sector (where creditworthiness is the primary factor in determining credit) this is an indication that banks, on average, are not a source of budget softness. There is weak evidence of a size effect, namely, the bigger the enterprise is, the more it can borrow from banks to finance a shortfall. There is also variation in industry, indicating that industry matters with respect to bank credit. As for trade credit from suppliers (late or otherwise), the results indicate no difference between profit-makers and loss-makers, i.e. the coefficients on profitability are insignificant. Profit- and loss-makers rely on (overdue) trade credit to the same extent. This is an indication that suppliers are not a source of budget softness.
10 I wish to thank a colleague who pointed out that size matters in the case of veksels, i.e. larger firms tend to use veksels more. Also, many firms are more likely to trade (3rd party) veksels rather than issue them.
Table 3.6 OLS Regression Results for Financing a Shortfall in Working Capital Independent Variables Profit/assets (Profit/assets)2 Size (Log L) Ownership shares:Outsiders State
ST assets
Trade credit
.180 (3.6) 7.46 (8.96) -.184 (.244)
Bank credit 78.1** (36.7) -150* (91.4) 4.25* (2.49)
-.00919 (.00636) -.00207 (.0146) F=0.72 (.6064) F=3.86*** (.0000) 0.187
-.0943 (.0648) -.247* (.149) F=3.78*** (.0025) F=.46 (.9387) 0.133
.0634 (.0477) .0295 (.11) F=.78 (.5639) F=1.88** (.0364) 0.101
-18.8 (27.1) 43.6 (67.3) -1.16 (1.83)
Dependent Variables Trade Veksels arrears 15.0 -7.43 (23.5) (9.01) -21.4 16.2 (58.5) (22.4) .904 .327 (1.59) (.611)
.0312 (.0415) .0106 (.0955) F=1.46 (.2038) F=1.49 (.1275) 0.095
-.0256 (.0159) -.0447 (.0366) F=1.73 (.1273) F=1.51 (.119) 0.100
Barter -5.89 (32.0) -16.8 (79.5) -1.09 (2.17)
Tax arrears -41.6* (22.2) 69.0 (55.1) -1.59 (1.50)
Wage arrears -42.6** (17.6) 90.2** (43.8) -.130 (1.19)
Budget subsidy -3.78 (2.41) 4.98 (6.00) .112 (.164)
-.0197 (.0564) .0353 (.13) F=1.82 (.1085) F=.67 (.7804) 0.067
.0233 (.0391) .167* (.09) F=1.68 (.1389) F=.58 (.8616) 0.096
-.0279 (.031) -.0105 (.0715) F=1.46 (.2024) F=1.63* (.0821) 0.129
-.00534 (.00426) -.00245 (.00981) F=.35 (.8825) F=.97 (.4815) 0.062
Industry dummies Region dummies R2 Marginal impact of profitability evaluated at profit/assets not not not not not ratio of: not significant significant significant significant significant -41.6 -51.6 -0.05 significant 93.1 (linear) -6.5 0.20 18.1 Notes: The table reports the estimated coefficients for the main independent variables; the standard errors are shown in parenthesis. For industry and region dummies, the p-values are in parenthesis. The levels of significance are indicated as follows: *** at one per cent, ** at five per cent and * at ten per cent. The number of observations in each case is 307.
54
Transition, Taxation and the State
The results for tax arrears are quite different and do suggest some evidence, albeit weak, of budget softness on behalf of the state. The coefficient on profitability (but not profitability squared) is weakly significant and suggests a linear negative impact; the larger a firm’s losses, the more likely it is to delay tax payments in order to finance its working capital. Interestingly, as the coefficient on state ownership share is both positive and (weakly) significant, it appears that privatised enterprises with a large residual state ownership share are more likely to delay payments to budgets (taxes) and extrabudgetary funds than enterprises with a small residual state ownership share. The results for a direct government subsidy are different. The coefficients on profitability are insignificant, implying that there is no difference between profit-makers and loss-makers with respect to the use of budgetary subsidies to cover temporary shortfalls in working capital. According to the results from our sample, toleration of tax arrears rather than explicit government subsidies is the instrument used by the state when it softens the budget constraint of the enterprise sector. With respect to wage arrears, the coefficients on profitability are significant. In a situation of a temporary shortfall in working capital, loss-making enterprises are more likely to delay payments to workers than are profit-making enterprises (but with diminishing marginal effects). As in the case of tax arrears, wage arrears appear, from our results, to be a route through which budget constraints are softened. 3.3.3 Debts and Overdue Debts Unlike in the previous two subsections where we were working with enterprises’ estimates, in this subsection we consider real transactions and the resulting volumes of and changes in debts and overdue debts. We examine levels of and changes in debts and overdue debts separately, for all three categories of creditors.11 In our regressions, all dependent variables are winsorised at 2.5 per cent. The means of these dependent variables, normalised by end-year total assets, are presented in Table 3.7. Although the levels of debts are high in some cases, notably tax debts and payables to suppliers, the real changes in the levels of debts do not appear to be very large. We note, in particular, that overdue tax debts are the single largest category of overdue debts. However, as the inflation rate was quite high during this period, in the order of 36.8 per cent (CPI, end-year), and where debts are not sufficiently indexed, inflation may have been wiping out some of these debts (see Chapter Four for more on this topic). We begin by examining levels of debts and overdue debts to the three categories of creditors.
11 In respect of the SBC syndrome, it is important to distinguish between stocks and flows of debts. This distinction is explained, in the context of tax debts, in Chapter Four.
Budget Softness and the Russian Enterprise Sector
Table 3.7
55
Levels of/Changes in Debts and Overdue Debts (Percentage of Assets) Mean (% share of assets)
Levels: Tax debts of which, overdue Payables to suppliers of which, overdue Payables to banks of which, overdue
12.2 7.5 13.0 4.2 3.2 0.1
Real changes: Tax debts of which, overdue Payables to suppliers of which, overdue Payables to banks of which, overdue
-1.6 -1.1 -1.5 -0.8 0.5 -0.1
Source: Author’s calculations.
Levels of debts and overdue debts Here, we examine levels of debts and overdue debts to the various creditors, that is, government, banks and suppliers. In particular, we are interested in establishing if the stock of debts and overdue debts is related to profitability, measured as profits over assets. Are levels of debts and arrears associated with (low) profitability? The appendix to the survey questionnaire (not reprinted here) provides balance sheet data and, in particular, the end-of-year levels of debts and overdue debts, to all creditors. We note here that tax debts are defined as the sum of payables to regional/local and federal budgets and payables to extrabudgetary funds. In our regression equations for 1999, shown in Table 3.8, the dependent variables are the stock of debts and overdue debts, normalised by total assets. We estimate the equations for all three categories of creditors. As we are interested in how profit affects levels of debts and overdue debts, profit is our main explanatory variable. As in the previous two cases, we have a quadratic formulation for profit. As usual, size (log employment), industry, region and ownership are controlled for. We do not report the results for payables to banks and suppliers.
Transition, Taxation and the State
56
Table 3.8 Regression Results for Levels of Debts and Overdue Debts Dependent Variables Independent Variables Profit/assets (Profit/assets)2 Size (Log L)
Total tax debts
Of which, Tax arrears
Overdue payables to suppliers
Overdue payables to banks
-0.641*** (0.105)
-0.655*** (0.0958)
-0.254*** (0.0655)
-0.0168*** (0.00475)
1.13*** (0.261)
1.21*** (0.238)
0.315* (0.164)
0.0289** (0.0118)
-0.0103 (0.00712)
-0.0108* (0.00643)
0.00791* (0.0044)
-0.000316 (0.000318)
0.000228 (0.000186) 0.00136*** (0.000438) F=3.11*** (0.0094) F=4.04*** (0.0000) 0.345
0.000237 (0.000174) 0.00132*** (0.000407) F=4.58*** (0.0005) F=2.75*** (0.0015) 0.338
-0.0000002 (.000119) 0.000495* (0.000278) F=1.19 (0.3156) F=1.11 (0.3525) 0.185
-0.0000009 (0.0000009) -0.0000186 (0.0000201) F=2.55** (0.028) F=2.09** (0.018) 0.186
318
305
303
302
-0.754 -0.189
-0.776 -0.171
-0.286 -0.128
-0.020 -0.005
Ownership shares: Outsiders State Industry dummies Region dummies R2 Number of observations Marginal impact of profitability evaluated at profit/ assets ratio of: -0.05 0.20
Notes: The table reports the estimated coefficients for the main independent variables; the standard errors are shown in parenthesis. For industry and region dummies, the p-values are in parenthesis. The levels of significance are indicated as follows: *** at one per cent, ** at five per cent and * at ten per cent.
Consider the case of goverment as creditor. The results indicate that the more profitable the enterprise is, the less likely it is to have (overdue) tax debts. The negative coefficients on profitability imply that a loss-maker, as compared with a profitable firm, is more likely to have (overdue) debts to the state. Not surprisingly, this result is repeated for all categories of creditors, although the scale of the impact varies significantly by creditor, i.e. the marginal impact of profitability for tax debts is about double that for trade credit arrears, and 20-40 times that for arrears to banks. Negative and significant coefficients indicate that in all cases, unprofitable firms are more likely, as compared with profitable firms, to have debts and overdue
Budget Softness and the Russian Enterprise Sector
57
debts. Interestingly, in the case of debts to the state (not overdue and overdue), the coefficient on the state ownership share is positive and significant, i.e. firms with a large residual state ownership share are more likely to have tax debts and run up tax arrears than firms with a small residual state ownership share. As for the other explanatory variables, size is weakly significant in the case of tax and trade credit arrears. There are variations in industry and region in all cases except overdue payables to suppliers. Real changes in debts and overdue debts Here, we are interested in the change in debts and overdue debts, i.e. flows of debts and overdue debts. Again, we want to establish whether the change in debts and overdue debts is related to profitability. As before, the survey questionnaire’s appendix provides the end-of-year levels of debts and overdue debts to government, banks and suppliers. If the stocks of debts and overdue debts are revalued using a price index (that is, December 1998–December 1999 CPI), we can calculate the real change in debts and overdue debts for 1999.12 Once again, all our dependent variables are normalised by total assets. We now turn to our explanatory variable. Here we simplify matters by defining profit as a dummy variable, pmaker, where pmaker is equal to 1 if the firm is a profit-maker and is equal to 0 otherwise, i.e. if firm is a loss-maker. This is useful in assessing whether there is any difference between profit-making and loss-making firms with respect to the change in debts and overdue debts. The results of our regression analysis are presented in Table 3.9. In the case of the change in tax debts, the coefficient on profitability is negative and (weakly) significant. In other words, we observe some evidence that on average a firm is accumulating more tax debts if the firm is a loss-maker as compared with a profit-maker. The coefficient on profitability is insignificant for all other categories of changes in debts. Whereas residual state ownership was correlated with larger overdue tax debts, we find no relationship between residual state ownership and changes in tax debts or other categories of debts. The findings here indicate that once a firm finds itself in trouble, it runs up debts to its various creditors. However, once a firm is established as a chronic loss-maker, not all creditors respond in the same way. It appears that trade creditors and banks impose hard budget constraints when faced with financially distressed firms. In contrast, the troubled firm manages to accumulate new tax debts with the state (willingly or not, in the case of the state). Non-payment of new tax obligations is the form that the SBC takes. This observation coincides with findings from empirical studies of enterprises’ tax payments in Russia and other transition countries (Alfandari and Schaffer 1996; Tóth and Semjén 1998). What Kornai observed as early as 1991 in 12 The use of an appropriate price index is an issue here and in the next chapter. Whereas industry or firm prices are more correct when dealing with the enterprise sector, data limitations result in the use of the CPI index. I wish to thank two colleagues who drew this point to the author’s attention.
Table 3.9 Regression Results for Real Changes in Debts and Overdue Debts
Independent Variables
Tax debts
Tax arrears
Pmaker
-.0126* (.00688) .00231 (.00346)
-.00658 (.00596) .00143 (.00287)
.0000292 (.0000882) .0000424 (.000211) F=1.75 (.1232) F=1.60* (.0908) 0.104 316
-.0000664 (.0000758) -.0000641 (.000179) F=2.15* (.0599) F=1.63* (.0840) 0.104 300
Size (Log L) Ownership shares:Outsiders State Industry dummies Region dummies R2 Number of observations
Dependent Variables Payables to Overdue suppliers payables to suppliers -.00705 -.00116 (.0113) (.00611) -.00957* -.00171 (.00561) (.00294)
-.000032 (.000145) .000165 (.000346) F=1.05 (.3909) F=.71 (.7447) 0.055 316
-.000101 (.0000777) -.000187 (.000183) F=.78 (.5677) F=1.10 (.3632) 0.070 298
Payables to banks .0104 (.00687) .00476 (.00341)
Overdue payables to banks -.000122 (.000547) .000468* (.000264)
-.000228*** (.0000878) .000336 (.000211) F=1.69 (.1365) F=1.14 (.3259) 0.112 315
-.000016** (6.99e-06) .0000114 (.0000165) F=2.33** (.0428) F=1.73* (.0597) 0.129 297
Notes: The table reports the estimated coefficients for the main independent variables; the standard errors are shown in parenthesis. For industry and region dummies, the p-values are in parenthesis. The levels of significance are indicated as follows: *** at one per cent, ** at five per cent and * at ten per cent.
Budget Softness and the Russian Enterprise Sector
59
Hungary, namely, unpaid tax obligations, has been a feature of the transition process for many ex-socialist countries.13 This chapter indicates that, despite the progress achieved in more recent years, the SBC did not completely disappear along with the socialist system in Russia. 3.4 Conclusions In this chapter, we set out to find evidence of financial discipline, or indiscipline, among a sample of privatised manufacturing enterprises in Russia. The usual sources or candidates for budget softness are the state, the banking sector or suppliers. In transition countries, budgetary subsidies, tax arrears, overdue payables to banks and to trade suppliers are the usual forms by which the SBC manifests itself. The results of our survey of 437 privatised enterprises in Russia appear to indicate that, on average, neither the banking sector nor trade suppliers are a source of budget softness. When goods are not paid for or when loans are not repaid, the average supplier and the average bank do not continue to assist firms that are in financial trouble. In contrast, subsidies of various forms (particularly tax arrears) from government are not uncommon, and we find weak evidence that these are more likely to be allocated to loss-making firms. Delayed wage payments to workers are an additional mechanism used by firms to finance shortfalls in working capital. Appendix D: Enterprise Restructuring and Capacity Utilisation Questionnaire Table D.1 Extracts from Questions on Enterprise Finance (29)
Imagine your enterprise will have to obtain a bank loan on commercial terms. How easy would it be for your enterprise to obtain a short-term bank loan on commercial terms? A long-term bank loan?
Very easy Fairly easy Fairly difficult Very difficult Impossible Difficult to answer Refuse to answer
Short-term bank credit on commercial terms 1 2 3 4 5 6 7
Q29_1
Long-term bank credit on commercial terms 1 2 3 4 5 6
Q29_2
7
13 In 1992, Kornai wrote that ‘the sum of …(unpaid taxes and social-security contributions) in mid-1991 was greater than the entire budget deficit planned for 1992’ (Kornai 1992b).
Transition, Taxation and the State
60
(30)
In the last three years, has your enterprise received the following kinds of financial assistance from the federal and/or regional state budget in terms of:
Credit Tax holiday Tax exemption Restructuring of tax debts Direct subsidy Other
(31)
Has not received
Difficult to answer
Refuse to answer
3 3 3
4 4 4
5 5 5
Q30_1n Q30_2n Q30_3n
1 1 1
Q30_1_1 Q30_2_1 Q30_3_1
Received from Regional / Local State Budget 2 Q30_1_2 2 Q30_2_2 2 Q30_3_2
1
Q30_4_1
2
Q30_4_2
3
4
5
Q30_4n
1
Q30_5_1
2
Q30_5_2
3
4
5
Q30_5n
1
Q30_6_1
2
Q30_6_2
3
4
5
Q30_6n
Received from Federal State Budget
Imagine your enterprise will have to obtain some financial assistance (out of the list above). How easy would you find it to obtain any kind of financial assistance from the federal state budget? From the regional/local state budget? Federal State Budget
Very easy Fairly easy Fairly difficult Very difficult Impossible Difficult to answer Refuse to answer
1 2 3 4 5 6 7
Q31_1
Regional and Local State Budget 1 2 3 4 5 6 7
Q31_2
Budget Softness and the Russian Enterprise Sector
(37)
61
Consider the following situation facing your firm. Your incoming cash flow for the quarter is ten per cent lower than you expected. This cash flow is not permanently lost, but merely unexpectedly delayed. However, your working capital needs and level of production remain the same. How would you finance the gap? I would finance the gap from the following sources – list top three or four sources only, summing to 100 per cent. Please give their percentage breakdown.
Liquidated short term financial assets Bank borrowing Credit from suppliers Delayed payment to suppliers/utilities Bills of exchange (veksels) Exchange of goods for goods Delayed payments to budget (taxes) and extrabudgetary funds Delayed payments to workers (wages) Subsidy from government Other It cannot be done without reduction in production by the firm Difficult to answer Refuse to answer
1
Q37_1
%
100% Q37p_1
2 3 4
Q37_2 Q37_3 Q37_4
% % %
Q37p_2 Q37p_3 Q37p_4
5 6 7
Q37_5 Q37_6 Q37_7
% % %
Q37p_5 Q37p_6 Q37p_7
8
Q37_8
%
Q37p_8
9 10 11
Q37_9 Q37_10
% %
Q37p_9 Q37p_10
1 2
Q37p_na
Q37n 12 13
This page intentionally left blank
Chapter Four
Tax Arrears and the Romanian Enterprise Sector 4.1 Introduction The relationship between the state and the enterprise sector in the socialist economies of Central and Eastern Europe was characterised by the SBC and financial indiscipline. Loss-making firms were accustomed to receiving aid from the state, in the form of arbitrary pricing and/or budgetary subsidies. As these countries have recently witnessed a transition from a planned to a market economy, enterprises have found new ways of extracting subsidies from what is often a compliant state. One mechanism by which the state channels funds to the enterprise sector, and in the case of the SBC to financially distressed firms, is by means of tax arrears. Tax arrears arise when firms’ tax liabilities are overdue. In this chapter we set out to measure the size of this tax arrears subsidy for one of the slow reforming transition economies, namely Romania. Using a stock/flow analysis on firm-level data (balance sheets and profit and loss statements), we develop a methodology that allows us to measure the extent of the tax arrears subsidy in the enterprise sector. The outline to the chapter is as follows. A brief explanation of tax arrears in the context of economic transition is given in Section 4.2. In Section 4.3, the methodology for measuring the tax arrears subsidy is outlined. Once developed, the methodology is applied to a database of over 9,000 medium and large non-financial firms. The results and policy prescriptions arising from our analysis are presented in Section 4.4. 4.2 Tax Arrears The tax system can be used to soften the budget constraint of firms. Under the socialist system enterprises, in addition to receiving direct budgetary subsidies, bargained for ‘tax’ concessions from the state. The arbitrary nature of pricing and the vertical relationship that existed between the state and the enterprise sector allowed the state to channel funds, through the tax system, to the enterprise. In postsocialist countries, market-economy style tax systems have been introduced. However this has not stopped enterprises continuing to extract funds from the state. One such way is by means of tax arrears. Tax arrears are tax liabilities that are known to the tax authorities, have come due but have not been settled, i.e. they are overdue tax
64
Transition, Taxation and the State
liabilities. As these tax liabilities are known to the authorities, the problem is one of collection rather than detection as in the case of tax evasion. Delayed payment of taxes can provide cheap working capital to firms. In a high inflation environment and assuming relatively low penalties/interest on overdue taxes, the real cost to the enterprise of this breach of obligation is reduced (in the case where the debt is eventually paid; in the case where the liability goes unpaid, the cost is eliminated). In this sense, tax arrears can be viewed as a subsidy from the state to the enterprise. Ultimately, tax arrears might prevent financially distressed firms from bankruptcy and, in doing so, ensure their survival. This is the case of the SBC. In the literature three cases, in general, of (tax) arrears are identified (Perotti 1994; Schaffer 1996; Alfandari and Schaffer 1996). The first case relates to ‘late payment’ by firms. In this case, firms that incur a tax liability do not pay in time but do eventually settle their tax debt. The reasons for the late payment can vary, from short-term liquidity problems encountered by firms, legitimate tax disputes or poor payments discipline. The second case relates to ‘financial distress’ and is analogous to the SBC. In this case, firms are often in severe financial difficulty, resulting in non-payment to creditors. The experience of some transition countries to date indicates that for many firms, the government is viewed as the ‘softest’ creditor. If this is the case, banks and suppliers get paid before the government. When the cost to the firm of late payment of taxes is lower than the cost of alternate sources of funds, the result is a build-up of tax and social security liabilities to the government; often they remain unpaid. Unlike in mature market economies where financially distressed firms turn to the courts for help, in transition countries these firms often remain in business because their creditors, for various reasons, do not pursue them into liquidation. The third case relates to strategic behaviour on the part of firms and is termed ‘collusive arrears’. In this case, tax arrears arise from strategic decisions taken by firms on the basis of likely government actions. For instance, firms may allow tax arrears to build up in the anticipation that a benevolent government will bailout recalcitrant firms. From the evidence on the different forms that the SBC can take, it appears that tax arrears are a major source of fiscal aid to the enterprise sector (Belka et al. 1995; Schaffer 1998; Kornai 2001). In the early years of transition, budgetary subsidies were cut, eventually falling close to levels found in mature market economies; those that remain are largely related to price controls or provision of public services, and, as in the industrial economies, are concentrated in a small number of sectors, namely agriculture, public transport, energy and housing (Schaffer 1995). As prices were liberalised, the ability of enterprises to extract subsidies from the state through price adjustments declined. The creation of commercial banks within a two–tiered banking system changed the relationship that had existed between banks and enterprises. As a result, enterprises were no longer assured of obtaining cheap bank credit. Unprofitable projects were less likely to be refinanced. Despite the rise in inter-enterprise arrears (trade credit arrears), there is evidence that firms largely have imposed hard budget constraints on each other, as judged by the use of credit control and other normal market mechanisms (Bonin and Schaffer 1995; Schaffer 2000;
Tax Arrears and the Romanian Enterprise Sector
65
Santarossa 2001).1 If there is any evidence of softness via inter-enterprise arrears, the source is often the state (in the Romanian case, the Ministry of Finance and/or the National Bank of Romania) and its clearing operations and not the enterprises themselves (see below). Romania is an appropriate transition country in which to investigate the presence or otherwise of tax arrears and, more especially, budget softness. Romania is generally regarded to be one of the laggard reformers in Central and Eastern Europe (EBRD 1999; World Bank 2002). This is evident in most dimensions of transition. Evidence of this slow pace of reform is given in Table 4.1 where the EBRD transition indicators for Romania are reported, for the period 1994–1998. As a benchmark, the average value for the respective transition indicators for the four Visegrad countries is also reported. Across many different dimensions of transition (enterprise reform, privatisation, banking sector reform) Romania’s score is less than the average score for the four Visegrad (rapid-reforming) countries, for the five-year period reported.
Table 4.1
EBRD Transition Indicators for Romania (and Visegrad Four), 1994-1998 1994
Dimensions of Reform Enterprise Reform Small-scale Privatisation Large-scale Privatisation Banking Sector Reform
1995
1996
1997
1998
R
V
R
V
R
V
R
V
R
V
2.0
3.0
2.9
3.0
2.0
3.0
2.0
2.9
2.0
3.0
2.3
3.9
2.7
3.9
3.0
4.2
3.3
4.3
3.3
4.3
2.0
3.3
2.0
3.5
2.7
3.5
2.7
3.8
2.7
3.8
2.0
2.9
3.0
2.9
3.0
2.9
2.7
3.2
2.3
3.3
Note: R = Romania; V = Visegrad countries (Czech Republic, Hungary, Poland and Slovak Republic). Source: EBRD 1999; Author’s calculations.
More specifically, Romania has witnessed severe problems in the area of financial discipline and payments arrears. In response, the Romanian authorities have attempted to tackle the problems (finance and otherwise) of the (state-owned)
1 The rise in trade credit and trade credit arrears in these ex-socialist countries should be treated with caution. For one, as trade credit was not common under the socialist system, the transition to a market economy where trade credit is normal was likely to produce a rise in inter-firm credit. Secondly, the levels of trade credit arrears in transition countries are not altogether different from the levels of trade credit arrears in some Western market economies. Of course, trade credit and arrears can become a problem if chains of interlocking arrears result, often leading to government bailouts.
66
Transition, Taxation and the State
financially-distressed enterprises by use of so-called enterprise isolation or reorganisation programmes in the hope, among other things, of preventing a wide-scale general payments problem (Djankov and Ilayperuma 1997). Although this chapter focuses on overdue tax liabilities (of all enterprises, profitable and loss-making), enterprises in Romania have, throughout transition, accumulated other payment arrears, most notably inter-enterprise arrears and overdue payables to banks. These delinquent firms have often been loss-making firms, with new bank credit being the main channel of budget softness (Claessens and Djankov 1998; Coricelli and Djankov 2001). These other breaches of obligation, trade credit arrears in particular, caused much anxiety in Romania in the early 1990s. With inter-enterprise arrears rising to close to 50 per cent of GDP (and with some firms extending new trade credit in the anticipation of a general government bailout), the government initiated a ‘global compensation’ scheme to clear the financial blockage in the payments system. As the experience shows from other transition countries, it was not long before the inter-enterprise arrears reappeared (Hildebrandt 2002). Finally, in the context of tax revenue and the government budget, it is noticeable that Romania, during the 1990s, had one of the lowest tax revenue shares of GDP for the transition countries of Central and Eastern Europe. Table 4.2 reports the tax revenue/GDP ratio for Romania and other CEE countries for the period 1994–1998.
Table 4.2 Tax/GDP Ratios, 1994-1998 1994
1995
1996
1997
1998
Bulgaria
33.8
30.9
27.7
28.6
31.2
Czech Republic
37.9
37.2
36.6
36.6
36.6
Hungary
41.6
39.4
37.6
39.2
39.1
Poland
38.1
37.3
37.4
36.6
35.9
Romania
28.2
28.8
26.9
26.5
28.2
Slovenia
41.9
41.4
40.6
39.9
40.2
Source: IMF 2000b, 2001b; Author’s calculations.
Revenue erosion, tax collection, compliance and evasion remained a serious concern for the authorities in Romania throughout the 1990s. The remaining part of the chapter deals with the tax collection problem and, in particular, (overdue) tax liabilities of enterprises. Before developing the methodology for measuring the tax arrears subsidy, we need to outline the different types of firms in the Romanian enterprise sector and the composition of the database relating to the different ownership-types. In the Romanian enterprise sector we can distinguish between four categories of enterprises, on the basis of ownership. They are (i) strategic SOEs, known as Régies Autonomes (RA); (ii) state-owned commercial firms; (iii) privatised firms that were once state-owned;
Tax Arrears and the Romanian Enterprise Sector
67
and (iv) de novo private firms. Since the legal status of SOEs was changed in 1990, the number of RAs has fallen significantly. As in other countries, these enterprises are to be found in defence, utilities, mining, railway, telecommunications and other ‘strategic’ sectors. As strategic enterprises, they were not slated for privatisation but were to remain in the public sector. In contrast, at the outset of transition in 1990, there were approximately 6,300 state-owned commercial firms listed for privatisation, under the aegis of the State Ownership Fund (SOF). The private sector comprises the former state-owned firms that were fully or partially privatised and the new private firms, established after 1990. According to the SOF, there were almost 4,000 privatised firms by Autumn 1997. In this chapter, we use a dataset of over 9,000 medium and large non-financial firms, comprising RA, commercial state-owned and privatised. The dataset was supplied by the Romanian Ministry of Finance. All registered firms in Romania are required to submit financial accounts to the state authorities. The dataset originates from the firm-level data obtained in these financial accounts (including balance sheets, profit and loss statements, tax returns). Some basic information on the 9,000 odd firms included in the dataset can be found in Table 4.9. 4.3 Methodology The methodology that we use to measure the tax arrears subsidy is based on the stocks and flows of tax liabilities.2 A stock is measured at a given point in time. A flow is simply the additions to, or subtractions from (i.e. changes in) the stock over a given period of time. We refer to these additions and subtractions as inflows and outflows, respectively. Applying these concepts to tax liabilities, we can say that the tax debt outstanding at the end of the year is a stock, taxes newly accrued during the year are an inflow and tax payments made during the year are an outflow. When distinguishing between stocks and flows, we will concentrate on the flows element as they are an indication of the size of the tax arrears subsidy. As for the tax liabilities, enterprises are liable for different types of tax. ‘Value-creating’ enterprises are liable for Value Added Tax.3 Enterprises with a salaried workforce are liable for wage tax and for social security contributions, the latter paid into the social budget fund. Profitable enterprises are liable for profit tax. The calculation of tax arrears is made more complicated by a number of factors, including some measurement problems. These include the changes in price levels and accounting for inflation, (exemptions from) penalties and interest charges accruing on overdue taxes and general problems associated with accounting standards (underreporting, treatment of depreciation, etc…). Our methodology for calculating 2 A similar framework, based directly on the inflows and outflows of tax arrears, can be found in Schaffer (1996). The advantage of the above methodology is the ease in which it can be applied to actual firm level data. 3 In most jurisdictions, some common activities are exempt from VAT. Examples include education and public administration (more on this in Chapter Five).
68
Transition, Taxation and the State
tax arrears accounts for these complications as best as possible. Moreover, changes in tax liabilities arising from tax disputes, write-offs, deferrals, holidays and tax reschedulings are omitted from our analysis due to inadequacies in the data. In addition, tax evasion is not addressed. As stated earlier, whereas evasion is primarily a problem of detecting unknown tax debts, tax arrears are a collection of known tax debts problems. In deriving a measure of the tax arrears subsidy, we work with firm-level data, that is, data from the financial accounts of enterprises, as reported to the relevant authorities. We use data from the balance sheets, profit and loss or income statements and tax tables to calculate the volume of tax arrears. Using the concepts of stocks and flows of tax liabilities, let us define T0 = Opening stock of tax liability, including social security contributions T1 = Closing stock of tax liability, including social security contributions ti = inflow of tax where ti = t1 + t2 = new tax obligations + penalties and interest to = tax payment, i.e. outflow of tax By definition, T1 = T0 + ti – to
(4.1) (Note: All expressed in current prices)
Using the methodology outlined above, we now examine the tax liabilities of a firm, called Romel.4 We use the firm accounts for the years 1996 and 1997, made available by the Romanian Ministry of Finance. With the financial accounts for 1996 and 1997 we have the stock of tax debt as at 1 January 1997 and at 31 December 1997, and the flow of tax debt for 1997. The stock of tax debt at 1 January 1997 and at 31 December 1997 and the flows (the inflow of new tax debt and the outflow of tax payments) of tax debt for 1997 can be extrapolated from the tax table that the enterprise is required to fill out and submit as part of their returns to the authorities. The presentation of the data in tabular form, as in Table 4.3 below, makes it easier to distinguish between taxes due and those overdue (arrears) and between stocks and flows. The officially required tax forms allow us to present the total tax liability (profit, salary, VAT) for Romel, to the state, the local budget and the social fund, both principal and interest. The figure of 23,062,572 thousand lei (US$3.8m) is Romel’s total tax liability as at 1 January 1997 to the government.5 It comprises the principal and the interest on overdue taxes. Liabilities to the state budget, the local budget and the social fund are all included. The flows are divided into inflows to and outflows from Romel’s initial stock of tax debt. The inflows comprise taxes newly accrued and new penalties and interest on 4 Romel is a real firm operating in the Romanian enterprise sector whose name has been changed. For reasons of confidentiality, we have also changed the numbers by a factor. 5 For the sake of simplicity, we will express all amounts from here on in billions of lei.
Tax Arrears and the Romanian Enterprise Sector
69
overdue taxes. In 1997, the new tax liabilities amounted to 124 billion lei. The penalties and interest amounted to approximately 5 billion lei. Tax payments made by Romel
Table 4.3
Stock/Flow Analysis of Romel’s Tax Liabilities in 1997, at Current Prices (in Thousand Lei) T0
Tax liabilities, principal and interest
Inflows
Outflows
T1 Stock 31.12.97
Stock 1.01.97
t1
t2
to
23,062,572
124,216,289
4,899,226
92,481,293
59,696,794
Overdue
37,597,433
Not Overdue
22,099,361
Source: Author’s calculations.
(including payments of interest and penalties) are the outflows from Romel’s stock of tax debt. They were 92 billion lei in 1997. The end-period stock of tax liabilities, as at 31 December 1997 is 60 billion lei, of which 38 billion lei is overdue and 22 billion lei is not overdue. CPI inflation in the 12-month period, 31 December 1996 to 31 December 1997 was 151 per cent. We now need to adjust our figures in order to account for this change in the price level. We do this by means of appropriate reflators, calculated from the Consumer Price Index.6 Once calculated, we end up with all amounts expressed in end-1997 prices. The appropriate reflators are calculated in Appendix E. Once the flows of tax debt are expressed in end-year prices, we can estimate a figure for the change in the stock of tax debt that is due to inflation. In effect, this is an outflow from Romel’s stock of tax debt. We call this the ‘inflation subsidy’ and denote it as s. In our analysis, the inflation subsidy s is as follows: s = T0 + ti – to – T1
(4.2) (Note: All expressed in constant or end-year prices)7
6 A colleague has pointed out that in general the CPI may not be the best index to use in the context of a firm and its prices. While acknowledging this (but with firm-specific output prices not available), we are reasonably satisfied that the CPI is appropriate to use when the exercise is extended to the enterprise sector. In estimating the real cost to the government of the tax arrears subsidy, the CPI or any other general price index is acceptable. Aside from this point, changes in producer and consumer prices in Romania were similar in 1997. 7 The bold type is used for constant prices, in order to distinguish between quantities in current prices and quantities in constant prices.
Transition, Taxation and the State
70
Table 4.4 shows Romel’s tax liabilities in constant prices. It incorporates a figure for Romel’s inflation subsidy for 1997.
Table 4.4 Stock/Flow Analysis of Romel’s Tax Liabilities in 1997, at Constant (End-Year) Prices (in Thousand Lei)
Tax liabilities, principal and interest
T0 Stock 1.01.97 57,979,307
Inflows
Outflows
t1
t2
to
s
158,499,984
6,251,412
118,006,130
45,027,780
T1 Stock 31.12.97 59,696,794
Overdue
37,597,433
Not Overdue
22,099,361
Note: T0 + ti – to – s = T1, in constant prices. Source: Author’s calculations.
For Romel, the inflation subsidy is estimated to be 45 billion lei, calculated as follows, s = 58 bn lei + (158.5 + 6.2) bn lei – 118 bn lei – 59.7 bn lei = 45 billion lei. We are now in a position to make further calculations, relevant to the tax arrears subsidy estimates. In the following two calculations, we are estimating real net flows.8 The flow of new financing to Romel from the non-payment of new taxes is the difference between tax payments made by Romel and taxes newly accrued. We call this ‘net new taxes’ (NNT) and it is expressed in equation (4.3).9 NNT = t1 – to
(4.3)
For Romel, the calculation for net new taxes is as follows: NNT = 158.5 bn lei – 118 bn lei = 40.5 billion lei
8 A figure for the real net flow can be calculated by subtracting the opening stock of tax debts in end-year prices from the closing stock of tax debts. Put it another way, it is simply total inflows minus total outflows. 9 The measure NNT as it relates to tax debt is analogous to the NBF (net bank financing) benchmark used in Bonin and Schaffer (1995), Schaffer (1998) and Croitoru and Schaffer (2002). The latter, using an inflation-adjustment framework, reports the flows of NBF and NNT for 5,000 firms in Romania for the period 1995–1999.
Tax Arrears and the Romanian Enterprise Sector
71
This flow of new financing can be considered as an increase in borrowing from the government. Tax concessions are been extracted from the government when Romel pays its taxes late in a high inflation environment. By subtracting the new interest charged on overdue taxes from the inflation subsidy, we get a measure of the effective real cost of this flow of financing to Romel. We call this the net inflation or ‘Tanzi’ effect (NTE).10 We state this as follows: NTE = t2 – s
(4.4)
For Romel, the net ‘Tanzi’ effect is – 38.8 bn lei, calculated as follows, NTE = 6.2 bn lei – 45 bn lei = - 38.8 billion lei In general, this is the amount that firms are subsidised by because of this period’s inflation rate exceeding the monetary penalties and interest charged on the late payment of taxes. In the case where the effective penalty rate charged on Romel’s stock of overdue taxes is lower than the inflation rate, the net ‘Tanzi’ effect is negative and Romel benefits from late payment as unpaid taxes are used as a source of cheap working capital. For Romel, the net ‘Tanzi’ effect (inflation erosion of tax debt net of interest and penalty charges) amounts to -38.8 billion lei. It is negative because the inflation rate for the period exceeded the penalties on late payment of taxes. Even with paying penalties on overdue taxes, Romel is benefiting at the cost to the government with the real value of its stock of tax debt being reduced by inflation. Table 4.5 shows the calculations for Romel. This methodology of measuring the tax arrears subsidy can now be applied to our database of firms operating in the Romanian enterprise sector. The 9,000 odd firms include nearly all medium and large non-financial firms in Romania. We use the same methodology as we used in the case of Romel. The calculations for the 9,000 odd firms are shown in Table 4.6. Unlike in the previous case of the single firm, here we report the actual data, for 1997. The stock of tax liabilities (and of tax arrears in the case of the end-year) is given for 1 January 1997 and for 31 December 1997, both in current prices and in constant prices. The table also includes the inflows to and the outflows from the stock of tax debt, at current and end-year prices. The inflation subsidy is a measure of the depreciation of tax debt due to inflation and amounts to 29,566 billion lei. The figure of 10,200 billion lei for NNT is the flow of new financing to the 9,000 odd firms from the non-payment of new taxes. In general, a positive sign for NNT indicates accumulating tax debt as firms are not keeping up with current tax obligations. In 1997 where taxes newly accrued were in excess of all tax payments, net new taxes is positive. 10 The well-known Olivera-Tanzi effect is named after the economist Vito Tanzi of the IMF who, along with Julio Olivera of the University of Buenos Aires, observed the phenomenon whereby inflation reduces the real value of tax revenue accruing to the government. In the context of tax arrears, extended collection lags and inflation, the government loses out. Firms with overdue tax liabilities gain as inflation reduces the real value of their tax debt.
Table 4.5 Flow Analysis of Romel’s Tax Liabilities in 1997, at Constant (End-Year) Prices (in Thousand Lei) T0 Tax liabilities, penalties and interest
Inflows
Outflows
T1
01.01.97
t1
t2
to
s
31.12.97
57,979,307
158,499,984
6,251,412
118,006,130
45,027,780
59,696,794
Overdue
37,597,433
Not Overdue
22,099,361
Note: T0 + NNT + NTE = T1, in constant prices. Source: Author’s calculations.
NNT 40,493,854
NTE - 38,776,368
Tax Arrears and the Romanian Enterprise Sector
73
Table 4.6 Stock/Flow Analysis of 9,000 odd Firms in the Romanian Enterprise Sector in 1997 (in Billion Lei) Current prices 1 Jan 97 1997 31 Dec 97 Taxes liabilities 17,065 of which, overdue
…
30,575 19,454
Constant (end-year) prices 1 Jan 97 1997 31 Dec 97 42,902
…
Inflows, of which - New taxes - Penalties / Interest charges
47,278 41,762 5,516
60,326 53,288 7,038
Outflows, of which - Tax payments - Inflation subsidyb
33,768 33,768
72,654 43,088 29,566
Net new taxes (NNT)c Net Tanzi effect (NTE)d
30,575a 19,454
10,200 -22,528
Notes: a. T0 + NNT + NTE = 42,902 + 10,200 – 22,528 = 30,575 = T1. b. As in equation (4.2), s = 42,902 + 60,326 – 43,088 – 30,575 = 29,566. c. A positive sign indicates accumulating tax debt as firms are not keeping up with current tax obligations. d. A negative sign indicates inflation eroding tax debt, with interest and penalties not sufficient to keep up with inflation.
The figure of -22,528 billion lei is the NTE. In general, a negative sign for NTE indicates inflation eroding tax debt, with interest and penalties not sufficient to keep up with inflation. As inflation was high in 1997 and interest charges on late payment of taxes were not indexed to inflation or market interest rates, the net ‘Tanzi’ effect is negative. Firms are benefiting at the cost to the government with the real value of the firms’ stock of tax debt being reduced by inflation. 4.4 Results, Analysis and Policy Prescriptions 4.4.1 Results and Analysis The results of our stock/flow analysis are presented in Tables 4.7 and 4.8. In Table 4.7 we report the stock of tax debt and overdue tax debt at December 1997, both in lei terms and expressed as a percentage of GDP.
74
Transition, Taxation and the State
Table 4.7 Stock of Tax Debt and Overdue Tax Debt at end-1997
All tax debt - principal - interest/penalties
Billion lei
In % of 1997 GDP
30,575 24.381 6,192
9.6 7.7 1.9
19,454 16,427 3,028
6.1 5.2 0.9
of which: Overdue tax debt - principal - interest/penalties
Note: All figures, including 1997 GDP of 318,757 billion lei, in December 1997 prices. Columns may not sum due to rounding. Source: Adam Smith Institute.
Outstanding tax liabilities at the end of 1997 amounted to 30,575 billion lei, or the equivalent of 9.6 per cent of GDP. Of this amount, 24,381 billion lei was principal and the remaining 6,192 billion lei was interest and penalties accruing on overdue taxes. In percentage terms, this is 7.7 per cent and 1.9 per cent of GDP respectively. Of the total tax debt, almost two thirds of it, amounting to 19,454 billion lei was overdue taxes, i.e. tax arrears. As a percentage of GDP, this is equivalent to 6.1 per cent. As indicated in the literature, the existence of a stock of tax arrears is not necessarily a cause for concern (Schaffer 1996; Alfandari and Schaffer 1996). Although the stock of tax arrears may appear high, what matters is the cause of the tax arrears and whether the stock is stable. For example, a stock of tax arrears that is stable and has arisen because of legitimate tax disputes between the tax authorities and a number of firms is less problematic that a rising stock of tax arrears caused by loss-making firms that are unable to pay their tax debt but still remain in business (what the literature refers to as the ‘financial distress’ case of tax arrears). In order to assess the tax arrears problem more closely we need to examine the change in the stock of tax debt in 1997. These flows are presented in Table 4.8, at end-year prices. Table 4.8 indicates that the firms in our database are not keeping up with their tax payments. Although tax payments amounted to 43,088 billion lei, taxes newly accrued amounted to 53,288 billion lei. The difference of 10,200 billion lei represents non-payment of new tax obligations and is, in effect, a flow of new financing from the government to the enterprise sector. As a percentage of GDP, this ‘net new taxes’ amounts to 3.2 per cent. For the state and the government budget, the problem of firms not keeping up with their current tax obligations (net new taxes is positive) is compounded by the
Tax Arrears and the Romanian Enterprise Sector
75
practice of charging interest and penalties on overdue taxes at rates not high enough to keep up with inflation (net Tanzi effect is negative). In 1997, the firms’ tax debts were heavily eroded by high inflation. Although interest on overdue tax debt was charged, the interest/penalty rate was not sufficient to compensate for the erosion of the tax debt due to inflation. As calculated above, the ‘inflation subsidy’ amounted to 29,566 billion lei, or the equivalent of 9.3 per cent of GDP. In contrast, interest and penalty charges amounted to 7,038 billion lei, or only 2.2 per cent of GDP. The net inflation effect, called the ‘Tanzi’ effect is therefore negative, a net reduction on tax debt of 22,528 billion lei, or 7.1 per cent of GDP. This negative real rate of interest charged means that, in effect, firms had access to cheap working capital.
Table 4.8 Flows of Tax Debt January-December 1997
New taxes accrued in 1997 Minus All tax payments made in 1997 Equals Net new taxes (NNT) accrued in 1997 Interest/penalties accrued in 1997 Minus Inflation erosion of tax debt in 1997 Equals Net Tanzi effect (NTE) in 1997
Billion lei
In % of 1997 GDP
53,288
16.7
43,088
13.5
10,200
3.2
7,038
2.2
29,566
9.3
-22,528
-7.1
This result, of a high and positive net new taxes and a negative net ‘Tanzi’ effect, contrasts with a typical analysis for mature market economies and their enterprise sectors where ‘net new taxes’ is generally flat or negative because firms pay their tax liabilities on time, and the net ‘Tanzi’ effect is positive because firms with tax arrears are paying positive real interest rates on their overdue taxes. In summary, it appears from our analysis that the scale of the tax arrears subsidy in Romania was large. The stock of tax arrears at the end of 1997 amounted to 6.1 per cent of GDP. Of greater concern is the flow of new financing, indicating that firms are not keeping up with current tax obligations. New taxes accrued in1997 exceeded tax payments made by an amount equivalent to 3.2 per cent of GDP. Moreover, the practice of charging low rates of interest in what was a high inflation environment meant that the recalcitrant firms benefited to the tune of 22,528 billion lei, equivalent to 7.1 per cent of GDP. These figures indicate a serious tax arrears problem in Romania. With respect to tax arrears, the data indicate both a stock problem and a flow problem. Moreover, an initial investigation into the different categories of firms with
76
Transition, Taxation and the State
tax arrears seems to indicate that the tax arrears problem is not concentrated solely in the unprofitable firms (the SBC syndrome), as the evidence from other transition countries would suggest (Alfandari and Schaffer 1996; Schaffer 1998). As Table 4.9 shows, a number of profitable firms have tax arrears, indicating that the problem has spread beyond loss-making firms. If so, the problem of loss-making firms extracting subsidies from the state by means of tax arrears is not the only problem. Profitable firms with overdue tax liabilities suggests a poor payments discipline problem. If this is true, we are confronted with two of the tax arrears cases that the literature has identified. The first is tax arrears relating to firms in ‘financial distress’; the case of the SBC originating from a paternalistic state and benevolent government. The second is tax arrears relating to ‘late payment’ and in particular, to poor payments discipline.11 In Table 4.9 we distinguish between three groups of firms that each account for a large part of the tax arrears. The three groups are large loss-makers, utility companies and profitable firms with large tax arrears. Although simple and somewhat arbitrary, this exercise is useful in trying to assess the nature and cause of the tax arrears problem in Romania. For example, a high concentration of tax arrears in financiallydistressed firms would indicate a SBC problem that has been evident in other CEE countries, including – in earlier days – the leading transition economies. In contrast, tax arrears among utilities and profitable firms might indicate a delay in restructuring or a general lax discipline problem, as is evident in some of the FSU republics. We define large loss-makers as firms with losses of over 10 bn lei in 1997 (at end-1997 prices). There are 176 such firms, that is, 2.4 per cent of all firms in the dataset, of which 85 per cent are state-owned. At end 1997, their overdue tax liability amounts to 4,758 bn lei, or 24.4 per cent of the tax arrears of all firms. As for the utilities, we report data on four, namely RENEL (electricity), SNCFR (railways), Romgaz (natural gas) and Petrom (oil refinery). Total tax arrears at end 1997 for these four was 4,806 bn lei. Profitable firms with large tax arrears are defined as firms with profits in excess of one bn lei and owing more than one bn lei in overdue tax at end 1997. There are 436 such firms, with tax arrears in Dec 1997 amounting to 5,231 bn lei. To get a measure of how these groups differ in other related respects, we report some key indicators for all groups, including the ‘all others’ group. Aside from debts (overdue and not overdue; to government and to banks) and profits/losses (operating profit and profit after deductions), we report group data for employment growth, average wages and interest costs. A priori, we might expect the large lossmakers to shed more labour and pay lower wages than any of the other groups. We begin with the large loss-makers. As is evident from the table, these firms are in serious financial trouble. As a group, they are making operating losses, indicating 11 The dual nature of the problem (‘late payment’ and ‘financial distress’) makes the Romanian tax arrears problem more similar in nature to the tax arrears problem in Russia than to the arrears problem evident in some of the other CEE transition countries. It also makes it different from mature market economies where tax arrears usually arise from firms that are either in bankruptcy or liquidation proceedings, or are in legitimate dispute with the tax authorities.
Table 4.9 Groups of Firms with Tax Arrears in 1997 Compared Group:
All firms
Definition of group:
No. of firms of which, state-owned of which, private Total employment, thous of which, state-owned of which, private Average employment of which, state-owned of which, private Total tax debts at end-97, billion lei Overdue tax debts at end-97, billion lei Overdue tax debts at end-97, as % of total assets Operating profit (EBITD) as % of assets Profit as % of assets Debt as % of assets, end-97 Bank debt as % of assets, end-97 Interest costs as % of operating profit Employment growth 1996-97, in % Average wage, thous lei per month
(1) Large loss-makers
(2) Utilities
Losses >10 bn lei
(3) Profitable firms with tax arrears
(4) Others
Profits > 1 bn lei and tax arrears > 1 bn lei
All others
9,329 4,284 5,045 3,134 2,124 1,010 336 496 200 30,575 19,454 6.4
176 150 26 349 325 24 1982 2164 933 5,819 4,758 12.5
4 4 0 342 342 0 85,622 85,622 0 9,593 4,806 5.8
436 247 189 772 570 202 1,771 2,306 1,071 7,455 5,231 8.4
8,713 3,883 4,830 1,672 888 784 192 229 162 7,705 4,660 3.9
12.1 0.7 46.7 12.7 29.2 -8.7 1,169
-1.4 -26.3 68.6 21.4 n.a. (operating profit < 0) -13.9 1,205
9.8 -0.9 41.3 8.5 7.9 -2.0 1,917
17.1 6.6 44.3 11.3 23.2 -6.3 1,206
15.3 7.4 44.7 13.7 26.2 -9.6 990
Note: All values expressed in December 1997 lei. Assets are at end-1997. Figures may not sum due to rounding.
78
Transition, Taxation and the State
that sales revenue is not even covering basic operating costs. Compared to all firms, their debts, at over 68 per cent of assets, are high. Of their 5,819 bn lei tax debt, over 80 per cent is overdue. As is the case in other transition countries, it appears that these firms remain in operation and are kept afloat by means of the tax arrears subsidy from the national and social security budgets.12 As in Russia and other FSU republics, the utility sector in Romania appears to be a key figure in the accumulation of tax (and other) arrears. The four utility companies comprising the utility group in Table 4.9 account for over 30 per cent of all tax liabilities and 25 per cent of all overdue tax liabilities (of medium and large firms in our dataset). During the transition era of the 1990s in Romania, the energy companies have accumulated large overdue receivables, participated in clearing operations or barter schemes, have been prevented from cutting off supplies to non-paying customers and, often in response to these politically-motivated actions, delayed paying taxes to the government budget. Budgetary subsidies to utility companies arising from price controls and the provision of public services are common in transition and mature market economies. Using utilities as conduits in order to allow loss-making firms or favoured firms to receive or extract (hidden) subsidies is not common in either leading transition or industrial economies. With respect to financial discipline, the evidence from the profitable firms with tax arrears group (despite the arbitrary criteria used) should be of greatest concern to the tax authorities. As defined, these 436 firms have both profits and tax arrears in excess of one bn lei. Although profitable, 70 per cent of their total tax liabilities are overdue, a figure that is higher than the corresponding figure for all the 9,000 odd firms in the dataset. Compared to the utilities, their operating profit is higher and their average wage is significantly lower. Possible reasons for why a group of profitable firms, that in other CEE transition countries is not a likely candidate for tax arrears, might accumulate overdue tax debt include political lobbying by favoured firms, expectation of a general government bailout (in the form of tax write-offs, reschedulings, waiving of interest/penalty, debt-equity swap) or the willingness to borrow from government at an effective interest rate charged on delayed tax payments that is less than the cost of alternate sources of funds i.e. bank credit (of course, low financial intermediation, as evident in many transition economies, tends to limit the availability of alternate sources). Another cause for concern is that of the 436 profitable tax debtors, almost half are private firms. One might expect that the state has less ties to private firms, thus, lowering the incentive to tolerate tax arrears.13 On the aggregate, how does this compare to other transition countries or to the Romanian experience in the earlier years of transition? The difficulty with any 12 An explanation for this (strategic) behaviour on the part of the state and the financiallydistressed firms can be found in Schaffer (1998). 13 Of course, the data for overdue tax liabilities in Table 4.9 are for stocks (rather than flows) of tax arrears and may very well reflect unpaid tax debts arising from pre-privatisation days.
Tax Arrears and the Romanian Enterprise Sector
79
comparative analysis of tax arrears, either across time or across countries, is twofold. One, measuring tax arrears, either by use of aggregate data or firm level data, is in itself a complicated and onerous exercise. Two, there are what we call ‘reporting’ problems. For example, the reporting of penalties, write-offs and reschedulings is not standardised across countries. These matters need to be considered before definite conclusions can be drawn from any comparative analysis. Table 4.10 below reports empirical evidence on tax arrears in a number of transition countries. Estimates for the stock of tax arrears vary from about two per cent of annualised GDP to over nine per cent of GDP. Estimates for the flow of tax arrears range from one to over three per cent of GDP per annum.
Table 4.10 Tax Arrears in Transition Economies Stock (end-year) Country Date of Total Estimate Czech Rep. 1993 3-4
Flow (annualised) Date of Total estimate 1993 1-2
Estonia Hungary
6.94-6.95 1993 1994
0.6 1.2 0.7
1994 1993
1.0 1-2 0-1 1.5 3.1
Latvia Lithuania Poland
Romania Russia
Slovakia
1994 1993 1994 1994 1993 1994 1993
5.7 6.9 7.5 7.0 3.7 4.4 7-9
1994 1993 1994 1993
5-7 1.5 4.6 1.5
1994 1993 1994
1994 4.0 1995 5.2 (6.5) 1996 12.0
1994 1995 1996
1993
2.1 1.4 (3.1) 7.3
Remarks Figures are estimates based on preliminary data
Interest component is an estimate
Figures in parenthesis includes 30:70 deferrals. Estimates are based on Goskomstat data on overdue liabilities of firms with adjustments for sample coverage.
4.5
Note: All figures in % of (annualised) GDP and are inclusive of principal, interest and late penalties. Source: Schaffer 1996, 1998.
80
Transition, Taxation and the State
With respect to these estimates, several remarks are worth noting. One, these figures are large by international standards. Moreover, in mature market economies, stocks of tax arrears are generally stable and reflect taxes that have been written off as uncollectible, arising from bankruptcy proceedings against delinquent firms. Two, for the leading transition countries the estimates for the flow of tax arrears are ‘…about the same size…’ as for budgetary subsidies to enterprises (Schaffer 1995). Three, where the breakdown of the stock into principal and interest is available, the interest component is large. This suggests that, for these countries, the greater part of their tax arrears is old. In general, the combination of a large interest component and relatively small flows of tax arrears suggests a stock problem rather than a flow problem. In Romania, the interest component of the stock of tax arrears at the end of 1997 was small (0.9 per cent interest compared to 5.2 per cent principal) but the net inflow due to non-payment of newly-accrued taxes was relatively large in 1997, amounting to 3.2 per cent of GDP. This suggests a sizable flow problem in Romania. Four, the figures for the other CEE countries are for earlier years in the transformation process where we might expect ‘transitional’ problems. As these countries continue to reform at a rapid pace (Hungary and Poland, in particular), we might expect to see a reduction in the tax arrears problem, to levels similar in Western economies. This appears not to be the case in Romania. 4.4.2 Policy Prescriptions A general improvement in the economic and financial environment in which enterprises operate is likely to increase tax collection in Romania. Continued depolitisation of the enterprise sector, improvements in corporate governance and a healthier banking sector are all likely to strengthen the enterprise sector and, in turn, contribute to a reduction in the tax arrears of recalcitrant firms. This is true of other CEE countries where enterprise restructuring and structural reforms have occurred at a quicker pace, thus contributing to a tax arrears problem of smaller proportions. More specifically to Romania, the tax arrears phenomenon is compounded by a number of problems in legislation and in the tax system itself. A number of these problems are addressed below. Recommendations are also included. Bankruptcy Law As in other transition economies, the threat of bankruptcy must be credible. Although the legislation regarding bankruptcy is viewed by many as adequate, problems arise in its enforcement. For instance, the pace is rather slow with inadequate resources available to judges and liquidators.14
14 Since first writing this, the Bankruptcy Law was indeed amended, in May 1999. Among other changes, procedures have been simplified allowing for quicker debt recovery (EBRD 1999).
Tax Arrears and the Romanian Enterprise Sector
81
Treatment of interest and penalties on overdue taxes As there is a single charge for late payment of taxes, no distinction in domestic law is made between interest on overdue taxes and penalties for late payment. It is customary to distinguish between interest, which is a charge for the use of government’s working capital, and penalties, which is a charge to punish and deter late payment of taxes. Tax legislation needs to be enacted (and enforced) in order to ensure that interest charges and penalties on overdue taxes are treated separately. Calculation of interest Interest charges are calculated on a simple interest rather than a compound interest basis. This means that interest is charged on the outstanding overdue principal only and not on overdue principal and interest. Not only is this cheaper on the firm, but it also encourages the firm to delay paying the overdue debt. In future, the authorities should calculate interest on a compound interest basis, i.e. interest should be charged on all overdue taxes, principal and interest. Index to inflation or interest rates The statutory interest charge on overdue taxes, at 91.25 per cent in 1997, is not updated in line with changes in the inflation rate, resulting in a negative real interest rate charged on overdue taxes. As a result of this, firms with overdue tax liabilities gain at the expense of the government budget in periods of high inflation. The interest rate charged on late payment of taxes should be tied to market interest rates and changed as market rates change. This would ensure that the interest rate charged would not differ significantly from interest rates charged by banks on loans. Uniform treatment for all firms and of all tax arrears It is not uncommon for some favoured firms to get their tax arrears formally rescheduled or to benefit from reduced interest charges on overdue taxes, or both. This is evident from our analysis where the effective interest rate charged on overdue taxes was only about half the statutory interest rate. The practice of rescheduling tax arrears or reduced charges encourages further financial indiscipline and bargaining on the part of delinquent firms. To prevent such behaviour by enterprises and the inevitable ‘moral hazard’ problems that arise, programmes that write off or reschedule tax arrears should be avoided. If such programmes are deemed necessary, the rules relating to reschedulings and interest charges on overdue taxes should be transparent and consistent, act as a sanction and be applied evenly.
82
Transition, Taxation and the State
Tax collection procedures The Romanian authorities depend on a means of ‘forced execution’ to collect tax arrears from recalcitrant firms. This involves using the banks as tax collection agencies by effectively ‘blocking’ debtors’ bank accounts and confiscating the amount owed to the authorities. The authorities also have the power to seize and sell off assets of the debtor firm. Although this system of tax collection is used in other transition countries (Russia, for example), it has a number of deficiencies, both internal to the procedures themselves and external to the actual operation of the procedures (like in other countries, it is subject to strong political pressures). This is reflected in the small amounts of monies that have actually been recovered. In principle, the forced execution procedure is a fairly powerful tool and, with some improvements can be used as a satisfactory tool for debt collection. One possibility is to broaden the range of assets that can be seized. Reform in the Utilities Sector As in Russia and other CIS countries, the utilities (energy, transport) play a significant role in the payments crisis and in the state’s toleration of arrears. Often, tax arrears and inter-enterprise arrears are linked to utility arrears. In Romania, the energy utility RENEL and the railways company SNCFR are two of the largest delinquent firms with respect to overdue tax liabilities. Reform of the utility companies, involving commercialisation, competition, changes in ownership and corporate governance, would go some way in assisting the authorities to tackle the tax arrears problem.15 Avoid Government-Led Bailout Schemes Since the emergence of the arrears problem in Romania, the authorities have, at different stages, intervened in the hope of resolving the problem. Either through extension of new credit (in response to rising trade credit arrears), isolation programmes (partly in response to loss-making firms accumulating non-payments) or reductions in interest and penalties (as an incentive for firms with tax arrears to pay overdue principal), different governments have sent the wrong signal to firms (both delinquent and complaint) in the enterprise sector. These schemes, although well-intended, should be avoided, if possible.
15 Again, as in the case of bankruptcy law, there has been some recent attempts to reform the utilities sector. A new energy law allowing for a restructuring of RENEL was enacted in late 1998. A year later an independent energy regulator was established. The restructuring of the state railways recommenced, again in 1998, with the unbundling of SNCFR into several joint-stock companies (EBRD 1999).
Tax Arrears and the Romanian Enterprise Sector
83
Putting the tax arrears subsidy explicitly onto the government budget Tax arrears that go unpaid are effectively a hidden government subsidy. This subsidy can be recognised as such and scrutinised and monitored in the same way as other government subsidies are if the tax arrears subsidy was to be explicitly put onto the government budget. This formal recognition of the tax arrears subsidy, as recommended by the World Bank, would in the future help to prevent the government from discretely channeling funds to certain enterprises and these enterprises lobbying officials and ministers for state funds.16 Other possible initiatives used in some transition countries include a registry of highly-indebted firms, a reduction in barter trade or non-cash tax payments (or, at least, legislating against in-kind payment of tax in arrears) and the introduction of a broad-based system of advance tax payments.17 Apart from these predominately legislative and tax changes, there is also the political aspect to tax arrears and, more generally, SBC. Enterprises allow overdue tax liabilities to build up in the anticipation that the government will bail them out, through tax write-offs, reschedulings and so forth (Appendix F). In a mature market economy, a persistent build up of these (overdue) debts and other enterprise arrears (to banks and to firms) would eventually lead to bankruptcy and liquidation proceedings. In Romania and elsewhere, many loss-making enterprises manage to survive because the government regards the alternative i.e. firm closures and job losses, as unacceptable. The evidence of tax arrears, bankruptcy proceedings and bailouts by the state would indicate that the budget constraints of firms are not enforced; indeed they are softened. If this is the case, the government needs to adopt a strategy that enhances the government’s position as a tough creditor (tax collector). Such a strategy might involve prioritising tax enforcement and collection, strengthening the threat of bankruptcy and avoiding tax reschedulings, tax write-offs and reduced penalties. Moreover, there needs to be a concerted effort on behalf of the authorities to avoid the tax arrears problem spreading (a phenomenon that is not common in leading transition countries) beyond loss-making firms, which in itself is a problem (and has been evident in many transition countries, both slow and rapid reformers). Poor payments discipline is a serious problem and can have grave implications for other areas of public policy. As a priority (and learning from the Russian experience), tax discipline needs to be restored in profitable firms. In the next chapter, we examine tax administration and the effectiveness of tax collection in transition countries, including Romania.
16 An example of this is the explicit budgetary subsidy to the railways, following a restructuring (see previous footnote), that was allowed for, in the order of 0.7 per cent of GDP, in 1999 (IMF 2001a). 17 For more on non-monetary transactions including cases of tax debts redeemed in goods see Gaddy and Ickes (1998) and Schaffer (2000).
Transition, Taxation and the State
84
Appendix E: Calculating Price Reflators We calculate the price reflators from the Consumer Price Index monthly data published in the Statistical Bulletin. For monthly values of the CPI for 1997 see Table E.1 (December 1996 = 100): The end-year CPI for 1997 is 251.4. To convert start-year prices into end-year prices the appropriate reflator is 2.514. The average (arithmetic mean) monthly CPI for 1997 is 197. The end-CPI/average-CPI is 251.4/197 = 1.276. To convert average-year prices into end-year prices the appropriate reflator is 1.276. End-year prices remain the same.
Table E.1 Consumer Price Index, January-December 1997 Jan 113.7
Feb 135
Mar 176.5
Apr 188.7
May 196.7
Jun 201.2
Jul 202.6
Aug 209.8
Sep 216.7
Oct 230.8
Nov 240.6
Dec 251.4
Source: Central Bank of Romania Quarterly Bulletins.
Appendix F: Hard Budget Constraints and Enterprise Restructuring in Romania This is the case study component of Chapter Four. In Spring 2000, several firms were shortlisted for a case study on the basis of their financial position i.e. highly-indebted, lossmaking firms are strong candidates for ‘budget softness’. A list of unprofitable firms was drawn up. Roman S.A. and Tractorul UTB S.A. were chosen from the list on the basis of a number of criteria. For one, both firms are in a poor financial position. Sales are falling and losses are mounting up. Two, both firms employ a large workforce. This can strengthen a firm’s bargaining position if it wishes to seek financial or state aid. Three, both firms have overdue tax liabilities to the state and have had debt rescheduled in the past. The appendix is in three parts. The first part is a case study of Tractorul UTB S.A., a firm that makes tractors. The second part is a case study of Roman S.A., the truck-making company. As both enterprises are based in Brasov, the third part of the study examines the importance of Tractorul and Roman to the local region. A comparison between the region of Brasov and Romania under a number of relevant headings is included. The appendix ends with some tentative conclusions. F.1 Tractorul UTB S.A.18 Based in Brasov, Tractorul UTB S.A. makes tractors. It was set up in 1925 as a Franco-Romanian aircraft company. The first tractor was manufactured in 1946. 18 In March 2000, a visit was arranged to the plant in Brasov where we were met by a technical director and a counsellor of Tractorul.
Tax Arrears and the Romanian Enterprise Sector
85
A licensing agreement with FIAT dates back to 1968. In 1991 it became a jointstock company. As of 2000, its corporate status is as a state-owned commercial company with 75 per cent of shares owned by SOF and the remainder owned by private shareholders (18 per cent in a POF and seven per cent owned by individual investors). As a firm in the automotive industry, the relevant ministry for Tractorul is the Ministry of Industry and Trade. As it produces tractors, Tractorul also has relations with the Ministry of Agriculture. Along with 63 other companies, Tractorul has been shortlisted for privatisation, with the support of the World Bank. At its peak in the mid-1980s, Tractorul was manufacturing close on 50,000 tractors per annum with a workforce of 20,000. Average production in the period 1990–1998 was 17,000 tractors per year. By 1999, the number of tractors sold had fallen to less than 5,000. The full-time workforce in 1999 was 10,112, down from just over 19,000 in 1990. Partly due to a new leasing arrangement put in place, Tractorul hoped to sell 15,000 tractors in 2000, of which 8,000–9,000 are for sales in the home market. A leasing scheme was to start in Spring 2000 and was co-financed by the Ministry of Agriculture, a special bank-financed fund and the farmers’ own funds. In addition, it was hoped that another financing scheme (involving goods for goods) to be launched later would help to boost domestic sales. In recent years, Tractorul’s sales of new tractors have almost exclusively been for the foreign market as domestic demand has been very weak. Instead of purchasing new machines, cash-poor farmers have repaired old machines. Whatever home market exists, Tractorul is virtually the sole producer. It is hoped that the new leasing arrangements will provide a new source of demand for Romanian tractors. Since 1994, the company has been making heavy losses. Management puts much of these losses down to external factors, such as the weak state of the home demand, exchange rate losses and certain high costs, such as electricity prices. The company did admit that layoffs had not been high enough to offset the fall in demand. Partly in response to these conditions and to prepare for the privatisation of the company under the World Bank’s PSAL (Private Sector Adjustment Loan) programme, Tractorul S.A. was split into 10 separate companies at the end of 1999. Tractorul UTB S.A. remains the largest, manufacturing tractors with a workforce of 5,600 workers. The activities of the other nine companies vary from the manufacturing of engines, machine tools and spare parts to the design of cabs and to general repairs. In addition, a new management team took over in September 1999, replacing the old management team that had been in place since 1990. It consists of a new general manager and nine managers responsible for specific functions ranging from finance to quality. On further investigation, it transpired that the change in management team was initiated because the outgoing general manager had asked to be replaced. The General Manager is the firm’s representative on the Administrative Council, a=body comprising three representatives from SOF, one from the POF and the firm’s representative. Tractorul’s foreign customers come from countries as far apart as the USA, Canada, Egypt, Pakistan, Turkey, France, Hungary, Slovenia and Croatia. In general, Tractorul has no payment problems with its clients. As its customer base is foreign, Tractorul, on
86
Transition, Taxation and the State
average, gets paid on time, in cash and in foreign currency. This is a fortunate position for Tractorul as many Romanian companies have overdue receivables and often get paid in kind rather than in cash. Like all companies that sell on world markets, they face competition. Earlier, their main competitors came from other CEE countries, including Poland and the Czech Republic, and from Belarus. Currently, it is the view of management that foreign competition for its product range is quite limited. The average monthly wage was 1.8 million lei ($100) in December 1999.19 This is net (after tax and social security contributions) and includes bonuses. It compares favourably with the average monthly wage for the sector. On the question of strikes and late payment of wages, management was insistent that there were no strikes at Tractorul and that wages were always paid on time. When questioned about the strikes in Brasov in the summer of 1999, the Tractorul management was adamant that the workers on strike in June 1999 were not employees of Tractorul and that the press reports were incorrect. Like other state-owned companies, Tractorul UTB has divested itself off most of its social assets, in this case to the local municipality. All that remains is a number of flats that are occupied by unmarried personnel. The next issue raised was finance and payments. When questioned about the firm’s creditors and the ordering of payment, we were told that in practice (and contrary to legislation where payments to the state have priority) the ordering is as follows: utilities (paid in cash) followed by banks (paid in cash), suppliers (paid in mix of cash and goods) and, finally, government. This, in turn, explains the breakdown of the firm’s outstanding debts. By January 2000, Tractorul UTB S.A. had debts amounting to 1,400bn lei ($76.3m), of which a high proportion was interest and penalties for late payment.20 Of the total, 745bn lei ($40.6m) was owed to the state (comprising state budget, social security and special funds). 500bn lei ($27.2m) of this were penalties and interest charges. An amount of 360bn lei ($19.6m) was debt to banks, of which two thirds was principal. Tractorul’s main bank is the Romanian Development Bank, subsequently bought by Société Generale. Close on 160bn lei ($8.7m) was owed to suppliers, who normally extend 30 days trade credit to Tractorul. It appears that Tractorul has little or no debts to the electricity or gas companies. In response to a question relating to deferrals or write-offs of debt, management admitted that the company has had debt rescheduled in the past. Tractorul’s precarious financial position is shown in Table F.1, for the period 1990–1998. The financial situation for Tractorul looks bleak. In every year since 1994, except one, that is 1997, the firm has been making operating losses. This suggests that Tractorul is having difficult covering basic costs, like materials and labour costs. In addition, the large losses accumulating since 1994 indicate a debt overhang problem. In 1990, Tractorul owed 9bn lei, of which the majority was owed to banks. Debts to suppliers (of power, materials and components) and to the state were insignificant, 19 The average lei/US$ exchange rate for December 1999 was 17,996 lei = 1 US$ (National Bank of Romania 2000). 20 Using the average lei/US$ exchange rate for January 2000 of 18,353 lei = 1 US$ (National Bank of Romania 2000).
Table F.1 Selected Indicators for Tractorul UTB S.A., 1990-1998 1990
1991
1992
1993
1994
1995
1996
1997
1998
pcs.
22938
21603
21003
26013
13846
15072
12598
10350
9436
for domestic market
pcs.
13340
11590
9788
16852
7993
5335
2738
1322
1005
for export
pcs.
9598
10013
11215
9161
5853
9737
9860
9028
8431
Turnover
mill. Lei
5668
19831
40242
118695
159694
209521
301772
573586
674993
No. of tractors produced of which
No. of employees
prs.
19050
18015
18064
18979
18738
16777
14780
13535
12397
Profit -
mill. Lei
275
1408
3268
2782
-32133
-78434
-98215
-193684
-312012
Operating
mill. Lei
432
1408
8023
7605
-17773
-63785
-30099
14845
-126075
Financial
mill. Lei
-88
0
-4755
-4823
-9509
-20786
-25288
-30486
-74077
Exceptional
mill. Lei
-69
..
..
..
-4851
6137
-42828
-178043
-111860
Total debts
mill. Lei
9057
13794
24292
86700
169857
250901
432991
727919
1123386
to suppliers
mill. Lei
317
548
10782
22155
38138
45970
71223
127947
134155
to state and other budgets
mill. Lei
20
916
2617
21347
56039
94006
197232
417962
536618
Principal
mill. Lei
20
916
2617
21347
44488
87358
146203
193017
281266
penalties for payment delay
mill. Lei
..
..
..
..
11551
6648
51029
224945
255352
Inflation rate, end-year
222.8
199.2
295.5
61.7
27.8
56.9
151.4
40.6
Inflation rate, annual average
161.1
210.4
256.1
136.7
32.3
38.8
154.8
59.1
of which
of which
Memo Items
Source: Report on Tractorul UTB S.A. Brasov, supplied by the management of Tractorul; National Bank of Romania 2000.
88
Transition, Taxation and the State
at less than five per cent of the total debt. By 1998, the debt owed to the state had ballooned. Of a total debt of over 1,100bn lei ($127m), almost 540bn lei ($60.5m) was owed to the state.21 Of this, almost half was penalties and interest on overdue debt. Tractorul does receive some aid from the state, largely in the form of low interest rate loans and injections of capital from SOF. On account of the difficult financial and trading situation that Tractorul finds itself in, management wants the government to dissolve the debts owed to the state. At a minimum, it wants the interest charges and penalties on late tax and contribution payments to be written-off and the principal to be rescheduled. On a follow-up visit to Tractorul, it was learnt that Tractorul was one of a number of debt-ridden firms whose debts were to be rescheduled by the government. Although agreed by cabinet, the plan was abandoned after unfavourable press coverage. F.2 Roman S.A.22 Roman S.A. is a truck-making company, based in Brasov. It is a state-owned commercial company, with 94 per cent of its shares owned by SOF and the remaining six per cent owned by local investors. It is one of the 64 Romanian companies that have been shortlisted for privatisation, with the support of the World Bank. It was founded in 1921, manufactured the first Romanian truck in 1954 and became a jointstock trading company in 1990. At its peak it employed over 26,000 workers. In 1980, it manufactured 30,000 trucks, mainly for the domestic market. In 1999, it employed 10,600 workers and sold 1,000 trucks, virtually all (in excess of 90 per cent) for the home market. On further investigation, it transpired that the fall in production was gradual between 1980 and 1990, with 12,000 sold in 1990. Between 1990 and 1999, the downward trend was more dramatic, with a large fall in sales reported in the earlier part of the decade. This period (1990–1992) coincides with the early years of economic transition in the Romanian economy when the fall in national output was substantial.23 Further reductions in the workforce are presently taking place, with a
21 Using the average lei/US$ exchange rate for 1998 of 8,864 lei = 1 US$ (Tractorul 1999). 22 Again, it was March 2000 when a visit was arranged to the plant and an interview took place with a counsellor to the General Manager. On arrival, we were informed that the majority of the workforce was on a ‘technical unemployment’ day. In response to market conditions and the deteriorating financial situation of the firm, an agreement had been reached between management and the trade unions whereby parts of the workforce, on a rotating basis, would be laid off for a number of days in the year. On these days, workers receive a percentage of their salary, depending on the number of years service in the company. There is a provision to allow for up to 30 days in the current year. Up to March, there had been four ‘technical unemployment’ days. 23 The output fall in Romania for the period 1990–1992 was as follows: 1990 (-5.6%); 1991 (-12.9%); 1992 (-8.8%) (Source: EBRD).
Tax Arrears and the Romanian Enterprise Sector
89
current workforce of 9,150 workers. All 9,150 workers are full-time employees and close on 90 per cent are unionised. The average monthly wage before tax and social security contributions is two million lei ($110). All wage payments are made on time and the firm has no wage arrears. Strikes have played an important part in Roman’s (and Brasov’s) history. Under Ceausescu’s regime, workers in Roman (known then as the Red Flag factory) and elsewhere went on strike in November 1987. Riots ensued in response to the workers’ demands for an end to shortages and pay-cuts. In the more recent past, there were demonstrations in June and, again, in November 1999. The November 1999 strike was resolved when a government military order for Roman trucks was agreed. In return for this, Roman had some of its tax debt reduced. By 1999, Roman had divested itself off all its social assets with the exception of two blocks of flats that are currently occupied by unmarried employees. Its other social assets were sold to local firms and investors. Historically, Roman sold primarily to the home market, with domestic sales accounting for over 90 per cent. Since the demise of the CMEA market and with Europe being dominated by large competitors (the likes of Scania and Volvo from Sweden, Man and Mercedes from Germany, Iveco from Italy), Roman’s main external market has been Asia. Sales to the Asian market were hit by the Asian crisis of 1997 and it is only in recent months (late 1999) that market conditions have reportedly improved. As for the composition of the domestic customers, 60 per cent of buyers are small, private firms and the remaining 40 per cent are state-owned. As ties and relations are important, Roman’s customer base has not changed much over the years, with one exception. In early times, the military sector was a reliable customer but this has changed with the cuts in military spending and the downsizing of government. In recognition of these market changes, Roman is seeking new markets and new ways of selling its product. One option is leasing and with a bank credit guaranteed by SOF, Roman has managed to lease 45 trucks since November 1999. Despite this, management feels that this market will be constrained by the buyers’ limited access to finance. Management also expressed its concern over the granting of favourable tax conditions by the Romanian government to foreign firms operating in Romania. As for Roman’s suppliers, the majority of them are Romanian and like its customers, are traditional trading partners of Roman. The components and engines that are imported come from different countries, namely Germany, Italy, France, UK and Hungary. By 1999, Roman’s debts amounted to close on 1,500bn lei ($97.8m), of which 85 per cent is owed to the state.24 In 1999 alone, the company made a loss (after exceptional items) of 750bn lei ($48.9m). 25 The debt to the state comprises both 24 Using the average lei/US$ exchange rate for 1999 of 15,333 lei = 1 US$ (National Bank of Romania 2000). 25 The profit and loss statement is divided into three sections. The first reports operating revenues, expenditures and profit or loss. The second reports financial results, including
90
Transition, Taxation and the State
principal and interest, with the latter accounting for a greater percentage. In 1999, Roman paid off the outstanding principal that was due to the state budget and to social security, leaving penalties and interest charges of 650bn lei ($42.4m). The total debt to the local budget and to the special funds by the end of 1999 amounted to close on 1,000bn lei ($65.2m).26 The management of Roman was insistent that interest charged on late payment to the state was excessive and could not be paid. It also complained that because of the mounting interest charges, it was not in a position to avail of the 1999 government scheme allowing firms to pay off their principal due as of August 1999 in return for a writing-off of interest charges. On further questioning, Roman admitted that in return for its payment of outstanding principal, the government did agree to a rescheduling of interest over a five-year period. In response to questions relating to government subsidies and regional aid, Roman was insistent that it was not in receipt of any such aid. It did admit, however, that the government (by tolerating late or non-payments) was helping the firm to avoid bankruptcy and in doing so, was ensuring (for the short-term, at least) its survival. In its defence, the management of Roman argued that a closure of the firm would cost the government billions in welfare and other payments. In addition, Roman is an important employer in Brasov, with one job in Roman effecting 2.5 jobs elsewhere in Brasov, it claimed. Unlike its debt to government, Roman has only a small amount of debt owed to its banks and no debt to the utility companies. As for its trading partners, amounts due on receivables and payables are similar, close on 120bn lei ($7.8m). There has been a deliberate policy on behalf of the company to keep trade receivables and payables in line with each other. Of the amount due to suppliers, the majority of it was less than three months old. As for the utilities (gas and electricity), payments were generally made on time and in cash. The reason for prompt payment, or the absence of late payment, was the threat of a power cut. Roman believed that late or non-payment would result in disconnection. In addition, any late payment to the utility company was subject to an interest charge. When settling with its creditors, Roman generally pays in cash. Sometimes, depending on the creditor, Roman pays in trucks, in other firms’ goods (received by Roman in exchange for trucks) or in vouchers issued by the utilities (received by Roman in lieu of payment for delivery of trucks/components to clients). The normal period of trade credit given by Roman to its customers is 15 to 30 days. Trade credit extended to Roman by its suppliers is normally 30 days. On average, Roman receives 60 per cent of its payment in cash and 40 per cent in goods/vouchers via the compensation scheme. This so-called compensation scheme is common in Romania and works as follows. Suppose Roman sells a truck to customer A. A has no cash to pay for the truck but he has a receivable from B. This receivable is paid to Roman, in exchange for the truck. Likewise, Roman’s domestic suppliers interest on loans. The third section reports exceptional results, including penalties on late payments. 26 These figures were quoted during the interview.
Tax Arrears and the Romanian Enterprise Sector
91
will accept payment from Roman in goods, either its own products or others. The utility companies also take part in this scheme, accepting payment in goods and/or vouchers. In this scheme, invoices are exchanged and taxes accrue (VAT). Although these transactions only require the trading parties involved, large transactions in excess of 100 million lei are centralised, involving an agency within the Ministry of Industry and Trade. This clearing of inter-firm arrears function by the Ministry is not unusual in Romania where, since transition began, various enterprise surveillance or isolation programmes have been initiated by government with the aim, among other things, of clearing the financial payments ‘blockage’ brought about by the build-up of trade credit arrears in the enterprise sector. As for banks, Roman has reasonably good relations with its two main banks, Banca Comerciala Romana (Romanian Commercial Bank) and the Romanian Development Bank (owned by Société Generale). Roman’s outstanding debt to these and others (including a German Bank) was close on 120bn lei as of the end of 1999. A small amount is overdue. Likewise, it has had a small bank debt rescheduled by the bank-restructuring agency, AVAB (Agentia pentru Valorificarea Activelor Bancare). In response to a question relating to overdue bank debt and penalties, management inferred that the penalties imposed by these (often state-owned) banks were large and acted as a deterrent. In addition, late or non-payments would threaten any extension of credit in the future and would harm its relations with the bank. Although the government did have the power to block the bank accounts of delinquent firms, in practice it was not in the banks’ interests to have their clients’ accounts seized. Roman indicated that, in terms of payment priorities, the banks were paid first, followed by workers. This is contrary to legislation, which states that wages should be paid first, followed by payments to the state budget. In response to a question relating to government assistance, Roman was insistent on the need for some state aid. It expected credits from government for R+D spending and general investment spending. In addition, it hoped that the government would provide finance for leasing and installment schemes. Roman felt that government aid provided to foreign investors (Renault-Dacia was mentioned) should also be made available to domestic producers. This assistance would help Roman to reach its target of 55-60 per cent of the domestic market. Roman currently has 45 per cent of the home market. When questioned about the marketing and possible identity of these new buyers, management did not appear to have many ideas, only commenting that it would be private buyers. Although it did not state it explicitly, there is a suspicion that it will look to the government, and the military in particular, for new orders. Roman feels that it will be unable to capture market share beyond 60 per cent because of foreign competition (in receipt of government assistance) and its own inability/unwillingness to supply certain types of trucks. Despite the relatively low labour costs in Romania, the price of trucks manufactured and exported by Roman is close to world levels. According to management, this is because the engines in
92
Transition, Taxation and the State
these trucks are imported (from France, Germany and the USA).27 Management also admitted that further cuts in wage levels and/or the workforce were unlikely due to trade union opposition. Without a change in market conditions, the breakeven point of 1,700 sales (with sales of components) or 2,700 sales (without sales of components) would not be reached. F.3 Brasov Region Brasov is Romania’s second largest city and is located in the central part of Romania. The county accounts for 2.3 per cent of the whole surface of Romania and with a population of 636,434 inhabitants, accounts for 2.8 per cent of the country’s population. The city (municipality) of Brasov in 1997 had a population of 317,772. Table F.2 shows a selected number of indicators for Brasov county, as compared to Romania as a whole. Apart from the government, the biggest single employers in the Brasov region in 1999 were Roman S.A. with 10,600 employees, Tractorul S.A. with 10,112 employees and Rulmentul S.A. Between them, they account for a very large percentage of total employment in the region. All three firms pay wages above the average wage level in Brasov. Other important firms in Brasov include Hidromecanica and IAR Ghimbav. Hidromecanica Brasov, a manufacturer of pumps, is one of the 64 companies (along with Roman and Tractorul) shortlisted for privatisation under the World Bank PSAL programme. IAR Ghimbav is the Romanian airplane and helicopter manufacturer that was embroiled in the Bell Helicopters affair. The proposed sale of IAR Ghimbav to Bell Helicopters fell through when the Romanian government refused to buy the military helicopters from Bell. More recently, the Franco-German company Eurocopter has expressed an interest in buying IAR Ghimbav. The task of enterprise restructuring in transition countries is not easy. Romania is no exception. Despite the economic benefits that accrue from restructuring the enterprise sector, policymakers are well aware of the costly social and political consequences that are associated with restructuring. Brasov with its truck and tractor-making factories is a good example. With its history of strikes and demonstrations by its workforce, combined with a small number of very large employers, in what are declining industries beset with over capacity worldwide, it makes for an interesting case study.
27 Roman can claim back the custom duties that it pays on imported components, under an export facility scheme.
Tax Arrears and the Romanian Enterprise Sector
93
Table F.2 Selected Indicators for Brasov (County) and Romania, 1997
Population, mid 1997
000
Brasov County 636.4
Labour Force
000
289.1
9904.4
Employment, end 1997 of which
000
265.6
9023
State-owned Private Other
000 000 000
n.a. n.a. n.a.
2633 5186 1204
Agriculture Services Construction Industry Manufacturing Other
000 000 000 000 000
46.1 77.4 22.9 115.1 108.2 4.1
3384 2580 439 2450 2079 170
Registered Unemployed Unemployment Rate
000 %
23.5 8.1
881.4 8.9
Ave. net salary/economy
000 lei/month 000 lei/month 000 lei/month
692.8
632.1
767.8
693.4
748.6
628.8
Ave. net salary/industry Ave. net salary/manufacturing
Romania 22545.9
n.a. = not available. Source: Romanian Statistical Yearbook 1998, National Commission for Statistics, Bucharest.
F.4 Summary and Conclusions In the socialist system, firms were subject to a SBC. Under that regime, fiscal subsidies and tax concessions were (aside from their role in the pricing mechanism) instruments by which budget constraints were softened. With a paternalistic state, the source of the budget softness was government. In the transition period, new instruments for budget softness emerged, as did new or alternate sources of budget softness. In some transition countries, tax arrears, overdue bank credit and inter-firm trade credit arrears were often
94
Transition, Taxation and the State
used as ways of softening the budget constraints of enterprises. In addition to the state, banks and other firms were sometimes the sources of budget softness.28 In this case study, we chose two firms operating in the Romanian enterprise sector that, on initial investigation, are strong candidates for budget softness. In order for there to be evidence of budget softness, firms must be (i) loss-makers and (ii) in receipt of ex-post subsidies i.e. the losses result in a net flow of financing to the deleterious firms. From our case study of Roman and Tractorul, we know that both firms are loss-makers. It would appear that they are also in receipt of government subsidies, either in the form of tax concessions or cheap loans. Are these subsidies conditional, paid over to guarantee the survival of these loss-making firms? We attempt to answer this question by examining the conduct and behaviour of government. It does appear from this case study that there is some evidence of a recurring government rescue or bailout of the two big Brasov firms. The government is well aware of the difficult political and social factors that surround the big Brasov firms. As for the position of the firms themselves, the likelihood of state aid (based on past experience and on current bargaining power) weakens the incentive or motivation to restructure. This is evident in the case of Roman and Tractorul where, despite the collapse of their markets, employment numbers are still high, organisational changes have been slow to evolve and the search for new markets, products and suppliers has been sluggish. Well-publicised strikes have resulted in much media attention, often ending with the government conceding to the demands of the workforce/management. In these cases, the government’s role of creditor is counter positioned by its role of owner and employer. These concessions, if repeated, tend to undermine the government’s commitment to enforce hard budget constraints on delinquent firms. The credibility of its ‘no bail-out’ commitment is weakened by recurring infringements of financial discipline. In the context of ex-post bailouts, there is little doubt that the two firms in our case study are in a strong bargaining position, because of their poor financial situation. In the case of both firms, the government has tolerated late payment of taxes/social security contributions (at favourable penalty charges to the delinquent firms) and has rescheduled debt (principal or interest). It is also true in both cases that it is the government that behaves (and is viewed by its debtors) as the softest creditor. Other creditors, whether it is banks or suppliers, are unlikely to accept late or non-payment to the same degree as does the government. For the two firms in question, the risks associated with late or non-payment to banks and/ or suppliers are considered too great. It is likely that non-payment to banks would result in high penalty charges and a loss of banking services. Non-payment to certain utilities might result in disconnection. Suppliers are likely to stop shipping goods if 28 Often, as already alluded to in Chapter One, banks and firms are simply channels through which budget constraints are softened – the real source of the budget softness is the state. This happens when state-owned or controlled banks are under instructions from the government to continue financing loss-making firms and when firms accumulate trade credit arrears in the expectation that the state will bail them out.
Tax Arrears and the Romanian Enterprise Sector
95
they are not paid. Delayed payment to workers is not considered as an option due to a combination of factors (militant workforce, strong trade union membership, little precedence). When in trouble, these state-owned firms turn to the softest creditor for help, namely, the government. Although the state purports to be tough in its commitment to ensuring financial discipline in the enterprise sector, its actions suggest otherwise. In the case of both Roman and Tractorul, government assistance in the form of state orders, injection of cash, toleration of late payment of taxes and social security contributions, debt relief (rescheduling of principal and a writing off of penalties) and favourable tax treatment (similar to the concessions given to foreign investors) is sought. The evidence, outlined in this case study and elsewhere, suggests that, in most cases, the external assistance is given.29 This summary completes our case study analysis. Whether Roman, Tractorul and the machine-building industry in general (no doubt constituting part of Romania’s financial black hole) can restructure sufficiently to meet the demands of a market economy remains to be seen.
29 It is the recurring expectation of the external assistance that has a detrimental influence on the behaviour, and performance, of firms. Among other things it can result in low effort, an undermining of economic incentives and a delay in restructuring (Coricelli and Djankov 2001; Djankov and Murrell 2002).
This page intentionally left blank
Chapter Five
Effective Tax Administration in Transition Countries 5.1 Introduction Tax erosion and poor revenue mobilisation have been features of many TEs since the start of transition. The SBC that enterprises often face manifests itself, through the tax system, in the form of arrears, exemptions, deferrals and offsets. Another feature of the tax system in TEs that hinders revenue collection, both from lossmaking (the SBC syndrome) and profit-making firms alike, is a poor and ineffective tax administration. Weaknesses in tax administration (and in a broader sense, fiscal institutions) during transition are hardly surprising given the inevitable change from the pre-transition revenue administration system that depended on the state monobank as the main fiscal agent, a small number of compliant taxpayers (namely enterprises) and, in effect, one large public sector to a Western-style tax administration system where confrontation and non-compliance are not uncommon, multiple payment systems prevail, millions of taxpayers exist (often very small, the majority in the private sector) and there is a proliferation of taxes (often leading to a dispersion of administrative resources). This chapter attempts to measure the effectiveness of tax administration in TEs and how it compares to a benchmark for the mature market economies. Tax administration effectiveness can be measured by the difference between the tax that should be paid according to the tax laws and statutes and the tax that is actually paid and collected, what is traditionally referred to as the tax gap (Silvani and Baer 1997).1 In this chapter, we measure the effectiveness of tax administration by comparing statutory tax rates with effective tax yields. This method of measuring the administrative capacity of tax systems has been alluded to in the literature but not systematically pursued in crosscountry aggregate comparisons. Alex Radian, in his inquiry into tax administration in poor countries, noted that effective tax rates are lower than legal tax rates (Radian 1980). David Newbery (1987) raised the issue of differences between statutory and effective tax rates, again in the context of developing countries, when he stressed the importance of examining the effective tax system rather than the legally defined tax 1 The problem of tax collection and non-compliance can be addressed from the tax collector side (viewed in the context of effective tax collection and the administrative capacity of tax systems, as addressed in this chapter) or, more commonly, from the taxpayer side (as addressed by the vast tax compliance literature).
98
Transition, Taxation and the State
system. In the same World Bank publication, Vito Tanzi (1987) suggested that the gap between the statutory tax system and the effective tax system might be large in developing countries.2 Elsewhere, Burgess and Stern (1993) argued that the wedge between the statutory and effective tax systems can be reduced by improvements in administrative capabilities. In this chapter, we take a methodology previously used for measuring fiscal or revenue capacity in federal states and adapt it to enable cross-country comparisons of effective versus statutory taxation (ACIR 1962). We then use actual fiscal and national accounts data from 25 TEs and, as a benchmark, the average for the 15 member countries of the European Union, to measure the effectiveness of tax administration.3 Tax exemptions, deferrals, write-offs and arrears that firms receive or extract from the state are widespread, not only in TEs but in market economies as well. In a broader sense, these tax concessions are often manifestations of a tax system that is politicised. One possible result of this bargaining and general politicisation of the tax system is a low level of tax compliance combined with a high incidence of tax avoidance. Measuring the extent of this financial aid (to the entire enterprise sector as opposed to only the loss-making sector; the latter being analogous to the SBC phenomenon) using firm-level information is difficult and faces obvious data difficulties, e.g., these concessions may not be widely known or may not show up in the government’s budget. By measuring the difference between effective and statutory taxation at the aggregate level, this methodology enables us to obtain aggregate measures of the degree of effectiveness of tax collection that can be compared across countries, for different taxes, and over time. In this chapter we develop indicators that allow us to measure how broadly and strictly valued added tax, payroll tax and corporate income tax are implemented and complied with in TEs. We use the average of the EU-15 countries as an appropriate benchmark for comparison. The outline of the chapter is as follows. Section 5.2 explains the changes in the tax system due to transition from plan to market. Our methodology is outlined in Section 5.3. The data coverage and sources are described in Section 5.4. Section 5.5 contains the results and analysis. Our conclusions are outlined in Section 5.6. 5.2 The Tax System in TEs The tax system that existed under the socialist command economy was different from a Western-style tax system. As opposed to the market-based system that 2 In an interview with the IMF Survey in 2000, Tanzi said that ‘A third characteristic is the quality of the tax administration, which is much better in industrial countries where the actual, or effective, tax system is not very different from the nominal, statutory one’ (IMF 2000c). 3 The EU-15 before the 2004 enlargement. The EU-15 are Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, The Netherlands, Portugal, Spain, Sweden and the UK.
Effective Tax Administration in Transition Countries
99
collects revenue to finance public spending, the tax system in socialist countries was an instrument, along with the monetary and banking system, to fulfill the plan. More specifically, there was no corporate income tax system, in the usual sense of the term. State enterprises were subservient to the various ministries and any ‘profits’ or surpluses made were expropriated by the state. Likewise, losses were made good by arbitrary pricing and subsidies. Often, in respect of taxation, enterprises were treated on an individual basis according to their strategic importance and powers of negotiation. Tax rates were numerous and non-parametric, tax structures were complex and differentiated, and tax liabilities were not based on tax law but were discretionary and negotiable. The main sources of tax revenue typically were enterprise profit tax (i.e. profit remittances or deductions), payroll taxes and turnover taxes;4 direct taxes on individuals were unimportant. Although taxation as a percentage of GDP was high in socialist countries, administrative costs were low and tax collection was straightforward as firms tended to be large, few in number, state-owned and closely monitored. The generally accepted principles of modern tax administration, namely voluntary compliance and self-assessment, were absent from the system of revenue administration in Soviet and pre-transition times.5 Once the socialist system collapsed, TEs, some lacking an explicit, codified tax system (or culture), had to build a market-oriented, rule-based tax system (including a market-type tax administration) from scratch (McKinnon 1991; Tanzi 1993). The creation of a new tax system involved the introduction of a corporate income tax system. Not only did this involve changes in how enterprises were treated by the state in terms of taxation, but it was also introduced in conjunction with other policies, such as price liberalisation, demonopolisation and privatisation. A VAT system to replace the turnover tax was introduced in the early years of transition and was in place, sometimes with little preparation, in most TEs by 1994.6 Tax on individuals 4 The turnover tax, essentially the difference between the administratively set wholesale and retail prices, was a highly differentiated, product-specific tax. The rates (both positive and negative – subsidies were paid in the form of negative turnover tax rates), often in the thousands, were not legislated and were often changed subject to negotiation between enterprise and planner. As in the case of the enterprise profit tax (and its analogous tax, namely, corporate income tax), the turnover tax had very little in common with its nearest equivalent (i.e. sales tax) in modern market-based tax systems. 5 The structure and profile of tax administration that existed in the USSR just before the start of transition, namely the STS, is discussed in Chapter III.I of A Study of the Soviet Economy, Vol. 1. One indication of how straightforward and unsophisticated tax collection was in pre-transition days is the number of staff employed in the Union Ministry of Finance’s Department of Main State Tax Inspectorate (MSTI). In 1990, of a total staff of 40,000 in the STS, only 91 persons worked for the MSTI (OECD 1991). This tendency to understaff central headquarters continued throughout the 1990s. For example, by the mid-1990s in Russia, less than one per cent of the total STS staff worked in headquarters (Lopez-Claros and Alexashenko 1998; Tanzi 2001). 6 A VAT system has many advantages over the turnover tax in the facilitation of tax administration. In particular, a properly designed invoice-credit VAT creates an audit trial and
100
Transition, Taxation and the State
accounted for a small proportion of the total tax take in socialist countries: the transition to a market economy meant that a personal income tax system as operates in market economies was also to be introduced. Other changes in tax policy and design include introduction of import duties, reduction in export taxes, reduction in excess wage taxes and the proliferation of other taxes (Ebrill and Havrylyshyn 1999; Martinez-Vazquez and McNab 2000; Dobrinsky 2002).7 In the context of TEs and taxation, we are interested in the budget constraints of firms. Hence, the firm is the unit of analysis in this chapter. As for the different taxes collected from firms, corporate income or profit tax, VAT and social security taxes are mostly linear, flat rate taxes. Neither sales taxes (because they are levied at one stage only), excise taxes (because they are product-specific) nor personal income taxes (because of the multiple tax rates) are considered in our study. We do, however, treat social security contributions as a payroll tax. As mentioned above, we will concentrate on three taxes collected from the firm: corporate income tax (CIT), value-added tax (VAT) and social security tax (SST). Although VAT is essentially a tax charged on final purchasers, it is imposed at different stages of production at the firm level. For these three taxes, we estimate the difference between average effective tax rates and tax yields that would result if statutory tax rates were strictly applied. The methodology is explained below. 5.3 Methodology Our methodology is based on one commonly used in measuring tax or fiscal capacity in federal states. The ACIR Representative Tax System (RTS) method was initially proposed in the early 1960s and has been modified on several occasions since then (ACIR 1962, 1988). Essentially, by applying national average or representative tax rates to member-state tax bases, the RTS method shows the amount of revenue that could be collected by the individual member states of a federal country, i.e., their fiscal capacity. With some modifications, we apply this methodology, in this and the next chapter, to sovereign states (and transition countries, in particular) rather than to states within a federal system. We begin with some definitions. Statutory tax rates are the rates that taxpayers are required to pay by law. Effective tax rates are the realised average tax rates. These are the same average tax rates as employed by Whalley (1975), Lucas (1990) and Mendoza et al. (1994).8 uses businesses and traders as tax collectors. 7 Of course, many industrialised and middle-income countries introduced radical tax reforms in the 1980s. Many of these changes were to influence policy in the transition countries throughout the 1990s. 8 These are aggregate average tax rates as distinct from the marginal effective tax rates that are commonly used in studies of household income, income distribution and taxation. Aggregate tax rates are normally used in macroeconomic modeling and in the taxation, fiscal policy and economic growth debate. For the problems associated with measuring effective tax rates (average and marginal), see Fullerton (1984).
Effective Tax Administration in Transition Countries
101
Let Y be the gross tax base and T be actual tax payments; hence income net of tax is Y-T. We denote by t the statutory tax rate applied to gross income. The effective tax rate e, also defined on a gross basis, is calculated by dividing actual tax payments T by the appropriate gross tax base, or T (5.1) Y Using the statutory and effective tax rates thus defined, we calculate two indicators that measure the effectiveness of tax administration. The first indicator is the ratio of effective tax to statutory tax. The effective/statutory (E/S) ratio is defined as follows: e {
Effective / statutory ratio {
e t
{
T tY
(5.2)
This indicator measures the extent of the wedge between the statutory tax rate and the realised average tax rate. A ratio close to one, or a small tax gap, indicates that the effective tax rate is close to the statutory tax rate. A ratio below one, or a large tax gap, indicates that the effective tax yield is falling short of what application of the statutory tax rate would yield. Differences across countries in the extent of this shortfall in revenue may be accounted for by tax breaks, tax arrears, evasion, avoidance and other forms of non-compliance. As approximations of the gross tax base Y, we use national accounts measures of income: for VAT, total national income (GDP); for SST, income from labour; and for CIT, income from capital.9 These are, of course, only rough approximations of the actual statutory tax bases. For example: even in industrialised countries, large portions of the economy are exempt from VAT (e.g. public administration, financial services and education); corporate income tax applies only to corporations and the usual tax base is net of depreciation and interest; for all three taxes, entities must usually be over a certain size threshold before becoming liable for taxation. Furthermore, the national accounts statistics of transition countries are generally regarded as less reliable than those of developed market economies, including their statistics on the division of GDP into labour and capital income. National accounts measures of income do, however, have the important advantages of being both readily available and, in principle, readily comparable across countries. Moreover, our focus is not on levels of effective taxation in countries but on comparisons of levels across countries, and as noted the reasons for these deviations between income measures and statutory tax bases are found in all countries.
9 Due to data limitations, it is difficult to acquire a suitable proxy for the VAT tax base. Alternatives to our GDP measure include final domestic consumption (OECD 2001), aggregate gross value absorbed (Dobrinsky 2002) and taxable value added (UNECE 2004). Also, Dobrinsky (2002) adjusts the calculation of the VAT E/S ratio, by applying a weighted average, to account for multiple tax rates.
Transition, Taxation and the State
102
Our second indicator entirely avoids these possible problems with national accounts measures of income by simply not attempting to match tax payments to the appropriate tax base. Instead, the normalised tax yield (NTY) relates tax payments adjusted for cross-country differences in statutory rates to GDP, and is defined as follows: Normalised tax yield {
T b GDP t
(5.3)
where t is the statutory tax rate and b is a benchmark rate. Put simply, the normalised tax yield tells us what the tax yield (for a particular tax) would be for a specific country if the statutory tax rate were the same for all countries. In the definitions above, we have used the convention of a tax base that is gross of tax. In practice, statutory tax rates are sometimes defined relative to a tax base that is inclusive of tax and tax liabilities are paid out of gross income; and sometimes statutory rates are defined relative to a tax base that is net of tax. Corporate income tax is an example of the former; tax liability is calculated by applying the corporate tax rate to gross profit. VAT and SST are examples of the latter. Firms calculate their gross VAT liability by ‘adding on’ VAT as a percentage of the pre-tax price, and SST is typically calculated as a percentage of wages and salaries paid. We use the former convention – the tax base is gross of tax – for our calculations for all three taxes considered. This requires adjustments to the statutory tax rate for both VAT and SST.10 Denote by tN the tax rate applied to net income Y(1-t) that would yield the identical tax revenue as the tax rate t applied to gross income Y. We then have, by definition,
{ t N >Y (1 t )@
tY
(5.4)
Rearranging equation (5.4) yields
t
{
tN 1 tN
(5.5)
10 We are therefore using gross income only and adjusting statutory rates as necessary. The alternative approach would be to use statutory rates in conjunction with gross or net income, as appropriate. This approach, however, encounters data availability problems with net income. As noted above, we use national accounts measures of gross income that are only approximations of the actual tax base. Calculating net income from these approximations and the actual tax yield introduces further measurement error into the tax base approximations. By contrast, our tax rate adjustment in Equation 5.5 is an identity and introduces no such measurement error.
Effective Tax Administration in Transition Countries
103
Equation (5.5) is used to convert a tax rate defined by statute as applying to net income into the equivalent tax rate on gross income. We illustrate this adjustment by reporting the benchmark rates used to calculate our normalised tax yield indicator. We denote by bN the benchmark net of tax rate equivalent to the gross rate b. We take as our benchmark rates the approximate (rounded to the nearest five or ten per cent) average statutory tax rates in use in the 15 member states of the European Union: bN=20 per cent for VAT, bN=40 per cent for SST, and b=35 per cent for CIT. The following table reports the equivalent gross and net rates for these benchmarks; the figures in bold are the statutory rates as legally defined by statute.
Table 5.1 Benchmark Tax Rates, Gross and Net Equivalents b bN VAT 16.7% 20% SST 28.6% 40% CIT 35% 53.8% Note: Rates defined by Statute are in bold.
5.4 Data Coverage and Sources Our primary interest is in examining the administrative capacity of tax systems in TEs and how it compares to levels in well-established market economies. We use the mean of the EU-15 countries as a benchmark, taking 1996 as the benchmark year. We have data for 25 ex-socialist countries. The TEs that are not included are those that are normally treated outside the CEE/FSU setting (Mongolia, Vietnam), where the tax system is highly complicated (China) or where war has recently occurred (Serbia and Montenegro, Bosnia and Herzegovina). There are ten CEE countries (Albania, Bulgaria, Croatia, Czech Republic, FYR Macedonia, Hungary, Poland, Romania, Slovak Republic and Slovenia) and 15 FSU countries (Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan). All these countries have a corporate income tax system, of sorts. All 25 countries with the exception of Croatia, FYR Macedonia and Slovenia had a VAT system in place by 1996.11 All TEs in our study have a social 11 Of course, not all VAT systems adopted by TEs were the same. In particular, the VAT system adopted by CIS countries had several drawbacks, including the treatment of capital inputs, imports and exports and the cash method of accounting; the latter compounding the payment arrears problem in many CIS countries (Summers and Sunley 1995). However, these and many other problems with the definition of the tax base have been corrected in recent years. Evidence indicates that the longer a VAT system has been in place, the higher is the compliance rate (Agha and Haughton 1996). This suggests that TEs with low VAT/GDP ratios, as reported in Table 5.2, might expect a rate close to the OECD average of about seven per cent in years to come.
104
Transition, Taxation and the State
security tax system mostly financed by payroll taxes, with contributions being made by employers and/or employees. For all countries in our study, tax coverage is for general government, comprising central and state, regional or provincial units of government and local government. An exercise in calculating statutory versus effective taxation depends on data pertaining to tax rates, tax takes and tax bases. As for statutory taxation, the basic tax rates are taken primarily from international tax handbooks. In particular, we use the International Bureau of Fiscal Documentation’s (IBFD) European Tax Handbook, Coopers & Lybrand’s International Tax Summaries and Ernst & Young’s Worldwide Tax Guides. We also use various EBRD Transition Reports. Tax payments were obtained from the governments’ fiscal accounts where taxes are reported on a cash basis, i.e., counting actual receipts rather than accrued liabilities. Where possible, tax payments data are from the IMF’s Government Finance Statistics Yearbook (GFSY) or the CIS Statistical Yearbook. Other publications used include the OECD’s Revenue Statistics, the IMF’s Staff Country Reports and statistical yearbooks for various countries (noting that the IMF and the OECD classification of taxes are very similar). As already mentioned, the proxy tax bases for the three types of taxes are taken from the national accounts. For VAT, we use GDP as a proxy for the VAT base. Although in all VAT systems there are some goods and services that are exempt from VAT, the most important of these exemptions are quite standard and hence GDP is a reasonable proxy.12 For CIT, we use gross operating surplus as a proxy for the tax base on corporate capital income. Operating surplus corresponds to value added after deducting compensation of employees and net taxes on production. Due to the difficulties in acquiring reliable estimates for the operating surplus of corporations for 25 countries, we instead use the operating surplus of the economy.13 For SST, we use compensation of employees as a proxy for the tax base. Compensation of employees is the sum of gross wages and salaries plus employers’ social contributions.14 12 As GDP is larger in size than the actual tax base, this leads to a downward bias in the VAT E/S estimates (Dobrinsky 2002). However, as stated earlier, our focus is not on levels of tax effectiveness in countries but on comparisons of levels across countries. 13 Likewise, we use gross rather than net operating surplus because the treatment of depreciation varies from one jurisdiction to another. Aside from depreciation provisions, corporate tax systems differ from country to country, in respect of, for example, treatment of interest payments and other normal business costs, stock appreciation provisions and integration with the personal income tax system. Accordingly, computations arising from cross-country comparisons of corporate tax systems should be treated with caution. This is particularly true when the cross-country comparisons include the likes of Russia and other CIS countries whose accounting rules are different to international standards. However, a common feature of CIT systems in transition countries in early years was the proliferation of exemptions, holidays and tax incentives. 14 I wish to thank an anonymous referee for comments on differences in the definition of tax bases. In the context of this exercise, this is a problem not only for transition countries but also for established market economies. However, it is acknowledged that many transition
Effective Tax Administration in Transition Countries
105
Table 5.2 Statutory Tax Rates and Tax/GDP Ratios for 25 TEs, 1997 Country
VAT Statutory VAT/ tax rate GDP
SST Statutory tax rate
SST/ GDP
CIT Statutory tax rate
CIT/ GDP
CEE Countriesa Albania Bulgaria Croatia Czech Republic FYR Macedonia Hungary Poland Romania Slovak Republic Slovenia
12.5 22 -22 -25 22 18 23 --
4.6 6.2 -7.1 -7.9 8.3 4.7 8.4 --
42.5 44 43.4 47.5 30.1 57 48.2 34 50 38
3.9 6.9 14.4 15.2 12.3 13.1 11.0 7.1 14.4 13.8
30 36 35 39 15 18 38 38 40 25
0.7 6.4 2.0 3.4 0.7 1.9 3.1 4.3 3.7 1.2
FSU Countriesa Armeniab Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistanb
20 20 20 18 20 20 20 18 18 20 20 20 20 20 17
3.3 3.8 9.4 10.4 3.2 3.5 5.6 8.8 8.7 9.4 7.2 1.5 7.2 8.1 6.1
38 39 37 33 34 32 37 37 31 40 39.5 38 31 40 43
2.9 2.5 10.1 10.7 2.2 6.2 5.9 10.5 7.0 7.2 9.9 1.6 4.5 11.1 6.7
30 32 30 26 20 30 30 25 29 32 35 40 25 30 37
2.5 2.8 4.7 1.9 0.6 2.4 1.1 2.4 1.6 2.4 4.2 1.2 5.2 6.1 7.9
Notes: a. The two country groups used were chosen on geographical and historical grounds rather than on any economic criterion. b. The figures for Armenia and Uzbekistan are for 1996. -- = not applicable. Sources: IBFD 1998; IMF 1999a, 1999b.
Although imperfect, these are reasonably good approximations given the omissions in the national accounts, the dubious nature of some transition countries’ data and the cross-country nature of the exercise. The main publications used for national accounts data are the IMF’s International Financial Statistics Yearbook (IFSY), the OECD’s National Accounts Main Aggregates, the CIS Statistical Yearbook and statistical yearbooks for various countries.
countries did implement, throughout the 1990s, comprehensive changes to the tax base for different categories of tax. For example, for corporate income tax, these included abolition of certain exemptions, inclusion of deductible expenses, implementation of loss carryforwards and simplification of depreciation rules (Ebrill and Havrylyshyn 1999). By the end of the 1990s, differences in the definition of the tax base across the transition countries had narrowed.
106
Transition, Taxation and the State
The data sources used for all 25 countries are listed in Appendix G. The statutory tax rates and the actual tax/GDP ratios for each of the three taxes, for our 25 countries are reported in Table 5.2, for the year 1997.15 As we can see from Table 5.2, many of the 25 transition countries in our study have the same statutory tax rates. This is particularly true for VAT where 11 of the FSU countries have a VAT rate of 20 per cent.16 Yet, the actual tax/GDP ratios across countries differ quite substantially. One interpretation of differences in these tax ratios is the quality of tax administration across countries. The methodology outlined in section 5.3 allows us to investigate this further. 5.5 Results and Analysis Table 5.3 is a cross-country comparison showing the two indicators for the three different taxes for the 25 transition countries. We report estimates for the TEs for 1997 as the 1998 data for some FSU countries are distorted by the August 1998 financial crisis in Russia. The benchmark is the 1996 average of the EU-15 countries. Our estimates are highly approximate and precise values should be treated accordingly. Nevertheless, they do provide a measure of effective tax administration and, in the case of some TEs, point to poor tax collection and weak tax administration. 17 As we can see from the table, the (unweighted) means of the E/S ratios in the EU15 for VAT, SST and CIT are 0.45, 0.88 and 0.24 respectively. When we normalise the EU-15 VAT, SST and CIT yields by our benchmark rates of 20 per cent (b=16.7 per cent), 40 per cent (b=28.6 per cent) and 35 per cent respectively, we get normalised tax yields of 7.4, 12.5 and 3.0 per cent of GDP respectively. How do the rates for TEs compare with these levels for the EU? The results indicate that the 25 transition countries on average are not as effective in tax collection or enforcement compared to the average for EU countries. The (unweighted) means of the E/S ratios for the TEs for VAT, SST and CIT are 0.40, 0.69 and 0.23 respectively. Moreover, the variations from the mean are larger for the TEs as compared to the EU countries. For the EU-15 countries, the standard deviations of the E/S ratio for the three tax categories are 0.06, 0.18 and 0.12 respectively. This compares to 0.15, 0.20 and 0.14 respectively for the TEs. 15 E/S estimates for previous years are not reported here (but are reported in Appendix H for reference) as official statistics for the transition countries (national accounts and government fiscal data) for the early years of transition are not very reliable. This precludes any time series analysis. 16 This is due to the Soviet VAT system introduced in 1991 and adopted in all CIS countries. With all CIS countries initially applying the origin principle for trade among themselves, as opposed to the normal destination principle, they followed the example of the main trading partner’s (Russia) choice of rate, that is, 20 per cent. 17 Treisman (1999) uses a ‘tax accounting’ method to measure the effectiveness of tax collection, in Russia. In his paper, the unexplained parts of the fall in tax revenues are attributed to the poor quality of tax administration, i.e. ineffective tax collection.
Effective Tax Administration in Transition Countries
107
For many transition countries, revenue erosion has become a serious obstacle in their attempts to embrace effective fiscal policy. Pre-transition, many ex-socialist economies had high government revenue shares of GDP. With most of economic activity taking place in the state sector, tax collection was a straightforward task. In contrast, for a market economy, the private sector dominates and confrontation between taxpayers and tax collectors is not uncommon. Thus, in the transition from a centrally planned to a market economy, a fall in revenue was not unexpected. Yet, for some transition economies, the fall in revenue has been excessive, with tax/GDP ratios currently at levels below what is considered normal in many market economies. Of course, for many TEs with income per capita levels below $1,000 per annum, tax capacity is low (see Chapter Six). Our research indicates that, in addition to having a low tax capacity, some TEs have relatively low tax effectiveness rates, as the evidence in Table 5.3 indicates.
Table 5.3 Statutory and Effective Taxation, 1997 Country
VAT
SST
CIT NTY
EBRD Transition Indicatora
E/S
NTY
E/S
NTY
E/S
CEE Countries Albania Bulgaria Croatia Czech Republic FYR Macedonia Hungary Poland Romania Slovak Republic Slovenia
0.42 0.34 -0.40 -0.40 0.46 0.31 0.45 --
7.0 5.7 -6.6 -6.6 7.7 5.1 7.5 --
n.a. 0.61 n.a. 0.94 n.a. 0.80 0.76 0.83 0.93 0.94
3.7 6.5 13.6 13.5 15.2 10.3 9.7 8.0 12.3 14.3
n.a. 0.31 n.a. 0.23 n.a. 0.26 0.20 0.21 0.22 0.15
0.8 6.3 2.0 3.0 1.6 3.7 2.9 4.0 3.3 1.6
2.58 2.75 3.00 3.46 2.63 3.67 3.42 2.67 3.25 3.21
FSU Countries Armeniab Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistanb
0.20 0.23 0.57 0.68 0.19 0.21 0.34 0.58 0.57 0.56 0.43 0.09 0.43 0.49 0.42
3.3 3.8 9.4 11.4 3.2 3.5 5.6 9.6 9.6 9.4 7.3 1.5 7.2 8.2 7.0
0.27 0.46 0.87 0.83 0.38 0.68 0.68 0.76 0.71 0.60 0.71 0.31 n.a. 0.80 0.61
3.0 2.6 10.7 12.3 2.5 7.3 6.2 11.1 8.4 7.2 10.0 1.6 5.5 11.1 6.4
0.16 0.12 0.39 0.22 0.04 0.15 0.06 0.23 0.13 0.18 0.33 0.04 n.a. 0.63 0.47
3.0 3.1 5.4 2.6 1.1 2.8 1.3 3.4 2.0 2.6 4.2 1.0 7.3 7.1 7.5
2.38 2.04 1.63 3.42 2.71 2.71 2.83 3.08 3.04 2.63 3.00 1.58 1.46 2.46 2.38
EU-15 Mean
0.45
7.4
0.88
12.5
0.24
3.0
Notes: a. EBRD Transition Indicator: average indicator for 1997. b. The estimates for Armenia and Uzbekistan are for 1996. -- = not applicable. n.a. = not available. Sources: Author’s calculations.
108
Transition, Taxation and the State
We now turn our attention to factors related to effective tax administration in transition countries. We begin by investigating the relationship between the effectiveness of tax collection and progress in transition. A priori, we might expect to find a positive relationship between progress in transition and effective tax administration. We take as our measure of progress in transition the average of the EBRD transition indicators for 1997;18 these are reported in Table 5.3 for all 25 TEs in our study. We present the relationship between progress in transition and the effectiveness of tax administration by means of a scatterplot in which the average EBRD transition indicator is plotted on the X -axis and the E/S ratio is plotted on the Y-axis. This exercise is carried out separately for the three categories of taxes covered in our study.19 The horizontal line in each of the scatterplots is the relevant EU mean, plotted to provide a benchmark for comparison. All three scatterplots are depicted in Figures 5.1a,b,c (at the end of the chapter). Our second indicator, the normalised tax yield (NTY) is plotted against the average EBRD transition indicator in Figures 5.2a,b,c. As the E/S ratio and the NTY measures are highly correlated, the scatterplots in Figures 5.1a,b,c and 5.2a,b,c are similar. The relationship between progress in transition and effectiveness of tax administration comes out quite clearly in two of the three taxes that we are examining, namely VAT and SST. From the scatterplots we can see that there are no countries advanced in transition that have low tax collection administrative capacities. The largest difference is between the so-called leading reformers (Poland, Hungary, Slovenia, Czech and Slovak Republics, the Baltic States) and the laggard reformers (Albania, Bulgaria, Romania and most of the CIS countries), but there are other differences of interest between the TEs. One interesting comparison is between Ukraine and Russia. Although Russia has made more progress in transition than Ukraine (as judged by the EBRD transition indicators), it fares worse in terms of the effectiveness of tax collection (as measured by our E/S ratio).20 Russia’s poor relative performance as regards tax collection may have something to do with the nature of 18 The EBRD transition indicators are a set of numerical indicators across a range of dimensions, under the headings – markets and trade, enterprises, and financial institutions. The purpose is to measure the progress of economic reforms (see Appendix A in the earlier part of the book for how the EBRD transition indicators are used to measure progress in institution building). A more detailed explanation can be found in any of the EBRD Transition Reports. 19 As we can see from Table 5.3 the results for CIT are different to those for either VAT or SST, both in terms of levels and relative positions. This observation applies to the two groups of countries in our study, namely EU countries and transition countries. Accordingly, in any analysis of our calculations, we treat the three categories of taxes separately. 20 As one colleague noted, the VAT E/S estimate for Russia (a country regarded as having a weak tax administration) is close to the EU-15 average. However, the VAT E/S estimates for all other years preceding 1997 are lower, averaging 0.38 for the period 1993-1996. The same is true for SST. The 1997 reported E/S estimate for SST is 0.71 but the average estimate for the preceding four years is 0.63.
Effective Tax Administration in Transition Countries
109
its federal tax system (see Chapter Two). In particular, perverse incentives arising from divided property rights between different levels of government exacerbate the tax collection problems in Russia. It is not uncommon for enterprises and regional governments to collude against the federal government and the tax collection agency, the Ministry of Taxation and Fees. Likewise, any improvements in tax collection at subnational levels are likely to be penalised or ‘taxed away’ by reductions in transfers (Shleifer and Treisman 2000). We examine the three categories of taxes separately. In the discussion that follows, we only refer to one set of scatterplots, namely Figures 5.1a,b,c. Value-Added Tax There is a strong positive correlation between progress in transition and the effectiveness of VAT collection.21 All the leading reformers that have a VAT system have E/S ratios that are close to the EU benchmark of 0.45. These advanced reformers are all clustered in the top right-hand corner of Figure 5.1a.22 In contrast, the majority of the laggard reformers are clustered in the bottom left-hand quadrant: the slow reformers have E/S ratios below the EU benchmark. An interesting feature is captured in the top left-hand quadrant where a small number of slow reformers (Moldova, Belarus, Turkmenistan and Ukraine) have high E/S ratios. Slow reformers with relatively high E/S ratios and normalised VAT yields may be accounted for by the observation that these TEs have maintained a functioning state, a feature that may have prevented the revenue erosion that is prevalent in most of the partial reforming TEs.23 The ability of some slow reforming TEs to collect taxes (Ukraine and Belarus, in particular) has been alluded to in the transition literature, in Murrell (1996), EBRD (1994, 1999) and elsewhere. Johnson, Kaufman and Shleifer (1997) argue that it is the repressive nature of states and their willingness to suppress the unofficial economy that explains high collection rates. Little or no reform in these countries has allowed these states to maintain revenues at relatively high levels. Social Security Tax As in the case of VAT, there appears to be a positive correlation between progress in transition and the effectiveness of SST collection. All the leading reformers occupy the top right-hand quadrant of Figure 5.1b. This group of TEs has an E/S ratio similar to the EU benchmark of 0.88. This is in sharp contrast to the bulk of the slow reformers that have E/S ratios and normalised tax yields below 21 The VAT effectiveness rate as applied here is sometimes referred to as the VAT efficiency ratio and is used to measure VAT productivity (Tanzi and Davoodi 2000; Stepanyan 2003). 22 The high E/S ratios for the three Baltic States may be due to the lower VAT rate and the size of their populations, two factors that are associated with high VAT compliance (Agha and Haughton 1996). 23 The case of Moldova is less clear. For example, it is true that VAT collection in Moldova has improved, due to the elimination of many exemptions. Nonetheless, its total tax/GDP ratio, at close to 30 per cent in 1997, seems very high for a country with a GDP per capita of less than $600 per annum. One possible explanation is that the GDP figure may be an underestimate, failing to adequately account for the unofficial economy.
110
Transition, Taxation and the State
the EU benchmark. Among this group, the lowest E/S ratios are concentrated in TEs that have suffered internal conflicts (Armenia, Azerbaijan and Tajikistan) or that have witnessed a collapse of the state (Georgia, for example).24 Again, we see that there are a small number of so-called slow reformers that have managed to maintain tax discipline, in this case with respect to payroll taxes. Belarus and Ukraine both have SST E/S ratios close to the EU benchmark. As with VAT collection, strong presidential leadership (Lukashenka and Kuchma, respectively) combined with functioning state institutions, albeit in need of reform, may explain these high E/S ratios and normalised tax yields. Corporate Income Tax The mean of the E/S ratio in the EU-15 for corporate income tax is 0.24. The normalised tax yield is three per cent of GDP. Again, we see from Figure 5.1c that the leading TEs are all tightly clustered around the EU benchmark. This contrasts sharply with the laggard reformers, some with E/S ratios above the EU level. One possible explanation for why some of the slow reformers have high normalised CIT yields is the upward inflation bias in profits arising from historical cost accounting. High inflation rates prevalent in TEs would generate large profits and high tax yields. This may explain why the leading TEs, in general, experience a fall in the E/S ratio for CIT during the transition period. The average E/S ratio for CIT in 1991, at the beginning of transition, for Hungary, Poland and Czechoslovakia was 0.42. By 1997 the average CIT E/S ratio for these (now four) advanced reformers had fallen to 0.23. To summarise, leading reformers (Central European countries and the Baltic States) and countries that have maintained a functioning state (Belarus, Ukraine and Uzbekistan) have higher levels of tax compliance and collection than either slow reformers (Romania, Bulgaria, Kazakhstan), countries with decaying or corrupt states (Russia, Georgia) or countries that have suffered internal conflicts (Armenia, Azerbaijan, Tajikistan).25 A somewhat similar result emerged from the EBRD/WB BEEPS section on progress in economic reform, quality of governance and state intervention. Whereas countries that have adopted partial reforms score badly in terms of the quality of governance, it is the most advanced and the least advanced reforming countries that score well in terms of governance. Likewise, the survey results indicate that ‘…progress in transition is not necessarily synonymous with a reduction in state intervention in enterprises.’ (EBRD 1999). Again, high levels of 24 Following the devastating period of the early 1990s, Georgia did manage to implement improvements to its tax administration structures and procedures during 1994–1997. Some of these changes are outlined in Tanzi (2001). This improvement is reflected in the change in the E/S ratio from 1994 to 1997. For example, the E/S ratio for VAT increased from 0.07 in 1994 to 0.19 in 1997. For SST, the E/S ratio increased from 0.09 in 1994 to 0.38 in 1997 (see Appendix H). 25 After the 1996/97 financial crisis in Bulgaria, the authorities introduced various reform measures, including changes to the tax administration system. These and other tax reform measures contributed to an improvement in tax collection in 1998, reflected in higher E/S ratios for both VAT and SST (see Appendix H).
Effective Tax Administration in Transition Countries
111
state intervention are evident in many of the least advanced transition countries and in the leading transition countries. These results raise serious issues, including the need to rethink, in the context of the transition experience and often in a political and institutional vacuum, the state’s capacity to govern and the need for market institutions to develop. The EBRD’s transition indicators summarise overall ‘progress in transition’. We now briefly explore one possible specific contributor to the ineffectiveness of tax administration, namely corruption and bribery. Tax administrations frequently figure near the top of public sector organisations with a high incidence of corruption (Gill 2003). It is not uncommon for enterprises in TEs (and in market economies, although presumably less so) to pay bribes to government officials in return for various services or favours. As these payments are direct private benefits to public officials, they do not turn up in the government fiscal accounts. Yet, in all other respects, they can be viewed as unofficial taxes that add to the tax burden of enterprises. Is it possible that TEs that report low ‘official’ tax revenue shares as a percentage of GDP have high ‘unofficial’ taxes, in the form of bribes?
Table 5.4 Bribe Tax and Corruption for TEs Country CEE Countries Albania Bulgaria Croatia Czech Republic FYR Macedonia Hungary Poland Romania Slovak Republic Slovenia
Ave. Bribe Tax
CPI
n.a. 3.5 2.1 4.5 n.a. 3.5 2.5 4.0 3.7 3.4
7.7 6.7 7.3 5.4 6.7 4.8 5.8 6.7 6.3 4.0
Country FSU Countries Armenia Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan
Ave. Bribe Tax
CPI
6.8 6.6 3.1 2.8 8.1 4.7 5.5 n.a. 4.2 6.1 4.1 n.a. n.a. 6.5 5.7
7.5 8.3 6.6 4.3 7.7 7.7 7.8 6.6 6.2 7.4 7.6 n.a. n.a. 7.4 8.2
Note: n.a. = not available. Sources: EBRD 1999; Transparency International’s website http://www.transparency.de/.
In Table 5.4 we reproduce a measure of the extent to which firms pay bribes to government officials. The measure was constructed from the BEEPS, conducted by the EBRD and the World Bank in over 3,000 firms in 20 countries and reported in Transition Report 1999 (EBRD 1999). To estimate a measure of bribes, firms were asked what percentage of annual revenues were made by ‘firms like yours’
112
Transition, Taxation and the State
in ‘unofficial payments’ to public officials. In the countries surveyed, the average bribe tax ranges from a low of 2.1 per cent of annual revenues in Croatia to a high of 8.1 per cent in Georgia. For comparison, Transparency International’s Corruption Perceptions Index (CPI), the most recognised measure of corruption, is also reported.26 Figures 5.3a and 5.3b plot our E/S ratio measure for VAT and SST against the average bribe tax as a percentage of annual firm revenues, for 18 and 19 transition countries respectively.27 Countries that have a high measure of effective tax administration (Estonia, Poland, Slovak Republic and Belarus, for example) also have a relatively low average tax bribe. In contrast, countries with ineffective tax administration (Armenia, Azerbaijan and Georgia, for example) have a relatively high average tax bribe. The correlation coefficients are -0.54 and -0.74 respectively.28 As we did before, we plot the normalised tax yield (NTY) in Figures 5.4a,b. Again, as we expect, the scatterplots in Figures 5.3a,b, and 5.4a,b, are similar. If we use Transparency International’s CPI as the measure of corruption, we get similar results. Although interesting, more evidence needs to be gathered before any strong conclusions can be drawn. 5.6 Conclusions In this chapter we adopted an existing methodology to measure the effectiveness of tax administration. Comparing effective with statutory taxation allows us to get a handle on the administrative capacity of tax systems. The results indicate that, on average, the 25 TEs are not as effective in tax collection as compared to the average of the EU countries. A more surprising result is the differences between TEs: in particular, the ability of some slow reformers to maintain high tax effectiveness rates. In general, tax administration in transition countries has been undermined by the use of tax arrears, amnesties and offsets, barter transactions, lobbying and negotiating tax liabilities, tax evasion and other non-compliance practices and activities. As for policy implications, tax administration reforms aimed at improving tax collection and compliance have lagged behind general tax policy reforms since 26 The organisation Transparency International ranks countries, on the basis of surveys, in terms of the degree to which corruption is perceived to exist among public officials and politicians. The CPI score ranges from 0 (highly corrupt) to 10 (virtually corrupt free). To allow for comparisons with the BEEPS bribery tax, we have adjusted the CPI scores, e.g. Bulgaria’s CPI score of 3.3 is adjusted to 6.7. The CPI scores recorded in Table 5.4 are for 1999; scores for previous years are unavailable for many of the TEs. 27 From the 20 countries in the EBRD/WB survey, we exclude Croatia and Slovenia from Figure 5.3a as neither country had a VAT system in place in 1997. Figure 5.3b does not include Croatia as we have no E/S ratio. 28 Interestingly, when examining the link between corruption and the public finances, Tanzi and Davoodi (2000) consider a similar exercise, plotting the relationship between the VAT efficiency ratio and Transparency International’s index of corruption, the CPI.
Effective Tax Administration in Transition Countries
113
transition began, often because administrative reforms (and consequently the benefits) take time (Martinez-Vazquez and McNab 2000; Mitra and Stern 2003). Market-oriented fiscal institutions (tax administration and treasury systems, for example) do not develop overnight. Administrative reforms involve changes in incentives and in the behaviour of taxpayers and public officials. In general, TEs with tax collection problems need to widen the tax base by subjecting previously exempt income to taxation, reduce exemptions and privileges and, where possible (subject to normal equity considerations and country-specifics), implement lower (to discourage tax avoidance and evasion) and single (to avoid the rent-seeking activities of producers) tax rates. Although the day-to-day administrative side of the tax system needs to be separated from political interference, ending the tradition of bargaining and negotiating between the authorities and the taxpayer might prove difficult given the political constraints in TEs. One reform that can be introduced quickly is the reduction in the discretionary power of tax officials and inspectors. Well-established tax administration measures that TEs might consider include strengthening the tax administration (assessment, audit, appeals, collection enforcement) agencies in terms of organisation (by function rather than by tax, taxpayers or economic sectors) and personnel (higher salaries, more graduates, staff training and a code of conduct), codification of the various tax laws, a credible system of penalties (monetary and otherwise) for non-compliance, use of third-party information, expansion of withholding taxes, simplifying the identification (by use of a unique tax identification number), registering, reporting and filing requirements, and greater efforts at improving the data on taxpayers through the deployment of modern IT systems. Other measures include the enactment of a tax administration law providing tax authorities with legal powers to undertake collection enforcement actions and legal provisions for taxpayers’ rights, establishment of large taxpayers units (LTU), improving taxpayers services through the creation of customer services units and information campaigns (emphasising the role of tax administrator as a facilitator as well as an enforcer while at the same time viewing the taxpayer as a client rather than an offender), avoiding multi-tax amnesties and simplify tax forms (Silvani and Baer 1997; Martinez-Vazquez and McNab 2000; Bird 2003). As in other countries that have adopted successful tax administration reforms, TEs with limited administrative and institutional capacities should initiate pilot projects in advance of any large-scale changes. Pilot sites can be used to design and test new procedures at limited cost and risk to the authorities.29 In the implementation of these reforms, it is important that the constraints (and the traditional values and practices) of TEs are recognised, as are the dangers of
29 The tax administration in Hungary is generally considered to be the leading tax agency in ex-socialist transition countries. For an interesting account of the tax administration reforms affecting the Tax and Financial Control Office (APEH), see Pitti and Vacquez-Caro (1998).
114
Transition, Taxation and the State
transplanting Western-style practices.30 It is recognised that improvements will come with the development of civil society and greater trust in government and its institutions. At present in transition countries, taxpayers’ perceptions of uneven enforcement and impartiality, equity and fairness, legitimacy and corruption contribute to the high levels of non-compliance. Other administration issues for transition countries that may impact on the effectiveness of tax collection include the independence of the revenue agency vis-à-vis the Ministry of Finance (and its relationship with the civil service), the level of co-operation with other collection agencies (customs, social security), and the regional structure and decentralisation of tax administration (Gill 2003). With respect to the latter, it is generally recognised that the system of dual subordination that confronts tax officials in Russia hinders effective (federal) tax collection. Greater allegiance must be required of regional and local tax offices to national tax administration as opposed to regional and local authorities. Separate tax collection agencies for different levels of government, as exists in many other federal countries, is an alternative solution. Possible suggestions for related work would be to extend the data coverage to a larger set of countries, and in particular, non-EU OECD countries and, where possible, developing countries (see next chapter). Inclusion of non-EU OECD countries would provide us with an alternate and possibly more suitable benchmark to the EU-15. The more interesting possibility would be to apply our methodology to tax data for a range of developing countries in Latin America, Asia and Africa and for the Newly Industrialised Countries of South East Asia. This would allow us to compare ex-socialist transition countries with countries that closely resemble TEs either in terms of initial conditions or in terms of a ‘transition’ experience. For example, a more suitable comparison for the fast growing countries like Poland, Estonia and Hungary might be the Tiger countries of South East Asia. Likewise, the FSU countries of Central Asia and the Caucasus states more resemble, in terms of GDP per capita and sectoral composition, developing countries than EU countries. One implication of this would be a need to extend the tax classifications to include foreign trade taxes. Although revenue from duties on international trade is tiny for EU countries, it is a major source of revenue for developing countries and for some transition countries. By extending the coverage to developing countries, it would also allow us to make more meaningful policy prescriptions as regards tax reform for transition countries. Alternately and in the context of EU enlargement, tax harmonisation and fiscal convergence, our effective tax rates can be used to examine the overall 30 Although tax administrations in transition countries and developing countries share many similar problems, the postsocialist legacy in TEs has some unique features. These include a general mistrust of government and public institutions, customised, opaque and negotiable taxes, no culture of voluntary taxpayer compliance or self-assessment, lack of awareness of ‘tax’ obligations and payment (Mitra and Stern 2003). Features of tax administration in the 15 former Soviet Union republics are listed in Ebrill and Havrylyshyn (1999).
Effective Tax Administration in Transition Countries
115
tax burden and the distribution of the tax burden across different tax bases in EU countries and in the transition accession countries. In this chapter, we briefly examined some factors that might impinge on tax administration and collection in transition countries. In particular, we explored the significance of two factors, namely progress in transition and corruption. Other possible explanatory factors may include the shadow economy and tax evasion, the initial conditions, political constraints and the distribution of power, GDP per capita and levels of development. Some of these factors are not unique to TEs, applying to developing and developed countries alike. With respect to fiscal and tax reform, the first decade of transition has focused primarily on tax policy. If further revenue erosion is to be avoided, the next decade must concentrate on administrative reform. In the Transition Report 1994, the EBRD called for a strengthening of tax administration, an issue that ‘…lies at the core of fiscal reform’ (EBRD 1994).31 For many TEs, this policy recommendation is as relevant today as it was in the early years of transition. This applies even in the leading TEs, as the United Nations Economic Commission for Europe noted that, for the EU accession states, the ‘…efficiency of tax collection reflects the organisational capability of the tax administration to enforce tax laws and regulations, and administrative capacity is still relatively underdeveloped in the acceding countries’ (UNECE 2004). In the next chapter, we use an IMF methodology to measure the tax capacity and effort of TEs. This allows us to measure differences between actual tax collection (our primary concern in this chapter) and potential tax collection, based on country characteristics (the subject matter in the next chapter).
31 This view is shared by many others including Ebrill and Havrylyshyn (1999), Martinez-Vazquez and McNab (2000) and Mitra and Stern (2003). This observation is not new. Stanley Surrey, in 1964, wrote ‘The concentration on tax policy – on the choice of taxes – may lead to insufficient consideration of the aspect of tax administration.’ (Surrey 1964). Of course, tax policy and tax administration are intrinsically linked so that any future change in tax policy must be cognisant of the administrative capacities of tax systems in transition countries.
116
Transition, Taxation and the State
Appendix G: Data Sources for Transition Countries Table G.1 Data Sources for the 25 TEs Country Albania Bulgaria Croatia Czech Republic FYR Macedonia Hungary Poland Romania Slovak Republic Slovenia Armenia Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Moldova Russia Tajikistan Turkmenistan Ukraine Uzbekistan
Tax Rates IBFD; EBRD IBFD; EBRD IBFD; ERBD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD IBFD; EBRD
Tax Payments IMF1 IMF1 IMF1 IMF1 IMF1 IMF1 IMF1 IMF1 IMF1 IMF1 IMF1; CIS IMF1; CIS IMF1; CIS IMF1 IMF1: CIS IMF1; CIS IMF1; CIS IMF1 IMF1 IMF1; CIS IMF1; CIS IMF1; CIS IMF1; CIS IMF1; CIS IMF1; CIS
GDP IMF2 IMF2 IMF2 IMF2 IMF2 IMF2 IMF2 IMF2 IMF2 IMF2 IMF2; CIS IMF2; CIS IMF2; CIS IMF2 IMF2; CIS IMF2; CIS IMF2; CIS IMF2 IMF2 IMF2; CIS IMF2; CIS IMF2; CIS IMF2; CIS IMF2: CIS IMF2; CIS
Capital and Labour Income n.a. NSO n.a. NSO n.a. NSO NSO NSO NSO NSO CIS CIS CIS NSO CIS CIS CIS NSO NSO CIS CIS CIS n.a. CIS CIS
Notes: IBFD = International Bureau of Fiscal Documentation’s European Tax Handbook. EBRD = European Bank for Reconstruction and Development’s Transition Reports. IMF1 = International Monetary Fund’s Government Finance Statistics Yearbook (GFSY)*. CIS = Statistical Committee of the Commonwealth of Independent States Statistical Yearbook*. IMF2 = International Monetary Fund’s International Financial Statistics Yearbook*. NSO = various National Statistics Offices’ Statistical Bulletin. n.a. = not available. * An alternate source used is the International Monetary Fund’s Staff Country Reports/ Recent Economic Developments for the various countries.
Effective Tax Administration in Transition Countries
117
Appendix H: Effective/Statutory Ratios 1992-1998 Table H.1 VAT Effective/Statutory Ratios, 1992-1998 Country
1992
1993
1994
1995
1996
1997
Albania
..
..
..
..
..
0.42
1998 0.39
Armenia
0.30
0.28
0.16
0.20
0.20
0.28
0.38 0.27
Azerbaijan
0.41
0.48
0.28
0.09
0.21
0.23
Belarus
0.55
0.51
0.50
0.50
0.47
0.57
..
Bulgaria
..
..
..
0.44
0.44
0.34
0.47
Croatia
..
..
..
..
..
..
0.81
Czech Rep.
..
0.41
0.40
0.39
0.40
0.40
0.37
0.69
0.60
0.73
0.66
0.66
0.68
0.57
Estonia Georgia
..
0.03
0.07
0.10
0.15
0.19
..
Hungary
0.30
0.40
0.39
0.38
0.37
0.40
0.39
Kazakhstan
0.27
0.24
0.15
0.20
0.23
0.21
0.28
Kyrgyzstan
0.24
0.26
0.26
0.27
0.35
0.34
0.35
Latvia
0.34
0.51
0.56
0.61
0.62
0.58
0.55
Lithuania
..
..
..
0.54
0.47
0.57
0.55
Moldova
0.33
0.26
0.31
0.46
0.42
0.56
0.67
Poland
..
..
0.40
0.40
0.43
0.46
0.46
Romania
..
..
0.30
0.34
0.32
0.31
0.37
0.48
0.39
0.37
0.37
0.40
0.43
0.35
..
0.40
0.42
0.51
0.45
0.45
0.41
Russia Slovak Rep. Tajikistan
0.30
0.54
0.79
0.14
0.11
0.09
..
Turkmenistan
0.23
0.30
0.30
0.36
0.33
0.43
0.32
Ukraine Uzbekistan
..
0.53
0.49
0.50
0.46
0.49
0.43
0.37
0.46
0.30
0.37
0.42
0.49
0.59
Source: Author’s calculations.
Transition, Taxation and the State
118
Table H.2 SST Effective/Statutory Ratios, 1992-1998 Country
1992
1993
1994
1995
1996
1997
Armenia
0.95
..
0.18
0.26
0.27
..
..
..
0.63
0.55
0.34
0.43
0.46
..
Belarus
0.80
0.65
0.68
0.80
0.82
0.87
0.86
Bulgaria
0.66
0.63
0.64
0.62
0.61
0.61
0.65
..
0.88
0.92
0.92
0.91
0.94
0.93
0.73
0.84
0.81
0.77
0.82
0.83
0.87
Azerbaijan
Czech Rep. Estonia
1998
Georgia
0.09
0.08
0.09
0.13
0.26
0.38
..
Hungary
0.80
0.77
0.78
0.76
0.78
0.80
0.84
Kazakhstan
0.52
..
0.55
0.76
0.78
0.68
..
Kyrgyzstan
0.51
..
0.53
0.67
0.60
0.68
..
..
..
0.70
0.74
0.68
0.76
0.75 0.79
Latvia Lithuania
0.87
0.66
0.81
0.70
0.73
0.71
Moldova
0.47
0.47
0.64
0.66
0.69
0.60
..
..
..
0.72
0.76
0.78
0.76
0.77
Romania
1.12
1.17
1.11
0.95
0.88
0.83
..
Russia
1.05
0.68
0.64
0.62
0.58
0.71
0.60
Poland
Slovak Rep.
..
0.74
0.75
0.90
0.94
0.93
0.91
Slovenia
..
0.98
0.95
0.92
0.92
0.94
0.94
Tajikistan
0.55
0.70
1.10
0.18
0.22
0.31
0.23
Ukraine
0.93
0.81
0.88
0.79
0.79
0.80
0.80
..
..
..
0.57
0.61
..
..
Uzbekistan
Source: Author’s calculations.
Effective Tax Administration in Transition Countries
119
Table H.3 CIT Effective/Statutory Ratios, 1992-1998 Country
1992
1993
1994
1995
1996
1997
1998
Armenia
0.46
0.31
0.35
0.27
0.16
0.14
0.10
Azerbaijan
0.31
0.58
0.21
0.17
0.18
0.12
0.09
Belarus
..
0.94
0.86
0.54
0.33
0.39
0.45
Bulgaria
0.39
0.16
0.22
0.22
0.22
0.31
0.23
..
0.37
0.32
0.30
0.27
0.23
0.28 0.27
Czech Rep. Estonia
0.35
0.40
0.48
0.34
0.20
0.22
Georgia
0.19
0.03
0.03
0.04
0.05
0.04
..
Hungary
0.18
0.13
0.15
0.28
0.26
0.26
0.29
Kazakhstan
0.30
0.36
0.22
0.22
0.18
0.15
0.14
Kyrgyzstan
0.27
0.21
0.21
0.17
0.09
0.06
0.08
Latvia
0.15
0.41
0.26
0.14
0.20
0.23
0.24
Lithuania
..
0.36
0.23
0.17
0.13
0.13
0.11
Moldova
0.46
0.25
0.45
0.41
0.31
0.18
..
Poland
0.27
0.25
0.21
0.18
0.19
0.20
0.20
Romania
0.21
0.16
0.15
0.19
0.16
0.21
0.16
Russia
0.44
0.68
0.55
0.51
0.35
0.33
0.29
Slovak Rep. Slovenia
..
0.32
0.44
0.40
0.35
0.22
0.21
0.08
0.06
0.10
0.07
0.12
0.15
0.16
Tajikistan
0.52
0.49
0.67
0.10
0.08
0.04
0.04
Ukraine
0.33
0.52
0.79
0.79
0.74
0.63
0.53
Uzbekistan
0.44
0.74
0.38
0.46
0.47
0.41
0.36
Source: Author’s calculations.
Transition, Taxation and the State
120
0 .70
E s t o n ia
0 .60 B e la r u s
M o ld o v a
0 .50
U k r a in e .
E/S Ratio
L a t v ia L it h u a n ia
0 .30 A z e r b a ija n 0 .20
C ze c h R e p.
H unga ry
K a z a k h s ta n G e o r g ia
A r m e n ia
0 .10
E U -15
P o la n d S lo v a k R e p .
R u s s ia A lb a n ia U z b e k is t a n B u lg a r ia K y rg y zs ta n R o m a n ia
T u r k m e n is t a n
0 .40
T a jik is t a n
0 .00 1 .0
1.5
2 .0
2.5
3.0
3 .5
4.0
A v e r a g e T r a n s itio n In d ic a to r
Figure 5.1a VAT Effective/Statutory Ratio 1997 vs Progress in Transition 1.00
S lo v e n ia
0.90
C ze c h R e p . S lo v a k R e p .
.
0.80
U k r a in e
0.70
E/S Ratio
E U -1 5
B e la r u s
U z b e k is t a n
0.60
R o m a n ia
E s t o n ia
L a t v ia R u s s ia K a za k h s ta n L it h u a n ia K y r g y zs ta n
H u n g a ry
P o la n d
B u lg a r ia M o ld o v a
0.50 A z e r b a ija n 0.40
G e o r g ia T a jik is t a n
0.30
A r m e n ia 0.20 0.10 0.00 1.0
1.5
2.0
2.5
3 .0
3.5
4.0
A v e r a g e T r a n s i t io n I n d i c a t o r
Figure 5.1b SST Effective/Statutory Ratio 1997 vs Progress in Transition 0 .7 0 U k r a in e 0 .6 0
0 .5 0
E/S Ratio
U z b e k is ta n 0 .4 0
B e la ru s R u s s ia
B u lg a r ia
0 .3 0 .
L a tv ia S lo v a k R e p .
R o m a n ia
0 .2 0 A rm e n ia
M o ld o v a K a z a k h sta n
H u n gary E s to n ia C zech R ep. P o la n d
E U -1 5
S lo v e n ia L ith u a n ia
A z e rb a ija n 0 .1 0
K y rg y z sta n G e o rg ia
T a jik is ta n 0 .0 0 1 .0
1 .5
2 .0
2 .5
3 .0
3 .5
4 .0
A v e r a g e T r a n s itio n In d ic a to r
Figure 5.1c CIT Effective/Statutory Ratio 1997 vs Progress in Transition
Effective Tax Administration in Transition Countries
121
12 .0 E s t o n ia 10 .0 B e la r u s U k r a in e
8 .0
VAT NTY
L a t v ia L it h u a n ia
M o ld o v a
.
T u r k m e n is t a n
6 .0
E U -1 5
P o la n d S lo v a k R e p .
R u s s ia
A lb a n ia U z b e k is t a n
H unga ry C ze c h R e p .
B u lg a r ia K y r g y z s ta n R o m a n ia
4 .0
A z e r b a ija n
K a z a k h s ta n G e o r g ia
A r m e n ia 2 .0 T a jik is t a n
0 .0 1.0
1 .5
2 .0
2.5
3.0
3 .5
4 .0
A v e r a g e T r a n s itio n I n d ic a to r
Figure 5.2a VAT Normalised Tax Yield (NTY) 1997 vs Progress in Transition 1 6.0 F Y R M a c e d o n ia S lo v e n ia
1 4.0
C r o a t ia
1 2.0
U k r a in e
E U -1 5
B e la r u s
L a t v ia H u n g a ry
R u s s ia
1 0.0
SST NTY
C ze c h R e p . S lo v a k R e p . E s t o n ia
.
P o la n d
L it h u a n ia
8.0 U z b e k is t a n
6.0
T u r k m e n is t a n
4.0
R o m a n ia K a za k h s ta n M o ld o v a B u lg a r ia K y r g y zs ta n
A lb a n ia A z e r b a ija n
2.0
A r m e n ia G e o r g ia
T a jik is t a n
0.0 1.0
1.5
2.0
2 .5
3 .0
3.5
4.0
A v e r a g e T r a n s itio n I n d ic a t o r
Figure 5.2b SST Normalised Tax Yield (NTY) 1997 vs Progress in Transition 8 .0 U z b e k is ta n U k ra in e
T u rk m e n is ta n 7 .0
B u lg a r ia 6 .0 B e la ru s
CIT NTY
5 .0
4 .0
R u s s ia
R o m a n ia
H u n gary L a tv ia
A z e rb a ija n
.
3 .0
A rm e n ia
S lo v a k R e p .
C zech R ep. P o la n d E s to n ia
K a z a k h s ta n M o ld o v a
E U -1 5
C r o a tia L ith u a n ia
2 .0
F Y R M a c e d o n ia K y rg y z s ta n G e o rg ia A lb a n ia
T a jik is ta n
1 .0
S lo v e n ia
0 .0 1 .0
1 .5
2 .0
2 .5
3 .0
3 .5
4 .0
A v e r a g e T r a n sitio n In d ic a t o r
Figure 5.2c CIT Normalised Tax Yield (NTY) 1997 vs Progress in Transition
Transition, Taxation and the State
122
0.70
E stonia
0.60 L ithua nia
B e la rus
M oldova
0.50
U kra ine
E/S Ratio
P ola nd 0.40
Slova k R e p. R ussia H unga ry
C ze c h R e p.
B ulga ria
U zbe kista n K yrgyzsta n
R om a nia
0.30
A ze rba ija n A rme nia
K a za khsta n
0.20
Ge orgia
0.10
0.00 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
Ave ra g e B ribe T a x
Figure 5.3a VAT Effective/Statutory Ratio and the Average Bribe Tax
1.00 Slove nia Slova k R e p C ze c h R e p
0.90
B e la rus E stonia R om a nia H unga ry U kra ine P ola nd R ussia L ithua nia K a za khsta n K yrgyzsta n U zbe kista n B ulga ria M oldova
0.80
E/S Ratio
0.70 0.60 0.50
A ze rba ija n 0.40
Ge orgia
0.30 A rme nia 0.20 0.10 0.00 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
A ve ra g e B ribe T a x
Figure 5.3b SST Effective/Statutory Ratio and the Average Bribe Tax
9.0
Effective Tax Administration in Transition Countries
123
12.0 E s tonia 10.0 L ithua nia
B e la rus
VAT NTY
M oldova U kra ine
8.0
P ola nd
Slova k R e p. R us sia H unga ry
6.0
U zbe kista n
C ze c h R e p.
B ulga ria
K yrgyzs ta n R om a nia
4.0
A ze rba ija n A rme nia
K a za khsta n
Ge orgia
2.0
0.0 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
A ve ra g e B ribe T a x
Figure 5.4a VAT Normalised Tax Yield (NTY) and Average Bribe Tax
16.0 Slove nia
14.0
C roa tia
10.0
SST NTY
C ze c h R e p. E stonia
12.0
P ola nd
Slova k R e p. U kra ine
B e la rus H unga ry R us sia L ithua nia R om a nia K a za khsta n
8.0 B ulga ria
6.0
M oldova U zbe kista n K yrgyzs ta n
4.0 A rm e nia A ze rba ija n
Ge orgia
2.0
0.0 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Ave ra g e B ribe T a x
Figure 5.4b SST Normalised Tax Yield (NTY) and Average Bribe Tax
9.0
This page intentionally left blank
Chapter Six
Tax Capacity and Effort in Transition Countries 6.1 Introduction Reform of the public finances is a major policy issue in transition countries. In the centrally planned economies of Central and Eastern Europe and the former Soviet Union both government revenue and expenditure shares of GDP were high, often in excess of 50 per cent. During transition, a fall in both tax revenue and public expenditure as a percentage of GDP was expected. Indeed, since transition began over a decade and a half ago, many ex-socialist countries have witnessed dramatic falls in tax revenue shares of GDP, as alluded to in previous chapters. This decline in the tax/GDP ratio, whether arising from the SBC, a more widespread poor payments discipline problem or an ineffective tax administration, has raised concerns regarding these countries’ ability to mobilise revenues sufficient to finance public expenditure, redistribute income and, at the same time, embrace effective fiscal policy. Fear of further revenue erosion and low tax potential raises the question of tax capacity levels for these countries in the post-transition era. The purpose of this chapter is to apply a tax capacity methodology, favoured by the IMF in the 1970s to study international differences in tax ratios among developing countries, and extend it to the ex-socialist TEs using data for the 1990s (Lotz and Morss 1967; Bahl 1971). This methodology is similar in principle to the state and local fiscal capacity/effort approach commonly used in the context of intergovernmental fiscal relations and federal revenue-sharing arrangements (ACIR 1962, 1988; Manvel 1971; Reeves 1986). In total, there are 65 countries in our study – 25 transition countries, 15 (non-transition) low- and middle-income countries and 25 developed countries. The methodology used by the IMF arose from dissatisfaction with the standard measure, in the 1960s and 1970s, used for international tax comparisons (and, indeed, of measuring tax capacity and gauging relative fiscal behaviour), namely, tax revenue shares of GDP. Despite its wide use, the simple tax/GDP measure has serious limitations, especially in relation to issues of policy. For example, it is difficult to tell whether a low tax/GDP ratio for a particular country reflects limited taxable capacity or simply a preference for low tax and a small public sector. To address this problem, the IMF Fiscal Affairs Department (FAD) undertook a study of taxation, focusing on developing countries with the purpose of examining variations in tax ratios and their limited taxable capacity as compared to industrial countries. Arising
126
Transition, Taxation and the State
from this research, a number of papers were published, in the IMF Staff Papers (Lotz and Morss 1967; Chelliah 1971; Bahl 1971, 1972). Other IMF papers on the same topic were published later, including Chelliah, Baas and Kelly (1975); Tait, Gratz and Eichengreen (1979) and Stotsky and WoldeMariam (1997). Our purpose is to use the same methodology and apply it to the ex-socialist transition countries of Central and Eastern Europe and the former Soviet Union republics. By using cross-country data from the transition period of the mid- to late-1990s, the chapter sets out to measure (i) cross-country variations in the tax ratio; (ii) tax capacity estimates of individual countries; and (iii) relative tax performance of countries and revealed tax preferences of governments, as inferred by a tax effort index. In its simplest form, this approach attempts to answer the question – cannot tax or will not tax? Likewise, it can be useful when disbursing international aid to low income countries. According to this approach, international aid is best disbursed to high tax effort index countries as these governments have utilised their capacity such that local resources have already been tapped; foreign aid is not a substitute for the mobilisation of domestic revenues. The so-called tax capacity/effort approach, previously applied to non-transition cross-country studies (Leuthold 1991; Teera 2002) or to interstate comparisons within federal countries (Manvel 1971; ACIR 1988) has been applied to transition countries in a very limited sense; first in the early years of transition (Barbone and Polackova 1996) and secondly as applied to the Russian Federation (Bahl 1994; Martinez-Vazquez and Boex 1997; MacFarquhar 1998). Cross-country comparisons of tax ratios are notoriously difficult to undertake and any interpretation of results needs to be treated with caution. Differences in the composition of GDP, the tax treatment of public transfers and the way the selfemployed are recorded in the tax accounts are just some of the reasons why making international comparisons of tax ratios is an inherently difficult task. Likewise, there has been much criticism of the tax capacity/effort methodology employed by the IMF and others, ranging from the pejorative connotation and the catch-all nature of the ‘tax effort’ term, to the strong assumption that the explanatory variables reflect only differences in tax capacity and not tax effort, to the erroneous assumption that cross-country comparisons can reveal lessons for individual countries (Bird 1976; Radian 1980; Newlyn 1985).1 Nonetheless, ex-socialist countries that are undergoing the transition from a centrally planned to a market economy have witnessed dramatic changes in their tax ratios. While recognising the limitations of the approach, the tax capacity/effort methodology allows us to examine the importance of both the economic and political factors that impinge on tax collection. The summary measures used in this chapter give us an insight into the different constraints that might bind tax revenues in transition and non-transition countries. The results, when used in conjunction with in-depth analyses of individual countries, can help to reveal whether 1 Despite his criticisms of the tax effort approach, Radian (1980) admits that ‘…the most important finding is the confirmation that such a thing exists. The process of resource mobilisation is not governed solely by economic factors.’
Tax Capacity and Effort in Transition Countries
127
a country is limited in its revenue collections by low tax capacity and potential or by a reluctance to use the available tax capacity. In Section 6.2, we outline the methodology used. Data coverage and sources are explained in Section 6.3 of the chapter. Section 6.4 presents the results of the analysis. The conclusions in Section 6.5 complete the chapter. 6.2 Methodology Our methodology is based on the tax capacity/effort approach used by the IMF in the 1970s to measure cross-country differences in tax ratios and applied to developing countries. We begin with some definitions. Taxes, as noted earlier, are defined as compulsory, unrequited nonrepayable payments imposed by a government for public purposes. For our purposes, as in Chapter Five, social security contributions are treated as tax revenues as they are usually compulsory, unrequited levies to general government, i.e. they have the same characteristics of taxes. The tax share or ratio is the ratio of tax revenue to national income. In our study, national income is measured by Gross Domestic Product (GDP). In the same way that (relative) fiscal capacity is defined, (relative) tax capacity, or the ability to tax, is the extent to which a country can mobilise resources through taxation (ACIR 1962; Goode 1984).2 As an indicator of a country’s revenue potential, it is the predicted tax share that would result if a country applied a set of average effective rates of taxation to its tax bases (for a given sample of countries). The rates are computed as regression coefficients from a single tax capacity regression equation.3 With respect to tax capacity, the binding constraints are economic in nature and relate to the structural characteristics of an economy and, more specifically, the availability of taxable resources, or tax bases. For example, a developing country may have a low tax capacity because of a high agricultural share of GDP where the ‘taxable surplus’ is relatively low. The residual between the actual tax ratio and the estimated or expected tax ratio namely, tax capacity - is a measure of tax effort. Tax effort, or the willingness to tax, is the extent to which a country actually utilises its tax capacity or potential (Bird 1964; Bahl 1971). Here, the binding constraints are of a political (non-economic) nature and relate to government preferences. For example, a developed country may have a relatively high tax effort because of a preference for high public expenditures, as revealed by a large residual, the tax effort measure (Denmark, for example). Of 2 Fiscal or revenue capacity, in the context of state-local financing and intergovernmental transfers, is defined as the amount of revenue states would raise if they used a representative (average) tax system (RTS), consisting of national average tax rates applied to commonly used tax bases (Reeves 1986; ACIR 1988). 3 This is similar to the RTS regression method used for calculating a state’s fiscal capacity. Other methods for measuring a state’s fiscal capacity are outlined in Reeves (1986) and Martinez-Vazquez and Boex (1997).
128
Transition, Taxation and the State
course, it also reflects an overstretched taxable capacity as revenue bases are taxed intensively. In contrast, an equally developed country may have a relatively low tax effort because of a preference for a low tax burden (Ireland, for example). Of course, it also reflects an under-extracted tax capacity.4 The tax effort index, often used as a measure of a country’s tax performance, can be expressed as the ratio of the actual tax ratio to the expected tax ratio (that is, the estimated taxable capacity). The methodologies used in Chapters Five and Six are similar. The tax effort ratio is the ratio of the actual tax yield to the expected tax yield (taxable capacity), given the tax bases and the average practice of the sampled countries. The effective/ statutory ratio, outlined in Chapter Five, is the ratio of the actual tax yield to the tax yield that would result if the statutory tax rates were strictly applied. Whereas the latter is a measure of tax effectiveness (reflecting the quality of tax administration), the former is a measure of the willingness to tax (reflecting the preferences of government). Both methodologies set out to explain variations in the tax ratio. In the case of the effective/statutory ratio approach, variations in the tax ratio are explained in terms of variations in the statutory tax rates and variations in the effectiveness at which a given rate of tax is collected. In the case of the tax effort index approach, variations in the tax ratio are explained in terms of variations in the taxable capacity and variations in tax effort. There are a number of economic factors determining a country’s taxable capacity. An assumption of the model is that these factors (see below) relate to taxable capacity only and are independent of the factors effecting tax effort; hence, no variable representing the demand for public expenditures is included. The taxable capacity and revenue determinants literature argues, a priori, for a number of capacity or structural factors, including economic development, foreign trade, monetisation and the economic structure of a country (Hinrichs 1965; Lotz and Morss 1967). As alternative measures exist for many of these factors, there is a need to choose the measures that are most appropriate. Previous studies have suggested various proxies for these ‘tax handles’ (Musgrave 1969; Bahl 1971). For the level of national output or economic development, it is either per capita income or the agriculture share in income.5 We use both, in separate sets of regressions, and we also allow for a non-linear relationship between the level of economic development and tax capacity by including the squares of these variables. For the index of openness or foreign trade, the preferred proxy is the export ratio. For the sectoral composition of national output or a country’s economic structure, a commonly used proxy is the mining share in GDP. Partly for reasons of data availability, however, we use a dummy variable that indicates whether or not the country is a petroleum exporter. A summary of the main explanatory and significant variables as used in earlier studies is included in Stotsky and WoldeMariam 4 These are confirmed by the tax effort estimates, reported in Appendix J. 5 It is widely recognised that income works well as an explanatory variable when developed countries are compared to developing countries. However, among the same income group (high-income developed or low-income developing countries) its explanatory power is less (Hinrichs 1965; Bahl 1971).
Tax Capacity and Effort in Transition Countries
129
(1997). Other possible determinants include demographic factors, income distribution, technological and, as included in Tanzi (1992), foreign and public debt.6 Aside from foreign trade and the level and composition of income, we also want to test both for transitional effects and unitary versus federal systems of government effects. We do so by including dummy variables, namely TE, for transition economies and F, for federal systems of government and tax collection.7 These dummy variables will allow us to test whether, for example, Russia’s well-publicised tax collection problem is associated with the transition from a centrally planned to a market economy or with fiscal federalism and its system of intergovernmental fiscal relations. We write the tax ratio function as follows: T Y
f (D,
X ,P,TE,F) u Y
(6.1)
where D
A measure of economic development Agriculture GDP
or GDP per capita
X Y P
Exports GDP Petroleum exporting dummy
TE Transition dummy F Federal dummy
and u is the error term. 6 Two determinants appropriate for transition countries are the size of the unofficial economy and the level of corruption. Although estimates exist for both variables – Johnson, Kaufman and Shleifer (1997) for the unofficial economy estimates and Transparency International for a measure of corruption – we do not include them in our regressions as the data for the 25 TEs in our study are incomplete. Johnson, Kaufman and Zoido-Lobaton (1998) consider various determinants of government revenue, including corruption. For a wider selection of countries, their findings indicate a highly significant effect of corruption on revenues. As for the unofficial economy, they find ‘…smaller unofficial economies in countries with higher tax revenue…’ (Johnson, Kaufman and Zoido-Lobaton 1998). 7 A priori, we might expect large, closed (low levels of foreign trade), multi-racial federal countries (US and Australia, for example) to generate low tax/GDP ratios whereas small, open (high levels of foreign trade), single race federal countries (Austria, for example) to generate high tax shares.
130
Transition, Taxation and the State
Cross-country regression analysis (using OLS) is used to estimate the predicted tax ratio or taxable capacity, given these tax base proxies or handles. The list of proxies and their expected signs is shown in Table 6.1.
Table 6.1 Determinants of Tax Capacity RHS variables or capacity factors Economic development
Proxies
Expected sign
Agriculture share
-
of GDP or
+
GDP per capita Foreign trade or openness Sectoral composition of income
Export ratio
+
Petroleum exporter
+
Federal economy
-/+
Transition economy
-/+
We expect, a priori, that taxable capacity is positively related to the level of economic development, the degree of openness to trade and the presence of petroleum exports. With respect to the agriculture share of GDP (one of our proxies for economic development), a high level of activity in agriculture is generally associated with a large subsistence sector, a low level of monetisation and a low per capita income. In addition, the agricultural sector is not as administratively easy to tax as other sectors in the economy.8 As for GDP per capita, it is associated with increased monetisation and urbanisation, high levels of literacy and improved law enforcement, all factors contributing to higher taxable capacity. In the case of foreign trade, a greater level of exports relative to national income suggests both a greater degree of monetisation and an industrial sector that is administratively easier to tax – foreign trade flows are administratively easier to tax than domestic transactions. As for petroleum exporting, the hypothesis here is that the extractive sector by its very nature requires a heavy fixed investment on behalf of its participants which, in turn, confines operations to a few large and readily identifiable firms, making it administratively easier to levy and collect taxes, especially since they are being exported. For the purposes of this chapter, we are particularly interested in the sign and significance of the transition economy dummy. As alluded to before, many of the slow 8 The farming sector is recognised as a powerful lobby group which governments tend to tax lightly. However, this is not a structural factor but is related to the government’s reluctance to tax. Thus, this cannot be considered as an explanation for the negative relationship between the agricultural sector and the tax ratio (or else, it will violate the assumption that the explanatory variables relate to taxable capacity only). In general, as with other similar exercises, it is difficult to separate demand and supply-side factors.
Tax Capacity and Effort in Transition Countries
131
reforming transition countries are poor at collecting taxes compared to the developed market economies of the EU; but are TEs poor at collecting taxes compared to other countries of a similar GDP per capita or level of economic development? Section 6.4 reports the analysis and results. Before this, we need to outline the data coverage and sources used in our study. 6.3 Data Coverage and Sources There are 25 ex-socialist transition countries in our study. As in Chapter Five, the TEs that are not included are Mongolia and Vietnam (generally considered outside the CEE/FSU setting), China (where the tax system is highly complicated) and Serbia and Montenegro and Bosnia and Herzegovina (where war has recently occurred). Ten of our 25 countries are CEE countries, namely Albania, Bulgaria, Croatia, Czech Republic, FYR Macedonia, Hungary, Poland, Romania, Slovak Republic and Slovenia. The remaining 15 are all FSU republics. These are Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine and Uzbekistan. To allow for comparisons, we include two other broad categories of countries. These are the (non-transition) member states of the OECD, excluding Switzerland,9 and a small sample of (non-transition) low- and middle-income countries whose selection was determined by data availability. We include OECD countries in order to generate a ‘market economy’ benchmark. Aside from the purpose of cross-country comparison, we include a selection of low- and middle-income countries in order to allow for comparisons with the earlier tax capacity work of the IMF. For all countries in our study, tax coverage is for general government, comprising central and state, regional or provincial units of government and local government. We include social security contributions in our tax coverage. Non-tax revenues, such as transfers from Central Bank and privatisation receipts, are excluded. In the study, there are both unitary and federal states – 54 are unitary, 11 are federal. The federal states are Argentina, Australia, Austria, Belgium, Brazil, Canada, Germany, Mexico, Russia, US and Venezuela. In most countries of our sample, the tax year coincides with the calendar year. For a small number of countries, including the OECD countries of Australia, Canada, Japan, New Zealand and US, the tax year is different to the calendar year. An exercise in calculating tax ratios relies on data pertaining to general government tax revenue and a measure of national income. Tax revenue data are taken from the governments’ fiscal accounts where taxes are reported on a cash basis. Where possible, tax payments data are from the IMF’s Government Finance Statistics Yearbook (GFSY). Otherwise, other publications used include the OECD’s Revenue Statistics and the IMF’s Staff Country Reports. As already stated, we use GDP as the measure of national income. GDP data are obtained from the IMF’s International 9 Switzerland is excluded on account of data limitations with respect to the sectoral shares of GDP.
132
Transition, Taxation and the State
Financial Statistics Yearbook (IFSY) and from the UN National Accounts database (both sources report official data, as submitted by national governments of countries). The GDP data were prepared in accordance with the concepts and definitions of the System of National Accounts 1993 (SNA 1993). Incidentally, a majority of exsocialist transition countries went directly from the Material Product System (MPS) to the SNA 1993 and did not implement the System of National Accounts 1968 (SNA 1968).10 A number of other publications were referenced in our study. These included the CIS Statistical Yearbook, yearly EBRD Transition Reports, the UNDP Human Development Report, the UN Statistical Yearbook for Latin America and the Caribbean and statistical yearbooks for various countries. As for the choice of period, many of the earlier tax capacity studies used a three- or five-year period (see Lotz & Morss 1967; Bahl 1971; Chelliah, Bass & Kelly 1975; Tait, Gratz & Eichengreen 1979). In our study, we use a five-year average; the period covered is 1995–1999 (unless otherwise stated, see Appendix I) as data for some of our TEs before 1995 are unreliable. In addition, a five-year average is preferred to a threeyear average as the five-year average lessens the effects of the 1998 Russian crisis as compared to the three-year average. Table 6.2 reports the tax ratios for all 65 countries in our study. The actual tax/ GDP ratio is calculated as a five-year average for the period 1995–1999, unless otherwise stated. Country data and sources are listed in Appendix I. Table 6.2 indicates that the 25 transition countries have on average a lower tax ratio than either the EU-15 countries or the 25 OECD countries. The (unweighted) mean of the tax/GDP ratio for the TEs is 28.0 per cent, as compared to 40.9 per cent for the EU-15 and 36.2 per cent for the 25 OECD member states. Moreover, the variations from the mean are large for the TEs, compared to the EU or OECD countries. For the 25 TEs, the standard deviation of the tax ratio is 10.2. This compares to 5.9 and 8.8 respectively for the EU-15 and 25 OECD member states. The TEs have, however, on average, higher tax ratios than the low/middle income countries in the sample; 28.0 per cent versus 20.8 per cent for the latter. In our model, we have data for a number of explanatory variables, including the export ratio, the agriculture share of income and GDP per capita.11 Where possible, the data for the export and agriculture shares of GDP were obtained from the UN National Accounts database. The GDP per capita data were taken from various EBRD Transition Reports. Table 6.3 reports the agriculture and export shares of GDP and the GDP per capita for the 25 transition countries in our study. As before, the estimated ratio is a five-year average for the period 1995-1999, unless otherwise indicated.
10 The national accounts of the socialist countries and, in particular, the Soviet Union were different in many respects to the system of national accounts used elsewhere (United Nations SNA). For example, Soviet accounts excluded depreciation and the output of most so-called nonmaterial services (OECD 1991). As for SNA 1968 and SNA 1993, there are numerous differences between the two systems as regards classifications, definitions and valuation. 11 Purchasing power parity-based GDP per capita is an obvious alternative.
Tax Capacity and Effort in Transition Countries
133
Table 6.2 Tax/GDP ratiosa Country
Tax
Country
ratio
Tax
Country
ratio
Tax ratio
Albania
15.0
Argentinab
Armenia
16.2
Bolivia
17.5
Belgium
45.3
Azerbaijan
14.5
Botswanab
30.9
Denmark
49.7
Belarus
43.5
Brazilb
24.7
Finland
46.1
20.6
Austria
43.2
Bulgaria
30.2
Chile
20.6
France
43.9
Croatia
41.5
El Salvador
11.0
Germany
37.7
Czech Republic
37.0
Jordan
19.3
Greece
35.5
Estonia
34.9
Lesotho
35.8
Ireland
32.5
FYR Macedoniac
33.0
Panamab
17.4
Italy
42.8
Georgia
12.0
Peru
15.2
Luxembourg
42.1
Hungary
39.0
Philippines
15.9
Netherlands
42.1
Kazakhstan
16.6
Thailandb
17.3
Portugal
33.4
Kyrgyzstan
13.1
Tunisia
25.4
Spain
33.7
Latvia
33.3
Uruguay
26.2
Sweden
50.5
Lithuania
30.3
Venezuela
13.6
UK
35.4
Moldova
27.5
Average Low- and
20.8
Average EU (15)
40.9
Middle-Income Countries (15) Poland
36.2
Romania
29.1
Australia
28.1
Russiab
29.1
Canada
37.4
Slovak Republicc
34.3
Iceland
33.9
Slovenia
40.7
Japan
27.7
Tajikistan
12.0
Korea
22.0
Turkmenistanb
17.4
Mexico
16.5
Ukraine
34.1
New Zealand
33.5
Uzbekistan
29.3
Norway
41.4
Average TEs (25)
28.0
Turkey
21.7
US
27.8
Average OECDd
36.2
(25)
Notes: a. Tax = general government tax revenue. b. 1994–1998. c. 1996–1999. d. Excludes Switzerland and the transition OECD countries. Source: Author’s calculations.
Transition, Taxation and the State
134
Table 6.3 GDP per capita, Agriculture and Export Shares of GDP for TEs Country
Agriculture/GDP (%)a
GDP per capita ($)
Exports/GDP (%)
Albania
53.8
848
10.7d
Armenia
30.9
436
21.1
Azerbaijan
21.5
463
27.5
Belarus
11.7
1248
61.1
Bulgaria
18.6
1395
50.7
Croatia
8.1
4422
40.1
Czech Republic
4.0
5268
57.3
Estonia
6.0
3159
75.6
10.5
1884
37.7
30.2
610
15.9
Hungary
5.1
4539
46.5
Kazakhstan
10.6
1275
36.3
Kyrgyzstan
38.8
346
37.0
Latvia
5.6
2287
48.7
Lithuania
9.8
2425
48.8
Moldova
25.7
418
52.4
Poland
4.6
3626
26.1
Romania
15.6
1576
27.8
Russiab
6.2
2368
26.9
Slovak Republicc
4.4
3775
59.3
Slovenia
3.6
9561
55.4
FYR Macedonia
c
Georgia
Tajikistan
21.8
173
66.4
Turkmenistanb
20.5
516
76.3e
Ukraine
12.4
811
45.8
Uzbekistan
27.2
432
24.8
Notes: a. Includes agriculture, hunting, forestry and fishing. b. 1994–1998. The (1995-1999) five-year averages for Russia were 6.4, $2,248 and 32.8 respectively. The (1995–1999) fiveyear averages for Turkmenistan were 22.3 (agriculture share) and $503 (GDP per capita) respectively; the average export share for 1995–1999 was unavailable due to problems with the 1999 data. c. 1996–1999. The (1995–1999) five-year averages for FYR Macedonia were 10.6, $1,960 and 36.8 respectively. The (1995–1999) five-year averages for the Slovak Republic were 4.5, $3,705 and 59.3 respectively. d. 1998–1999. e. 1994–1997. Source: Author’s calculations.
Tax Capacity and Effort in Transition Countries
135
With respect to the level and composition of GDP, we can see from Table 6.3 that the set of TEs is not a homogeneous group. Of the 25 TEs, there are six (Albania, Armenia, Georgia, Kyrgyzstan, Moldova and Uzbekistan) whose agricultural sector accounts for over one quarter of total national value added. In contrast, seven TEs, namely, Czech Republic, Estonia, Hungary, Latvia, Poland, Slovak Republic and Slovenia have an agriculture share of GDP of six per cent or less. With respect to income per head, ten of our 25 transition countries have an annual GDP per capita of $1,000 or less. Six of these, namely Armenia, Azerbaijan, Kyrgyzstan, Moldova, Tajikistan and Uzbekistan have a GDP per capita of less than $500. In contrast, six TEs have an income per capita greater than $3,500 per annum with Slovenia having a GDP per capita in excess of $9,500. As regards the foreign trade sector and exports, eight TEs export less than one third of total GDP. They are Albania, Armenia, Azerbaijan, Georgia, Poland, Romania, Russia and Uzbekistan. In contrast, there are four transition countries with an export ratio of 60 per cent or more – Belarus, Estonia, Tajikistan and Turkmenistan. The methodology outlined in Section 6.2 allows us to investigate this further. The following section reports the results. 6.4 Results and Analysis Our regression results are presented in Table 6.4. We present two specifications.12 In (1), the level of economic development is proxied by ln(GDP per capita). The square of ln(GDP per capita) was included in alternative specifications (not reported here) and was found to be insignificant. In (2), the share of agriculture in GDP, and its square, proxy for the level of economic development. Our priors are partially confirmed. As expected, the level of economic development is strongly related to tax capacity: richer, more developed countries collect more in tax. The low-income transition countries, namely the Central Asian states (the exception of Uzbekistan is discussed later) and the three Caucasus countries, have relatively low tax ratios as compared with the higher tax ratios found in the middleincome countries of Central Europe and the Baltic states. The export share is also positively related to tax capacity, though in specification (1) the relationship appears to be non-linear – the boost to tax revenues from additional exports falls as exports rise. A federal tax system is unrelated to tax capacity. Curiously, the sign on the petroleum exporter dummy is negative although not significant.13
12 An alternative specification would be the inclusion of non-export income per capita and the deduction of the share of agricultural exports from the total share of exports since the value added in the agricultural sector is a component of total exports. Data limitations for the TEs and the low- and middle-income countries in our study preclude this specification. 13 Other recent studies have reported a similar negative sign for mining (Stotsky and WoldeMariam 1997; Eltony 2002). The explanations given are that the countries in the study that are heavily dependent on oil revenues do not classify this revenue as tax revenue in
136
Transition, Taxation and the State
Table 6.4 Tax Capacity Regression Results
ln(GDP per capita)
(1) 10.99** (1.28)
Agriculture/GDP (Agriculture/GDP)2 Exports/GDP (Exports/GDP)2 Petroleum exporter TE dummy Federal dummy Constant N R2
0.42** (0.15) -0.003** (0.001) -6.16 (3.28) 8.73** (2.39) -2.31 (2.53) -79.3** (11.66) 65 0.63
(2)
-1.65** (0.29) 0.022** (0.006) 0.36* (0.18) -0.003 (0.002) -7.30 (3.76) 5.20* (2.61) -3.25 (2.95) 32.4** (4.59) 65 0.53
Notes:*Significant at five per cent level. **Significant at one per cent level. Standard errors in parenthesis.
What is interesting for our purposes is the sign and size of the TE dummy. It is positive and significant in both specifications, although less so when the level of economic development is proxied by the agriculture/GDP share. In other words, holding constant the level of economic development, transition economies have relatively high rates of tax capacity compared to market economies. Whereas the analysis of the effectiveness of tax collection (see Chapter Five) suggested that some transition economies, particularly the laggard reformers where the state apparatus had deteriorated or collapsed, were relatively poor at tax collection compared to the developed market economies of the EU, we find here evidence that they are nevertheless no worse at tax collection than other low- or middle-income countries. The results in Table 6.4 (taking specification 2) allow us to estimate the tax capacity for our sample of countries. This is the tax share that would result, for a given country, if it used its tax handles to the average extent used by countries in the sample. A priori, we expect the predicted tax ratio to be higher for countries at
the budget, or may not require a sophisticated tax revenue system, resulting in a negative coefficient.
Tax Capacity and Effort in Transition Countries
137
Table 6.5 Tax Capacity Estimatesa Taxable Capacity (in per Rank cent) 16.4 63 (23)
Country Albania Armenia
15.2
64 (24)
Country Panamab
Taxable Capacity (in per Rank cent) 56.1 3
Peru
25.7
54
Azerbaijan
22.2
57 (20)
Philippines
25.8
53
Belarus
43.3
16 ( 7)
Thailandb
32.8
43
Bulgaria
32.7
44 (17)
Tunisia
30.4
47
Croatia
40.1
25 (10)
Uruguay
28.6
48
Czech Republic
51.9
7 ( 3)
Venezuela
34.5
39
Australia
34.5
39
Estonia
55.7
4 ( 1)
FYR Macedoniac
36.3
34 (14)
Georgia
13.6
65 (25)
Austria
43.9
15
Hungary
46.5
10 ( 5)
Belgium
56.7
2
Kazakhstan
35.6
37 (16)
Canada
42.9
18
Kyrgyzstan
20.0
61 (21)
Denmark
41.0
20
Latvia
46.5
10 ( 5)
Finland
40.4
22
Lithuania
41.1
19 ( 8)
France
36.7
33
Moldova
28.6
48 (18)
Germany
40.3
23
Poland
39.9
26 (11)
Greece
27.7
51
Romania
27.2
52 (19)
Iceland
37.4
29
37.9
27 (12)
Ireland
54.6
5
52.1
6 ( 2)
Italy
37.2
30
Russia
b
Slovak Republicc Slovenia
51.9
7 ( 3)
Japan
33.3
42
Tajikistan
36.0
36 (15)
Korea
37.6
28
Turkmenistan
b
40.5
21 ( 9)
Luxembourg
70.7
1
Ukraine
37.0
31 (13)
Mexico
36.1
35
Uzbekistan
17.9
62 (22)
Netherlands
49.2
9
New Zealand
32.3
45
Norway
43.1
17
Argentinab
27.8
50
Bolivia
21.4
58
Portugal
37.0
31
Botswanab
45.8
12
Spain
35.3
38
Brazilb
23.9
55
Sweden
45.0
13
Chile
32.3
45
Turkey
20.3
60
El Salvador
23.7
56
UK
40.2
24
Jordan
45.0
13
US
34.0
41
Notes: a. Tax capacity estimates (for the period 1995-1999) are the expected tax ratios given the tax base proxies and the average practice of the sampled countries. b. 1994–1998. c. 1996–1999. Source: Author’s calculations.
138
Transition, Taxation and the State
a higher stage of economic development and where the size of the foreign trade sector is larger. Table 6.5 reports the taxable capacity estimates (that is, the predicted tax shares) for the 65 countries. It also reports each country’s tax capacity ranking where the country with the highest taxable capacity estimate is ranked number one and so on (the 1–25 rankings for TEs only are included in parenthesis). As indicated in Table 6.4, the richer, less dependent on agriculture, more exportoriented transition countries, such as Czech Republic, Estonia, Hungary, Slovak Republic and Slovenia have all high estimated tax ratios. These five countries are all ranked in the top ten, in terms of tax capacity, of our 65 countries. In contrast, TEs with low income per capita, low export and high agriculture shares of GDP, such as Albania, Armenia, Georgia, Kyrgyzstan and Uzbekistan have estimated tax ratios that are low by international standards. These five countries fill the bottom five rankings, in terms of tax capacity, among our 65 countries. From the estimated tax ratio predicted by our equation, we can infer tax effort comparisons. As previously stated, tax effort is measured by the difference between the actual tax ratio and the estimated tax ratio predicted by our taxable capacity equation. When the actual tax ratio is less than the predicted ratio, tax effort is lower (than average); tax effort is higher (than average) when the actual tax ratio exceeds the predicted ratio. A tax effort index (defined as the ratio of actual tax collection to predicted tax collection) less than one indicates a preference for a lower than ‘average’ level of taxation – the country is exploiting its estimated tax potential less than the average; greater than one indicates a preference for more than ‘average’ level of taxation. In the case of a low tax effort index, for example, the main obstacle to a higher tax ratio is not the country’s taxable capacity but the unwillingness of government to impose and collect higher taxes. Accordingly, countries with a low tax ratio and a low tax effort have the potential for tax increases. A ranking of countries on the basis of the tax ratio and the tax effort ratio – as ranking is more meaningful and less problematic than the use of the absolute measure of tax effort – is reported in Table 6.6 (see Appendix J for the tax effort estimates). In the tax effort column, countries are ranked on the basis of the actual tax ratio expressed as a fraction of the expected tax ratio, namely, taxable capacity. Countries recording higher tax effort rankings are taxing closer to their capacity than countries with lower rankings. Unlike the simple tax ratio rank, ranking countries according to tax effort takes into account international differences in taxable capacity. As before, the rankings for the TEs only are included in parenthesis.14
14 In interstate comparisons of tax effort, high capacity-low effort and low capacityhigh effort combinations are considered in the context of intergovernmental fiscal relations, revenue-sharing and equalisation transfers. We do not consider such combinations here.
Tax Capacity and Effort in Transition Countries
139
Table 6.6 Tax Ratio and Tax Effort Rankingsa Country
Tax Ratio 1
Sweden
Rank Tax Effort 8
Country Bulgaria
Tax Ratio 34 (13)
Rank Tax Effort 21 ( 7)
Denmark
2
4
Uzbekistan
35 (14)
Finland
3
7
Romania
36 (15)
9 ( 2)
Belgium
4
39
Russiab
36 (15)
41 (15)
5
2 ( 1)
France
5
Australia
38
36
Belarus
6 (1)
15 (5)
US
39
36
Austria
7
16
Japan
40
35
Italy
8
6
Moldova
41 (17)
17 ( 6)
Luxembourg
9
54
Uruguay
42
21
Netherlands
9
32
Tunisia
43
33
Croatia
11 (2)
13 (4)
Brazilb
44
13
Norway
12
17
Korea
45
56
Slovenia
13 (3)
46
9
40 (14)
Turkey
Hungary
14 (4)
33 (13)
Argentina
Germany
15
20
Chile
b
47
42
47
50
Canada
16
31
Jordan
49
61
Czech Republic
17 (5)
44 (17)
Bolivia
50
36
Poland
18 (6)
Lesotho
19
1
Turkmenistanb
Greece
20
3
Thailand2
53
57
UK
21
29
Kazakhstan
54 (19)
58 (23)
22 (7)
51 (22)
Mexico
55
60
Estonia Slovak Republic
c
25 (10)
Panama
b
51
65
51 (18)
61 (24)
23 (8)
47 (19)
Armenia
56 (20)
11 ( 3)
Ukraine
24 (9)
21 (7)
Philippines
57
52
Iceland
25
25
Peru
58
54
Spain
26
17
Albania
59 (21)
21 ( 7)
New Zealand
27
12
Azerbaijan
60 (22)
47 (19)
Portugal
28
28
Venezuela
61
63
Latvia
29 (10)
44 (17)
Kyrgyzstan
62 (23)
49 (21)
FYR Macedoniac
30 (11)
25 (10)
Georgia
63 (24)
29 (12)
Ireland
31
53
Tajikistan
63 (24)
64 (25)
Botswanab
32
46
El Salvador
65
58
Lithuania
33 (12)
42 (16)
Notes: a. Tax effort estimates are the difference between the actual and the expected tax ratios. b. 1994–1998. c. 1996–1999. Source: Author’s calculations.
140
Transition, Taxation and the State
As with the majority of non-transition countries in our study, most of the TEs have an observed tax ratio less than the estimated tax ratio (resulting in a low tax effort estimate), and, by implication, are taxing below potential. This implies that for the TEs there is room to increase tax revenue in line with taxable capacity. This can be seen in Figure 6.1 (at the end of the chapter), where the tax ratio is plotted against the estimated tax ratio (taxable capacity). Low effort index countries, the majority in the case of transition (and non-transition) countries, correspond to those above the 45 degree line while high effort index countries correspond to those below. One of the most interesting observations from the results is for the Central Asian states. For Kazakhstan, Tajikistan and Turkmenistan, the actual tax ratio is far below the expected tax ratio (positioned in the top-left hand quadrant of Figure 6.1), given the tax bases and the average effective tax rates for the countries included in the study. These countries do not seem to be exploiting their taxable capacity fully. The implication here is that the main constraint binding tax collection for these low tax countries may not be economic but an unwillingness or reluctance to increase tax. This is in sharp contrast to Uzbekistan (an outlier and not for the first time) where the actual tax ratio far exceeds the estimated tax ratio. Uzbekistan’s ability to tax above its potential may be related to the authoritarian nature of its government (see Chapter Five for similar results). Barbone and Polackova (1996) findings with respect to the Central Asian states (with the exception of Tajikistan) were similar. Whereas Uzbekistan raised double the amount of revenue than predicted, Kazakhstan and Turkmenistan fell well short of their taxable potential. Our results (although tentative) for these low-income countries confirm the well-established findings for the traditional developing countries. Chelliah (1971) and Radian (1980) concluded that for the majority of poor countries the constraints for mobilising financial resources are political rather than economic. Another country, according to the data, that is exceeding its tax capacity is, surprisingly, Romania. The indication here is that Romania’s relatively low tax ratio (compared to other CEE countries) maybe due to its low tax capacity (arising from a high agricultural share and a low export share of GDP) rather than any (above average) reluctance or political resistance to taxation. This is in contrast to Russia whose tax ratio is below its estimated taxable capacity, indicating a reluctance or unwillingness to fully utilise its tax potential. 6.5 Conclusions The result that we found in Chapter Five, namely, the ability of some slow reformers to maintain high tax collection rates is reconfirmed here. On average, compared to market economies at similar levels of economic development, transition economies are no worse at collecting taxes. As with other studies, our analysis highlights the heterogeneous nature of the countries normally referred to as the transition countries. In terms of government revenue and tax ratios, many TEs, including Albania and the former FSU republics of Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan and Tajikistan resemble
Tax Capacity and Effort in Transition Countries
141
more closely the low- and middle-income countries of Asia and Latin America than their fellow transition countries of Central and Eastern Europe and the Baltic States. Of course, many of the TEs are indeed poor with very low incomes per capita. In 1999, the Caucasus states of Armenia, Azerbaijan and Georgia had an annual (unweighted) average GDP per capita of US$525 approximately. The five Central Asian states (Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan) had an annual average GDP per capita of approximately US$470 (EBRD 2001). The average tax ratio for these eight TEs, calculated as a three-year average (1996– 1998), was 16.6 per cent. This is much closer to the average tax ratio for the lowand middle- income countries in our sample, estimated at 20.8 per cent, than the average tax ratio of the other 17 transition countries in our study, estimated at 33.4 per cent. The implication here is that, in respect of revenue and tax collection, not all transition countries are the same. As with other economic, political and social aspects of transition, it is important to differentiate between the CEE countries / the Baltic States (countries that closely resemble mature market economies) and the FSU republics of the Caucasus and Central Asia (countries that, in many respects, resemble low-income, developing countries). Yet, according to our findings, for the Central Asian republics (with Uzbekistan as an exception), it may be political resistance to taxation that binds tax collection. Related work into the future would look to refine the methodology, expand the database to include a greater number of non-transition low- and middle-income countries and, as in similar studies, notably Leuthold (1991) and Tanzi and Davoodi (2000), examine the tax composition in a disaggregated model to assert whether direct or indirect taxes are been utilised given that many developing counties depend more on indirect taxation (that is, on domestic sales and foreign trade). This would allow us to attribute a country’s high tax effort, for example, to a particular tax category, direct or indirect. With respect to policy implications and tax changes, the question is one of whether tax increases are feasible in terms of a country’s taxable capacity and the practices of other countries. Likewise, transition countries with low tax effort and large or rising budget deficits may wish to consider tax increases as opposed to expenditure reductions as a way of correcting fiscal imbalances, compared to TEs with higher tax effort indices (Tait, Gratz and Eichengreen 1979; Stotsky and WoldeMariam 1997). Although the tax capacity/effort method has its weaknesses, by measuring tax collection capabilities and tax effort in these countries, we can provide information that will help to answer questions relating to the feasibility and scope for tax changes, the tax potential, the taxable surplus and the administrative capacity of tax systems.
45
Estonia Slovak Rep. Slovenia Czech Rep.
50
Latvia
Hungary Belarus
Lithuania
Turkmenistan
40
Croatia
Poland
Russia
Tax capacity
Tajikistan
Ukraine FYR Macedonia Bulgaria
Kazakhstan
30
Moldova Romania Azerbaijan Kyrgyzstan
20
Georgia
Albania Armenia
Uzbekistan
10
0
. 0
10
20
30
Tax ratio Figure 6.1 Tax Ratio and Est. Tax Capacity
40
50
Tax Capacity and Effort in Transition Countries
143
Appendix I: Country Data Table I.1 Country Data Country
Period
Local
Average
Average Tax
Average
Currency
GDP
Revenue
GDP per
(local
(local
capita
currency)
currency)
(US$)
363869
54500
848
786
127
436
15014
2183
463
843
366
1248
12807
3865
1395
122095
54714
4422
1652
611
5268
61408
21435
3159
186443
61487
1884
4577
548
610
8506
3315
4539
1546
257
1275
30631
4008
346
3188
1061
2287
35931
10873
2425
10114
2782
418
467
169
3626
264796
77016
1576
1903
553
2368
715
245
3775
2917
1186
9561
676
81
173
6719
1168
516
Transition Countriesa Albania
1995 – 1999
Armenia
1995 - 1999
Azerbaijan
1995 – 1999
Belarus
1995 – 1999
Bulgaria
1995 – 1999
Croatia
1995 – 1999
Czech Republic
1995 – 1999
Estonia
1995 – 1999
FYR Macedonia
1996 – 1999
Georgia
1995 – 1999
Hungary
1995 – 1999
Kazakhstan
1995 – 1999
Kyrgyzstan
1995 – 1999
Latvia
1995 – 1999
Lithuania
1995 – 1999
Moldova
1995 – 1999
Poland
1995 – 1999
Romania
1995 – 1999
Russia
1994 – 1998
Slovak Republic
1996 – 1999
Slovenia
1995 – 1999
Tajikistan
1995 – 1999
Turkmenistan
1994 – 1998
Ukraine
1995 - 1999
In millions of lek In billions of drams In billions of manats In billions of roubles In billions of levas In millions of kuna In billions of koruna In millions of kroons In millions of denars In millions of lari In billions of forints In billions of tenge In millions of soms In millions of lats In millions of litai In millions of leis In billions of zlotys In billions of lei In billions of roubles In billions of koruna In billions of tolars In millions of roubles In billions of manats In billions of hryvnias
92.4
31.5
811
Transition, Taxation and the State
144
Table I.1 (continued) Uzbekistan
1995 - 1999
In billions of soms
1061
311
1994 - 1998
Pesos
275886
56780
Bolivia
1995 - 1999
Bolivares
41405
7249
Botswana
1994 - 1998
Pula
15082
4659
Brazil
1994 - 1998
Reais
711752
175510
Low- and Middle-Income Countriesb Argentina
Chile
1995 - 1999
Pesos
30753
6328
El Salvador
1995 - 1999
Colones
96799
10692
Jordan
1995 - 1999
Dinars
5264
1017 1543
Lesotho
1995 - 1999
Maloti
4010
Panama
1994 - 1998
Balboas
8359
1457
Peru
1995 - 1999
Soles
151493
23024
Philippines
1995 - 1999
Pesos
2436
388
Thailand
1994 - 1998
Bhat
4364
755
Tunisia
1995 - 1999
Dinars
20892
5313
Uruguay
1995 - 1999
Pesos
192950
50647
Venezuela
1995 - 1999
Bolivares
40305
5468
OECD Countriesc Australia
1995 - 1999
Dollars
566
159
Austria
1995 - 1999
Schillings
2532
1094
Belgium
1995 - 1999
Francs
8739
3957 328
Canada
1995 - 1999
Dollars
878
Denmark
1995 - 1999
Krona
1117
555
Finland
1995 - 1999
Marka
640
295
France
1995 - 1999
Francs
8267
3632
Germany
1995 - 1999
Marks
3687
1389
Greece
1995 - 1999
Drachmas
32859
11657
Iceland
1995 - 1999
Krona
532
181
Ireland
1995 - 1999
Punts
53887
17534
Italy
1995 - 1999
Liras
1980
848
Japan
1995 - 1999
Yen
511754
141541
435243
95920
Korea
1995 - 1999
Won
Luxembourg
1995 - 1999
Francs
625
263
Mexico
1995 - 1999
Pesos
3193396
525967
Netherlands
1995 - 1999
Guilders
737
311
432
Tax Capacity and Effort in Transition Countries
145
Table I.1 (continued) New Zealand
1995 - 1999
Dollars
97373
32606
Norway
1995 - 1999
Krona
1069
443
Portugal
1995 - 1999
Escudo
18747
6265
Spain
1995 - 1999
Peseta
82677
27893
Sweden
1995 - 1999
Krona
1829
923
Turkey
1995 - 1999
Liras
36194
7845
UK
1995 - 1999
Pounds
804
285
US
1995 - 1999
Dollars
8306
2308
Notes: a. Excludes Serbia and Montenegro & Bosnia and Herzegovina. b. Selection based on data availability. c. Excludes Switzerland (due to data limitations) and the transition OECD countries. Sources: UN National Accounts database; IMF 2000b, 2001b; OECD 2000b; EBRD 2001.
Appendix J: Tax Effort Estimatesa Table J.1 Tax Effort Estimates Transition Countries Albania
Tax Effort Estimate 0.92
Armenia
1.06
Low- and Middle-Income Countries Argentinab Bolivia
Azerbaijan
0.66
Botswana
Belarus
1.00
Brazilb
b
Tax Effort Estimate 0.74
OECD Countries Australia
Tax Effort Estimate 0.82
0.82
Austria
0.98
0.67
Belgium
0.80
1.03
Canada
0.87
Bulgaria
0.92
Chile
0.64
Denmark
1.21
Croatia
1.03
El Salvador
0.47
Finland
1.14
Czech Republic
0.71
Jordan
0.43
France
1.20
Estonia
0.63
Lesotho
1.69
Germany
0.93
FYR Macedoniac
0.91
Panamab
0.31
Greece
1.28
Georgia
0.88
Peru
0.59
Iceland
0.91
Hungary
0.84
Philippines
0.62
Ireland
0.60
b
Kazakhstan
0.47
Thailand
0.53
Italy
1.15
Kyrgyzstan
0.65
Tunisia
0.84
Japan
0.83
Latvia
0.71
Uruguay
0.92
Korea
0.59
Lithuania
0.74
Venezuela
0.39
Luxembourg
0.59
Moldova
0.96
Mexico
0.46
Poland
0.91
Netherlands
0.86
Romania
1.07
New Zealand
1.04
Russiab
0.77
Norway
0.96
Transition, Taxation and the State
146
Table J.1 (continued) Slovak Republicc
0.66
Portugal
Slovenia
0.78
Spain
0.96
Tajikistan
0.33
Sweden
1.12
Turkmenistanb
0.43
Turkey
1.07
Ukraine
0.92
UK
0.88
Uzbekistan
1.63
US
0.82
0.90
Notes: a. Tax effort estimates (for the period 1995-1999) are expressed as the ratio of the actual tax ratio to the expected tax ratio. b. 1994-1998. c. 1996-1999. Source: Author’s calculations.
Chapter Seven
Conclusion 7.1 Death and Taxes There is an old adage that says that the only things certain in life are death and taxes. For transition countries, the evidence and experience since 1991 suggests otherwise. Loss-making enterprises that would normally exit if operating in a mature market environment have managed to survive through external forms of financial assistance. One form of this financial aid is late, in-kind or non-payment of taxes (of course, not exclusively to loss-making firms as the evidence in Chapter Four indicates). Taxes are suppose to be rigid (‘hard’) obligations, not ‘soft’, subject to negotiation. Likewise, tax liabilities are suppose to be parametric, not customised to individual taxpayers. Revenue erosion and lax payments discipline, manifesting itself in low revenue mobilisation, ineffective tax collection and widespread non-compliance, have been a problem in many TEs since the start of transition. This book examines the problem of tax payments discipline and collection in the context of the SBC and other legacies from the socialist era. Much of the literature on the revenue collapse in transition countries to date has focused on either economy-wide transitional phenomena (recession, unofficial economy) or taxpayers’ non-compliance (tax evasion, tax culture). In contrast, this monograph examines the problem from the position of the tax collector, that is, the state, and its ability and/or willingness to collect taxes. 7.2 Tax, Transition and the State: The Case of Russia Here, the problems of taxation are discussed in the context of transition from plan to market, the transformation from a socialist state to a capitalist state, the symbiotic relationship between government and business, and the persistence of the SBC. For many reasons, Russia is an appropriate setting for this discussion. Throughout the first decade of transition, Russia has witnessed a weakening of the state, poor governance and a general economic decline. Central to this malaise is the role of taxation in postsocialist Russia. As taxes are compulsory payments imposed by government on taxpayers, we examine the relationship between the tax collector and the taxpayer, that is, between the state and (for our purposes) the enterprise sector. A legacy from Soviet days is the nexus that exists between the state and the enterprise sector, and, in the transition period, between subnational regional government and incumbent (state or
148
Transition, Taxation and the State
privatised) firms, some of which are loss-making. In postsocialist Russia, there has often been evidence of Kornai’s SBC and the helping-hand state, concerned with job losses and social unrest. Likewise, we have seen evidence of Shleifer and Vishny’s predation and the grabbing-hand state, concerned with enrichment and maximisation of politicians’ political base. In transition, with the liberalisation of prices and the decline in direct budgetary subsidies, the tax system can be used as a conduit to soften the budget constraint of loss-making firms. Tax exemptions, arrears and offsets, non-payments and payments-in-kind are channels through which budget constraints are softened. The proliferation of these soft supports to enterprise, often with the approval of the state and sometimes at the instigation of the state, weakened, in an environment where tax awareness was not very high, tax collection and fiscal discipline. Tax obligations, in theory set by statutory tax legislation, were, in practice, negotiable and subject to bargaining. From the (federal) tax collector’s perspective, the inability to collect taxes (due to ineffective tax administration, rising tax evasion and growing taxpayers’ noncompliance), the toleration of tax arrears (often but not exclusively to loss-makers), the payments of tax obligations in the form of bribes and the diversion of federal taxes collected to subnational budgets and extrabudgetary funds (due to fiscal incentives and autonomy problems) contributed to falling revenues, and, ultimately, the fiscal crisis of 1998. By 1998, federal tax revenue in Russia was only 60 per cent of total tax revenue, compared to rates of 69, 77, 77 and 80 per cent for federal jurisdictions in Brazil, Australia, Mexico and Austria respectively (IMF 2000b). As the former chief economist of the EBRD wrote ‘…there can be no doubt that the capacity to tax is a defining characteristic of a viable state…’ (Buiter 1997). Russia’s failure in respect of tax collection is a manifestation of a weak state (as creditor) and poor public governance. Aside from specific revenue measures to improve tax collection, changes in political, civic and government institutions are required in order to strengthen the Russian state’s capacity to govern. 7.3 Budget Softness and the Russian Enterprise Sector In surveying the vast empirical literature on the determinants of enterprise restructuring in transition countries, Djankov and Murrell remarked that there has been a ‘…dearth of research focused on the soft budgets issue’ (Djankov and Murrell 2002). Conceptually, this may be due to the lack of a precise definition for the SBC. Empirically, this may be due to the endogeneity problem associated with establishing evidence of budget softness. In Chapter Three, we keep to Kornai’s paternalistic interpretation of the SBC. It is an empirical study that attempts to capture the basic tenets of the SBC syndrome, namely, expectations of a future bailout or rescue based or contingent on the financial difficulties of the firm. The chapter examines a number of instruments (tax concessions, inter-enterprise arrears, overdue bank credit, delayed wage payments) and sources (government, suppliers,
Conclusion
149
banks, workers) of budget softness. The expectational aspect to the SBC is captured in the first two empirical tests where managers of enterprises are asked to judge the ease of obtaining credit and the source and form of the financial assistance. The financial health of the firm is captured by the profitability measure. The contingency feature of the SBC phenomenon, namely that the bailout is conditional on losses incurred by ailing firms, is captured, as best as possible given the measurement difficulties, in the regression equations. The 437 surveyed enterprises, for the period 1997–1999, were randomly sampled from a stratified population of Russian firms. Although not representative of the Russian enterprise sector, the sample is representative of mid-sized manufacturing firms in Russia’s industrial heartland. The evidence suggests that on average banks (and suppliers) are unwilling to extend credit to unprofitable firms and hence are imposing hard budget constraints on firms. In contrast, there is evidence, albeit weak, that the state refinances loss-making firms. In this case, the financial assistance is not in the form of direct state credits as was popular at the outset of transition but in the form of non-payment of new tax obligations. Evidence of tax and wage arrears in our survey conducted in 2000 confirms the results of the earlier 1994 World Bank survey (Alfandari and Schaffer 1996). Both surveys are of medium and large-sized industrial enterprises but with only privatised enterprises included in our study. As for differences in results, our 2000 survey found no evidence of budget softness between enterprises and banks in contrast to the earlier findings that ‘…banks are still soft in their dealings with enterprises.’ (Fan et al. 1996). This difference may be due to the time period (and the progress, albeit uneven, in transition) that has elapsed between the two surveys. Notwithstanding the role of the utility companies in softening the budget constraint of firms, this chapter appears to confirm the findings of others relating to the experience of postsocialist Russia, that is, of unviable firms being kept afloat by lax government policy with respect to tax and other obligations (Kornai 2001; World Bank 2002). A more upbeat conclusion arising from the study is the apparent absence of any payments crisis post the 1998 financial crash. Predictions of a deterioration in payments discipline and an acceleration of non-cash payments did not materialise, as is evident from this chapter and other studies arising from the survey (see Angelucci et al. 2002). 7.4 Tax Arrears and the Romanian Enterprise Sector Since the transition process began in the early 1990s, arrears to different claimholders have become a common phenomenon. Overdue payables to all types of creditors, that is, government, trade suppliers, banks and workers have accumulated in many transition countries. These cases of breaches of financial obligations are not all the same and often require separate investigation. In Chapter Four, we examine the case of overdue liabilities to the government budget and social funds, i.e. tax and contributions arrears. Tax arrears are tax liabilities that are known to the tax authorities, have come due but have not been paid. Whereas the problem with tax evasion is one of detection,
150
Transition, Taxation and the State
the problem with tax arrears is one of collection. As known taxes remain unpaid, the ‘capped’ amount can be characterised as a form of subsidy. We develop a framework that allows us to measure the extent of this tax arrears subsidy. The methodology is based on the stock and flow of tax liabilities, of enterprises. Here, we focus on the flow measure as this is a better indicator of the tax arrears problem. In particular, we use two measures of the flow of financing arising from unpaid taxes by enterprises. Net new taxes, defined as new tax obligations minus all tax payments, measures the flow of new financing from unpaid taxes newly accrued. A positive net new taxes reflects an accumulation of tax debt as firms are not keeping up with current tax obligations. The net Tanzi effect, defined as interest and late penalties minus the inflation erosion of tax debt, measures the effective real cost of this flow of financing to delinquent firms. A negative net Tanzi effect reflects a negative real interest rate charged on unpaid taxes. In transition countries where there is a flow (of tax arrears) problem, net new taxes is positive and the net Tanzi effect is negative. The reverse is the case in mature market economies where tax arrears for operating firms are not common. This is a useful framework for measuring the tax arrears subsidy using firm-level data. More particularly, when extending the analysis to other creditors, net new taxes can be viewed as the parallel measure to NBF and the real growth in trade credit received (Bonin and Schaffer 1995; Croitoru and Schaffer 2002). The literature identifies three cases of arrears, namely, late payment, financial distress (bad debts) and collusive arrears. In the case of financial distress, ailing firms are kept afloat by means of tax and social security contributions arrears. This is analogous to the SBC. As transition progressed and where the source of the SBC is the government, the decline in direct budgetary subsidies coincided with the emergence and, in some cases, persistence of tax arrears. The 1997 data for medium and large non-financial firms in Romania indicates that this slow-reforming transition country is no exception. In the dataset of over 9,000 firms, the 25 largest loss-makers account for over 25 per cent of the overdue tax liabilities (stock) and over 20 per cent of net new taxes (flow). Moreover, chronic loss-makers, defined as firms with profits as a percentage of sales of less than -25 per cent, account for 15 per cent of all firms but over 30 per cent of the tax arrears (both stock and flow measure). Another feature of tax arrears in slow-reforming transition countries is the role that the utility sector often plays. In Romania, the big utility companies have figured largely in both the inter-enterprise arrears problem and the tax arrears problem. In the past, it has been the case that the utilities had large overdue receivables from non-paying customers. In the scenario where the utilities pass this on to the government budget and the fiscal system by late or non-payment of taxes, it can be viewed as a hidden subsidy to the non-paying and often ailing enterprises. In 1997, the utilities comprised some of the biggest tax debtors. Three of the four largest delinquent firms, in terms of the stock of tax arrears at end-1997, were utilities. When measured by net new taxes, the position is the same, that is, utilities comprise three of the four largest delinquent firms. The most alarming observation from the dataset is that the tax arrears problem appears to have spread to profitable firms, a phenomenon that is not common in
Conclusion
151
the leading transition countries. Toleration of tax arrears among profitable firms undermines an already fragile fiscal system. As a first priority the government, as creditor (i.e. tax collector), needs to restore tax discipline to profitable firms. Post1997 government revenue data for Romania indicate an increase in the tax share of GDP. However, this may be due to tax policy and design changes (particularly in the areas of personal and corporate income tax and VAT) rather than any improvement in tax administration or in the financial discipline of firms. As the rapidly-reforming transition countries accede to the EU and as the laggard countries try to catch-up, it is important that, in the face of variable tax revenues, mounting expenditure pressures and fragile fiscal systems, toleration of tax arrears is limited to loss-making firms. The tax arrears subsidy needs to be transparent and subject to the normal scrutinies of public policy. Others areas where the tax arrears subsidy has important implications for transition countries is in tax administration, bankruptcy and creditor protection and the state aids to industry debate. 7.5 Effective Tax Administration in Transition Countries Since the collapse of the socialist system, transition countries have witnessed dramatic changes to tax policy and tax design. The tax system that existed pretransition was very different to the Western-style (and indeed developing countries) tax systems. The vast majority of taxes, namely enterprise profit tax, turnover tax and payroll tax, were all collected from enterprises. These SOEs were either used as cash cows or as conduits for indirectly taxing households (McKinnon 1991). Tax liabilities were not decided by tax laws but by negotiation and lobbying. As for their collection, tax payments were made through the banking system, usually by transfers from the taxpayer’s (enterprise) account to the government budget. Hence, new tax systems had to be designed, and administered. However, given the nature of taxation in market economies (where legitimate disputes, tax evasion and avoidance are common practices), building new fiscal institutions, and, in particular, a tax administration that collects taxes in an efficient and effective manner, was never going to be a simple task. In some transition countries, where exemptions and privileges, arrears and offsets, evasion and bribery are prevalent, non-compliance with the statutory tax system is widespread. By calculating the difference between the statutory tax system, as per prescribed tax rates and the effective or actual tax system, as per effective tax rates, we construct a measure of the quality of tax administration. The effective/statutory ratio (E/S) measures the difference between the tax collected according to tax laws and regulations and the tax actually collected. It provides us with an indicator of the effectiveness of tax collection. Similar exercises measuring tax administration effectiveness, or the so-called tax gap, have been done elsewhere for income tax systems and for taxes of member states of federal countries. Our methodology is extended to value added tax and payroll tax, and applied to the ex-socialist transition countries of Central and Eastern Europe and the former Soviet Union. Our results indicate that, as expected, the transition countries are not as effective in collecting tax as the EU-15 benchmark. Although the estimates are highly
152
Transition, Taxation and the State
approximate and should be treated with caution, the cross-country comparisons indicate a greater non-compliance problem in transition countries as compared with mature market economies. A more surprising result is the difference between TEs and, in particular, a similarity between the leading reformers (Poland, Hungary, Slovak and Czech Republics, the Baltic States) and the least reform-minded countries (Ukraine, Belarus, Uzbekistan) as opposed to the so-called partial reformers (Russia, Bulgaria, Romania). Whereas the latter group of countries has a low level of tax administration effectiveness, as judged by our E/S estimates, the other two groups of countries have relatively high tax effectiveness rates. The leading reformers, all EU members since 2004, have the administrative and institutional capacities capable of collecting and enforcing taxes. Likewise, countries that have maintained a functioning state, albeit in need of reform, have been able to maintain high rates of effective tax collection throughout transition. In contrast, transition countries with weak and ineffective states, often associated with partial reforms, have low levels of effective tax collection and high levels of non-compliance. High levels of bribery, the sister activity of taxation, are also common in these countries. What are the policy prescriptions arising from this analysis? Tax administration systems need to be strengthened. The experience learnt from administrative reforms in developing and middle-income countries alike suggests that, despite the limited capacities of many of the transition countries, improvements in tax collection are possible. Reforms such as the establishment of large taxpayer units, organisational changes along functional lines, educational programmes for taxpayers, simplification of tax procedures and greater deployment of IT systems can prove effective in terms of improved tax collection. For the CIS countries where tax effectiveness rates are particularly low, a new social contract between a more honest state and a more compliant taxpayer is required. Moreover, international donors that continue to assist in building fiscal institutions must be cognisant of the social, legal and cultural factors that affect tax administration. Finally, second generation tax policy changes require matching administrative reforms if improved tax collection is to result. 7.6 Tax Capacity and Effort in Transition Countries What is the capacity of TEs to generate revenues? How much of that revenue capacity are they actually using? What is the feasibility and scope for tax changes? Can the recent revenue falls be reversed? These and other issues relating to the capacity to tax are addressed here. Tax capacity, defined as the ability to tax, is the extent to which a country can mobilise revenues through taxation. Tax effort, defined as the willingness to tax, is the degree to which the tax capacity or potential is used. The frequently used tax ratio, defined as the tax share of national income, reflects both tax capacity and tax effort. In Chapter Six, we use the capacity/effort methodology, commonly employed both for cross-country and interstate comparisons, to measure the tax potential and performance in transition countries. The traditional approach to measuring tax capacity, and, by inference, tax effort, is based on regressing the
Conclusion
153
tax ratio on variables that serve as proxies for a country’s tax handles or bases. The residual between the actual tax ratio and the predicted tax ratio – the estimated tax capacity – is taken as a measure of tax effort. The tax effort index is simply the ratio of the tax share to the estimated tax share. As the regression equation is designed to estimate taxable capacity in terms of economic structure, the choice of explanatory variables is limited, by assumption, to the factors reflecting the structural characteristics of an economy. The tax capacity and revenue determinants literature have identified a number of key factors, including the level and composition of income and the degree of openness. Suitable proxies for each variable were identified, including GDP per capita and the export share of GDP. Dummy variables for petroleum exporting countries, transition economies and federal countries are employed. Data and measurement problems prevent the inclusion of other suitable, a priori, determinants, namely the level of corruption and the size of the unofficial economy. The results indicate that, with respect to the ability and willingness to tax, TEs are not altogether dissimilar when compared to countries of similar income and structure. Tax capacity estimates, for transition countries, range from a low of 13.6 (per cent of GDP) in the case of Georgia to a high of 55.7 (per cent of GDP) in the case of Estonia, with an average of 35.4 per cent of GDP. In terms of tax effort, the majority of TEs have a tax effort index of less than one, as is the case for most of our 65 countries in the study. This indicates that these countries are taxing below their capacity and that there is some potential for tax increases, if required. These tentative findings are similar to the evidence found in developing countries (Chelliah 1971; Radian 1980). This appears to be particularly true for most of the Central Asian states, with the exception of Uzbekistan. In terms of the EU accession states, the average tax effort index for the eight transition countries that joined the EU in 2004 is 0.75 compared to an average tax effort index of 0.97 for the EU-15 countries. This implies that these transition countries have the potential for further tax increases, if necessary. As many of these countries have high public expenditure/GDP levels (some close to Western European standards) and fiscal deficits in excess of the limits set by the Stability and Growth Pact, tax rises, post EU membership, may very well be required. 7.7 Concluding Remarks In the introduction to this book, the motivation for this line of research was outlined. Since transition began over a decade and a half ago, one of the more noticeable features of the period was the decline in tax revenue shares of GDP. In some cases, particularly in the Caucasus and the Central Asian states (with the notable exception of Uzbekistan), tax revenues fell to alarmingly low levels, often below levels observed in developing countries. Aside from internal strife and the transformational recession experienced by many of these TEs, a weakened state appears to be a contributing factor (although not in all cases as some of the transition countries in Central Asia are
154
Transition, Taxation and the State
renowned for their authoritarian rule). In examining the revenue erosion phenomenon in the context of a weak or soft state, we drew upon János Kornai’s paternalistic state and the SBC syndrome (Kornai 1980, 1992a). Although the SBC can explain, from the perspective of the tax collector (the state), some of the tax collection problems prevalent in ex-socialist transition countries, it is certainly not the ‘whole story’. Again, from the state’s position, tax collection problems in transition (and nontransition) countries can arise for a number of other reasons. Factors outlined in this monograph include corruption and bribery, powerful vested interests, ineffective tax administration and other institutional weaknesses, poor intergovernmental fiscal arrangements and low tax effort reflecting political resistance to taxation. The lesson from other non-transition countries is for taxation to be used to raise revenues for the provision of public goods, for income redistribution and for macroeconomic stabilisation rather than, as currently used in some TEs, as an instrument to carry out various economic and social objectives, particularly, in the enterprise sector – often resulting in low and insufficient tax revenues. This will allow revenues to increase to ‘minimum’ levels observed in non-transition countries. For other TEs where tax levels are higher (many of the EU accession countries), better tax collection rates from state-owned and privatised firms will allow the promotion of the new smalland medium-sized enterprise sector to develop further. In turn, a larger SME sector in TEs can contribute to, among other things, the diminution of the SBC problem and a larger tax base from which to mobilise revenues, thus helping to prevent a repeat of the revenue erosion of the previous decade.
Bibliography Advisory Commission on Intergovernmental Relations (ACIR). (1962), ‘Measures of State and Local Fiscal Capacity and Tax Effort’, ACIR, USA. Advisory Commission on Intergovernmental Relations (ACIR). (1988), ‘State Fiscal Capacity and Effort’, ACIR, USA. Agha, Ali and Jonathan Haughton. (1996), ‘Designing VAT Systems: Some Efficiency Considerations’, Review of Economics and Statistics, 78 (2), pp. 303-308. Aitken, Brian. (2001), ‘Falling Tax Compliance and the Rise of the Virtual Budget in Russia’, IMF Staff Papers, Special Issue 48, International Monetary Fund, Washington DC. Alfandari, Gilles and Mark E. Schaffer. (1996), ‘“Arrears” in the Russian Enterprise Sector’ in Simon Commander, Q imiao Fan and Mark E. Schaffer (eds.), Enterprise Restructuring and Economic Policy in Russia, World Bank, Washington DC., pp. 87-139. Anderson, James H., Georges Korsun and Peter Murrell. (2000), ‘Which Enterprises (Believe They) Have Soft Budgets? Evidence on the Effects of Ownership and Decentralisation in Mongolia’, Journal of Comparative Economics, 28 (2), pp. 219-246. Angelucci, Manuela, Alan Bevan, Saul Estrin, Julian A Fennama, Boris Kuznetsov, Giovanni Mangiarotti and Mark E Schaffer. (2002), ‘The Determinants of Privatised Enterprise Performance in Russia’, CEPR Discussion Paper No. 3193, Centre for Economic Policy Research, London. Åslund, Anders. (1994), ‘Lessons of the First Four Years of Systemic Change in Eastern Europe’, Journal of Comparative Economics, 19 (1), pp. 22-38. Åslund, Anders. (1999), ‘Why has Russia’s Economic Transformation been so Arduous?’, World Bank Annual Conference on Development Economics, World Bank. Åslund, Anders. (2002), Building Capitalism. Economic Transformation of the Former Soviet Bloc, Cambridge University Press, Cambridge. Bahl, Roy W. (1971), ‘A Regression Approach to Tax Effort and Tax Ratio Analysis’, IMF Staff Papers, 18, pp. 570-610. Bahl, Roy W. (1972), ‘A Representative Tax System Approach to Measuring Tax Effort in Developing Countries’, IMF Staff Papers, 19, pp. 87-124. Barbone, Luca and Hana Polackova. (1996), ‘Public Finances and Economic Transition’, MOCT-MOST, 6 (3), pp. 35-61. Barbone, Luca and Rossana Polastri. (1998), ‘Hungary’s Public Finances in an International Context’ in Lajos Bokros and Jean Jacques Dethier (eds.), Public
156
Transition, Taxation and the State
Finance Reform during the Transition: The Experience of Hungary, World Bank, Washington DC., pp. 155-173. Belka, Marek, Saul Estrin, Mark E. Schaffer and I. J. Singh. (1995), ‘Enterprise Adjustment in Poland: Evidence from a Survey of 200 Private, Privatised and State-Owned Firms’, CEP Discussion Paper No. 233, London School of Economics, London. Berglof, Erik and Romesh Vaitilingam. (2000), Stuck in Transit: Rethinking Russian Economic Reform, Centre for Economic Policy Research, London. Berkowitz, Daniel and Wei Li. (1999), ‘Tax Rights in Transition Economies: A Tragedy of the Commons?’, Journal of Public Economics, 76 (3), pp. 369-397. Bevan, Alan A., Saul Estrin and Mark E. Schaffer. (1999), ‘Determinants of Enterprise Performance during Transition’, CERT Discussion Paper No. 99/03, Heriot-Watt University, Edinburgh. Bird, Richard. (1964), ‘A Note on “Tax Sacrifice” Comparisons’, National Tax Journal, 17, pp. 303-308. Bird, Richard. (1976), ‘Assessing Tax Performance in Developing Countries: A Critical Review of the Literature’, Finanzarchiv, 34 (2). Bird, Richard. (2000), ‘A Perspective on Fiscal Federalism in Russia’, Mimeo, University of Toronto. Bird, Richard. (2003), ‘Administrative Dimensions of Tax Reform’, Working Paper 0302, International Tax Programme, Joseph L. Rotman School of Management, University of Toronto. Blasi, Joseph, R., Maya Kroumova and Douglas Kruse. (1997), Kremlin Capitalism: Privatising the Russian Economy, Cornell University Press, New York. Bonin, John P. and Mark E. Schaffer. (1995), ‘Banks, Firms, Bad Debts and Bankruptcy in Hungary 1991-94’, CEP Discussion Paper No. 234, London School of Economics, London. Bönker, Frank, Klaus Müller and Andreas Pickel (eds.). (2002), Postcommunist Transformation and the Social Sciences: Cross-Disciplinary Approaches, Rowman & Littlefield, Maryland. Boycko Maxim, Andrei Shleifer and Robert W. Vishny. (1996), ‘A Theory of Privatisation’, Economic Journal, 106 (435), pp. 309-319. Brabant, Josef M. van (1998), The Political Economy of Transition: Coming to Grips with History and Methodology, Routledge, New York. Brana, Sophie, Mathilde Maurel and Jerome Sgard. (1999), ‘Enterprise Adjustment and the Role of Bank Credit in Russia: Evidence from a 420 Firms Q ualitative Survey’, Comparative Economic Studies, 41 (4), pp. 47-69. Broadman, Harry G. and Francesca Recanatini. (2001), ‘Is Russia Restructuring? New Evidence on Job Creation and Destruction’, World Bank Working Paper No. 2641, World Bank, Washington DC. Broadman, Harry G. (ed.). (2002), ‘Unleashing Russia’s Business Potential: Lessons from the Regions for Building Market Institutions’, World Bank Discussion Paper No. 434, World Bank, Washington DC. Brown, Archie. (1999), ‘The Russian Crisis: Beginning of the End or End of the
Bibliography
157
Beginning?’, Post-Soviet Affairs, 15 (1), pp. 56-73. Buiter, Willem. (1997), ‘Aspects of Fiscal Performance in some Transition Economies under Fund-Supported Programs’, IMF Working Papers No. 97/31, International Monetary Fund, Washington DC. Burgess, Robin and Nicholas Stern. (1993), ‘Taxation and Development’, Journal of Economic Literature, 31, pp. 762-830. Carlin, Wendy, Steven Fries, Mark E. Schaffer and Paul Seabright. (2001), ‘Competition and Enterprise Performance in Transition Economies: Evidence from a Cross-Country Survey’, EBRD Working Paper No. 63, European Bank for Reconstruction and Development, London. Cheasty, Adrienne. (1996), ‘The Revenue Decline in the Countries of the Former Soviet Union’, Finance and Development, 33 (2), World Bank. Chelliah, Raja J. (1971), ‘Trends in Taxation in Developing Countries’, IMF Staff Papers, 18 (2), pp. 254-327. Chelliah, Raja J., Hessel J. Baas and Margaret R. Kelly. (1975), ‘Tax Ratios and Tax Effort in Developing Countries 1969-1971’, IMF Staff Papers, 22 (1), pp. 187-205. Claessens, Stijn and Simeon Djankov (1998), ‘Politicians and Firms in Seven Central and Eastern European Countries’, World Bank Policy Research Working Paper No. 1954. World Bank, Washington DC. Clague, Christopher and Gordon C. Rausser. (1992), The Emergence of Market Economies in Eastern Europe, Blackwell, Oxford. Commander, Simon and Christian Mumssen. (1999), ‘Understanding Barter in Russia’, EBRD Working Paper No.37, European Bank for Reconstruction and Development, London. Coricelli, Fabrizio and Simeon Djankov. (2001), ‘Hardened Budgets and Enterprise Restructuring: Theory and an Application to Romania’, Journal of Comparative Economics, 29 (4), pp. 749-763. Croitoru, Lucian and Mark E. Schaffer. (2002), ‘Measurement and Assessment of Soft Budget Constraints in Romania’, National Bank of Romania Occasional Papers No. 2, Bucharest. Daianu, Daniel. (2000), ‘Structure, Strain and Macroeconomic Dynamic in Romania’, RCEP/WP No. 2, Bucharest. De Rosa, Donato. (2001), ‘Firm Power and Financial Survival Strategies in Russian and Ukrainian Industry’, Trinity College Dublin, draft. Dewatripont, Mathias and Eric Maskin. (1995), ‘Credit and Efficiency in Centralised and Decentralised Economies’, Review of Economic Studies, 62 (213), pp. 541555. Dewatripont, Mathias, Eric Maskin and Gerard Roland. (2000), ‘Soft Budget Constraints and Transition’ in Eric Maskin and András Simonovits (eds.), Planning, Shortage and Transformation: Essays in Memory of János Kornai, MIT Press, Cambridge, pp. 143-155. Djankov, Simeon and Kosali Ilayperuma. (1997), ‘The Failure of the GovernmentLed Program of Corporate Reorganisation in Romania’, William Davidson
158
Transition, Taxation and the State
Institute Working Paper No. 139, University of Michigan. Djankov, Simeon and Peter Murrell. (2002), ‘Enterprise Restructuring in Transition: A Q uantitative Survey’, Journal of Economic Literature, 40 (3), pp. 739-92. Dobrinsky, Rumen. (2002), ‘Tax Structures in Transition Economies: A Comparative Perspective vis-à-vis EU-Member States’, Paper presented at East-West Conference, draft. Easter, Gerald. (2003), ‘Building State Capacity in Postcommunist Russia: Tax Collection’, Mimeo, Boston College. Ebrill, Liam and Oleh Havrylyshyn (eds.). (1999), ‘Tax Reform in the Baltics, Russia, and other Countries of the Former Soviet Union’, IMF Occasional Paper No. 182, International Monetary Fund, Washington DC. Eltony, Nagy, M. (2002), ‘The Determinants of Tax Effort in Arab Countries’, Arab Planning Institute, draft. Ericson, Richard. E. (1999), ‘The Post-Soviet Russian Economic System: An Industrial Feudalism?’, Working Paper No.140, Institute of Transition Economics and East European Economies, Stockholm. European Bank for Reconstruction and Development. (1994), Transition Report 1994, EBRD, London. European Bank for Reconstruction and Development. (1999), Transition Report 1999: Ten Years of Transition, EBRD, London. European Bank for Reconstruction and Development. (2000), Transition Report 2000: Employment, skills and transition, EBRD, London. European Bank for Reconstruction and Development. (2001), Transition Report 2001: Energy in Transition, EBRD, London. Evans, Peter. (1992), ‘The State as Problem and Solution: Predation, Embedded Autonomy, and Structural Change’ in Stephan Haggard and Robert R. Kaufman (eds.), The Politics of Economic Adjustment: International Constraints, Distributive Conflicts and the State, Princeton University Press, New Jersey. Everaert, Greetje and Antje Hildebrandt (2001), ‘On the Causes of Soft Budget Constraints: Firm-Level Evidence from Bulgaria and Romania’, LICOS Discussion Paper No. 109, Katholieke Universiteit Leuven. Fairbanks, Charles H. Jr. (1999), ‘The Feudalisation of the State’, Journal of Democracy, 10 (2), pp. 47-53. Fan, Q imiao, Une J. Lee and Mark E. Schaffer. (1996), ‘Firms, Banks, and Credit in Russia’ in Simon Commander, Q imiao Fan and Mark E. Schaffer (eds.), Enterprise Restructuring and Economic Policy in Russia, World Bank, Washington DC., pp. 140-165. Freinkman, Lev, Daniel Treisman and Stepan Titov. (1999), ‘Subnational Budgeting in Russia, Preempting a Potential Crisis’, World Bank Technical Paper No. 452, World Bank, Washington DC. Frydman, Roman, Cheryl Gray and Andrzej Rapaczynski (eds.), (1996), Corporate Governance in Central Europe and Russia, Central European University Press, Budapest. Frye, Timothy and Andrei Shleifer. (1997), ‘The Invisible Hand and the Grabbing
Bibliography
159
Hand’, American Economic Review Papers and Proceedings, 87 (2), pp. 354358. Fullerton, Don. (1984), ‘Which Effective Tax Rate?’, National Tax Journal, 37, pp. 23-41. Gaddy, Clifford G. and Barry W. Ickes. (1998), ‘Russia’s “Virtual Economy”’, Foreign Affairs, 77 (5), pp. 53-67. Gaidar, Yegor. (1999a), ‘The Legacy of the Socialist Economy: the Macro- and Microeconomic Consequences of Soft Budget Constraints’, Institute for the Economy in Transition, draft. Gaidar, Yegor. (1999b), ‘Lessons of the Russian Crisis for Transition Economies’, Finance and Development, 36 (2), International Monetary Fund. Gao, Shumei and Mark E. Schaffer. (1998), ‘Financial Discipline in the Enterprise Sector in Transition Countries: How does China Compare?’, CERT Discussion Paper No. 98/1, Heriot-Watt University, Edinburgh. Gekker, Ruvin. (1998), ‘Paternalism and the Credibility Problem in a Centrally Planned Economy’, Journal of Economic Behaviour and Organisation, 34, pp. 189-192. Gill, Jit B. S. (2003), ‘The Nuts and Bolts of Revenue Administration Reform’, World Bank Europe and Central Asia Region, World Bank. Goode, Richard. (1984), Government Finance in Developing Countries, The Brookings Institution, Washington, DC. Gustafson, Thane. (1999), Capitalism Russian-Style, Cambridge University Press, Cambridge. Hare, Paul. (2001), ‘Institutional Change and Economic Performance in the Transition Economies’, Mimeo, Heriot-Watt University, Edinburgh. Heady, Christopher, Najma Rajah and Stephen Smith. (1994), ‘Tax Reform and Economic Transition in the Czech Republic’, Fiscal Studies, 15 (1), pp. 64-80. Hedlund, Stefan. (1999), Russia’s ‘Market’ Economy: A Bad Case of Predatory Capitalism. UCL Press, London. Hellman, Joel S. (1998), ‘Winner Takes All: The Politics of Partial Reform in Post Communist Transitions’, World Politics, 50 (2), pp. 203-234. Hellman, Joel S. and Mark Schankerman. (2000), ‘Intervention, Corruption and Capture: the Nexus between Enterprises and the State’, EBRD Working Paper No.58, European Bank for Reconstruction and Development, London. Hildebrandt, Antje. (2002), ‘Too Many to Fail? Inter-Enterprise Arrears in Transition Economies’, BOFIT Discussion Paper No. 11, Bank of Finland. Hinrichs, Harley H. (1965), ‘Determinants of Government Revenue Shares among Less-Developed Countries’, The Economic Journal, 75, pp. 546-556. Hobbes, Thomas. (1651), Leviathan. Holzman, Franklyn, D. (1962), Soviet Taxation: The Fiscal and Monetary Problems of a Planned Economy, Harvard University Press, Cambridge. Illarionov, Andrei. (1999), ‘The Roots of the Economic Crisis’, Journal of Democracy, 10 (2), pp. 68-82. International Bureau of Fiscal Documentation. (1998), European Tax Handbook,
160
Transition, Taxation and the State
IBFD, Amsterdam. International Monetary Fund. (1999a), Government Finance Statistics Yearbook 1999, IMF, Washington DC. International Monetary Fund. (1999b), International Financial Statistics Yearbook 1999, IMF, Washington DC. International Monetary Fund. (2000a), Russian Federation: Selected Issues, IMF Staff Country Report No.00/150, IMF, Washington DC. International Monetary Fund. (2000b), Government Finance Statistics Yearbook 2000, IMF, Washington DC. International Monetary Fund. (2000c), IMF Survey, 29 (13), IMF, Washington DC. International Monetary Fund. (2001a), Romania: Selected Issues and Statistical Appendix, IMF Staff Country Report No. 01/16, IMF, Washington DC. International Monetary Fund. (2001b), International Financial Statistics Yearbook 2001, IMF, Washington DC. International Monetary Fund. (2002), Russian Federation: Selected Issues and Statistical Appendix, IMF Staff Country Report No. 02/75, IMF, Washington DC. Interstate Statistical Committee of the Commonwealth of Independent States. (1999), Commonwealth of Independent States in 1998: Statistical Yearbook, Statistical Committee of the CIS, Moscow. Johnson, Simon, Daniel Kaufman and Andrei Shleifer. (1997), ‘The Unofficial Economy in Transition’, Brookings Papers on Economic Activity, 2, pp. 159221. Johnson, Simon, Daniel Kaufman and Pablo Zoido-Lobaton. (1998), ‘Corruption, Public Finances and the Unofficial Economy’, Presentation at the ECLAC Conference, draft. Kaldor, Nicholas. (1963), ‘Will Underdeveloped Countries Learn to Tax?’, Foreign Affairs, 41 (2), pp. 410-419. Kaminski, Bartlomiej (ed.), (1996), Economic Transition in Russia and the New States of Eurasia, M. E. Sharpe, New York. Karpov Commission (Interdepartmental Commission on Enterprise Balance Sheets). (1998), ‘On the Causes of Low Rate of Tax Collection, Origin of the Payment Crisis and Prospects for Restoration of Solvency in the Russian Enterprise Sector.’ Russian Economic Barometer. Kornai, János. (1979), ‘Resource-Constrained Versus Demand-Constrained Systems’, Econometrica, 47 (4), pp. 801-819. Kornai, János. (1980), Economics of Shortage (Vols. 1 and 2), North Holland Publishing Company, Amsterdam. Kornai, János. (1986), ‘The Soft Budget Constraint’, Kyklos. 39 (1), pp. 3-30. Kornai, János. (1992a), The Socialist System: The Political Economy of Communism, Clarendon Press, Oxford. Kornai, János. (1992b), ‘The Postsocialist Transition and the State: Reflections in the light of Hungarian Fiscal Reforms’, American Economic Review, Papers and
Bibliography
161
Proceedings, pp. 1-21. Kornai, János. (1993), ‘The Evolution of Financial Discipline under the Postsocialist System’, Kyklos, 46 (3), pp. 315-336. Kornai, János. (1998), ‘Legal Obligation, Non-Compliance and Soft Budget Constraint’ in Newman, Peter (ed.), The New Palgrave Dictionary of Economics and the Law, Macmillan, pp. 533-539. Kornai, János. (1998), ‘The Place of the Soft Budget Constraint Syndrome in Economic Theory’, Journal of Comparative Economics, 26 (1), pp. 11-17. Kornai, János. (2001), ‘Hardening the Budget Constraint: The Experience of the Postsocialist Countries’, European Economic Review, 45, pp. 1573-1599. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert Vishny. (1999), ‘The Q uality of Government’, Journal of Law, Economics and Organization, 15 (1), pp. 222-279. Lavrov, Aleksei, John M. Litwack and Douglas Sutherland. (2000), ‘Fiscal Federalist Relations in Russia: A Case for Subnational Autonomy’, OECD, Paris. Leuthold, Jane H. (1991), ‘Tax Shares in Developing Economies: A Panel Study’, Journal of Development Economics, 35, pp. 173-185. Levi, Margaret. (1988), Of Rule and Revenue, University of California Press, Berkeley. Lopez-Claros, Augusto and Sergei V. Alexashenko. (1998), ‘Fiscal Policy Issues during the Transition in Russia’, IMF Occasional Paper No. 155, International Monetary Fund, Washington DC. Lotz, Jorgen R. and Elliot R. Morss. (1967), ‘Measuring “Tax Effort” in Developing Countries’, IMF Staff Papers, 14, pp. 478-497. Lucas, Robert E. Jr. (1990), ‘Supply-Side Economics: An Analytical Review’, Oxford Economic Papers, 42 (2), pp. 293-316. MacFarquhar, Rory. (1998), ‘Can’t Tax or Won’t Tax?: Russia’s Tax Effort and the August 17 Financial Crisis’, Presented at the RECEP Conference, draft. Malleret, Thierry, Natalia Orlova and Vladimir Romanov. (1999), ‘What Loaded and Triggered the Russian Crisis?’, Post-Soviet Affairs, 15 (2), pp. 107-129. Manvel, Allen D. (1971), ‘Differences in Fiscal Capacity and Effort: Their Significance for a Federal Revenue-Sharing System’, National Tax Journal, 24, pp. 193-204. Martinez-Vazquez, Jorge and L.F. Jameson Boex. (1997), ‘Fiscal Capacity: An Overview of Concepts and Measurement Issues and Their Applicability in the Russian Federation’, AYSPS Working Paper No. 97-3, Georgia State University. Martinez-Vazquez, Jorge and L. F. Jameson Boex. (1999), ‘Fiscal Decentralisation in the Russian Federation during the Transition’, AYSPS Working Paper No. 993, Georgia State University. Martinez-Vazquez, Jorge and Robert McNab. (2000), ‘The Tax Reform Experiment in Transitional Countries’, National Tax Journal, 53 (2), pp. 273-298. Maskin, Eric and Chenggang X u. (2001), ‘Soft Budget Constraint Theories: From Centralisation to the Market’, Economics of Transition, 9 (1), pp. 1-27. Mau, Vladimir. (2000), ‘Russian Economic Reforms as Seen by an Insider: Success
162
Transition, Taxation and the State
or Failure?’, The Royal Institute of International Affairs, draft. McFaul, Michael. (1995), ‘State Power, Institutional Change and the Politics of Privatisation in Russia’, World Politics, 47 (2), pp. 210-243. McGee, Robert W. and Galina Preobragenskaya. (2003), ‘Taxation and Public Finance in a Transition Economy: A Case Study of Russia’, Mimeo, Andreas School of Business, Barry University. McKinnon, Ronald. I. (1991), ‘Financial Control in the Transition from Classical Socialism to a Market Economy’, Journal of Economic Perspectives, 5 (4), pp. 107-122. McKinsey Global Institute. (1999), Unlocking Economic Growth in Russia, McKinsey & Company Inc., Moscow. Mendoza, Enrique G., Assaf Razin and Linda L. Tesar. (1994), ‘Effective Tax Rates in Macroeconomics: Cross-Country Estimates of Tax Rates on Factor Incomes and Consumption’, Journal of Monetary Economics, 34 (3), pp. 297-323. Mitchell, Janet. (1998), ‘Strategic Creditor Passivity, Regulation and Bank Bailouts’, CEPR Discussion Paper No. 1780, Centre for Economic Policy Research, London. Mitchell, Janet. (2000), ‘Theories of Soft Budget Constraints and the Analysis of Banking Crises’, Economics of Transition, 8 (1), pp. 59-100. Mitra, Pradeep and Nicholas Stern. (2003), ‘Tax Systems in Transition’, World Bank Policy Research Working Paper No. 2947, World Bank, Washington DC. Murrell, Peter. (1996), ‘How Far Has the Transition Progressed?’, Journal of Economic Perspectives, 10 (2), pp. 25-44. Musgrave, Richard A. (1969), Fiscal Systems, Yale University Press, New Haven. Myrdal, Gunnar. (1968), The Asian Drama. An Inquiry into the Poverty of Nations. Periodical Service Co., New York. Nagy, Piroska M. (2000), The Meltdown of the Russian State, Edward Edgar, Cheltenham. Newbery, David. (1987), ‘Taxation and Development’ in David Newbery and Nicholas Stern (eds.), The Theory of Taxation for Developing Countries, Oxford University Press for the World Bank, New York, pp. 165-204. Newlyn, W. T. (1985), ‘Measuring Tax Effort in Developing Countries’, Journal of Development Studies, 21 (3), pp. 390-405. North, Douglass. (1990), Institutions, Institutional Change and Economic Performance, Cambridge University Press, Cambridge. Olson, Mancur. (1965), The Logic of Collective Action, Harvard University Press, Cambridge. Olson, Mancur. (2000), Power and Prosperity. Outgrowing Communist and Capitalist Dictatorships, Basic Books, New York. Organisation for Economic Co-operation and Development. (1991), A Study of the Soviet Economy, International Monetary Fund/World Bank/European Bank for Reconstruction and Development/Organisation for Economic Co-operation and Development, Paris. Organisation for Economic Co-operation and Development (1997), OECD Economic
Bibliography
163
Surveys: The Russian Federation, OECD, Paris. Organisation for Economic Co-operation and Development. (1999), OECD National Accounts Main Aggregates 1960-1997, OECD, Paris. Organisation for Economic Co-operation and Development. (2000a), OECD Economic Surveys: The Russian Federation, OECD, Paris. Organisation for Economic Co-operation and Development. (2000b), OECD Revenue Statistics 1965-1999, OECD, Paris. Organisation for Economic Co-operation and Development. (2001), National Accounts of OECD Countries Volume II 1988-1999, OECD, Paris. Ott, Katarina. (1998), ‘Tax Administration Reform in Transition: The Case of Croatia’, Occasional Paper No.5, Institute of Public Finance, Zagreb. Perotti, Enrico. (1994), ‘Collusive Arrears in Transition Economics’, Financial Markets Group Discussion Paper No. 198, London School of Economics, London. Perotti, Enrico C. and Octavian Carare. (1997), ‘The Evolution of Bank Credit Q uality in Transition: Theory and Evidence from Romania’, CERT Discussion Paper No. 97/02, Heriot-Watt University, Edinburgh. Pinto, Brian and Sweder van Wijnbergen. (1994), ‘Ownership and Corporate Control in Poland: Why State Firms Defied the Odds’, World Bank Policy Research Working Paper No. 1308, World Bank, Washington DC. Pinto, Brian, Vladimir Drebentsov and Alexander Morozov. (2000), ‘Give Macroeconomic Stability and Growth in Russia a Chance: Harden Budgets by Eliminating Non-payments’, Economics of Transition, 8 (2), pp. 297-324. Pipes, Richard. (1974), Russia under the Old Regime, Charles Scribner, New York. Pitti, Zoltan and Jaime Vacquez-Caro. (1998), ‘Tax Administration during the Transition: Coping with Legal Changes and a Shrinking Base’ in Lajos Bokros and Jean Jacques Dethier (eds.), Public Finance Reform during the Transition: The Experience of Hungary, World Bank, Washington DC., pp. 511-532. Q ian, Yingyi and Gerard Roland. (1998), ‘Federalism and the Soft Budget Constraint’, American Economic Review, 88 (5), pp. 1143-1162. Radian, Alex. (1980), Resource Mobilisation in Poor Countries: Implementing Tax Policies, Transaction Books, London. Rawski, Thomas G. (1997), ‘Who has Soft Budget Constraints?’, Global Economic Review, 26 (1), pp. 29-49. Reeves, H. Clyde (ed.). (1986), Measuring Fiscal Capacity, Oelgeschlager, Gunn & Hain in association with the Lincoln Institute of Land Policy, Boston. Rodden, Jonathan, Gunnar S. Eskeland and Jennie Litvack (eds.). (2002), Fiscal Decentralisation and the Challenge of Hard Budget Constraints, MIT Press, Cambridge. Roland, Gerard. (2000), Transition and Economics: Politics, Markets, and Firms, MIT Press, Cambridge. Roland, Gerard, János Kornai and Eric Maskin. (2003), ‘Understanding the Soft Budget Constraint’, Journal of Economic Perspectives, 41 (4), pp. 1095-1136. Sachs, Jeffrey D. (1995), ‘Russia’s Struggle with Stabilisation: Conceptual Issues
164
Transition, Taxation and the State
and Evidence’, World Bank Annual Conference on Development Economics 1994, World Bank. Santarossa, Luoana, D. (2001), ‘Arrears as a Sign of Financial Repression in Transition Economies: The Case of Romania’, CERT Discussion Paper No. 2001/04, Heriot-Watt University, Edinburgh. Sapir, Jacques. (1999), ‘Russia’s Crash of August 1998: Diagnosis and Prescription’, Post-Soviet Affairs, 15 (1), pp. 1-36. Schaffer, Mark E. (1989), ‘The Credible Commitment Problem in the CentreEnterprise Relationship’, Journal of Comparative Economics, 13 (3), pp. 359382. Schaffer, Mark E. (1990), ‘State-owned enterprises in Poland: taxation, subsidisation and competition policies’, European Economy 43, pp. 183-201. Schaffer, Mark E. (1995), ‘Government Subsidies to Enterprises in Central and Eastern Europe: Budgetary Subsidies and Tax Arrears’ in David. M. G. Newbery (ed.), Tax and Benefit Reform in Central and Eastern Europe, Centre for Economic Policy Research, London, pp. 115-144. Schaffer, Mark E. (1996), ‘Tax Arrears in Transition Economies’, Mimeo, HeriotWatt University, Edinburgh. Schaffer, Mark E. and Lorand Ambrus-Lakatos, eds. (1997), Fiscal Policy in Transition, Forum Report of the Economic Policy Initiative, No.3. Centre for Economic Policy Research/Institute for EastWest Studies, London. Schaffer, Mark E. (1998), ‘Do Firms in Transition Economies Have Soft Budget Constraints? A Reconsideration of Concepts and Evidence’, Journal of Comparative Economics, 26 (1), pp. 80-103. Schaffer, Mark E. (2000), ‘Should We be Worried about the Use of Trade Credit and Non-Monetary Transactions in Transition Economics? Discussion’, Economic Systems, 24 (1), pp. 55-61. Schumpeter, Joseph A. (1911), The Theory of Economic Development, Duncker & Humblot, Leipzig, 1912. Schumpeter, Joseph A. (1918), ‘The Crisis of the Tax State’ in Richard Swedberg (ed.) Joseph A. Schumpeter: The Economics and Sociology of Capitalism, Princeton University Press, New Jersey, 1991. Seabright, Paul (ed.), (2000), The Vanishing Rouble. Barter Networks and NonMonetary Transactions in Post-Soviet Societies, Cambridge University Press, Cambridge. Shevtsova, Lilia. (1999), Yeltsin’s Russia: Myths and Reality, Carnegie Endowment for International Peace, Washington DC. Shleifer, Andrei and Robert W. Vishny. (1992), ‘Pervasive Shortages under Socialism’, RAND Journal of Economics, 23 (2), pp. 237-246. Shleifer, Andrei and Robert W. Vishny. (1993), ‘Corruption’, Quarterly Journal of Economics, 108, pp. 599-617. Shleifer, Andrei and Robert W. Vishny. (1994), ‘Politicians and Firms’, Quarterly Journal of Economics, 109 (4), pp. 995-1025. Shleifer, Andrei. (1997), ‘Government in Transition’, European Economic Review,
Bibliography
165
41 (3-5), pp. 385-410. Shleifer, Andrei and Robert W. Vishny. (1998), The Grabbing Hand: Government Pathologies and Their Cures, Harvard University Press, Cambridge. Shleifer, Andrei and Daniel Treisman. (2000), Without a Map: Political Tactics and Economic Reform in Russia, MIT Press, Cambridge. Silvani, Carlos and Katherine Baer. (1997), ‘Designing a Tax Administration Reform Strategy: Experience and Guidelines’, IMF Working Paper No. 97/30, International Monetary Fund, Washington DC. Skoog, Gun Eriksson. (2000), The Soft Budget Constraint – The Emergence, Persistence and Logic of an Institution, Kluwer Academic Publishers, Dordrecht. Slemrod, Joel (ed.), (1992), Why People Pay Taxes: Tax Compliance and Enforcement, University of Michigan Press, Ann Arbor. Stepanyan, Vahram. (2003), ‘Reforming Tax Systems: Experience of the Baltics, Russia, and Other Countries of the Former Soviet Union’, IMF Working Paper No. 03/173, International Monetary Fund, Washington DC. Stiglitz, Joseph. (1994), Whither Socialism? MIT Press, Cambridge. Stotsky, Janet G. and Asegedech WoldeMariam. (1997), ‘Tax Effort in SubSaharan Africa’, IMF Working Paper No. 97/107, International Monetary Fund, Washington DC. Summers, Victoria P. and Emil M. Sunley. (1995), ‘An Analysis of Value-Added Taxes in Russia and Other Countries of the Former Soviet Union’, IMF Working Paper No. 95/1, International Monetary Fund, Washington DC. Surrey, Stanley. (1964), ‘Tax Administration in Underdeveloped Countries’ in Richard Bird and Oliver Oldman (eds.), Readings on Taxation in Underdeveloped Countries, John Hopkins Press, Baltimore, pp. 503-533. Tait, Alan A., Wilfrid L. M. Gratz and Barry J. Eichengreen. (1979), ‘International Comparisons of Taxation for Selected Developing Countries, 1972-76’, IMF Staff Papers, 26, pp. 123-56. Tanzi, Vito. (1987), ‘Q uantitative Characteristics of the Tax Systems of Developing Countries’ in David Newbery and Nicholas Stern (eds.), The Theory of Taxation for Developing Countries, Oxford University Press for the World Bank, New York, pp. 205-241. Tanzi, Vito (ed.).(1993), Transition to Market: Studies in Fiscal Reform, International Monetary Fund, Washington DC. Tanzi, Vito. (1999), ‘Transition and the Changing Role of Government’, Finance and Development, 36 (2), International Monetary Fund. Tanzi, Vito and George Tsibouris. (2000), ‘Fiscal Reform Over Ten Years of Transition’, IMF Working Paper No. 00/113, International Monetary Fund, Washington DC. Tanzi, Vito and Hamid R. Davoodi. (2000), ‘Corruption, Growth and Public Finances’, IMF Working Paper No. 00/182, International Monetary Fund, Washington DC. Tanzi, Vito. (2001), ‘Creating Effective Tax Administrations: The Experience of Russia and Georgia’ in János Kornai, Stephen Haggard and Robert Kaufman
166
Transition, Taxation and the State
(eds.), Reforming the State: Fiscal and Welfare Reform in Postsocialist Countries, Cambridge University Press, Cambridge. Teera, Joweria M. (2002), ‘Tax Performance: A Comparative Study’, Working Paper No. 1, Centre for Public Economics, University of Bath. Tóth, István János and András Semjén. (1996), ‘Tax Behaviour of Small and Medium Enterprises’, Review of Sociology of the Hungarian Sociological Association, Special Issue, pp. 67-87. Tóth, István János and András Semjén. (1998), ‘Tax Behaviour and Financial Discipline of Hungarian Enterprises’ in László Csaba (ed.), The Hungarian SME Sector Development in Comparative Perspective, KOPINT-DATORG Foundation of Economic Research, pp. 103-134. Treisman, Daniel. (1995), ‘The Politics of Soft Credit in Post-Soviet Russia’, Europe-Asia Studies, 47 (6), pp. 949-976. Treisman, Daniel. (1998), ‘Fighting Inflation in a Transitional Regime: Russia’s Anomalous Stabilisation’, World Politics, 50 (2), pp. 235-265. Treisman, Daniel. (1999), ‘Russia’s Tax Crisis: Explaining Falling Revenues in a Transitional Economy’, Economics and Politics, 11 (2), pp. 145-169. United Nations. (2001), Statistical Yearbook for Latin America and the Caribbean 2000. United Nations, New York. United Nations Development Programme. (2000), Human Development Report 2000. United Nations, New York. United Nations Economic Commission for Europe. (2004), ‘Tax Reforms in the EU Acceding Countries’, Economic Survey of Europe, 1, United Nations, Geneva. Wade, Robert. (1990), Governing the Market. Economic Theory and The Role of Government in East Asian Industrialisation, Princeton University Press, New Jersey. Whalley, John. (1975), ‘Distortionary Factor Taxation - A Calculation of Effective Tax Rates in the UK 1968-70’, The Manchester School, 43, pp. 1-30. Wildasin, David E. (1997), ‘Externalities and Bailouts: Hard and Soft Budget Constraints in Intergovernmental Fiscal Relations’, Mimeo, Department of Economics, Vanderbilt University. Winiecki, Jan. (1992), ‘Resistance to Change, Large Enterprises and Soft Budget Constraint: A Property Rights Perspective’ in Carlo Frateschi and Gianni Salvini (eds.), A Comparative Analysis of Economic Reforms in Central and East Europe, Dartmouth, Aldershot. World Bank. (2002), Transition, The First Ten Years: Analysis and Lessons for Eastern Europe and the Former Soviet Union, World Bank, Washington DC. Zhuravskaya, Ekaterina V. (2000), ‘Incentives to Provide Local Public Goods: Fiscal Federalism, Russian Style’, Journal of Public Economics, 76 (3), pp. 337-368.
Index
Albania bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134-5 per capita 134 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 statutory/effective taxation 107 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Argentina, tax revenue/GDP ratio 133 tax effort 145 Armenia bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134-5 per capita 134-5, 141 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Åslund, Anders 15-16, 18, 23, 27 Australia tax effort 145 tax revenue/GDP ratio 133 Austria tax effort 145 tax revenue/GDP ratio 133
Azerbaijan bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134-5 per capita 134-5, 141 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 105, 133 transition indicator 107 bailouts 6, 46, 66, 78, 82, 91 bank credit 47-9 difficulty of obtaining 47 long-term 47, 48 short-term 47, 48 bankruptcy law, Romania 80 and SBC 4 barter economy 12, 15, 51-3 BEEPS 22, 34-6, 110-2 Belarus bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134-5 per capita 134 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105
168
Transition, Taxation and the State
state capture 35 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Belgium tax effort 145 tax revenue/GDP ratio 133 Bolivia tax effort 145 tax revenue/GDP ratio 133 Botswana tax effort 145 tax revenue/GDP ratio 133 Brasov Region (Romania) 92-3 Brazil tax effort 145 tax revenue/GDP ratio 133 bribe tax CEE countries 111 effective/statutory ratios 122-3 extent 111-12 FSU countries 111 NTY 122-3 Buiter, William 148 Bulgaria bribe tax 111 CIT effective/statutory ratio 119 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134 per capita 134 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 66, 133 transition indicator 107 bureaucrats, corruption 15, 38 Burgess, Robin 98 Canada tax effort 145
tax revenue/GDP ratio 133 CBR (Central Bank of Russia) 22,31 CEE (Central and Eastern European) countries 2, 6, 103, 131 bribe tax, 111 CIT/GDP ratio 105 effective statutory ratio 107 SST/GDP ratio 105 VAT/GDP ratio 105 NTY 107 statutory tax rate 105 tax /GDP ratio 7, 133 transition indicator 107 Chile tax effort 145 tax revenue/GDP ratio 133 CIS (Commonwealth of Independent States) 10, 25, 42, 82, 108, 152 CIT (Corporate Income Tax) 100, 101, 103 base 104 effect collection and transition 110, 120 effective/statutory ratios 107, 119 GDP ratios CEE countries 105 FSU countries 105 NTY 107 and transition 110 CMEA (Council for Mutual Economic Assistance) 18, 89 Commander, Simon 12, 29, 52 corruption bribery 111-2 bureaucrats 15, 38 Russia 32 small business sector 37 tax administration 111, 115 see also state capture CPI (Corruption Perceptions Index) 37, 111-2 credit supply logit procedure 45-6, 48 Russian enterprises 45-9 see also bank credit Croatia bribe tax 111 CIT/GDP ratio 105 CPI 111 effective statutory ratio 107,117-9 GDP
Index agriculture share 134 exports share 134 per capita 134 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Czech Republic bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134 per capita 134 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax arrears 79 tax effort 145 tax revenue/GDP ratio 7, 66, 133 transition indicator 65, 107 Denmark tax effort 127, 145 tax revenue/GDP ratio 133 Dewatripont, Mathias 4, 41 Djankov, Simeon 20, 66, 95, 148 Dobrinksy, Roman 100, 101, 104 Easter, Gerald 11, 22 EBRD (European Bank for Reconstruction and Development) 29, 33-35, 64 BEEPS 22, 34-6, 110-2 on tax administration 115 transition indicators 65, 107-8 economy depoliticization 3 planned, transition to market economy 2, 8, 13 El Salvador tax effort 145 tax revenue/GDP ratio 133 enterprise sector, tax arrears 6, 9
169
Romania 66-95, 149-51 Russia 27-8, 51-4, 56-9, 149 enterprises, loss-making 28-9, 41, 76-7 Estonia bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134 per capita 134 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax arrears 79 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 EU (European Union) 98, 153-4 effective/statutory ratio 106-7 tax revenue/GDP ratio 133 Evans, Peter 18 Fan, Q imiao 46 financial-industrial groups 32-43 Finland tax effort 145 tax revenue/GDP ratio 133 France tax effort 145 tax revenue/GDP ratio 133 FSU (Former Soviet Union) countries 2, 6, 103, 131 bribe tax 111 CIT/GDP ratio 105 effective/statutory ratio 107 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax/GDP ratio 7, 133 transition indicator, 107 VAT/GDP ratio 105 Gaddy, Clifford 12, 15 28, 83 Gaidar, Yegor 21, 23 Gazprom 27 GDP (Gross Domestic Product)
170
Transition, Taxation and the State
agriculture share, TEs 134-5 exports share, TEs 134-5 per capita, TEs 134-5, 141 tax revenue ratios, CEE/FSU 7, 23-4, 66, 125, 133 Georgia bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134-5 per capita 134, 141 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 105, 133 transition indicator 107 Germany tax effort 145 tax revenue/GDP ratio 133 Goldman, Marshall 16 governance index comparisons 35 and enterprise reform 19 ‘grabbing-hand’ model, state 4-5, 13-16, 20, 28-30, 148 gradualism 3 Gray, John 16 Greece tax effort 145 tax revenue/GDP ratio 133 ‘helping-hand’ model, state 13-14, 16, 20, 29-30, 148 Hobbes, Thomas, Leviathan 22 Hungary bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134-5 per capita 134, 141 governance index 35
intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax arrears 79 tax effort 145 tax revenue/GDP ratio 7, 66, 133 transition indicator 65, 107 Iceland tax effort 145 tax revenue/GDP ratio 133 Ickes, Barry 12, 15, 28, 83 IMF (International Monetary Fund) 2, 10, 30, 31 Fiscal Affairs Department 125-6 inflation subsidy 69-71, 73, 75 institutions 3, 15-17, evolutionary paradigm 3, 5, index 34 reforms 31-4 intervention index, comparisons 35-6 invisible-hand model 14, 16 Ireland tax effort 128, 145 tax revenue/GDP ratio 133 Italy tax effort 145 tax revenue/GDP ratio 133 Japan tax effort 145 tax revenue/GDP ratio 133 Jordan tax effort 145 tax revenue/GDP ratio 133 Kazakhstan bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134 per capita 134, 141 market-supporting institutions 34 NTY 107
Index SST/GDP ratio 105 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Kiriyenko, Sergei 11-12, 30 Korea tax effort 145 tax revenue/GDP ratio 133 Kornai, János 1, 4-5, 13-14, 17, 20, 28, 41, 46, 57, 59, 148, 154 Kyrgyzstan bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134 per capita 134 governance index 35 intervention index 35 NTY 107 SST effective/statutory ratio 118 SST/GDP ratio 105 state capture 35 statutory tax rate 105 statutory/effective taxation 107 tax effort 145 tax revenue/GDP ratio 7, 105, 133 transition indicator 107 Latvia bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-119 GDP agriculture share 134-5 exports share 134 per capita 134 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax arrears 79 tax effort 145 tax revenue/GDP ratio 7, 105, 133 transition indicator 107 Lesotho
tax effort 145 tax revenue/GDP ratio 133 levelling effect 4, 14, 28, 46 Leviathan 18 Thomas Hobbes 22 Lithuania bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134 per capita 134 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax arrears 79 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 logit procedure, 45-6, 48 Lucas, Robert E. 100 Luxembourg tax effort 145 tax revenue/GDP ratio 133 Macedonia (FYR) bribe tax 111 CIT/GDP ratio 105 CPI 111 GDP agriculture share 134 exports share 134 per capita 134 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Maskin, Eric 4, 41 Mau, Vladimir 18 McKinnon, Eric 4, 41 McKinsey 20, 21 Mendoza, Enrique G. 100
171
172
Transition, Taxation and the State
Mexico tax effort 145 tax revenue/GDP ratio 133 Ministry of Taxation and Fees 32, 109 Mitchell, Janet 5, 6 Moldova bribe tax 111 CIT/GDP ratio 105 CPI 111 GDP efftective/statutory ratio 107, 117-9 agriculture share 134 exports share 134 per capita 134-5 market-supporting institutions 34 NTY 107 SST effective/statutory ratio 118 SST/GDP ratio 105 statutory tax rate 105 statutory/effective taxation 107 tax effort 145 tax revenue/GDP ratio 7, 133 transition indicator 107 Mumssen, Christian 12, 29, 52 Murrell, Peter 20, 95, 148 Netherlands tax effort 145 tax revenue/GDP ratio 133 New Zealand tax effort 145 tax revenue/GDP ratio 133 Newbery, David 97 nomenklatura 15, 16 non-monetary payments 4, 12, 27-9, 90-1, 149 Norway tax effort 145 tax revenue/GDP ratio 133 North, Douglass, 33 NTY (Normalised Tax Yield) bribe tax 123 CEE countries 107 EU-15, Mean 107 formula 102 FSU countries 107 and transition 121
OECD countries, tax revenue/GDP ratio 131-133 Olson, Mancur 11, 16 Olivera-Tanzi 71 Panama tax effort 145 tax revenue/GDP ratio 133 paternalism SBC 4, 21, 29, 41, 46, 50, 76, 148, 154 state 13-14 Peru tax effort 145 tax revenue/GDP ratio 133 Philippines tax effort 145 tax revenue/GDP ratio 133 Pinto, Brian 20, 46 Poland bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134-5 per capita 134 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax arrears 79 tax effort 145 tax revenue/GDP ratio 7, 66, 133 transition indicator 65, 107 Portugal tax effort 146 tax revenue/GDP ratio 133 privatisation 2, 20, 65, 67, 85, 88 RA (Régies Autonomes) 66, 67 Radian, Alex 97, 126 rent-seeking behaviour 14, 15, 16, 21, 30, 113 Roland, Gerrard 3, 5, 12, 26 Roman S.A. 84, 88-92 debts 89-90, 91 markets 89 non-monetary payments 90-1
Index state aid 91 strikes 89 Romania bankruptcy law 80 Brasov region 92-3 bribe tax 111 CIT/GDP ratio 105 CPI 111 economic indicators 65-6, 92 effective/statutory ratio 107, 117-9 enterprise sector case studies 84-95 Régies Autonomes 66, 67 SBC 63-4, 76, 83, 94-5 tax arrears 64-95, 149-51 types of firms 66-7 GDP agriculture share 134 exports share 134-5 per capita 134 government bailout schemes 66, 82 NTY 107 price reflators, calculation 84 privatisation 65, 67, 85, 88 SOF 67 SST/GDP ratio 105 statutory tax rate 105 tax arrears budget 51-4 definition 63-4, 149 interest 81 profitable firms 76-8, 150-1 subsidy, identification 13, 83, 150 uniform treatment 81 tax arrears survey analysis/findings 73-80, 149-51 dataset 67, 71, 73 loss makers 76-8 methodology 67-73 policy prescriptions 80-3 stock/flow analysis 9, 63, 67-80, 150 tax liabilities 63-4, 67-80 utility sector 76-78, 82, 86, 90, 150 tax collection procedures 140, 82 tax effort 145 tax revenue/GDP ratio 7, 66, 133 transition indicator 65, 107 VAT effective/statutory ratio 117 RTS (Representative Tax System) 100
173
Russia 1998 crisis 9, 11-12, 21, 30-1, 41, 148, 149 bribe tax 111 budget deficit 30 CIT/GDP ratio 105 corruption 32, 38 CPI 111 effective/statutory ratio 107, 117-9 enterprise survey analysis/findings 45-59 conclusions 59, 149 credit supply 45-8 debts 54-9 firm ownership 43-4 firm size 42-3 government assistance, obtaining 49-50 industries 42, 44, 45 questionnaire 59-61 regions 43, 44, 45 sample/data 42-5 working capital, financing 51-4 enterprises 18-20 and SBC 20-1, 28-9, 148-9 state, relationship 20-9 subsidies 28-9 tax arrears 27-8, 51-4, 59, 79, 149 fiscal federalism 25-6, 29 GDP agriculture share 134 exports share 134-5 per capita 27, 134 revenue share 24 governance 8-18, 148 index 35 institutional reforms 32-3, 33-4 intervention index 35 market economy, transition 16 market-supporting institutions 34 NTY 107 political reforms 31-2 regions 42-5, 50 disparities 25 governments 29 small business sector 36-9 banking system 38 bureaucracy/corruption 37-8 economic conditions 39
174
Transition, Taxation and the State tax system 37 SST/GDP ratio 105 state 11, 33 distrust of 22 irresponsibility 18 role 17-18 state capture 35 statutory tax rate 105 tax collection problems 8-9, 23, 24-7 reforms needed 32-3 and transition 147-8 tax effort 140, 145 tax payments 12-13 non-monetary 12, 27-8 tax revenue/GDP ratio 1, 7, 11, 16, 23-4, 133 transition indicator 107 wages, arrears/non-payment 28-9, 51-4, 59, 148-9
Sachs, Jeffrey 12 SBC (soft budget constraint) 1, 4-6, 8-10, 154 and bankruptcy 4 definition 4, 41 institutional-evolutionary approach 5 interpretations 4-5 models bureaucratic hierarchical 4, 13-14 dynamic commitment 4 gambling banks 4 politicians and firms 4, 14-15 paternalism 13-14, 41, 46, 50, 148 Romania 63, 76, 94-5 Russia 1, 16, 20-1, 28, 41, 50, 148-9 and SOEs 8-9, 13, 41, 57-9, 63 state role 6, 17-18, 20-1 in transition countries 1,5-6, 97 Washington Consensus view 5 Schaffer, Mark 5-6, 41-2, 57, 64, 67, 70, 74, 80, 149-50 Schumpeter, Joseph creative destruction theory 4 on the tax state 11, 33 Shleifer, Andrei 4, 5, 13-16, 20, 24, 27, 41, 109, 148 shock therapy 2-3 Slovak Republic
bribe tax 111 CIT effective/statutory ratio 119 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134 per capita 134 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax arrears 79 tax effort 146 tax revenue/GDP ratio 7, 133 transition indicator 65, 107 Slovenia bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134 per capita 134-5 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax effort 146 tax revenue/GDP ratio 7, 66, 133 transition indicator 107 small business sector and banking system 38 and bureaucracy 37-8 and corruption 37 and economic conditions 39 Russia 36-9 and tax system 37 SOEs (State-Owned Enterprises) 1, 18-20, 23, 154 and SBC 8-9, 13, 41, 63 SOF (State Ownership Fund) 67, 88 soft taxation, in TEs 7-10, 13, 147
Index Soros, George 16 Spain tax effort 146 tax revenue/GDP ratio 133 SST (Social Security Tax) 100, 101, 103 base 104 bribe tax 112, 122-3 effective collection and transition 10910, 120 effective/statutory ratios 107, 118 GDP ratios CEE countries 105 FSU countries 105 NTY 107 state ‘grabbing-hand’ model, 4-5, 13-16, 20, 28-30, 148 ‘helping-hand’ model 13-16, 20, 29-30, 148 paternalism 13-14 role in SBC 6, 20-1 socialist, transition to capitalist state 11 as tax collecting body 1, 11, 33, 113-5, 148 state capture 18, comparisons 35 State Tax Service (STS) 22, 32, 99 Stern, Nicholas 98 Stiglitz, Joseph 4-5 stocks and flows, tax liabilities 67-80 Sweden tax effort 146 tax revenue/GDP ratio 133 Tajikistan bribe tax 111 CIT/GDP ratio 105 CPI 111 statutory/effective ratio 107, 117-9 GDP agriculture share 134 exports share 134-5 per capita 134-5, 141 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax effort 140, 146 tax revenue/GDP ratio 7, 105, 133 transition indicator 107 Tanzi effect 71, 75, 150
175
Tanzi, Vito 98, 99, 109, 110, 129, 141 tax avoidance 12, 98, 101, 113, 151 concessions 63, 98 definition 12, 127 evasion 1, 12, 29, 64, 68, 112-5, 147-8, 151 gap 97, 151 tax administration EBRD on 115 and transition 97-112 tax administration study conclusions 112-15 data 103-6, 116 methodology 100-3 policy implications 112-14, 152 results/analysis 106-12 scatterplots 108, 120-3 sources 104, 105, 116 tax arrears 6, 9, 13-14 calculation 67-73 definition 63-4, 149 enterprise sector Romania 63-95, 149-51 Russia 27-8, 51-4, 59, 79, 149 as fiscal aid 64 subsidy 83 TEs 79-80 cases 64 tax capacity constraints 127 definition 127 economic factors 128-30 TEs 9-10, 125, 152-3 tax capacity study conclusions 140-1 data 131-5, 143-86 methodology 127-31 regressions 135-6 results/analysis 135-40 sources 131-2 tax effort constraints 127-8 country estimates 138-40, 145-6 definition 127 ratio 128 tax liabilities 2, 8, 12-3, 27, 147, 149-51 calculation 63-73 response to 9, 63
176
Transition, Taxation and the State
stocks and flows 67-80, 150 tax rankings 138-9 tax rates statutory CEE/FSU countries 105 effective/statutory 97-8, 100, 112, 151-2 calculation 101-3 CEE countries 107, 117-9 EU-15, Mean 107, 117-9 FSU countries 107, 117-9 tax revenue decline 1, 7, 125, 154 function 129 GDP ratios CEE/FSU 7, 23, 66, 133 share 125, 127 tax system politicization 98 and small business sector 37 in TEs 98-100 TEs (Transition Economies) 1 GDP agriculture share 132, 134-5 exports share 132, 134-5, 141 per capita 134 market-supporting institutions 34 SBC in 5-6, 97, 147, 154 soft taxation 7-10,13, 147 study, data sources 103-6, 116, 131-2 tax administration, effectiveness 97-123, 151-2 tax arrears 79-80 tax capacity 9-10, 125-46, 152-3 tax system 98-100 tax/GDP ratios 7, 133 see also transition countries Thailand tax effort 145 tax revenue/GDP ratio 133 Tractorul UTB S.A. 84-8 debt 88 exports 85-6 financial indicators 87 losses 86 privatisation 85 SOF payments 88 wage payments 86 transition 118, 147, 154 effective tax administration 97-123 paradigms 2-3
progress in 107-8, 12-3 SBC 5-6, 97 tax system 98-100 transition indicators CEE countries 107 EBRD 65, 107-8 effective tax administration 120-3 FSU countries 107 Romania 65 Visegrad countries 65 Treisman, Daniel 14-5, 20, 24, 27, 106, 109 Tunisia tax effort 145 tax revenue/GDP ratio 133 Turkey tax effort 146 tax revenue/GDP ratio 133 Turkmenistan bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134-5 per capita 134, 141 market-supporting institutions 34 NTY 107 SST/GDP ratio 105 statutory tax rate 105 tax effort 140, 146 tax revenue/GDP ratio 7, 133 transition indicator 107 turnover tax 7, 23, 99 UK tax effort 146 tax revenue/GDP ratio 133 Ukraine bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134 exports share 134 per capita 134 market-supporting institutions 34 NTY 107
Index SST/GDP ratio 105 statutory tax rate 105 tax effort 146 tax revenue/GDP ratio 7, 105, 133 transition indicator 107 Uruguay tax effort 145 tax revenue/GDP ratio 133 US tax effort 146 tax revenue/GDP ratio 133 utilities 4, 6, 26, 51-2, 76-8, 82, 149-50 energy sector 28-9, 86, 90 Uzbekistan bribe tax 111 CIT/GDP ratio 105 CPI 111 effective/statutory ratio 107, 117-9 GDP agriculture share 134-5 exports share 134-5 per capita 134-5, 141 governance index 35 intervention index 35 NTY 107 SST/GDP ratio 105 state capture 35 statutory tax rate 105 tax effort 140, 146 tax revenue/GDP ratio 7, 133 transition indicator 107
177
van Wijnbergen, Sweder 46 VAT (Value-Added Tax) 7, 67, 91, 99, 100, 101, 103, 104, 105 base 104 effective collection and transition 109, 120 effective/statutory ratios 107, 117 and bribe tax 112, 122-3 veksel 29, 51-2 Venezuela tax effort 145 tax revenue/GDP ratio 133 virtual economy 28 Visegrad countries, transition indicator 65 Vishny, Robert 4, 5, 13, 14-15, 16, 20, 41, 148 wages, arrears/non-payment 28-9, 42, 51-4, 59, 148-9 Washington Consensus 2, 3 view of SBC 5 Whalley, John 100 World Bank 2, 30, 83, 85, 88, 92 Russian enterprises, survey 46, 149 Yavlinksy, Grigory 16 Yeltsin, Boris 16