WORLD BANK INSTITUTE Promoting knowledge and learning tor a better world Russia's Transition to a New Federalism 22026 February 2001 00ow ' 7.,, , - 1 [E u s~~~~P 4KRUN" Jorge Martinez-Vazquez Jameson Boex Cr. , * S~~~ SS 3 S Recent Books from WBI Case Studies in Participatory Irrigation Management David Groenfeldt and Mark Svendsen, editors 1999. 169 pages. ISBN 0-8213-4540-0. Stock No. 14540. Price code S22 The Challenge of Urban Government: Policies and Practices Mila Freire and Richard Stren, editors 2001. 458 pages. ISBN 0-8213-4738-1. Stock No. 14738. Price code S30 Chile: Recent Policy Lessons and Emerging Challenges Guillermo Perry and Danny M. Leipziger, editors 1999. 437 pages. ISBN 0-8213-4500-1. Stock No. 14500. Price code S35 Corrupt Cities: A Practical Guide to Cure and Prevention Robert Klitgaard, Ronald MacLean-Abaroa, and H. Lindsey Parris 2000. 175 pages. ISBN 0-8213-4600-8. Stock No. 14600. Price code S20 Curbing Corruption: Toward a Model for Building National Integrity Rick Stapenhurst and Sahr Kpundeh, editors 1998. 264 pages. ISBN 0-8213-4257-6. Stock No. 14257. Price code S25 Economic Analysis of Investment Operations: Analytical Tools and Practical Applications Pedro Belli withJock R. Anderson, Howard N. Barnum, John A. Dixon, andJee-Peng Tan 2001. 287 pages. ISBN 0-8213-4850-7. Stock No. 14850. Price code 535 Economic Development and Environmental Sustainability Jose L. dos R. Furtado and Tamara Belt with Ramachandrajammi, editors 2000. 125 pages. ISBN 0-8213-4573-7. Stock No. 14573. Price code S22 Korea and the Knowledge-based Economy: Making the Transition Carl Dahlman and Thomas Andersson, editors 2000. 152 pages. ISBN 0-8213-4881-7. Stock No. 14881. Price code S30 Preventing Bank Crises: Lessons from Recent Global Bank Failures Gerard Caprio, Jr, William C. Hunter George G. Kaufman, and Danny M. Leipziger editors 1998. 392 pages. ISBN 0-8213-4202-9. Stock No. 14202. Price code S40 Principles of Health Economics for Developing Countries William Jack 1999. 305 pages. ISBN 0-8213-4571-0. Stock No. 14571. Price Code S30 Resetting Price Controls for Privatized Utilities: A Manual for Regulators Richard Green and Martin Rodriguez Pardina 1999. 116 pages. ISBN 0-8213-4338-6. Stock No. 14338. Price code S35 Social Funds and Reaching the Poor: Experiences and Future Directions Anthoniy G. Bigio, editor 1998. 258 pages. ISBN 0-8213-4209-6. Stock No. 14209. Price code S25 The complete list of Institute publications is available in the WBI Publications Catalog 2000. Stock No. 31708 Order World Bank Institute publications by phone or fax: Phone: 1-800-645-7247 or (703) 661-1580: Fax: (703) 661-1501 by mail: Write to The World Bank, PO. Box 960, Herndon, VA 20172-0960, U.S.A. hy e-mail: books@aworldbank.org through the World Wide Web at: http://w-w.worldbank.org/publications or visit the WtBI website at: http://www.worldbank.org/whbi WBI LEARNING RESOURCES SERIES Russia's Transition to a New Federalism Jorge Martinez-Vazquez and Jameson Boex Fiscal Decentralization Series Editors: Robert Ebel Blanca Moreno-Dotson Victor Vergara Serdar Yilmaz The World Bank Washington, D.C. Copyright © 2001 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing February 2001 The World Bank Institute was established by the World Bank in 1955 to train officials concerned with development planning, policymaking, investment analysis, and project implementation in member developing countries. At present the substance of WBI's work emphasizes macroeconomic and sectoral policy analysis. Through a variety of courses, seminars, workshops, and other learning activities, most of which are given overseas in cooperation with local institutions, WBI seeks to sharpen analytical skills used in policy analysis and to broaden understanding of the experience of individual countries with economic and social development. Although WBI's publications are designed to support its training activities, many are of interest to a much broader audience. This report has been prepared by the staff of the World Bank. The judgments expressed do not necessarily reflect the views of the Board of Executive Directors or of the governments they represent. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use, is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978-750-8400, fax 978- 750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the World Bank at the address above or faxed to 202- 522-2422. The backlist of publications by the World Bank is shown in the annual Index of Publications, which is available from the Office of the Publisher. Jorge Martinez-Vazquez is professor of economics and director of international studies at the Andrew Young School of Public Policy, Georgia State University. Jameson Boex is a senior research fellow in the international studies program at Georgia State University. Library of Congress Cataloging-in-Publication Data Martinez-Vazquez, Jorge. Russia's transition to a new federalism / Jorge Martinez-Vazquez, Jameson Boex. p. cm -- (WBI learning resources series) Includes bibliographical references. ISBN 0-8213-4840-X 1. Intergovernmental fiscal relations--Russia (Federation) 2. Russia (Federation)--Appropriations and expenditures--Statistics. 3. Federal government--Russia (Federation) 4. Russia (Federation)--Politics and government--1991- I. Boex. Jameson, 1971- II. Title. III. Series. H1J1211.52.Z7 M37 2000 336.47--dc2l 00-063317 Contents Foreword vii Preface ix Acknowledgments xiii 1 The Dynamics of Decentralization in the Russian Federation I The Forces behind Decentralization in the Russian Federation 1 A Historical Overview of Intergovernmental Fiscal Relations, 1992-98 5 The Role of Politics in the Decentralization Process 7 Dimensions of Fiscal Decentralization in the Russian Federation 9 2 Expenditure Responsibilities 11 How Have Expenditures Been Assigned? 11 How Has the System Performed? 19 Pending Policy Issues in Expenditure Assignments 24 3 Revenue Assignments and Tax Administration Issues 33 How Have Revenues Been Assigned? 33 Trends in Revenue Collections 39 How Has the System Performed? 43 The Role of Perverse Incentives and Soft Budget Constraints 50 The Role of Tax Administration 51 Pending Reforms in Revenue Assignments 52 4 The System of Transfers 55 Evolution of the Transfer System 55 The Current System of Transfers 58 How Has the Transfer System Performed? 61 Pending Policy Agenda for Transfer Reform 68 5 Subnational Borrowing, Debt, and Financial Controls 75 Evolution of the Legal Framework for Borrowing 75 Analysis of Subnational Deficits and Debts 76 Pending Policy Agenda for the Reform of Subnational Borrowing 80 6 Implementing Reform 83 Piecemeal Versus Unified Reform 83 Uniform Versus Asymmetric Federalism and Addressing Fiscal Disparities 84 Regional Compliance with Federal Laws 85 Communications between Federal and Regional Governments 86 Determining the Form of Fiscal Federalism 86 7. Postscript: Recent Reforms in Intergovernmental Relations in the Russian Federation 89 Reform of the Vertical Power Structure 90 Expenditure Assignments and Budgetary Responsibilities 91 iii iv Contents Revenue Assignments and Tax Reforms 92 Changes in Intergovernmental Transfers 94 Concluding Remarks 95 References 97 Tables 1.1 Chronology of Major Events in Russia during the Transition 3 2.1 Assignment of Expenditure Responsibilities, 1997 12 2.2 Distribution of Expenditures between Government Levels, 1992-97 13 2.3 Distribution of Expenditures between Government Levels, 1992-97 14 2.4 Distribution of Expenditures by Function between Government Levels, 1992-97 15 2.5 Average Regional Expenditures by Category, 1994-97 17 2.6 Composition of Expenditures for Subnational Governments in Real Terms, 1992-97 18 2.7 Regional Economic Disparities: Per Capita GRP, Selected Years, 1992-96 19 2.8 Measures of Horizontal Fiscal Imbalance: Per Capita Consolidated Regional Expenditures, 1992-97 20 2.9 Equalization Level at Different Stages of Revenue Assignment, 1992-97 21 2.10 Regional Socioeconomic and Demographic Variables: Descriptive Statistics 22 2.11 Determinants of Per Capita Expenditures, 1994-97 22 2.12 Determinants of Per Capita Expenditures on the National Economy, 1994-97 23 2.13 Determinants of Per Capita Expenditures on Housing and Public Utilities, 1994-97 24 2.14 Determinants of Per Capita Expenditures on Education, 1994-97 25 2.15 Determinants of Per Capita Expenditures on Public Health and Physical Education, 1994-97 25 2.16 Expenditure Assignment in the Budget Code, 1998 26 3.1 Legislated Sharing Rates of Major Taxes, 1992-98 34 3.2 Status of Taxes in the 1998 Tax Code 39 3.3 Own-Source Revenues and Shared Revenues as Percentages of Consolidated Subnational Collections (Excluding Transfers and Subventions), 1992-97 40 3.4 Revenue Distribution between Government Levels, 1992-97 41 3.5 Consolidated Subnational Own-Source Revenue Collections, 1997 42 3.6 Summary of Legislative Control over Subnational Own-Source Revenues, 1997 44 3.7 Revenue Distribution between Government Levels, 1992-97 45 3.8 Composition of Revenues for Subnational Governments in Real Terms, 1992-97 46 3.9 Measures of Vertical Fiscal Imbalance for the Subnational Sector, 1997 48 3.10 Measures of Vertical Fiscal Imbalance for the Subnational Sector: An International Comparison 48 3.11 Measures of Regional Fiscal Disparities: Per Capita Revenue Collections (including Federal Share), 1993-97 48 3.12 Concentration of Total Revenue Collections (including Federal Share) among Regions, 1993-97 49 3.13 Concentration of Regional Revenue Coliections among Regions, 1992-97 49 3.14 Analysis of Regional NRT Collections, 1994-97 50 4.1 Fund for Financial Support of the Regions 57 4.2 Formula for Aliocating FFSR Transfers, 1999 59 4.3 Funding Sources for the Regions, 1992-97 62 4.4 Federal Transfers to the Regions, 1992-97 62 4.5 Transfer Composition for Subnational Governments in Real Terms, 1992-97 63 4.6 Regression Analysis: Per Capita FFSR Transfers, 1994-97 64 Contents v 4.7 Regression Analysis: Per Capita Other Transfers (excluding Budget Loans), 1994-97 64 4.8 Regression Analysis: Allocation of (Per Capita) Regional Resources, 1994 65 4.9 Regression Analysis: Allocation of (Per Capita) Regional Resources, 1995 65 4.10 Regression Analysis: Allocation of (Per Capita) Regional Resources, 1996 66 4.11 Regression Analysis: Allocation of (Per Capita) Regional Resources, 1997 66 4.12 Explaining Tax Effort for Regional Collections, 1994-97 67 4.13 Explaining Tax Effort for Overall Collections (including Federal Share), 1994-97 67 4.14 Executed Share of FFSR in Each Region, 1995-97 69 5.1 Consolidated Budget Execution, 1992-97 77 5.2 Relative Size of Subnational Debt, Comparison of Selected Countries 77 5.3 Regression Analysis: Per Capita Other Sources of Financing, 1994-97 78 7.1. Legislated Sharing Rates of Major Taxes, 1999-2001 93 7.2. Federal and Subnational Shares of Consolidated Tax Revenues, 1998-2001 94 7.3. Budget Proposal 2001: Federal and Regional Shares of Consolidated Revenues 95 Foreword Throughout the 1970s and 1980s, economic decentralization-the shift toward private enterprise and markets-was clearly underway in the Soviet Union. Although a not insignificant amount of private activity such as moonlighting and small agricultural enterprises had always been entrenched in a large informal economy, the mid-1980s witnessed the important development of a central government-led policy of privatization. This was the era of Mikhail Gorbachev and perestroika. The political and public sector landscape was also undergoing fundamental changes as pressure grew throughout the entire Soviet sphere to increase political independence and localize fiscal decisionmaking. Then, in November 1991, the Soviet state ended. Almost immediately the focus turned to issues of gover- nance and especially to the issue of intergovernmental relationships, both among the former republics, now newly independent states, and within these new states. At this point, fiscal decentralization and central sharing of public sector roles had not only become issues as important as economic decentraliza- tion, but they also emerged as keys to successful achievement of the nation-state's overall reform goals of macroeconomic stabilization, poverty alleviation, privatization of state-owned enterprises, and nationbuilding. Indeed, determining the right set of intergovernmental fiscal arrangement is among the most dominant policy issues facing Russia and the states of the former Soviet Union. As Christine Wallich points out in her preface to this volume, Russia in Transition to a New Federalism, and to which authors Jameson Boex and Jorge Martinez-Vazquez so forcefully attest, "history hovers... .over any discussion of fiscal federalism in Russia" (p. x). This volume not only reviews that history, but also follows it to the present situation with respect to the difficult policy choices facing the new Vladimir Putin-led government and neatly links Russia's choices with the broader issues facing any country un- dergoing intergovernmental reform. Thus, the authors weave into the discussion the dynamics of Russia's options with respect to deciding who delivers what services and how to finance these services with the question of the fiscal politics of change, or the intergovernmental and inter-regional balancing and coun- terbalancing of power. Furthermore, the authors tie these questions together with those of how to imple- ment good governance: should it be piecemeal or unified, asymmetric or uniform, and centrally driven or locally controlled? This book also initiates the first in a series of studies that will be published in the World Bank Institute's Learning Resource Series on the issues of governance and decentralization as they are framed by the World Bank Institute's broader set of programs in public finance and financial management. Three themes frame this set of World Bank Institute programs: * Globalization, which reflects the progressive integration of the world's economies, and localiza- tion, which involves demand for greater self-determination and influence in government among a country's citizens, stem from the same source and reinforce one another. Both reflect the impact of advanced information and communications technologies, and both look to the establishment of new and, again, reinforcing institutions at the international, national, and local levels. Thus, glo- balization has circumscribed the power of nation-states to not only control markets, but also the ability to manage some once easily controlled fiscal decisions. This trend has led many central governments to grant fiscal, political, and administrative powers to local governments; however, governments have often not granted these powers at a pace that matches the growth in desire for citizen empowerment, or the rate of decentralization in the economy itself. * The shift from government, the institutions through which a political authority exercises power, to governance, the more dynamic and participatory process by which people collectively determine vii viii Foreword the level and mix of the tax-public service packages, is fundamentally tied to the question of the decentralization of the state. Decentralization is both a cross-cutting discipline and a theme that ranges from questioning the role and design of the public sector to the analysis of which sphere of government(s) should manage and monitor the delivery of, and the payment for, public services. The proximity of local governments to the poor and the familiarity and understanding of the vary- ing institutional environments that the poor inhabit in different regions and communities provide distinct advantages to decentralized governmental units for designing and implementing anti- poverty policies. To successfully address the goals implicit in these themes, investing in human capital through effec- tive knowledge development and dissemination is key. At the heart of the World Bank Institute's pro- grams is the determination to help people help themselves by providing the resources needed to share knowledge, build capacity, and forge partnerships. We are pleased to initiate this fiscal series with such a timely and important contribution. Vinod Thomas Vice President World Bank Institute Preface Whither Russia? has been a key question since the Russian Federation was born out of the dissolution of the USSR in December 1991. A diverse country with 11 time zones, 88 regions, and many nationalities, Russia remains heavily burdened by the overhang of the socialist system. The political and economic challenges facing Russia in its transition to a more democratic and market-oriented economy have many dimensions. This book looks at one of the most important of these dimensions: the decentralization of the Soviet state that is Russia, and the implications of federalism for the growth of Russia's economy, for the equity and well-being of its diverse peoples, and for its national cohesion. This is the third in a series of World Bank books on fiscal decentralization in Russia and the transition economies. The first, Russia and the Challenge of Fiscal Federalism (Wallich 1994) explored Russia's early steps in decentralization, and concluded that as Russia makes the difficult transition from a command to a market economy, the design of a well-functioning, intergovernmental, fiscal system is key to achieving the broader policy goals of macroeconomic stability, the equitable provision of social services, a social safety net, and especially, political cohesion. It outlined the nature of these center-regional issues and tensions and how a well-designed intergovernmental system can can help reduce them. Focused more broadly on Central Europe and the Commonwealth of Independent States, The Decen- tralization of the Socialist State (Bird, Ebel, and Wallich 1995) systematically reviewed and analyzed inter- governmental fiscal issues in seven transition economies. Key themes were that local finance is not a local matter and that the ongoing reforms in subnational finance taking place in virtually every transition economy are very different from those of market economies, very similar to each other, and more impor- tant than is generally recognized. That book provided a picture of the emerging process of fiscal decen- tralization in such countries as Hungary and Poland, which have moved down the path of local govern- ment reform and may now be embarking on a second generation of reforms, as well as those that were beginning the process, such as Albania, Romania, and Ukraine. This volume is a companion piece to these earlier studies, exploring many of the same themes in a wider country and comparative context, with a far richer empirical base than was possible in those early works. Rich in fieldwork at the oblast and rayon levels, this volume also provides a clearer policy roadmap than was practicable in the early days. The early literature on the transition from command to market economy tended to focus on the big picture issues of rapid price reform, enterprise privatization, and macroeconomic stabilization policies. More recently, the focus has been on second generation reforms of, for example, governance and civil service, judicial and legal systems, and bankruptcy and the public sector. The fiscal decentralization tak- ing place in Russia and other countries of the former Soviet Union and Soviet sphere and the transition economies of Eastern and Central Europe has, by comparison, not received as much scrutiny. Yet how the ensuing issues in intergovernmental finance are resolved continues to influence the transition, macroeco- nomic stabilization, and the maintenance of the social safety net that is often critical to support the re- form. This theme and these books thus fill an important gap in the literature on transition. Fiscal decentralization took place early in the transition, motivated by economic efficiency and politi- cal demands for greater autonomy from the centralized structures. Characteristic of the early decentrali- zation reforms was the breadth of spending responsibilities assigned to subnational and regional govern- ments. In many countries, macroeconomic strains on central governments led them to off-load spending functions, including social welfare, education, and health, to subnational governments. However, re- gions were rarely given taxing powers and, generally, transfer systems remained limited, ad hoc, ix x Preface negotiated, and sometimes counterequalizing. For macroeconomic reasons, central governments across the region typically kept the major national tax bases for their own budgets. The general lack of fiscal discretion, fiscal responsibility, and accountability, along with a general absence of revenue autonomy and ability to set local tax rates, squeezed subnational budgets, threatening the provision of subnational public services and their efficient and equitable delivery. Early research suggested that a broad analytical framework was needed to analyze decentralization issues in the transition economies. That framework should do the following. First, it should into take account the subnational governments' continued ownership of enterprises on a significant scale and factor this into the incentive framework for privatization. Second, it should focus more closely on the subnational role in macroeconomic stabilization. While subnational spending and deficits cannot on their own create macroeconomic problems (subnational governments cannot print money), the soft budget constraints of many government agencies mean that subnational deficits in the transition economies can be, and often are, monetized. Third, to the extent that the social safety net has been assigned to subnational governments (or off-loaded for stabilization purposes), the framework must pay careful attention to dis- tributional issues, especially in countries where the revenue base is not broadly distributed. History hovers over this book and over any discussion of fiscal federalism in Russia. As the USSR entered the 1990s, fiscal relationships between the central government and its 15 republics were strong and the Soviet Union was intact. But by late 1991, the powers and activities of the USSR's central govern- ment had eroded. An important catalyst for dissolution was the Union republics' decision, led by Russia and Ukraine, to halt revenue collections and transfers to the Union budget. This culminated in the fiscal death of the Union and in the dissolution of virtually all central bodies. With pressure for political au- tonomy also irresistible, Russia declared its independence in November 1991. Fiscal federalism in the Russian Federation-Russia and its 88 oblasts-is as important as in the USSR. It was apparent early on that the decentralization challenges facing Russia, such as potentially huge revenues from natural resource taxes, could prove divisive in terms of who gets what share, and could become explosive if prices rose to world levels. Obvious also was that ethnicity, regional differences, and demands for fiscal sovereignty by secession- ist regions within Russia were growing and that, as in the Union, the intergovernmental system could play an important role in increasing (or reducing) political tensions. Thus, given the considerable centrifugal forces facing Russia, the contribution of fiscal federalism to nation building could not be overlooked. Also evident was the huge challenge of providing equitably for Russia's population, which is spread out over 17,075 square kilometers of territory, that is, nearly twice that of China and the United States and five times that of India. The contrasts between regions are startling. Five of Russia's 88 regions collect 40 percent of all revenue and 52 percent of federal resource-related fees and revenues. One autonomous region in Siberia, Khanty-Mansisk, is the size of France, has 1/50th of the population of Russia, and produces more than 80 percent of the oil-about as much as Texas or Kuwait. However, the regional capital has only one paved street, pedestrians mingle with livestock, and houses have walls of bare wood and windows of paper in an area where temperatures fall to minus 40 degrees centigrade or below. Many homes have no running water. Environmental damage from the exploitation of oil and gas gave one resource-rich city the unwelcome distinction of being among the United Nations Environment Programme's top 10 environmental disaster zones, with off-the-charts levels of birth defects and other health emergencies. Russia's system of fiscal federalism and transfers contributed only marginally to enabling such regions to raise their social service standards to a fraction of Russia's average. This book analyzes the emerging process of fiscal federalism and the effects of recent legislation on intergovernmental relations in Russia. It evaluates the strengths and weaknesses of new structures and outlines possible directions for further reform. The book-which examines the political dynamics of de- centralization, expenditure responsibilities, revenue assignments and tax administration, transfers, and borrowing-represents an integrated view of fiscal federalism in Russia. The analysis focuses on 1992-98, and much of the analytical and empirical work draws on data begin- ning as early as 1992. The book begins by examining the dynamics of decentralization in Russia, including a Preface xi historical overview and the politics behind the decentralization process. Russia's centrifugal forces are strong and have significant fiscal implications. Given the growing disparities in the regions' economic base, the federal government is vulnerable to political pressures from better-off regions that are opposed to paying subsidies to redistribute to the poorer ones. Some regions, such as Dagestan and Tatarstan, have even opted out of the system through bilateral treaties, creating a de facto asymmetric system of federalism in which better-off regions define federalism in their own terms, while the formal system applies to their weaker brethren. Growing economic and fiscal disparities have led to a pronounced divergence of interests among regions and to a lack of consensus in national politics. This is especially true of the natural resource- producing regions and those relatively few regions with industries integrated into world trade. The fed- eral government must either engage in more redistribution among the regions or achieve faster, but geographically uneven, economic growth. The authors raise the question of whether a permanent underclass of regions is being created, thereby providing a basis for further division. They argue rightly that these tendencies need to be contained in the interests of national cohesion, balanced growth, and macroeconomic stabilization. The authors make a strong case for greater clarity and more efficient assignment of spending respon- sibilities. They highlight the growing disparities in spending across regions: in 1992 Russia's richer re- gions spent 12 times as much per capita as the poorest; by 1997 this gap was 22 times. The authors observe that such growing differentials cannot be good for national cohesion. A transfer system, absent in the early days of decentralization, was introduced in 1993, but has yet to have a major impact on regional equity. While the system is mildly equalizing (transfers do appear to be inversely related to regional per capita gross domestic product), it is insufficient to offset the other, more powerful, tendencies creating greater concentrations of economic well-being in those regions that are benefiting from reforms, integra- tion into world trade, or natural resource endowments. In their fiscal distress, many of Russia's regions have borrowed, and subnational debt has grown significantly. The authors do not link this directly to the ruble crisis of 1997, but they note that the federal government's attempt to transfer its deficit to subnational governments by off-loading responsibilities and expenditure mandates was only partially successful, and that in the end the federal government still had to borrow to finance part of the subnational deficit. To the extent that these deficits were monetized, they may have contributed indirectly to the crisis. The authors conclude that, while the needed fiscal decentralization reforms are all interrelated, a piecemeal but systematic approach must be taken. This reflects the difficulties in reaching national con- sensus on the general objectives of decentralization, on the extent and nature of federalism in Russia, on the desirable degree of equalization, on the efficient use of budgetary resources, on the creation of incen- tives for revenue mobilization, and even on the preservation of fiscal discipline and responsibility among subnational governments. Because the process of defining an overall strategy for fiscal federalism will be difficult and slow, the authors recommend institutional vehicles-such as the National Consultative Commission-for gaining consensus and generating a vision of the system for the longer term. Although this book focuses on Russia, similar tendencies and problems are manifest in the other former republics of the Soviet Union and comparable countries elsewhere. The COMINTERN technical assistance provided by the USSR in the 1950s transplanted variants of the USSR system of intergovern- mental finance across a range of countries, including the former Yugoslavia and China, the Democratic People's Republic of Korea, and Vietnam. This volume should therefore be of interest to those working not only on the former USSR states, but others where the Soviet model of intergovernmental finance was transplanted and where new forms of and approaches to decentralization must also be found. While the authors are academics and the subject matter offers much for researchers to address, this volume presents the issues in intergovernmental fiscal relations in Russia to a broad audience. Aimed at general economists, practitioners of public finance, academics, policymakers, and their advisers, this volume should also be of great interest to those working on other areas of transition. Those who are xii Preface interested in social safety nets, macroeconomic stabilization, privatization, and the role of the state and rule of law in facilitating the transition to a market economy should also find this book relevant. In sum, this book is important for all those who wish to enter the Russian labyrinth. Together with its companion books, it alerts policymakers to the critical role of fiscal federalism in facilitating a successful transition to a market economy. Important subnational fiscal changes will continue in Russia for some time. Those concerned with ensuring successful transition neglect them at their peril. Fiscal federalism in Russia is not a trivial matter that can be tidied up after the "big" decisions are made on such matters as privatization and stabilization policy. On the contrary, how the nation deals with fiscal federalism is likely to be key to its achievement of broader reforms. Christine I. Wallich World Bank and Asian Development Bank October 2000 Acknowledgments We would like to thank Roy Bahl, Michael Bell, Richard Bird, Lev Freinkman, Albert Igudin, Galina Kourliandskaia, John Mikesell, Yelena Nikolayenko, and Sally Wallace for their comments and many helpful suggestions. We are grateful to Lev Freinkman, Alexander Morozov, and Stepan Titov for sharing various data sets with us. This research was funded by the U.S. Agency for International Development under the Georgia State University Russian Fiscal Reform Project and the World Bank Institute. The opinions expressed in this study are strictly the authors' own and should not be attributed to either the agency or the World Bank. xiii 1 The Dynamics of Decentralization in the Russian Federation Intergovernmental fiscal relations may be the most contentious issue in the Russian Federation today. Complex, intertwined problems that affect the federal fiscal system include the recent disintegration of the Soviet Union; the war in Chechnya; ethno-linguistic, religious, and historical differences across the vast territory of the Russian Federation; great variations in the endowment of mineral resources; increas- ingly large economic and fiscal disparities across regions and local governments; and the spreading sen- timent that some regions would benefit by seceding from the Federation. The survival of the Russian Federation depends on a functional and appropriate system of intergovernmental fiscal relations. At the time of independence in 1991 it was generally accepted that a federative government was the only political system that would fit the size and diversity of the Russian territory (Polishchuk 1996). Since its conception, the Russian Federation has been overwhelmed with continued pressure to decentralize. The motivations behind this push for greater subnational autonomy have been both positive and negative; they include the desire to achieve a more efficient and fair government, on one hand, and a lack of national solidarity on the other hand, causing some regions to pursue policies that conflicted with common national interests. In an environment of political instability and economic transition, the negative motives seem to have dominated the evolution of intergovernmental relations within the Russian Federation. A significant difference exists between the legislated system of intergovernmental relations and real- ity. On paper, transparent and symmetric components of intergovernmental relations were developed in the years after the disintegration of the Soviet Union. Previously differentiated sharing rates for shared taxes were standardized, and a formula-driven equalization mechanism was introduced to replace a system of negotiated transfers, while Articles 71 and 72 of the 1993 Constitution of the Russian Federa- tion assigned expenditure responsibilities between the federal government and the "subjects of the Fed- eration." However, in reality general confusion, dissatisfaction, and even open political conflict between the federal government and the regions, and in some cases, between the regions and local governments, mar intergovernmental relations. Since independence from the Soviet Union, the federal government has struggled for power with the regional governments. Ambiguity with which expenditure responsibilities were assigned in the 1993 Constitution and the large number of bilateral fiscal arrangements between the federal government and individual regions together weakened the designed transparency and symme- try of intergovernmental fiscal relations within the Russian Federation. The Forces behind Decentralization in the Russian Federation According to the 1993 Constitution, the Russian Federation comprises 89 subnational regions or "sub- jects of the Federation," including 49 oblasts, 21 republics, 6 krais, and 2 cities (Moscow and St. Peters- burg). In addition, the Russian Federation contains 11 okrugs, which are subunits within larger oblasts but have independent budgets. Discounting Chechnya, which is not a participating member of the Federa- tion, the national government maintains regular fiscal relations with 88 regions. The basic economic argument for decentralization in Russia, identical to that in other countries in transition, is the failure of the socialist economic model and the centralized government control that it represented. Fiscal decentralization in Russia has been fueled by the belief that decentralization is an effective way to increase government efficiency and strengthen democratic institutions. Until recently, policymakers paid little attention to the risks associated with decentralization, such as more unequal resource distribution and the possibility of compromising macroeconomic stability. I 2 Jorge Martinez-Vazquez and Jameson Boex Fiscal Power Struggles The decentralization dynamics of the Russian Federation involved a top-down process by which the federal government delegated a number of expenditure responsibilities to the regions. In addition, the Russian Federation subjected subnational governments to strict expenditure guidelines and granted limited revenue-raising opportunities (Igudin 1998). As a means of decentralization, delegation pro- vided continued centralized control over expenditures and federal budget relief. However, the 1993 Constitution granted regional governments a constitutional right to govern their own affairs and to formulate and execute their own budgets, with little meddling from the federal authorities. This devo- lution, or bottom-up approach to decentralization, arguably provided the Russian Federation with features of a federal system. This mix of decentralization models is conspicuous in the 1993 Constitution, which deliberately ob- scures the origination point of fiscal power (Teague 1996). In reality, power is shared, but its division is far from having been decided. A safety valve in the 1993 Constitution allowed the federal government to enter into bilateral treaties with regional governments. With the first treaties in 1994, it became clear that the regions could circumvent the 1993 Constitution (OECD 1997a; Teague 1996). Disregard for the federal laws was evidenced by the constitutions and statutes approved by the regional legislatures. McLure, Wallich, and Litvack (1996) remarked that the Russian state is a federation in name only. They call atten- tion to the few legal powers and the virtual lack of revenue autonomy of the regions. At the same time, many regions have claimed powers akin to those of a confederation, such as refusing to remit tax collec- tions to the central government or taking over the responsibilities of the federal government in the re- gions. Fiscal relations, characterized by confrontation and at times chaotic claims to power, have been described as fiscal feudalism in the Russia Federation (Lavrov 1998). In addition, the favorable treatment given by the Russian Federation to ethnic republics, such as Tatarstan and Bashkortostan, created resentment and imitators among many other regions.' Federal def- erence to ethnic regions led to the evolution of an asymmetric federalism in which some regions retained more resources than others. Political Factors The process of reform of the system of intergovernmental fiscal relations in the Russian Federation in- volves many economic and technical issues, but this decentralization continues to be foremost a political process. The political nature of fiscal decentralization in the Russian Federation is unquestionably con- nected to its vast territory, significant disparities in industrial development and resource endowments, and ethno-linguistic diversity. These factors exert centrifugal forces on the country that can only be coun- tered with an integrated economy that allows regions to be more prosperous as part of the Russian Fed- eration than as independent states. As of 1999, the transition to a market economy has failed to bring the economic growth and prosperity to all or even to a majority of the regions. The Russian Federation was created in 1991 when Russia and other former Soviet republics refused to remit tax revenues to the Soviet government in Moscow, proclaimed that their laws took precedence over Soviet laws, and ultimately declared themselves independent. This disintegration process created a weak political start for the Russian Federation because it set a precedent for the Russian regions to assert high degrees of autonomy (Teague 1996). The aggressiveness of the regions, especially the ethnic republics, to grab more authority grew bolder during 1992 and 1993 when President Boris Yeltsin and Parliament (the Supreme Soviet) were involved in a power struggle of their own. As a stopgap measure, the Federation Treaty was signed in March 1992 to regulate the relationships between the federal and regional governments, but Tatarstan and Bashkortostan refused to sign it. As 1. Of the 21 ethnic republics, only 6 can claim a majority ethnic population. The Dynamics of Decentralization in the Russian Federation 3 with the subsequent 1993 Constitution, the Federation Treaty left the issues of the division of power between the federal and regional govermments unresolved. Although the ethnic republics were given some additional powers in the Federation Treaty, these were taken away by the 1993 Constitution. For a brief chronology of major events during the Russian transition, see Table 1.1. Regional positions on federation issues and policies increasingly split between have and have-not regions. Donor regions contribute more in overall taxes than they get back in revenue-sharing and trans- fers, while subsidy regions receive a net inflow of funds from Moscow. The group of donor regions has varied with transfer and revenue-sharing policies, but included Moscow, St. Petersburg, Nizhny Novgorod, Samara, Sverdlovsk, Volgograd, Tyumen, Khanty-Mansi, Krasnoyarsk, and Sakha (Yakutia). A formal association of these regions was promoted by the mayor of Moscow, Yuri Luzhkov. Although regions entered large geographical associations to bargain with the federal government, such as the Northwest, the Volga basin, the Central-Black Earth, the Urals, and the Siberian associations, it is the donor versus subsidy divide that increasingly dominates the political process. Devolution gained importance in the decentralization process as democratic institutions developed at the subnational level. A significant shift in the balance of power between the federal government and the regions came with the 1993 Constitution and a subsequent law that required all subjects of the Russian Federation to have popularly elected heads of the executive by the end of 1996. All regions but five complied with the deadline; all adhered to the mandate by 1997. Before the elections, many regions had been amenable to following orders; however, after the election of regional governors, the federal authori- ties were forced to negotiate with each of the regions. Negative forces gained momentum in part because federal authorities did not provide regional gov- ernments with enough funding for basic services, and regional constituents believed that abuse, fraud, and cronyism abounded in Moscow. Separatist demands and open opposition to federal government Table 1.1. Chronology of Major Events in Russia during the Transition Date Event June 1991 Boris Yeltsin becomes Russia's first popularly elected president. October 1991 Law on the Basic Principles of the Budgetary System and the Budgetary Process enacted. December 1991 Russia, Ukraine, and Belarus dissolve the Soviet Union and form the Commonwealth of Independent States. December 1991 Law on the Basic Principles of the Tax System in the Russian Federation enacted. March 1993 Congress of People's Deputies votes to give itself the power to remove the president from office for unconstitutional behavior and to suspend presidential decrees. April 1993 Law on the Foundations of Budgetary Rights and Rights to Form and Use Extrabudgetary Funds enacted. December 1993 New constitution adopted in referendum; parliamentary elections held; nationalists gain support. August 1995 Law on the General Principles of Organization of Local Self-government in the Russian Federation enacted. December 1995 Parliamentary elections held; major gains for the Communist Party. July 1996 Presidential elections held; Yeltsin defeats Communist challenger Zyuganov. September 1997 Law on Financial Foundations of Local Self-governments enacted. April 1998 Yeltsin dismisses Chernomyrdin and appoints Kiriyenko as prime minister. July 1998 Government reaches agreement with International Monetary Fund on major anticrisis loan package. State Duma approves the Budget Code but fails to pass other components of the anticrisis plan. August 1998 Government announces ruble devaluation and suspends payment on government bonds. September 1998- Kiriyenko dismissed; a number of prime ministers are successively appointed and August 1999 dismissed. December 1999 Yeltsin resigns; prime minister Putin becomes acting president. Source: Authors. 4 Jorge Martinez-Vazquez and Jameson Boex policies grew after 1993 and received a significant boost by the power vacuum created during the 1998 financial crisis when Kiriyenko's administration was dismissed. This time, as in the earlier years of the transition, several regions declared they would stop remitting funds to Moscow. Several others intro- duced price controls and prohibited the export of food supplies from the region. Kaliningrad even in- tended to declare a state of emergency. Asymmetric Federalism In response to regional opposition and centrifugal forces, the federal government systematically of- fered more advantages and privileges to politically difficult regions than to cooperative ones, creating an asymmetric federalist system (Wallich 1994). Many researchers have documented this pattern, which clearly emerged in 1992, of favoring the 21 ethnic republics in the form of larger subsidies, permission to retain a higher share of tax collections, and special decrees granting economic benefits (Lavrov 1995; Treisman 1996, 1998b). Asymmetric federalism, taking the form of particular political and economic arrangements, is fre- quently regarded as a flexible policy that accommodates regional diversity and has been effective in Canada and Spain (McLure, Wallich, and Litvack 1996). Asymmetric federalism has also been heralded as a positive way to address the growing conflict of interest and divergence between richer regions, with stronger economic bases and more incentives to resist interregional distributions, and poorer regions, which are becoming increasingly dependent on federal transfers (Freinkman, Titov, and Treisman 1998). The ambiguity in Articles 71 and 72 of the 1993 Constitution concerning the division of powers between the federal government and the regions, as well as provisions allowing the federal authori- ties to enter into bilateral agreements with separate regions, all point to the legality of asymmetric federalism. Bilateral agreements provide a contractual framework within which detailed functional agreements are stipulated between the federal government and a region, including budgetary rela- tions; expenditure and revenue assignments; state property; sectoral regulations; and mandates about ownership and use of natural resources, environmental protection, and regional migration. Until 1995, these treaties were used to clarify intergovernmental fiscal relations between the Russian Fed- eration and the ethnic republics. While legal, such bilateral agreements created a good deal of resentment among monoethnic re- gions. Starting in 1995, the use of negotiated arrangements was expanded to cover other types of regions, such as oblasts, krais, okrugs, and cities. Consequently, the federal government became in- volved in a simultaneous bargaining game in which concessions to one region affected those granted to other regions. Asymmetric results followed because not all participants had the same degree of credibility and bargaining power (Solnick 1995), while Moscow rewarded, rather than punished, those regions that were most defiant. As a result, all regions had an incentive to deviate from federal laws and press for special treatment. Asymmetric federalism led to a weaker government and worsened economic situations for all regions (Treisman 1996). Why did federal authorities choose a mechanism for disunion and resentment among the regions that may eventually bring about the disintegration of the Russian Federation? Solnick (1995) ar- gued that the inability of the federal government to impose and enforce universal rules may have caused the federal government to consider independent bilateral negotiations a superior solution. That is, the pur- chase of regional consent may have been imposed on the federal government by political reality. Prior to the implementation of bilateral treaties, some subnational governments constantly resisted federal fiscal disci- pline. The subsequent bilateral arrangements may have weakened any coordinated actions and demands by the ethnic republics and may have prevented the secession of Tatarstan. As a result, asymmetric federal- ism fostered, but also helped control, the centrifugal forces in Russia. The bilateral agreements appear to be only one of several causes of the poor fiscal incentives and revenue collections at the subnational level. Since 1993, the Russian Federation has lacked a coherent vision for decentralization. The different regional viewpoints have made achieving a consensus impossible. The Russian Federation will have The Dynamics of Decentralization in the Russian Federation 5 continued difficulty implementing an overall strategy for reform and to find an appropriate balance among key decentralization objectives: fairness in the allocation of resources across regions, the promo- tion of growth, and the preservation of macroeconomic stability and fiscal and monetary discipline. A Historical Overview of Intergovernmental Fiscal Relations, 1992-98 Decentralization of government functions has been an integral part of the Russian Federation's transition to a market-based economy and can be distinguished by three phases: (a) the early years of the transition (1992-93), (b) the reform process started in 1994, and (c) recent reform policies (1997-98). The role of politics, which has been central in defining each phase of decentralization policy, will be discussed fol- lowing the historical overview. The Early Years of the Transition, 1992-93 This early period of decentralization can be best characterized as a continuation of the approaches and practices inherited from the former Soviet Union when customized revenue-sharing rates and bargained transfers were used to fund the budgets of all subnational governments. However, during the transition, expenditure responsibilities in many policy areas were shifted down to the subnational government level, including subsidies for food, medicine, local transportation, and public utilities, which in the past had been the responsibility of the federal government. This meant a significant increase in the involvement of subnational governments in the economic sphere. In addition, the Russian Federation delegated respon- sibility for many social welfare policies and various capital investments to the regions. The greater expenditure responsibilities of subnational governments were accompanied by an in- crease in the subnational governments share in overall government expenditures and an increase in the importance of intergovernmental transfers. These changes occurred against a background of a drastic decrease in the involvement of the public sector in the economy; overall public sector expenditures de- creased from 52 percent of gross domestic product (GDP) in 1992 to 42 percent by 1994 (Freinkman, Titov, and Treisman 1998). Regional governments received funding from three different sources: revenues generated from re- gional and local taxes; shared revenues from the major taxes, enterprise profit tax (EPT), individual in- come tax, value added tax (VAT), and excise taxes; and transfers. Only limited fiscal autonomy was granted to the regions, which remains the most significant factor hindering full decentralization.2 Not only were regional and local governments given little revenue-raising authority, they were also under dual subordi- nation to both federal and subnational authorities. Through the entire Soviet period, sharing rates for the major taxes were kept low enough so that most regions received an annual subvention from the federal budget. During the early transition pe- riod, the federal government retained its longstanding power to regulate the rate of at least one of the major taxes, effectively reducing funds to those oblasts with large tax bases and increasing funds to those with smaller tax bases. In 1992 and 1993, tax sharing rates were increased to the point that fewer oblasts required a general subvention from the federal government. However, in 1993 alone, sharing rates were changed three times. Transfers took the form of negotiated subventions in 1992-93, which were used when the other two sources of revenue were not enough to cover a negotiated, minimum, required expenditure budget. At the local level, rayons (local government districts, similar to U.S. counties) and cities had the same general structure of funding sources as the regions, but they received revenue sharing and subventions from regional governments in similar ways to federal government funding of the regions. Each local 2. This was granted in 1991 by the Law on Basic Principles of the Budget System and the Budgetary Process and in 1993 by the Law on the Foundations of Budgetary Rights. 6 Jorge Martinez-Vazquez and Jameson Boex government had its own customized sharing rate for at least one major tax in which revenues were shared between the region and local governments. This regulating principle in tax sharing gave upper- level government the ability to redistribute funds to poorer jurisdictions. The ability to "regulate" rev- enues in this fashion came at the expense of lower-level government budget autonomy while also intro- ducing perverse fiscal incentives. In particular, regional governments commonly recaptured or "clawed back" the local budget surplus of the previous year or offset any increase in local revenues, such as that from a higher local tax effort, simply by reducing the local share of the shared taxes. Between 1991 and 1993 a series of laws enacted to regulate fiscal issues would also serve as precursors to the 1993 Constitution: the Law on the Basic Principles of the Budgetary System and the Budgetary Process of October 1991, the Law on the Basic Principles of the Tax System in the Russian Federation of December 1991, and the Law on the Foundations of Budgetary Rights and Rights to Form and Use Extrabudgetary Funds of April 1993. Many of their provisions, especially those outlined in Basic Prin- ciples of the Tax System, were ignored (Igudin 1998; Lavrov 1998; Wallich 1994). While regional noncom- pliance with federal legislation was not unusual under the Soviet regime, continued resistance converted some of these regulations into mere proclamations. This created a precedent for the continued pattern of selective compliance by regions of federal laws. Not long after the disintegration of the Soviet Union, some of the ethnic republics, most notably Sakha (Yakutia), Tatarstan, and Bashkortostan, stopped tax remittances to Moscow and insisted on a single- channel arrangement by which the regions would retain all tax collections and make a single negotiated payment to the federal government. Either because it was absorbed in an intense fight between the ex- ecutive and the legislative branches or because it lacked effective means of enforcement, the federal gov- ernment could only respond to these demands by making these regional governments responsible for all federal programs. In summary, the 1992-93 period was a continuation of the Soviet system. Regional and local budgets were basically extensions of the federal government, and the system of intergovernmental relations was characterized by tightly centralized finances. Revenue sharing and intergovernmental transfers, the size of which were determined nominally on the basis of expenditure norms but primarily by negotiation, continued to be used as accounting tools to balance subnational budgets. Of the few changes to the fiscal system during the early transition period, the most important was the major shift in expenditure respon- sibilities from the federal to subnational governments. The responsibilities were redistributed, in part, to bring balance to the federal budget and in an effort to discontinue the subsidies and services that the public sector could no longer afford. From the 1994 Reforms to 1997 Between 1994 and 1997, the budgetary autonomy of regional governments was strengthened, and the federal government stopped the ad hoc off-loading of expenditure responsibilities to subnational governments that characterized the earlier transition period. However, responsibilities of different government levels were still poorly defined, and the federal government continued the practice of unfunded mandates to subnational governments. In revenue assignments, sharing rates for all major taxes shared between the federal and re- gional governments were standardized across regions and remained virtually unchanged through the de- cade. To offset the discontinuation of regulated sharing rates, a formula-driven mechanism of equalization transfers was introduced in 1994. Subnational governments also experienced some increase in the level of revenue autonomy through federal legislation that authorized the introduction of new subnational taxes. In addition, regions were allowed to lower the share of the EPT anywhere from 22 to 0 percent. The reforms implemented in 1994 stopped short of introducing a transparent, stable, and rational system of intergovernmental fiscal relations. The uniform revenue-sharing arrangements were not al- ways respected by the regions, and the power granted to subnational governments to introduce new taxes led to the introduction of a myriad of small taxes that created significant economic distortions. In addition, fundamental problems with revenue assignments were not addressed. For example, money The Dynamics of Decentralization in the Russian Federation 7 from the VAT continued to be shared based on the region where the tax was paid instead of the location where the productive value was added, and revenues from the EPT continued to be shared based on the location of company headquarters. Because of restricted funding and fundamental problems with the equalizing formula, the ability of the new formula-driven transfer system to equalize expenditures across regions was limited. The new formula was designed to fill the gap between revenue collections and normative expenditure needs rather than to equalize differences in fiscal capacity and actual expenditure needs. Regional dissatisfaction with the formula caused legislators to increasingly subject the formula's results to political negotiations. The federal government never complemented the formula-driven equal- ization transfers with a system of categorical or conditional grants to pursue sectoral objectives. Instead the federal government maintained the ad hoc and nontransparent mechanism of mutual settlements. These nonbudgeted and primarily negotiated funds are determined during budget execution. Also, the federal government continued to extemporaneously issue loans to regional governments, which were often not paid back. The 1993 Constitution and the 1994 reforms strengthened the position of the federal government in relation to the regions and brought some order and stability to intergovernmental relations. Bilateral treaties, initially signed with the ethnic republics of Tatarstan, Sakha (Yakutia), and Bashkortostan, ex- tended in an orderly fashion some power to the regions. Reforms made between 1992 and 1994 success- fully influenced all dissenting regions but Chechnya to remain part of the Russian Federation. During this midphase of the transition, reform efforts concentrated almost exclusively on the fiscal relations between the federal government and the regions. The regional governments were left on their own to structure their relationships with local governments. For the most part, intraregional fiscal rela- tionships perpetuated the former Soviet approach. The Reforms of 1997 and 1998 Starting in late 1997 and continuing until the ruble devaluation and financial crisis of August 1998, the government introduced significant reforms based on the draft Tax Code, the Budget Code, the Law on Financial Foundations of Local Self-governments, and the Concept of Reform of Intergovernmental Fis- cal Relations in the Russian Federation. The Budget Code approved in 1998 addressed expenditure and budget issues, and the draft Tax Code dealt with revenue assignments. The Concept of Reform, drafted by an interagency working group led by the Ministry of Finance, demonstrated a willingness by the federal government to provide greater structure to the intraregional fiscal systems. In particular, regional governments were required to meet average minimum sharing rates for the major taxes with local gov- ernments. The Concept of Reform also called for an important change in the system of transfers, but its agenda, though ambitious, was uncoordinated and incomplete. The Role of Politics in the Decentralization Process Politics have played a pervasive role in the design and workings of intergovernmental fiscal relations in the Russian Federation. However, observers hold widely different opinions regarding the impact that politics have had on the overall performance of the system. In one view, the failure of the political system to accommodate widely diverse economic and political interests in the regions fed conflict and height- ened the political and economic instability in Russia during the transition period (Ordeshook 1995; Polishchuk 1996). In a different view, the political system effectively prevented national disintegration by facilitating practical solutions, such as awarding special fiscal status to subnations, that convinced all regions, except Chechnya, not to pursue secession from the Russian Federation (Freinkman, Titov, and Treisman 1998; Treisman 1996, 1998a, 1998b). One-half of the 450 state Duma deputies were elected from nationwide party lists. Consequently, the state Duma failed to provide adequate representation for regional interests, which were subsequently addressed, almost exclusively, by regional administrations through the Federation Council. It has also 8 Jorge Martinez-Vazquez and Jameson Boex been alleged that deputies from poorer regions have been consistently bought by national forces control- ling large purses of resources and interests (Ordeshook 1995; Polishchuk, 1996). The effects of the political process on the substance of intergovernmental fiscal relations can be di- vided into three distinct periods: (a) pre-Constitution transition (1992-93), (b) post-Constitution policies (1993-96), and (c) growing political concerns with the cost of the bilateral tax treaties (1997-98). The simultaneous timing of the political periods and the phases of intergovernmental relations underscores the importance of politics in the evolution of intergovernmental fiscal relations in Russia. The Pre-Constitution Period, 1992-93 Before the 1993 Constitution was adopted, political relations between the federal and regional govern- ments were marked by subnational political defiance of federal authorities. Several ethnic republics started political campaigns, (so-called sovereignty parades) to gain special status and subsequently signed bilat- eral agreements with the federal government. During this period, the central government unilaterally shifted expenditure responsibilities and man- dates on subnational governments without, in the view of the regions, providing adequate additional funding. Regional governments responded by threatening to withhold tax revenues collected in their territories for the federal government. While at the start of 1992 only Tatarstan, Chechnya, Sakha (Yakutia), and Bashkortostan had stopped or greatly reduced remittances to the federal government, by the end of the year, 20 regions refused to pay taxes, and by 1993 10 more regions withheld tax payments to the federal government (Wallich 1994). The collective power of the regions relative to the federal government was also evidenced by the effective rejection by the regions of the revenue assignments contained in the Law on the Basic Principles of the Tax System of December 1991. Among other measures, this law pro- vided an exclusive assignment of VAT, taxes on international trade, and certain excise taxes to the federal government and assigned the personal income tax and enterprise profit tax to the regions. Post-Constitution Policies, 1993-96 The second period started with the enactment of the new (1993) Constitution and the federal government crackdown of the conservative Supreme Soviet and the dissenting regions. The 1993 Constitution gave considerable powers to the federal executive with regard to the legislative branch and the regions but stopped short of concretely defining powers and settling past disputes between the federal government and the regions. It also gave the federal government the ability to reach bilateral agreements with a large number of regions. While confrontations between the federal government and some regions continued to test the strength of the Russian Federation during the post-Constitution period, the nature of their dis- putes changed from centrifugal tensions between levels of government toward competition between regions for special recognition and advantageous fiscal status. Increasingly significant economic and fiscal disparities between regions heightened tensions between them. While poorer regions favored higher federal taxes and more transfers, richer regions strongly op- posed both. Although economic and political conflict is not particular to Russia, the Russian case is par- ticularly troublesome because the conflicting economic and political interests can be directly identified with regional boundaries, providing fertile ground for secessionist forces (Polishchuk 1996). In response to the continued pressure from dissenting regions, a pattern of more organized and long-term bargaining developed between the federal and subnational governments. The federal authorities, empow- ered by the 1993 Constitution, entered into a series of bilateral treaties with the regions in an effort to balance regional objectives that conflicted with national interests. The regions, first based on ethnic claims and later on threats and demands, bargained for and obtained preferential fiscal treatments, including negotiated tax arrangements, free economic zones, and regional development programs. At the same time, poorer regions demanded increased regional redistribution, and other regions pressed the federal government to resume funding a wide array of public goods and services that it had earlier sought to off-load onto the regions. The Dynamics of Decentralization in the Russian Federation 9 These regional demands were met most clearly through the allocation of transfers to the regions. Lavrov (1995) and Treisman (1996, 1998a, 1998b), among other researchers, found that federal transfers were more often given in response to political pressures than to meet objectives such as equalization and fairness. The federal government systematically used transfers to appease regional opponents and dis- senters. For example, those regions that withheld support from Boris Yeltsin and his supporters in elec- tions or declared sovereignty early in the transition received more money than cooperative regions. Concerns about Bilateral Agreements, 1997-98 During the post-Constitution period, inflation was contained by a restrictive monetary policy, but the federal government resolved conflicting budget demands by borrowing domestically and later interna- tionally to finance a growing deficit. High interest rates and short-term maturities ballooned the debt service requirements and ultimately led to the ruble devaluation and debt default of August 1998. Be- sides creating mounting budgetary pressures, the granting of special fiscal treatments to many regions carried other costs. These nonmonetary costs included the resentment among regions, the compromise of the objective of a unified national market, and the creation of conditions for demands for special treat- ment by practically all regions, where the executive and the legislatures, now democratically elected, had little excuse not to pursue the narrow interests of their constituencies. However, between 1997 and 1998, political awareness was mounting at the federal level with regard to the negative effects of these bilateral agreements on federal budget policy. In addition to the large number of tax holidays and exemptions granted by the Russian Federation in the earlier years of the transition, bilateral agreements were considered a contributing factor to lackluster federal tax collections. However, recovering from the damage caused by bilateral agreements is proving much more difficult than eliminating tax holidays and exemptions. Bringing bilateral treaties in line with the general legislative framework would provide all regions with standard sharing rates for the major taxes. Whether this goal has been accomplished to any extent is unclear. The main means of enforcement for the federal authorities is the withholding of Fund for Finan- cial Support of Regions equalization money and other transfers. However, even if the government tried to restrict regional funding, a significant number of regions, including some of the most politically prob- lematic, would be largely unaffected, because they do not receive any equalization transfers and only modest amounts of mutual settlements. Dimensions of Fiscal Decentralization in the Russian Federation An in-depth analysis of the various dimensions of the Russian system of intergovernmental fiscal rela- tions follows. Each chapter contains the discussion of a particular element of fiscal decentralization in the Russian Federation. The exposition of the Russian system of intergovernmental fiscal relations in the following chapters is supported extensively by empirical analyses. Most of the data used for this purpose originate from the Ministry of Finance of the Russian Federation. As is common to transition economies, the reliability of the fiscal data is limited due to the heavy reliance in Russia on methods of government financing that are not reflected in reported budget figures, which include the pervasive use of nonmon- etary fiscal transactions. For example, governments frequently engage in swaps in which budget arrears are settled with tax arrears. In addition, the ruble was redenominated in January 1997: One new ruble was issued for every 1,000 old ones. Unless otherwise mentioned, all amounts are stated in old rubles. 2 Expenditure Responsibilities Though expenditure assignments in Russia were inherited from the Soviet Union, lower-level govern- ments have been able to exercise considerable discretion in spending decisions since the beginning of the transition period. This chapter looks at the formal assignments of expenditure responsibilities as well as the actual spending between 1992 and 1997. In addition, the performance of the system of expenditure assignments is appraised. This chapter concludes with a summary and analysis of pending policies re- garding expenditure assignments. How Have Expenditures Been Assigned? Several things should be considered when examining the assignment of expenditure responsibilities in Russia during the economic transition. First, how have the expenditure responsibilities been assigned (both legally and in practice) among the various levels of government? Second, have there been any changes in the general expenditure pattern over time? Third, what trends or patterns can be detected when looking at expenditures by main functional classification over the transition period? Legislative Assignments of Expenditure Responsibilities Neither the Constitution nor any other legislation has clearly defined the assignment of expenditure responsibilities in the Russian Federation. During the transition period, only Articles 71 and 72 of the 1993 Constitution made explicit references to expenditure responsibilities, exclusively assigning some to the federal government while other responsibilities are assigned jointly to the federal and regional levels. There are no responsibilities specifically assigned to regional governments. Yet, these constitutional pro- visions are quite vague. The 1993 Constitution provided no guidance for resolving conflicts between federal and regional governments in joint policy areas or assignments for primary responsibility for the shared responsibilities listed in Article 72. The closest attempt to make a more explicit assignment of expenditure responsibilities is contained in the Budget Code, which was approved by the State Duma in 1998 but which did not come into effect until January 1, 2000. Although the 1993 Constitution states that the federal or regional governments can delegate func- tions to local governments, it does not enumerate any direct responsibilities for local governments. Therefore, the Law on General Principles of Organization of Local Self-government in the Russian Federation of August 1995 contained a list of local government expenditure responsibilities. Article 6 of this law stated that the federal and regional authorities have no direct competence in local spending responsibilities (Igudin 1998). The assignment of expenditure responsibilities that has taken shape over the past six years bears little resemblance to those stated in the 1993 Constitution. The de facto expenditure assignment that prevails in Russia today largely complies with the general principle that each level of government should be responsible for expenditures within its geographical scope (see table 2.1). The federal gov- ernment was responsible for services affecting the entire nation, such as defense and national security. Services affecting many communities, such as tertiary hospitals, were the responsibility of regional governments, and those affecting exclusively the residents of a community, such as sanitation, were the responsibility of local governments. 12 Jorge Martinez-Vazquez and Jameson Boex Table 2.1. Assignment of Expenditure Responsibilities, 1997 Expenditure Federal Regional Local Defense 100 percent (except military Military housing None housing) Justice/internal security 100 percent None None Foreign economic relations National programs Some regions maintain their None own relations Education All university and research All technical and vocational Wages, operation construct- institute expenditures schools ions, and maintenance of all primary and secondary schools Culture and parks National museums and Regional museums None theaters Health Medical research institutes Tertiary hospitals, psychiatric Secondary hospitals hospitals, veteran hospitals, Primary health clinics diagnostic centers, and spec- Medicines ial service hospitals (cardio- logy, etc.) Roads Construction and mainten- Construction and mainten- Maintenance of rayon and ance of federal roads ance of oblast roads city roads Public transportation None Most public transportation Some transportation facilities facilities, including subway systems Fire protection None Most fire protection services Voluntary, military, and enterprise services Libraries Special libraries (for example, Special library services Most local library services the Lenin library) Police services National militia Traffic police and national Local security police militia Sanitation (garbage collection) None None Garbage collection Sewage None Infrastructure capital Most of the operational investment expenditures Public utilities None Subsidies to enterprises Subsidies to enterprises (gas, electricity, and water) Housing None Part of housing construction Part of housing construction, and subsidies subsidies, and maintenance Price subsidies Part of food and medicine None Fuels; mass transport; food subsidies (bread, milk); medicines Welfare compensation Part central government res- Part oblast government res- Managing programs funded ponsibility ponsibility by upper-level governments Environment National environmental issues Regional environmental prob- None lems, (for example, the preser- vation of forests) Public enterprises Federal investment programs, Subsidies to particular sec- Various explicit and implicit subsidized credits, subsidies to tors (agriculture), investments, subsidies and benefits particular sectors (mining, agri- grants, tax benefits culture, tax benefits) Source: Updated from Martinez-Vazquez (1994). Expenditure Responsibilities 13 There are many expenditure areas in which two or three levels of government shared responsi- bilities (table 2.1). The constitutional and legal framework introduced further ambiguity by not dif- ferentiating expenditure assignments between the responsibility to fund, regulate, and implement expenditure programs. The major changes in expenditure assignments took place in 1992 and 1993 when the federal govern- ment shifted a wide array of responsibilities down to the subnational levels of government (Martinez- Vazquez 1994; World Bank 1996). Social welfare responsibilities for consumer subsidies, cash aid for vulnerable groups, welfare programs for pensioners and the disabled, family and child allowances, and support for the homeless all became regional expenditure responsibilities. Responsibilities for other kinds of subsidies were shifted down as well, including those for agriculture, transport, and many types of investment in infrastructure. In addition, regional governments were made partially responsible for hous- ing for the military. While these massive spending reforms helped bring the federal budget deficit under control, they simply shifted the funding problem to the subnational level, crowding out expenditures on other categories, especially social services (Freinkman and Titov 1994; Titov 1997). Although later in the transition the federal government resumed a few of the responsibilities relegated to regions during the early transition, the bulk of these added responsibilities remained at the subnational level. After 1994, responsibility for many law enforcement activities was transferred to regional gov- ernments, and Russian Federation assignments of expenditure responsibilities became more stable (Titov 1997). Yet, in practice expenditure assignments in the Russian Federation have continued to evolve, as many shared responsibilities have been adjusted in the annual budgets (Igudin 1998). This has been especially true for the division of responsibilities between the regional and local govern- ments where many functions are shared between these two levels of government and the actual division of subnational responsibilities differs from region to region. The issuance of expenditure mandates by the federal government has also contributed to the redefinition of shared responsibili- ties after 1994. Special arrangements introduced in bilateral treaties between federal and regional governments contributed to nonuniform expenditure assignments. Many treaty agreements allowed the regional government to take over some or all responsibilities of the federal government, includ- ing expenditures in overall administration, education, and construction, in exchange for a larger portion of shared revenues. Changes in Expenditure Patterns After 1992, the amount of economic resources available to the public sector in real terms declined substantially because of the public sector's disengagement from many aspects of economic activity, the implementation of fiscal policies, and a drop in gross domestic product (GDP) of more than 40 percent. When measured as a percentage of GDP, total government expenditures, excluding extrabudgetary funds, dropped from 52.0 percent of GDP in 1992 to 36.4 percent in 1997 (table 2.2). Table 2.2. Distribution of Expenditures between Government Levels, 1992-97 (percentage of GDP) Level of government 1992 1993 1994 1995 1996 1997 Federal budget 39.7 23.8 23.4 18.5 21.0 19.0 Consolidated region budget 12.3 16.1 18.2 15.2 16.0 17.4 Regional budgets 6.0 8.0 9.7 7.8 8.0 Local budgets 6.3 8.1 8.5 7.3 8.0 Total 52.0 39.8 41.6 33.6 37.8 36.4 - Not available. Note: Transfers to other levels of government may result in double counting. Eighty-eight regions were included in the study. Source: Freinkman, Titov, and Treisman (1998); Ministry of Finance data. 14 forge Martinez-Vazquez and Jameson Boex Subnational government expenditures grew the most between 1992 and 1993, peaked in 1994 at 18.2 percent of GDP, and by 1997 were 17.4 percent of government distributions. During the transition years the subnational share in total government expenditures increased steadily and significantly from 23.7 percent in 1992 to 47.9 percent in 1997 (table 2.3), with the largest change coinciding with the initial expenditure decentralization of 1992-93. No precise estimates are available of the additional fiscal burdens on subnational governments that followed from the 1992 divestiture of social responsibilities, including state-owned enterprise expenditures for facilities such as housing, kin- dergartens, and health clinics (Martinez-Vazquez 1993; World Bank, 1996). However, enterprises reported a drop in expenditures on social activities from 4.1 percent of GDP in 1992 to 3.3 percent of GDP in 1994 (Martinez-Vazquez 1993). The impact on subnational governments varied significantly across regions because each experienced a different level of exposure to the problem; those regions that had state-owned enterprises under the Soviet Union were most affected. Taking over the responsibility for housing im- pacted local budgets significantly because of steep increases in energy prices over time and regional governments' refusal to implement full-cost recovery policies for the cost of housing and public utilities. In addition, subnational governments periodically took over federal government expenditures such as arrears in federal wages for university personnel, the military, the State Tax Service (STS), and law enforcement agencies when nonpayment led to social tensions (Freinkman, Titov, and Treisman 1998). Subnational budgets were compensated with additional receipts from privatization of enterprises, par- tial compensation from the federal budget through mutual settlements, and increased tax collections. For example, a local turnover tax, at a rate of 1.5 percent, was introduced in 1993 to help regions fund the additional expenditures caused by decentralization. The overall pattern of expenditure decentralization is depicted in table 2.3. One-half of all expendi- tures were included in the federal budget, and the other half was spent through the subnational govern- ments. Roughly one-half of subnational expenditures was the responsibility of local governments. How- ever, actual decentralization spending cannot be inferred from this observed pattern of expenditure allocations. In many cases substantive decisions were made by higher-level government officials. For example, federal authorities played significant roles in setting educational policies, such as determining curricula and selecting textbooks. The degree of true discretion at the regional and local levels is a com- plex issue that requires close study. Expenditure Patterns by Functional Classification The shares at each level of government for major expenditure categories are shown in table 2.4. Although the federal government paid 100 percent of national defense expenses, the costs of all other functions were shared by different levels of government. Subnational governments paid more than four-fifths of all expenditures for education and health in 1996 and 1997, with the local government picking up most of the tab. Health services spending remained practically the same during the entire 1992-97 period, though regional governments increased their role in this sector. Table 2.3. Distribution of Expenditures between Government Levels, 1992-97 (percentage of total) Level of government 1992 1993 1994 1995 1996 1997 Federal budget 76.3 59.7 56.2 54.9 57.7 52.1 Consolidated region budget 23.7 40.3 43.8 45.1 42.3 47.9 Regional budgets 11.5 20.2 23.4 23.2 21.2 - Local budgets 12.1 20.2 20.3 21.8 21.2 - Total 100.0 100.0 100.0 100.0 100.0 100.0 - Not available. Note: Transfers to other levels of government may result in double counting. Eighty-eight regions were included in the study. Source: Freinkman, Titov, and Treisman (1998); Ministry of Finance data. Expenditure Responsibilities 15 Table 2.4. Distribution of Expenditures by Function between Government Levels, 1992-97 (percent) Level of government 1992 1993 1994 1995 1996 1997 Administration and justice 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 80.5 79.6 81.4 63.1 63.7 62.0 Consolidated subnational budget 19.5 20.4 18.6 36.9 36.3 38.0 Regional budgets 5.5 6.1 6.2 - 19.0 - Local budgets 14.0 14.3 12.4 - 17.3 - Defense 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 100.0 100.0 100.0 100.0 100.0 100.0 Consolidated subnational budget 0.0 0.0 0.0 0.0 0.0 0.0 Regional budgets 0.0 0.0 0.0 0.0 0.0 0.0 Local budgets 0.0 0.0 0.0 0.0 0.0 0.0 National economy 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 80.8 48.6 30.4 26.3 27.5 21.6 Consolidated subnational budget 19.2 51.4 69.6 73.7 72.5 78.4 Regional budgets 8.7 24.6 34.8 35.4 32.4 - Local budgets 10.5 26.8 34.9 38.2 40.1 - Education 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 33.8 19.5 19.9 15.9 14.5 13.2 Consolidated subnational budget 66.2 80.5 80.1 84.1 85.5 86.8 Regional budgets 14.5 18.9 19.1 19.7 18.0 Local budgets 51.8 61.6 61.0 64.4 67.5 - Culture and mass media 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 51.2 34.8 39.6 31.6 34.7 16.5 Consolidated subnational budget 48.8 65.2 60.4 68.4 65.3 83.5 Regional budgets 21.2 30.2 28.9 30.8 29.3 Local budgets 27.6 35.0 31.5 37.6 36.0 Health and physical education 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 11.3 10.5 11.8 9.9 10.2 11.8 Consolidated subnational budget 88.7 89.5 88.2 90.1 89.8 88.2 Regional budgets 28.5 30.2 33.4 33.4 33.1 Local budgets 60.2 59.3 54.8 56.7 56.8 Social protection 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 71.8 52.0 50.3 18.5 31.2 41.3 Consolidated subnational budget 28.2 48.0 49.7 81.5 68.8 58.7 Regional budgets 20.4 34.9 34.9 31.1 26.8 Local budgets 7.8 13.2 14.8 50.3 42.0 - Not available. Note: Transfers to other levels of government may result in double counting. Eighty-eight regions were included in the study Source: Freinkman, Titov, and Treisman (1998); Ministry of Finance data. The large shift in expenditure assignments between 1992 and 1993 reflects the evolution of re- gional government shares in the national economy category, which consists mainly of subsidies for housing, agriculture and industry, and social protection. While the federal government represented 80.8 percent of all expenditures in the national economy in 1992, reflecting a large involvement in pure market activities, by 1997 its share was reduced to 21.6 percent. Regional expenditures in the national economy category were concentrated in subsidies to industry and transportation while local governments provided subsidies for housing and communal services. Decentralization of social protection funding was less dramatic than for the national economy items, but still pronounced. The consolidated regional share of total expenditures on social protection, 16 Jorge Martinez-Vazquez and Jameson Boex excluding pensions, payments for disability, or the federal unemployment fund, reached a high of 81.5 percent in 1995 from 28.2 percent in 1992; in 1997, regions were taking responsibility for only 58.7 percent. On average the composition of regional consolidated budgets remained stable between 1994 and 1997 (table 2.5). The highest deviation among the yearly average shares of regional health budgets was only 0.9 percent. For education, the average share in the consolidated regional budget increased from 21.7 percent in 1992 to 24.4 percent in 1997. Education and health expenditures did not vary much across the regions, as shown by the small coefficients of variation in table 2.5. (The coefficient of deviation is defined as a variable's standard deviation divided by its mean.) On average, the highest share of regional consoli- dated budgets was delegated to national economy expenditures, and although the variation across re- gions has tended to increase somewhat over time, the variation has remained relatively low. The most important component of national economy expenditures has been for housing and public utility subsi- dies. On average, during 1994-97 regions spent more on subsidies for housing and public utilities than on health services. They funded education to a slightly greater extent than housing and utilities. The importance of housing subsidies in subnational budgets remained relatively constant over time. Subsi- dies to agriculture decreased from 7.1 percent in 1994 to 4.9 percent in 1997. The higher coefficient of variation associated with subnational budgets for agriculture reflects greater differences in spending among regions. The importance of social protection expenditures in regional consolidated budgets did not change much between 1994 and 1997; it remained a relatively small part of nonfederal budgets. A general concem in the Russian Federation, and in the international community, has been the impact of unfunded expenditure mandates and overall fiscal austerity on the ability of subnational governments to provide basic services in health, education, and social support. The Russian government declared that the social sectors-education, health, and social protection-would remain immune from budgetary re- allocations in 1992 and 1993, thereby sparing them from the drastic cuts suffered by other sectors, such as defense. In addition, the Law on Budgetary Process gave many expenditure items in the social sectors protected status under a provision called budget "sequestering," in which expenditure budgets were adjusted during the fiscal year to account for inadequate revenue collections. Yet, significant cuts were made in social services in the early years of the transition. Approximately one-third of all kindergartens in Russia closed down from 1992 to 1994. Existing health and education infrastructures were not main- tained, and practically no construction was started. The funds dedicated to supplies and operations, other than wages, were cut to a minimum or eliminated. Many schools around the country reported having no books, and hospitals lacked basic supplies and medicines (Martinez-Vazquez 1994). However, after the difficulties of the initial years, funding for the social sectors at the subnational level did not plummet. Real per capita expenditures peaked in 1994, as did education and health spending per person in real terms (table 2.6). Though real regional funding in education and health fell between 1994 and 1997, per capita real expenditures on social protection grew slightly over the period. Even though the composition of expenditures in the consolidated regional budgets did not change much after 1993, the relative shares of regional and local governments varied substantially across regions (Lavrov and others 1996; Freinkman, Titov, and Treisman 1998). The degree of centralization of regional finances was markedly higher in those regions most dependent on federal transfers. In addition, regions that had allocated greater expenditure responsiblities to local governments were more likely to have adopted market reforms. The imposition of expenditure requirements on subnational governments without adequate fund- ing from the federal authorities, unfunded mandates, remains one of the most vexing problems in intergovernmental budgetary relations in Russia at the present time. Two specific problems are caused by unfunded mandates: subnational budget autonomy is restricted, and regions are financially strapped by federal strictures. The history of unfunded mandates can be traced back to 1992 when, early in the transition process, the federal government shifted expenditure responsibilities for many social welfare programs to subnational governments. Among others, the reassignment of expenditure responsibilities included cash subsidies for vulnerable groups, welfare payments for the disabled and pensioners, family and child allowances, support for the homeless, and general consumer subsidies for housing and public utilities. Table 2.5. Average Regional Expenditures by Category, 1994-97 (percentage of total regional expenditures) National economy sector Housing Health and National and physical Social Year economy utilities Agriculture Transportation Education education protection Other 1994 36.0 19.9 7.1 2.9 21.7 15.8 6.6 19.0 (0.17) (0.35) (0.54) (0.57) (0.17) (0.17) (0.32) (0.25) 1995 34.8 20.4 6.4 3.9 22.6 16.1 8.1 18.3 (0.25) (0.34) (0.60) (0.56) (0.22) (0.19) (0.30) (0.49) 1996 41.2 21.9 5.4 4.1 24.0 15.7 8.9 10.3 (0.24) (0.38) (0.66) (1.20) (0.23) (0.20) (0.34) (0.18) 1997 32.2 20.9 4.9 3.3 24.4 15.2 7.9 20.4 (0.25) (0.34) (0.62) (0.65) (0.21) (0.19) (0.36) (0.29) Note: The figures in parentheses are the coefficient of the variation. Eighty-eight regions were included in the study. Source: Authors' calculations based on data from the Ministry of Finance. Table 2.6. Composition of Expenditures for Subnational Governments in Real Terms, 1992-97 In real terms: per capita 1995 rubles In real terms: index (1994 = 100.0) Expenditures 1992 1993 1994 1995 1996 1997 1992 1993 1994 1995 1996 1997 Total expenditures 1,671.6 2,017.2 2,023.7 1,625.8 1,590.3 1,761.3 82.6 99.7 100.0 80.3 78.6 87.0 National economy 559.4 662.7 821.7 664.4 735.5 648.9 68.1 80.7 100.0 80.9 89.5 79.0 Housing - - 529.9 407.1 409.5 414.9 - - 100.0 76.8 77.3 78.3 Agriculture - - 120.1 97.0 78.5 76.5 - - 100.0 80.7 65.3 63.7 Transportation - - 57.6 84.7 74.2 77.2 - - 100.0 147.2 128.9 134.2 Education 261.2 313.4 401.2 322.0 335.8 372.5 65.1 78.1 100.0 80.3 83.7 92.8 Health and physical education 241.6 266.8 320.1 250.2 242.0 252.7 75.5 83.3 100.0 78.2 75.6 78.9 Social protection 33.5 34.0 120.3 114.8 125.1 128.3 27.8 28.2 100.0 95.4 104.1 106.7 - Not available. Note: Eighty-eight regions were included in the study Source: Authors' calculations based on data from the Ministry of Finance. Expenditure Responsibilities 19 While regional governments may have been able to deliver these services more efficiently than the fed- eral government, the redistributive nature of social welfare programs justifies their full funding by the federal budget. Yet, the federal government justified this delegation of responsibilities by arguing that the general funding scheme for subnational governments, revenue sharing and transfers, took into ac- count the additional costs of providing these newly assigned services. The subnational governments disagreed, and continue to dissent, with this assessment and claimed that the federal government was imposing unfunded mandates on the regions to solve its own budget problems. As the 1990s came to an end, awareness and discussions of unfunded mandates increased in the Russian Federation because of increasing fiscal austerity at all levels of government and the dimension of the problems associated with the mandates. In response, The Law on the Financial Foundations of Local Self-governments of September 1997 prohibited regional governments from issuing unfunded mandates to local governments. The 1998 Budget Code, which will come into effect in 2000, will prohibit the use of unfunded mandates at every government level. Despite the formal bans on unfunded mandates, general skepticism abounds about the effectiveness of legislation to end the practice. How Has the System Perfonned? The significance of fiscal disparities across regions grew markedly during the transition years. The in- creasing differences in regional expenditure levels were caused in large part by growing regional eco- nomic inequalities. Table 2.7 shows the evolution of gross regional product (GRP) in per capita terms from 1992 to 1996, the latest year for which data are available. GRP was computed by Goskomstat, the government statistical agency, for nonautonomous regions of the Russian Federation. Data for autonomous regions were imputed based on the assumption that they made the same tax effort as the region in which they are contained. Because some of the autonomous regions had relatively high per capita revenue collections, any analysis using tax collections could be biased if the attributed data are a poor reflection of actual GRP levels. Fiscal Disparities Horizontal imbalances or fiscal disparities among the regions grew sharply over the transition pe- riod. As shown in table 2.8, the coefficient of variation for per capita expenditures across the regions increased substantially between 1992 and 1997. These figures do not reflect variations across indi- vidual regions over time. For example, Kostroma's expenditures per capita increased by 90 percent in real terms between 1992 and 1996, while those in the Tuva Republic decreased by 56 percent dur- ing the same period (Freinkman, Titov, and Treisman 1998). The fundamental driving force for these fiscal disparities is the fast growing economic inequalities across regions. Economic disparities among Russian regions were significant at the time of independence. An Table 2.7. Regional Economic Disparities: Per Capita GRP, Selected Years, 1992-96 (thousands of rubles) Coefficient Year Mean of variation Minimum Maximum 1992 124.7 0.87 24.6 786.0 1994 3,712.3 0.75 618.1 20,909.6 1995 9,487.9 0.84 1,881.9 59,004.5 1996 13,554.6 1.04 2,628.0 108,443.5 Note: Eighty-eight regions were included in the study Source: Authors' calculations based on data from Goskomstat. 20 Jorge Martinez-Vazquez and Jameson Boex Table 2.8. Measures of Horizontal Fiscal Imbalance: Per Capita Consolidated Regional Expenditures, 1992-97 (thousands of rubles) Coefficient Year Mean of variation Minimum Maximum 1992 18.9 0.83 7.2 85.1 1993 219.7 0.78 100.6 1,189.2 1994 959.7 1.19 359.8 8,000.7 1995 1,918.2 1.01 720.2 13,004.3 1996 2,861.5 1.05 1,050.3 16,521.1 1997 3,762.6 1.17 1,336.7 30,543.5 Note: Eighty-eight countries were included in the study. Source: Authors' calculations based on data from tie Ministry of Finance. additional factor in the increase of regional expenditure disparities could be increases in price variations across regions. Price levels should be expected to have become more dissimilar over time as federal subsi- dies for transportation of goods to the northern regions declined. High interregional price differences were also sustained by a lack of interregional market integration (Berkowitz and DeJong 1998). During the years of economic transition, the coefficient of variation for GRP grew from 0.87 in 1992 to 1.04 in 1996, and the difference between the poorest and richest region jumped from 31 times in 1992 to 41 times in 1996. After adjusting for the cost of living, the difference in product per capita was fourfold between the average product per capita in the lowest and highest quintiles of regional distribution in 1994 (Working Center for Economic Reform 1997). The increase in regional inequality can largely be explained by the growing concentration of economic activity in just a small group of regions (Working Center for Economic Reform 1997); the concentration of economic wealth was more pronounced for certain types of enterprises. During the transition, most oil and gas production was concentrated in two autonomous okrugs, Khanty-Mansi and Yamalo-Nenets. In 1995, approximately three-fourths of all metal was produced in 10 regions. Moscow received 20 percent of all income, but it represented only 7 percent of the population. With regard to income, distant second and third richest regions were the Tyumen oblast with 4.5 percent of all income and 1 percent of the population and St. Petersburg with 4.3 percent of all income and 3 percent of the population. The increase in economic disparities has become more acute because many of the industrial-location decisions made by the former Soviet Union were nonsustainable in a competitive market environment and because the military-industrial complex collapsed (Polishchuk 1996). Although in 1998, Russia's national GDP was approximately one-half of its early transition value, regions performed at widely dif- fering levels, reflecting structural changes in the economy. Regions that had natural resources, possessed metallurgic and chemical industries, or developed a service industry based on international trade and finance were relatively prosperous. Foreign direct investment was concentrated in a small number of regions, especially in Moscow. Agricultural areas and those with military industry and other manufac- turing plants did not thrive in the post-Soviet economy (Sutherland and Hanson 1996). The Fiscal System's Effect on Regional Budget Disparities The lack of market development and economic growth outside a few regions combined with labor immo- bility required the public sector to take an enhanced role in addressing regional fiscal disparities. In general, the fiscal system, comprising the assignment of revenues and transfers of all kinds, mostly helped to reduce budgetary disparities variability among regions. Yet, the equalizing effect of the fiscal system as a whole has not been enough to keep up with the increasing disparities in revenue collections. Expenditure Responsibilities 21 The impact of the fiscal system on regional budget disparities is illustrated in table 2.9. The table's first column reflects that variation of revenue collection, including the federal share of joint taxes, increased over time from 0.68 in 1993 to 1.52 in 1997. At the second stage of the allocation process, regional shared revenues (excluding the federal share), own source revenues, and other regional revenues (before transfers) are considered. This second stage shows a substantial decrease in the coefficient of variation compared to the first stage. Thus the system of revenue assignments has resulted in a more equal distribution of re- sources across the regions. All transfers, including those designed to equalize regional disparities, con- tributed to unequal distribution of fiscal resources across regions during the transition. However, the coefficient of variation of per capita expenditures was substantially lower than that for the collections of all revenues during 1995-97. In the early years of the transition, regional expenditure variation was higher than variation of regional revenue collection. Regression analysis was used to explore the role of conventional determinants of variations in per capita expenditures across regions.' Independent variables included fiscal capacity, the ability to pay for government services as evidenced by GRP; the need for public services as measured by socioeconomic characteristics as shown in table 2.10; and a less conventional determinant, the level of tax effort exerted by a region measured as the ratio of tax collections and fiscal capacity. Unlike in Western market-based economies, subnational governments have limited statutory discretion in tax collection. They can levy few taxes, change few rates, and grant few tax advantages. Though Russia uses a centralized tax admin- istration apparatus, the actual collection effort of the regional branches of the STS and the Tax Police varies significantly across regions and local governments. Tax administrators often show de facto alle- giance to subnational government interests that do not always coincide with those of the federal government. In addition, subnational government fiscal policy decisions and the extensive use of barter and underground economies cause further regional vexations in the ability to collect taxes. Table 2.9. Equalization Level at Different Stages of Revenue Assignment, 1992-97 (coefficient of variation of cumulative per capita allocation) Own, shared, and other Revenues after Total revenue regional Revenues after all transfers and Regional Year collections' revenues b FFSR c settlements d expenditures 1992 - 0.88 0.88 0.77 0.83 1993 0.68 0.67 0.67 0.79 0.78 1994 0.85 0.89 0.92 1.14 1.19 1995 1.12 0.86 0.89 0.94 1.01 1996 1.33 1.11 1.03 1.00 1.05 1997 1.52 1.39 1.22 1.18 1.17 - Not available. a. Defined as all shared and own source tax collections, including the federal share of shared revenues. b. Excluding transfers and other source of financing. c. The Fund for Financial Support of Regions was introduced in 1994. In earlier years equalization was achieved through variable sharing rates and other equalizing transfers. d. Excluding budget loans. Note: Eighty-eight regions were included in the study Source: Authors' calculations based on data from the Ministry of Finance. 1. The dependent variables in the regression analyses presented in tables 2.10-2.15 are per capita consolidated regional and local expenditure levels. Because price levels differ substantially among the regions, a set of similar analyses was estimated using per capita subnational expenditures, adjusted by cost of living, as the dependent vari- ables. The results generated by this alternative set of analyses were identical to those presented in tables 2.10-2.15. 22 Jorge Martinez-Vazquez and Jameson Boex Table 2.10. Regional Socioeconomic and Demographic Variables: Descriptive Statistics Coefficient Independent variable Mean of variation Minimum Maximum Cost of living index 113.3 0.44 68.0 357.0 Percentage of the population younger than working age 24.0 0.16 18.2 35.5 Percentage of the population older than working age 18.5 0.29 4.4 27.1 Percentage of the population in poverty (1995) 30.6 0.38 16.1 73.2 Note: Data are for 1996 unless noted otherwise. Source: Authors' calculations based on data from Goskomstat. In other countries, regions with greater fiscal capacity or greater expenditure needs typically have higher spending levels. Also, regions with an intrinsic higher demand for public services, as revealed by the level of collection effort, have higher regional expenditure levels. Because expenditures and revenues are used to define tax effort, one could argue that tax effort should not be included in the regression analysis. However, the results from the analysis remained essentially unchanged when tax effort was excluded from the regression calculations. Analysis using measures of fiscal capacity, service need, and tax collection effort explained more than 90 percent of the variation in per capita subnational expenditures across regions from 1994 to 1997 (table 2.11). Per capita expenditures were higher in those regions with higher per capita GRP for all the years. The regression coefficients for per capita GRP were positive, significant, and larger for subsequent years, an indication of the important role played by variations in the regional economic base or fiscal capacity in explaining regional public expenditures per capita. Per capita regional expenditures were also higher in regions with higher expenditure needs, as measured by the cost of living and the per- centage of the population that is either under or over the working age. These groups have greater need for government services, particularly in the areas of education and health care. The regression coefficients for these variables were positive and almost always statistically significant. A fourth vari- able considered for expenditure needs, the percentage of the regional population that lives in poverty, did not have any significant effect on the variation of expenditures per capita. Either no relationship existed between poverty levels and regional expenditures, or poverty level was not measured cor- rectly. Those regions that received greater transfers or exerted a higher tax effort also experienced higher per capita expenditures. Table 2.11. Determinants of Per Capita Expenditures, 1994-97 Independent variable 1994 1995 1996 1997 Intercept -2,516.2 -2,115.7 -4,607.7 -13,088.3 Per capita GRP (thousands) 117.9 81.3 131.0 215.9 Cost of living index 6.52 17.8 13.2 28.1 Percentage of the population younger than working age 37.5 11.6 53.7 202.4 Percentage of the population older than working age 29.0 -7.0 64.1 201.4 Percentage of the population in poverty -1.1 -5.0 -9.8 0.0 Per capita transfers (rubles) 0.84 0.80 1.86 0.97 Tax effort 30.2 63.4 46.8 44.8 R2 0.98 0.91 0.97 0.95 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Eighty-eight regions were included in the study. Source: Authors' calculations based on data from the Ministry of Finance. Expenditure Responsibilities 23 Composition of Subnational Expenditures Subnational government expenditures differed not only in level but also in composition. Regression analy- sis was again used to analyze the determinants of expenditure variations across the most important catego- ries: national economy spending, education, and health care (tables 2.12-2.15). Analyses were generated to address two questions. Are expenditure levels for certain government functions influenced by the type of funding received? What is the role of spending needs in expenditure variations across regions? Several consistent patterns arise in the determinants of education, health care, and national economy spending, which primarily consists of suboptimal, untargeted subsidies for housing and public utilities that the federal authorities unsuccessfully tried to eliminate. The regressions explaining the variation across regions for expenditures on the national economy in table 2.12 show that from 1994 through 1997 national economy spending was higher in regions with higher overall revenues, but also in regions that received more transfers from the federal government. These results coincide with those found in previ- ous research. Titov (1997) and Freinkman and Haney (1997) tested different hypotheses on the determi- nants of the propensity by subnational governments to spend on subsidies (using expenditures on "na- tional economy" as a proxy for subsidies). Titov (1997) found that national economy expenditures were closely related to regional GDP in 1992 but not in 1994. Freinkman and Haney (1997) discovered that the propensity to spend more in subsidies decreased in regions that are economically depressed and in- creased in fiscally wealthier regions and in those that received more in federal transfers. Also significant is that those regions spending more on subsidies were also borrowing more and incurring higher budget deficits (see "other sources of finance" in table 2.12). Expenditure needs, mea- sured by the population under or over working age, cost of living, and the poverty rate, did not affect regions' subsidy expenditures, except in 1996 and 1997 when the regression coefficients for cost of living and population under working age were negative and statistically significant. The regression results explaining the variation of expenditures on subsidies to housing and public utilities, as a subcategory of national economy expenditures, showed the same basic patterns (see table 2.13). In addition, analyses revealed that regions with higher education expenditure needs, as measured by the percentage of the population under working age, spent less on housing subsidies. The regression results explaining the variation of expenditures on education (table 2.14) and health (table 2.15) also show some interesting pattems. Regions with higher revenues are able to spend more on education and health, but for each additional revenue ruble, regions spent less than for national economy expenditures (table 2.12). For the most part equalization transfers from the Fund for Financial Support of Table 2.12. Determinants of Per Capita Expenditures on the National Economy, 1994-97 Independent variable 1994 1995 1996 1997 Intercept 121.03 -612.00 798.41 2,129.58 Regional collections 0.44 0.45 0.62 0.52 FFSR transfers 0.63 0.08 0.58 0.40 Other transfers 0.26 0.66 1.18 0.61 Other sources of finance 0.77 0.19 0.36 0.52 Cost of living index -0.31 1.37 -5.34 -8.61 Percentage of the population younger than working age -4.18 4.35 -28.57 -56.40 Percentage of the population older than working age -0.33 16.27 -8.43 -27.22 Percentage of the population in poverty -0.32 -0.05 4.26 4.50 R 2 0.98 0.93 0.98 0.97 Note: Italics indicate statistical significance at the 5 percent level. Other transfers include mutual settlements, subventions, and other minor transfers, but exclude budget loans. Other sources of finance is defined as the difference between expenditures and all revenues and transfers. Eighty-eight regions were included in the study. Source: Authors' calculations based on data from the Ministry of Finance. 24 Jorge Martinez-Vazquez and Jameson Boex Table 2.13. Determinants of Per Capita Expenditures on Housing and Public Utilities, 1994-97 Independent variable 1994 1995 1996 1997 Intercept 447.14 550.64 2,495.37 4,985.91 Regional collections 0.31 0.23 0.23 0.14 FFSR transfers 0.60 0.32 0.65 0.39 Other transfers 0.22 0.20 0.98 -0.11 Other sources of finance 0.51 0.06 0.16 0.36 Cost of living index -1.11 0.24 -4.98 -5.64 Percentage of the population younger than working age -12.88 -23.32 -57.79 -100.96 Percentage of the population older than working age -5.35 -4.44 -38.97 -90.99 Percentage of the population in poverty 0.16 0.86 -2.68 -5.69 R2 0.94 0.86 0.87 0.86 Note: Italics indicates statistical significance at the 5 percent level. Other transfers include mutual settlements, subventions, and other minor transfers, but exclude budget loans. Other sources of finance is defined as the difference between expenditures and all revenues and transfers. Eighty-eight regions were included in the study. Source: Authors' calculations based on data from the Ministry of Finance. Regions also led to higher expenditures in education and health, but other forms of transfers did not affect education or health spending. This result coincides with the interpretation that mutual settlements address mostly issues of national economy expenditures. Several other patterns were revealing. In 1996 and 1997 regional governments attracted other sources of financing (such as budget loans) to increase their expenditures on education and health, and those regions with higher poverty levels spent less on education, at least in 1995 and 1997. In addition, the regression analyses suggest that no systematic relationship exists between spending on education and health and the number of final service consumers, children and seniors, respectively. Pending Policy Issues in Expenditure Assignments A govemment strategy for the reform of intergovernmental fiscal relations emerged with the Concept of Reform of Intergovernmental Fiscal Relations drafted by the Working Group on Intergovernmental Fis- cal Relations in 1998. Earlier drafts focused almost exclusively on reforming the system of intergovernmental transfers and the need to reform fiscal management at the subnational level. The fol- lowing discussion complements the analyses in the 1998 Concept of Reform document by highlighting areas in expenditure assignments that will require the attention of policymakers in the immediate future. Better Delineation between the Public and Private Sectors Although the extent of government involvement in strictly market activities declined dramatically from former Soviet Union-planned economy standards, regional and local governments continued to partake in the private economy. In 1997, subnational governments, on average, still spent 32.2 percent of their budgets on private-consumption or producer subsidies (see table 2.5). There also has been a tendency among subnational governments to become entrepreneurs to stimulate economic activity and employment. The 1993 Constitutional mandate prohibiting the federal government from meddling with regional government budget decisions limited federal authority to force subnational expenditure rationalization in housing and public utility expenditures. According to the 1998 Concept on Reform, the most effective approach to reduce subnational spending on subsidies will likely be a program of conditional grants that requires participating subnational governments to reduce or eliminate these types of expenditures. Expenditure Responsibilities 25 Table 2.14. Determinants of Per Capita Expenditures on Education, 1994-97 Independent variable 1994 1995 1996 1997 Intercept -318.08 -510.39 36.14 -723.65 Regional collections 0.15 0.13 0.10 0.08 FFSR transfers 0.14 0.26 0.17 0.15 Other transfers 0.17 -0.05 -0.11 -0.07 Other sources of finance 0.06 -0.05 0.30 0.16 Cost of living index 0.93 3.04 3.53 7.37 Percentage of the population younger than working age 8.47 12.76 3.97 23.94 Percentage of the population older than working age 4.20 5.18 -6.64 0.43 Percentage of the population in poverty -0.54 -2.17 -3.19 -6.23 R 2 0.99 0.96 0.94 0.94 Note: Italics indicate statistical significance at the 5 percent level. Other transfers include mutual settlements, subventions, and other minor transfers but exclude budget loans. Other sources of finance is defined as the difference between expenditures and all revenues and transfers. Eighty-eight regions were included in the study. Source: Authors' calculations based on data from the Ministry of Finance. Clearer Expenditure Responsibilities A stable and transparent system of intergovernmental fiscal relations will require a well-defined institu- tional framework to resolve conflicts in the assignment of expenditure responsibilities among the differ- ent government levels. The assignment of expenditure responsibilities in the Budget Code was more explicit than those outlined in the 1993 Constitution (table 2.16), but more clarity is needed. Chapter 11 of the Budget Code identified exclusive responsibilities for each government level and outlined those to be jointly provided. However, important responsibilities were not clearly assigned. For example, only edu- cation and health institutions owned by local governments were mentioned as local responsibilities. Con- sequently, local governments cannot distinguish which school levels-primary, secondary, and so forth- or which educational products and services they are obligated to fund. For example, regional and local officials in the Leningrad oblast disagreed about the assignment of responsibility for teacher salaries. As a result, each year teachers in this region faced wage arrears (Bahl and others 1998). In fact, the assignments in the Budget Code are much less clear than the reality that has taken shape over the past years. The expenditure assignments in the Budget Code appear to be another exercise in Table 2.15. Determinants of Per Capita Expenditures on Public Health and Physical Education, 1994-97 Independent variable 1994 1995 1996 1997 Intercept 181.46 119.41 534.25 534.49 Regional collections 0.10 0.11 0.09 0.06 FFSR transfers 0.08 0.18 0.16 -0.16 Other transfers 0.14 0.06 -0.05 -0.24 Other sources of finance -0.13 -0.01 0.16 0.07 Cost of living index -0.08 0.54 0.37 2.24 Percentage of the population younger than working age -3.61 -2.34 -9.97 -6.34 Percentage of the population older than working age -2.68 -2.10 -9.37 -13.84 Percentage of the population in poverty -0.14 -0.26 -0.79 -3.10 R 2 0.97 0.95 0.94 0.90 Note: Italics indicate statistical significance at the 5 percent level. Other transfers include mutual settlements, subventions, and other minor transfers but exclude budget loans. Other sources of finance is defined as the difference between expenditures and all revenues and transfers. Eighty-eight regions were included in the study. Source: Authors' calculations based on data from the Ministry of Finance. Table 2.16. Expenditure Assignment in the Budget Code, 1998 A. Expendituresfinanced exclusivelyfrom thefederal budget B. Jointlyfinanced expenditures (federal, regional, and/or local) Support of the activity of the President of the Russian Federation, Federal Assembly of the State support of industries (except for nuclear power generation), construction and Russian Federation, Accounting Chamber of the Russian Federation, Central Election construction industry, agriculture, road and river transport, communications, Commission of the Russian Federation, federal executive authorities and their regional roads and metro systems branches, other expenditures on general government administration in accordance Law enforcement activities with the list established in the process of approval of the federal law on the federal Fire safety activities budget for the next fiscal year Research, design, and surveys in support of scientific and technological progress Functioning of the federal judiciary system Social protection of the population International activities in the interests of the Federation: financial support of the implem- Environmental protection, protection and reproduction of natural resources, entation of interstate agreements and agreements with international financial institut- hydrometeorological activities ions; international cultural, scientific, and information cooperation of federal Prevention and liquidation caused by emergency situations and natural disasters of executive authorities; quotas of the Russian Federation in international organizations; interregional significance other expenditures related to international cooperation determined by federal law on Development of the market infrastructure the federal budget for the next fiscal year Development of federal and interethnic relations Expenditures on national defense and national security, conversion of defense industries Operation of Election Commissions of regions pursuant to the federal legislation of the Fundamental research and promotion of scientific and technological progress Russian Federation State support of railway, air, and sea transport Support of mass media State support of nuclear power generation Financial assistance to other budgets Liquidation of consequences of emergencies and natural disasters of federal significance Other expenditures to be jointly financed by the Russian Federation, regions Space research and use and municipalities Maintenance of federally owned institutions or institutions managed by public authorities of the Russian Federation Formation of federal property Service and repayment of sovereign debt of the Russian Federation Compensation of expenditures incurred by state extrabudgetary funds in relation to state pensions, allowances, and other social benefits to be financed from the federal budget pursuant to the legislation of the Russian Federation Replenishment of state reserves of precious metals, gems, and state material reserves Elections and referendums in the Russian Federation Federal investment program Implementation of decisions taken by federal public authorities, which increased budgetary expenditures or reduced budgetary revenues of budgets of other levels Execution of individual public responsibilities delegated to other levels of power Financial support of subjects of the Russian Federation Official statistics Miscellaneous (table continues onfollowing page) Table 2.16 continued C. Expendituresfinanced exclusively from budgets of regions of the Russian Federation D. Expendituresfinanced exclusivelyfrom local budgets Operation of legislative (representative) and executive authorities of subjects of the Operation of local self-government authorities Russian Federation Formation and management of municipal property Servicing and repayment of the sovereign debt of regions Organization, maintenance, and development of education, health care, social and Elections and referenda among subjects of the Russian Federation physical culture institutions and sports facilities, mass media and other institutions Implementation of regionally targeted programs owned by municipalities or managed by local self-government authorities Formation of regional property Operation of municipal law enforcement agencies International and foreign economic relations for regions Organization, maintenance, and development of municipal housing and utilities Maintenance and development of enterprises, institutions, and organizations subordinate Municipal road construction and maintenance to public regional authorities Amelioration of municipally owned areas and plantation of trees Activities of mass media in regions Organization of domestic waste utilization and recycling (save for radioactive waste) Financial assistance to local budgets Maintenance of burial places managed by municipalities Execution of individual public responsibilities delegated to the municipal level Organization of transport service for the population and institutions owned by Compensation of additional expenditures resulting from decisions taken by public municipalities or managed by local self-government authorities authorities, which increased budgetary expenditures or reduced budgetary revenues Fire safety of local budgets Environmental protection in municipalities Other expenditures associated with the execution of responsibilities of subjects of the Implementation of targeted programs adopted by local self-government authorities Russian Federation Service and repayment of municipal debt Targeted subsidization of households Maintenance of municipal archives Municipal elections and local referendums Implementation of other decisions taken by local self-government authorities and other expenditures classified as local issues to be determined by representative bodies of local self-government in accordance with budget classification of the Russian Federation Source: 1998 Budget Code. 28 Jorge Martinez-Vazquez and Jamneson Boex legal ambiguity that has plagued Russia in the transition. The 1998 Concept on Reform of Intergovernment Fiscal Relations does not offer any solution either; it basically reproduces the expenditure assignments in the Budget Code. Both the Concept and the Budget Code require revision. The proper starting point for introducing clarity in the assignment of expenditure responsibilities is with understanding the actual assignments of expenditure responsibilities that have been in place over the past years. An inventory of actual expenditure assignments ought to be conducted at different levels of detail. Lavrov (1998) recommended conducting an exhaustive survey of actual expenditure assign- ments as a preliminary step to reforming those in the Budget Code. The next step would be to identify problematic de facto assignments. At a minimum, the clarification of fiscal relations between the Russian Federation and the regions will require the assignment of primary responsibility for the concurrent ex- penditures listed in Article 72 of the 1993 Constitution. Regional expenditures should include primary responsibility for government functions that are traditionally administered by local or regional govern- ments (such as costs for local administration, law enforcement, and elementary and secondary educa- tion). The federal government should be assigned the primary responsibility to fund policy areas of national importance or those that are redistributive in nature, such as civil rights and social policies. In addition, priorities among expenditures should be established along the lines of primary responsibility to reinforce the separation of federal and regional activities. The assignment and prioritization of pri- mary expenditure responsibilities and priorities among expenditures could either take place through the conventional legislative process or by amending the Constitution. Reform of Current Expenditure Assignments There are no clearcut rules for assigning expenditure responsibilities to different levels of government. Adequate assignments are reached typically as a compromise among often conflicting objectives pur- sued by decentralization policy. Several general principles, however, can help guide the design and reform of the assignment of expenditure responsibilities. An efficient provision of government services requires that the needs and preferences of taxpayers are satisfied in the least costly way possible. Because local governments are relatively close to taxpayers, efficiency can typically be increased by keeping the provision of services at the lowest level of govern- ment. This rationale is known as the "principle of subsidiarity." However, higher levels of government should be assigned expenditure responsibilities for funding public services with substantial spillover benefits across smaller jurisdictions, because it is likely that smaller governments will fail to internalize the spillover benefits adequately Higher levels of government should further be assigned expenditure responsibilities for those services with significant economies of scale and scope, although these produc- tion efficiencies may still be captured by smaller governments by contracting for services with either nongovemment enterprises or larger government units. The realization of efficiency gains in expendi- ture assignments requires effective democratic representation; proper budget institutions, including those that allow subnational governments to vary the composition of the budgets; and the ability to link costs of provision to fees, service charges, or local taxes. Expenditures undertaken by government for equity, or income equalization, and for macroeconomic stabilization should be the responsibility of the federal government. Attempts by subnational jurisdic- tions to redistribute income can easily backfire in the presence of significant taxpayer mobility. More important, the need for social expenditures is typically higher in relatively poorer jurisdictions, which by themselves are less able to afford such programs. Subnational governments may have an advantage in the implementation of redistributive programs, but with few exceptions, programs should be financed at the national level. The federal government further has an advantage in addressing the larger dimension of macroeconomic problems. However, subnational governments can play an effective role in outlining and implementing regional development policies. These general principles for assigning expenditure responsibilities can identify poor assignments, but do not lead to a unique or optimal blueprint for expenditure allocations. Balancing the objectives of Expenditure Responsibilities 29 efficiency, equity, and macroeconomic stability can lead to quite different, but equally acceptable, expenditure assignments. Overall, most of the defacto expenditure assignments in Russia do not violate the general principles of expenditure assignment. These assignments could be maintained in many policy areas because the man- dates were clear to all parties involved. However, even in these areas a formal and explicit statement of responsibilities would reduce the possibility of arbitrary reassignments of expenditure functions and increase stability of intergovernmental fiscal relations. In addition, there are some expenditure areas in which the current (de facto) assignments should be reformed. The most important of these is social wel- fare. A major defect of the current arrangement involves the dichotomy between financial and implemen- tation responsibilities. While financial responsibility for most public expenditures with a redistributional purpose should be allocated at the federal level, many regional and local governments still have a com- parative advantage in delivering many of these welfare programs because of proximity to the population that requires these services. Redistributive programs that should be provided locally but funded feder- ally include school meals for poor children or shelter for the elderly poor. Other areas require reassign- ment of responsibilities. Based on a clearer division of responsibilities, the sharing rates for the main taxes and the system of transfers should be reevaluated so they serve as appropriate sources of funding for those policy areas for which regions are given primary responsibility. Given the current budgetary constraints faced by the Rus- sia Federation, during the transition the level of funding for regional governments will likely fall short from providing an ideal minimum level of government services. Nonetheless, subnational governments will be expected to fulfill their constitutional responsibilities to the best of their ability with the available resources. In the realm of regional public policy, the role of the Russian government should be limited to setting national goals, like in Germany, or to serving as a facilitator to harmonize regional policies, as in Canada. Because regions should have primary responsibility for regional policy issues, the federal government should be prevented from issuing unfunded mandates in these regional policy areas. The harmonization of the many laws, decrees, and departmental orders that impact and overlap expendi- ture assignment issues will likely prove to be a more difficult exercise. The role of federal agencies, such as the Ministry of Education, must be defined in the context of regional and local delivery of public services, such as primary and secondary education. This will require distinguishing between the need for the decentralization of regulatory powers, responsibility for financing, and delivery of the public services. Resolve Unfunded Mandate Problems Unfunded mandates are not an exclusive problem of the Russian Federation. They are practically univer- sal issues among federal countries. As the respective roles of federal and regional governments evolve over time in a federal system, areas of ill-defined jurisdiction are often revealed. These ambiguities arise despite the fact that the assignment of expenditure responsibilities is generally contained in the constitu- tions of federal countries. For example, when federal authorities seek to set national standards for air and water pollution, it often becomes a point of contention whether the federal government mandate to pro- tect the environment supersedes the regional governments' rights to regulate their internal affairs with- out federal interference. Tensions between the various government levels are naturally heightened if unfunded compliance costs are associated with the federal mandate. Some degree of overlap or concurrence between federal and regional responsibilities is unavoidable in any decentralized system of government. One way to avoid political stalemate when conflict arises in areas of shared expenditure responsibility is to establish in the constitution or through legislation whether federal policies preempt regional policies. The dominance of one level of government over another is often referred to as "paramountcy." While paramountcy can be invoked to solve conflicts regarding the assignment expenditure responsibilities and disputes about unfunded mandates, this solution is not nec- essarily comprehensive or desirable. In addition, problems with unfunded mandates may arise even when government responsibilities are clearly and exclusively assigned. 30 forge Martinez-Vazquez and Jameson Boex Boex and Martinez-Vazquez (1998b) reviewed the policy responses to unfunded mandates in Canada, Germany, and the United States. The experiences of these countries constitute a spectrum of policy alter- natives from which to formulate a response to the issue in the Russian Federation. A number of valuable lessons can be drawn from the policies of these countries. CLARIFY EXPENDITURE ASSIGNMENTS. While expenditure responsibility assignments are relatively clearly assigned in the constitutions of Canada and Germany,-they are much less clear in the United States Con- stitution. Within the structure of the German federation, responsibility for virtually all policy areas is assigned to the federal government, largely limiting the role of the regional governments (Lander) to administering federal policies in accordance with regional customs. The German federal government is under no obligation to provide funds to regions when imposing new mandates. In contrast, in Canada the prohibition on unfunded mandates is implicit in the clearly delineated and balanced assignment of government responsibilities between the federal government and the provinces. While disputes about expenditure assignments and federal mandates occur less frequently in either Canada or Germany, they are more likely to be resolved in the political arena. Federal mandates in the United States are frequently at the center of legal disputes between the states and federal authorities. Although the U.S. Constitution enumerates the jurisdiction of the federal government, courts have interpreted these powers broadly despite the Tenth Amendment to the U.S. Constitution, which assigns any powers not delegated to the U.S. government "to the States respectively, or to the people." PROHIBIT NEW UNFUNDED MANDATES. For budgeting purposes, the policy areas for which the regional governments are solely or primarily responsible should lie outside the area of federal jurisdiction. Conse- quently, unfunded federal mandates in these areas should be prohibited by law to avoid off-loading federal responsibilities to the regions without corresponding funding. A provision to this effect was in- cluded in the 1998 Budget Code. Prohibiting unfunded mandates seems to offer a simple solution to the issue of unfunded mandates. However, when taken to an extreme, this approach also presents limitations. Some mandates, even un- funded, may be quite acceptable to subnational governments. For example, mandates that ensure civil and voting rights or provide equal access and opportunity to citizens clearly serve the national interest. In practice, accountability mandates, such as reporting requirements for judicial and financial adminis- tration of regional and local governments, are also appropriate directives that are rarely resisted in other countries, even when money is not provided to fund them. Strict prohibition of all unfunded mandates would also cause practical problems. Such blanket legis- lation could result in an administrative nightmare. A review of state mandates in the United States re- veals that reimbursement programs for mandates are administratively complicated, politically contro- versial, enormously expensive, and often do not work (Kelly 1995). In addition, outright prohibition of unfunded mandates could be circumvented by a creative legislature. Alternatively, the legislative branch could post a ceiling on the cost that can be imposed on the regions by federal policies before a mandate statute is triggered; thus at the very least, this will ensure that federal policies will be closely scrutinized as to whether they will impose an excessive burden on regional governments. IDENTIFY AND REFORM EXISTING UNFUNDED MANDATES. A large number of federal laws in Russia mandate subnational governments to provide social services according to federally established norms, such as the levels of cash assistance provided. However, subnational governments are not given additional funding to cover the expenses. The first step in effectively dealing with existing unfunded mandates is to identify which federal programs, based on the clarified assignment of expenditure responsibilities, contain provi- sions that impose unfunded mandates. Recent reforms in the U.S. approach toward unfunded mandates may serve as a reference for the needed reforms in Russia. Under the Unfunded Mandate Reform Act of 1995, the U.S. Advisory Commission on Intergovernmental Relations (1996) developed two sets of criteria to evaluate policy Expenditure Responzsibilities 31 proposals: one provided defining criteria for mandates of significant concern to state and local govern- ments, and another outlined policy recommendations for the reviewed mandates. A joint committee of Federation Council, State Duma, Ministry of Finance delegates, and possibly representatives from re- gional and local governments, could be formed to review the federal programs identified by subnational governments as burdensome. When the joint committee identifies an unfunded mandate, it should evaluate the size of the burden imposed by the mandate and recommend eliminating, funding, or modifying it. Though the repeal or modification of unfunded mandates would give regions much greater flex- ibility in addressing policy issues, the State Duma is unlikely to take these actions, as it originally enacted the mandates to reduce the federal budget deficit. Yet, under the 1998 Budget Code the federal government will be prohibited from requiring subnational governments to fund these fed- eral program from regional funds (own revenues, revenue sharing, and equalization transfers) after January 2000. Consequently, the Ministry of Finance should develop a system of earmarked grants to fund the delivery of federally mandated regional services. This approach forces the State Duma to evaluate the cost of its impositions realistically and to assume financial responsibility for the federal programs it adopts. INCREASE THE WATCHDOG POWERS OF THE FEDERATION COUNCIL. The Russian political system already incor- porates a powerful safeguard against the excessive use of unfunded mandates through the Federation Council, which consists of representatives of the Federation from the regions. The Federation Council can block federal legislation containing unfunded mandates, although its decisions may be overruled by a two-thirds majority of the State Duma; thus a legislative super-majority must approve an unfunded man- date before it can be passed into law. To further restrict the imposition of unfunded mandates, a constitu- tional amendment could require an even greater majority vote from the State Duma to pass legislation that affects regions fiscally. The U.S. practice of having the Congressional Budget Office evaluate each legislative proposal and apply objective criteria to determine whether legislation contains unfunded mandates could be applied to unfunded mandate propositions in Russia. INCREASE THE COOPERATION BETWEEN FEDERAL AND SUBNATIONAL AUTHORITIES. By establishing a working partnership between the various levels of government, intergovernmental tensions in a federal country could be relieved. The system of fiscal relations in Russia lacks an underlying framework of trust and basic cooperation between the federal and subnational governments. In a cooperative atmosphere, fed- eral and regional governments are more likely to jointly develop creative solutions to the problem of unfunded mandates. Intergovernmental cooperation in working groups and advisory councils fosters greater understanding, while increased communication allows regional governments to identify federal mandates that have outlived their usefulness and to present the federal legislature with a set of recom- mendations for reform or repeal. The German consensus-oriented approach to politics provides an important safeguard against the overuse of unfunded federal mandates. The federal government consults extensively with the regional governments through a variety of councils and carefully evaluates the impact on horizontal and vertical equity of federal policy proposals. Similar cooperation is evident in most other federal countries. For example, in Australia the Natioiial Fiscal Outlook is prepared on an annual basis by a working group of the federal and regional treasuries. This document presents medium-term projections of federal and regional government finances and serves as background information for the annual meeting between the Australian prime minister and the pre- miers of states and territories. The conference presents an opportunity for the heads of the federal and regional governments to negotiate the allocation of public resources, discuss the general direction of the national economy, and address other issues concerning intergovernmental fiscal relations in Australia. Australian intergovernmental relations are also solidified through the Loan Council, comprised of the federal treasurer and professional counterparts from the states and territories, which coordinates public sector borrowing by the federal and regional governments. 3 Revenue Assignments and Tax Administration Issues Sharing arrangements between the federal and subnational governments for the major taxes-value added tax (VAT), enterprise profit tax (EPT), personal income tax (PIT), and excise taxes-have been the focus of positive reforms during the transition. Table 3.1 summarizes the history of revenue-sharing arrange- ments during 1992-98. While tax sharing remains the dominant source of revenues for most regional governments, during the transition nonshared regional and local taxes have become an increasingly im- portant source of subnational funding. This chapter reviews the most important issues in the assignment of revenues among the different government levels in the Russian Federation. It includes an analysis of revenue collection trends and a discussion of pending revenue reforms. In addition, it addresses the role of incentives and the structure of tax administration. How Have Revenues Been Assigned? The main dichotomy in revenue assignments in the Russian Federation has been between shared rev- enues and own-source revenues. "Shared revenues" refers to the proceeds from four main taxes, VAT, EPT, PIT, and excise revenues, which are shared at temporary rates between federal and regional govern- ments. Regional governments also generate their own-source revenues. However, only taxes permanently assigned to regional and local governments are considered own-source revenue, even though subnational governments have little or no discretion over the rates or bases for the vast majority of these taxes. None- theless, own-source revenues form an increasingly important source of subnational revenues and may be seen as the predecessors for regionally and locally controlled taxes. In general terms, three periods, characterized by different approaches to revenue sharing and policy toward subnational tax autonomy, can be identified in the practice of revenue assignments since inde- pendence. Although on the whole revenue assignment practices have improved over the years, the Rus- sian Federation still needs to develop an adequate system of revenue assignment. Revenue Assignments in the Early Transition Years During 1993, the VAT and in some cases other major taxes, such as the EPT and excises, were shared between the federal government and the regions on a derivation basis. That is, the region that collected the taxes shared them with the federal government. Some minor taxes were allocated entirely to subnational governments. However, the most important taxes were regulated, just as they had been under the Soviet system. By adjusting the sharing rates for each region, the federal government provided just enough money to finance the region's minimum expenditure budget, which had been negotiated with the Minis- try of Finance. Forecasted own-source revenues for subnational governments were subtracted from the minimum expenditure budgets to arrive at the amount to be financed with shared revenues. Despite considerable bargaining between the Ministry of Finance and the regional governments, sharing rates were less important than the estimated financing gap. In poorer regions where overall revenue collec- tions were low, the budget gap remaining after combining own-source and shared revenues was covered by "subventions" or lump-sum transfers from the federal budget. Local government budgets were funded by regional authorities in an identical process. 33 Table 3.1. Legislated Sharing Rates of Major Taxes, 1992-98 (percentage of total) Law on Basic Principles of Taxation, 1991 1992, Q1 1992, Q2-Q4 1993 1994, QI Tax Federal Regional Federal Regional Federal Regional Federal Regional Federal Regional VATa 100 0 Ad hoc negot. 80 20 80-50 20-50 75 25 Profit tax 0 100" 47 53 41 59 31 69 37-34c 63-66' PIT 0 100b 0 100 0 100 0 100 0 100 Excise on alcohol 50 50 50 50 50 50 50 50 50 50 Energy excises - - 100 0 100 0 100 0 100 0 Excises on domestic productions - - - - - - - - 0 100 (table continues onfollowing page) Table 3.1 continued 1994, Q2-Q4 1995 1996 1997 1998 Tax Federal Regional Federal Regional Federal Regional Federal Regional Federal Regional VAT- 75 25 75 25 75 25 75 25 75 25 Profit tax 37-34c 63-66c 34 66 34 66 34 66 34 66 PIT 0 100 10 90 10 90 0 100 40 60 Excise on alcohol 50 50 50 50 50 50 50 50 50 50 Energy excises 100 0 100 0 100 0 100 0 100 0 Excises on domestic productions 0 100 0 100 0 100 0 100 0 100 Q Quarter. - Not available. a. VAT on imports, precious metals, and stones was assigned 100 percent at the federal level. b. These taxes could be shared with the federal government. c. The federal profit tax rate was 13 percent; regional profit tax rates ranged up to 22 (for most enterprises) and 25 (for banks and insurance companies) percent of profits. The sharing percentages in the table were calculated based on the 13/22 and 13/25 profit tax rate. Source: Modified and updated from World Bank (1996). 36 Jorge Martinez-Vazquez and Jameson Boex The system of revenue assignments during this early period evolved spontaneously in open contra- diction with the law. In late 1991, the Law on the Basic Principles of the Tax System had come into effect assigning the VAT, customs duties, excises, and taxes on financial institutions exclusively to the federal government, which was not to share them with any other level of government. The Basic Principles assigned the PIT and EPT to the federal government for use in regulating the flow of revenues to subnational budgets. The Basic Principles further assigned 21 minor taxes to subnational governments, and more important, prohibited the introduction of any tax not listed within its pages. The revenue-sharing provisions of the Law on the Basic Principles were never implemented. The federal government could not overcome regional opposition, especially the opposition to not sharing VAT collections. In addition, the general rules did not apply to all regions. In 1992, Tatarstan, Bashkortostan, Sakha (Yakutia) illegally declared fiscal sovereignty and benefited from a single-channel regime, whereby the regional government retained all tax collections and negotiated a single payment with the federal authorities. Chechnya never paid any taxes, and 24 other regions were at least partially retaining addi- tional revenues in 1993. Regions such as Ingushetia were declared free trade zones and were practically exempt from paying taxes. The revenue assignments of 1992-93 presented several important problems. A lack of revenue au- tonomy for subnational governments denied the system of intergovernmental fiscal relations the effi- ciency gains associated with ability of subnational governments to increase or decrease at the margin the funds available for regional public services. Shared taxes, in particular the VAT, proved unwieldy. Enterprises operating in more than one region received VAT credits and debits in those different re- gions; no mechanism was available to sort out the monetary flow among regions. Also, the EPT was paid in the region where the business was registered and not apportioned among regions according to any criteria of economic activity. More important, the bargained approach to revenue assignments created perverse incentives for the regions. Because any increase in regional tax effort would be offset by a reduction in the regulated shar- ing rates, subventions, or transfers, subnational governments remained unmotivated to increase tax col- lections, either through subnational taxes or through better enforcement. Due to the perverse incentives caused by this approach to revenue assignments, extrabudgetary funds came to play a much more sig- nificant role in subnational government finances. The use of extrabudgetary funds proved to be the only means to hide additional resources from the Ministry of Finance during negotiations on minimum bud- gets for the purpose of increasing overall funding levels. The same perverse incentives, of course, applied to local governments, which used the same means as regional authorities to prevent upper-level govern- ments from clawing back their additional revenues. Because the regulated sharing rates were subject to annual changes, uncertainty regarding the overall level of funding also prevented subnational governments from engaging rational expenditure planning. To address this problem, the Law on Budgetary Rights of April 1993 required the federal government to assign strictly fixed shares of the major taxes, with the exception of VAT, for at least five years. This requirement was only partially respected in subsequent years. Revenue Assignment Reform during 1994-97 In 1994 revenue-sharing rates for the major taxes were standardized across regions. As shown in table 3.1, these tax sharing arrangements remained practically unchanged throughout the decade. In addition, subnational governments gained some measure of revenue autonomy during this period. MAJOR SHARED REVENUE SOURCES. The 1994 federal budget law stated that the uniform sharing rates applicable that year would remain unchanged until 1998: 0/100 for the PIT, 75/25 for the VAT, 37/63 for the EPT, 50/50 for excises on alcohol, 100/0 for excises on energy products, and 0/100 for excises on domestic productions. The federal government retained all revenues from customs duties; the VAT on imports, precious metals and stones, hydrocarbons, and excise goods and license fees for alcohol production. Contrary to the government's earlier commitment, a 10 percent federal retention of the PIT Revenue Assignments anid Tax Administration Issues 37 in 1995 and 1996 was earmarked for distribution among regions and local governments for maintenance of residential housing and social and cultural assets. However, in most cases, the redisbursed funds were roughly equivalent to those the regions could have retained if they had not relinquished the 10 percent of the PIT. As a result, this practice was discontinued in 1997 with the PIT fully assigned again to the subnational level. The draft budget for 1999 recommended an alternative arrangement in which a federal PIT is collected at a flat rate of 3 percent and a regional PIT is collected with a progressive rate structure for incomes exceeding Rb 60,000. The introduction of standard homogeneous sharing rates in 1994 represented a notable achievement for the system of intergovernmental fiscal relations. As sharing rates could no longer be used for equaliz- ing purposes, a formula-driven system of equalization transfers was also introduced. These two mea- sures helped eliminate the negative incentives against increasing collections by subnational governments and provided enough stability to permit regional governments to better budget their resources. However, the perverse incentives to reduce collections did not disappear completely. The federal gov- ernment still left the door open for negotiation and bargaining with the regions through mutual settle- ments, which were unbudgeted and nontransparent forms of transfers, giving it an avenue to claw back some of the additional revenues raised by the subnational governments. Starting in 1994, the federal government also negotiated a series of bilateral treaties with many regions. These treaties broke away from the standard sharing rules by allowing some of the regions to retain a higher share of revenues. Although revenue assignments improved because of the use of standard sharing rates, the adminis- trative problems and inefficiencies in revenue assignments persisted. The VAT was shared with regional governments on a collection basis without adjustments, and the EPT was still shared by place of business registration because an apportionment formula based on payroll and property was hardly ever applied. Also the full or almost full assignment of the progressive PIT at the subnational level on a derivation basis reduced the redistributional and equalizing abilities of this tax. OWN-SOURCE REVENUES. In the period between 1994 and 1997, subnational governments gained some measure of revenue autonomy. The presidential decree "On Forming Republican Budget and Its Rela- tionship to the Subject of the Russian Federation" of December 1993 lifted the 1991 Law on the Basic Principles prohibition on subnational introduction of new taxes. The edict led to the introduction of all sorts of nuisance taxes and other whimsical fees, especially on enterprises. Some of these taxes were removed after taxpayers took legal action. The power for regions to introduce new taxes at the subnational level was revoked, again by presidential decree, in 1996. Other changes in tax legislation provided subnational governments with some degree of tax autonomy, contributing to the ability of subnational governments to exercise higher or lower tax efforts. Though the federal EPT rate of 13 percent cannot be modified by regional governments, important legislation gave regional governments the freedom to apply a regional EPT rate from 0 to 22 percent of the tax base. With the advent of post-1993 reforms, many regional governments granted tax holidays and advan- tages to particular investors. Regional governments also set the rates for the taxes on enterprise property and personal property up to a federally determined maximum. Following federal guidelines, regions also set the rate, and at times the base of other, often minor, revenue sources, such as the forest tax, water tax, ecological tax, the resort fee, the transport tax, regional license fees, and advertising tax. Local gov- ernments were free to introduce two turnover taxes on the sale of products: the Housing Maintenance Tax, with a rate of up to 1.5 percent, to fund housing and social service infrastructure divested by enter- prises and the Road Tax, at a rate of up to 2.5 percent, to fund local infrastructure. Local governments could also levy an education tax of up to 1 percent of business payrolls. Some other minor local taxes were available for local governments. REGIONAL-LOCAL FISCAL RELATIONS. Despite improvements in revenue assignments between the federal and regional governments, fiscal relations at the regional-local level for the most part remained unchanged throughout the transition. Although most regions continued to use negotiated sharing rates, practices varied by region. Some regions reformed their budgetary relations with local governments along the 38 Jorge Martinez-Vazquez and Jameson Boex lines of the federal-regional reforms of 1994 (Lavrov and others 1996). For example, 10 regions in 1995 were using uniform sharing rates with local governments, but only two of these, Sverdlovsk and the Jewish Autonomous Region, instituted a system of transfers based on formulas. Igudin (1998) discussed how various regions structured revenue sharing with their local governments. Some regional govern- ments did not change their sharing rates with local governments for years, while others used standard uniform sharing rates for rayons but regulated them for cities. Because local governments felt that they were completely at the mercy of regional governments, local governments lobbied the federal authorities for the imposition of sharing standards from the regions to the local governments. Of course, most regional governors opposed this kind of restriction, and the Law on the Financial Foundations was originally turned down by the Federation Council. Subsequently, the Duma introduced several changes, and the amended legislation was approved by both the Duma and the Federation Council. The 1997 Law on the Financial Foundations of Local Self-government was a compro- mise between the local and regional positions. This law mandated some minimum sharing rates with local governments, but only as an average for all local governments in the region. On average, local governments should retain a minimum of 50 percent of PIT revenues, 5 percent of EPT, 10 percent of VAT, 5 percent for excise duties on alcohol, and 10 percent for excises on other goods. The worst consequence of this state of affairs in regional-local relations is the continued lack of incen- tives for local governments to raise their own revenues. Despite the 1993 Law on Budgetary Rights and the 1997 Law on the Financial Foundations of Local Self-government that clearly prohibited the claw- back of local revenues, regional governments systematically recouped increased local collections by re- ducing sharing rates or transfers in the subsequent fiscal period (Zhuravskaya 1998). Reforms Contained in the Draft 1998 Tax Code Perhaps is it is appropriate to speak of a third phase or period in revenue assignments, which will begin with implementation of the draft Tax Code. In 1998 the State Duma approved the general part of the draft Tax Code that deals with tax administration issues, taxpayer rights and obligations, and tax assignments to different levels of government (see table 3.2). The VAT, PIT, EPT, and excises were kept as federal taxes to be shared with subnational governments. Regional and local governments were assigned several new taxes but relieved of many old ones. Whether or not the new taxes will be implemented depends on the passage of the part of the draft Tax Code that develops the structure of all taxes. In the draft 1998 Tax Code, regional governments were provided with a significant potential increase in revenue autonomy. If the draft Tax Code is adopted, regional governments will be able to introduce a final sales tax of up to a 5 percent rate for general goods and up to a 10 percent rate on excise goods. Before the legislature could pass the measure, a new regional sales tax, with a maximum 5 percent rate and a narrower base than that outlined in the 1998 Tax Code, was introduced in 1998 as part of the economic emergency measures precipitated by the ruble crisis. In December 1998, the Ministry of Fi- nance sent the State Duma a proposal to increase the regional sales tax rate to a maximum of 10 percent and to reduce the general VAT rate from 20 to 14 percent. The proposal to partially substitute the VAT with the higher rate sales tax was based on the erroneous belief that a retail sales tax would be easier to enforce and that it would bring more cash revenues than the VAT. However, neither sharing the federal VAT revenues nor implementing a regional VAT will achieve vertical balance between the federal and subnational governments or revenue autonomy for subnational governments. The 1998 draft Tax Code also provided for the introduction of a Western-style real estate tax at the regional level, which was to be shared with local governments. If passed, the regions will have the choice of using the real estate tax as a substitute for three existing property taxes (land tax, indi- vidual property tax, and tax on property of organization), and they will have some discretion on rates. The revenue potential of the real estate tax is questionable because setting up its proper ad- ministration will prove difficult. Though two pilot cities, Novgorod and Tver, have been working on Revenue Assignments and Tax Administration Issues 39 Table 3.2. Status of Taxes in the 1998 Tax Code Federal taxes Regional taxes Local taxes Value added tax Tax on property of organizations Land tax Excises on specific goods Tax on real estate Individual property tax Excises on raw materials Road tax Tax on advertising Enterprise profits tax Transport tax Inheritance or gift tax Capital gains tax Sales tax Local license fees Personal income tax Tax on gambling enterprises Contributions to state off- Regional license fees budget funds Customs duties State duties Tax on subsoil use Tax on rehabilitation of mninerals Tax on additional income from hydrocarbons production Fee for the right of use of fauna and water biological resources Forest tax Water tax Ecological tax Federal license fees Note: The assignments are in the general part of the Tax Code, which was approved by the Duma and signed by the president. The specific part of the Tax Code that develops each of the taxes had not been approved as of June 1999. Source: 1998 Tax Code. its introduction since 1993, probably only the Novgorod real estate tax on residential structures, but not land, will be implemented. Trends in Revenue Collections While throughout the transition the bulk of subnational revenues came from revenue sharing and trans- fers, there has been a steady increase in the role of own-source revenues in subnational government budgets. Subnational own-source revenues represented just 15.5 percent of collections in 1992, but by 1997, had increased to 43.0 percent (see table 3.3). These figures reflect the increased revenue significance of the new subnational taxes introduced between 1994 and 1996 and the poor collection performance for major shared taxes. Major Shared Revenue Sources In the turbulent early years of the transition, revenue shares were regulated, multiple budgets were created per year, and federally off-loaded expenditure responsibilities produced fiscal difficulties for regional and local governments. Yet, subnational tax revenues jumped from 39.7 percent in 1992 to 54.4 percent in 1993 (see table 3.4). With the 1994 revenue sharing and transfers reforms, the subnational government's share of overall tax revenues settled close to 50 percent and remained in this neighbor- hood through 1996. The figures available for 1997 show that the balance of revenue distribution grew to 57.5 percent in favor of subnational governments. Throughout the transition, actual revenue sharing between the federal and subnational governments has always differed from the revenue-sharing provisions stated in the annual federal budget law. Some 40 Jorge Martinez-Vazqtnez and Jameson Boex Table 3.3. Own-Source Revenues and Shared Revenues as Percentages of Consolidated Subnational Collections (Excluding Transfers and Subventions), 1992-97 Revenue source 1992 1993 1994 1995 1996 1997 Own-source revenues 15.5 15.2 27.8 32.0 40.1 43.0 Shared revenues 84.5 84.8 72.2 68.0 59.9 57.0 Source: Ministry of Finance data (1998). inconsistencies between annual tax budget laws (table 3.1) and the actual shares of collections (table 3.4) is apparent in most taxes but is most pronounced for the VAT. Several reasons for this lack of correspon- dence between statutory and actual sharing rates are clear. In 1994 regions started unilaterally and in later years were legally allowed to withhold extra amounts of VAT collections in their territories in lieu of budgeted transfers. In the early transition, some regions refused total or partial remittance of tax collec- tions to the federal government. Later, bilateral treaties with the federal government granted a number of regions the opportunity to keep larger revenue shares. Tax arrears accrued faster to the federal share than to the regional share of most taxes. In addition, loans outstanding at the end of the year were offset against the transfers due to those regions in the following year. For more discussion on major shared revenue sources, see Freinkman, Titov, and Treisman (1998) and World Bank (1996). The reasons for the different profiles for federal and regional tax arrears are also complex. First, only a fraction of tax payments to the State Tax Service (STS) was in cash, mainly due to the pervasive- ness of barter and other nonmonetary forms of payments among Russian businesses. In this environ- ment, regional governments were more willing and able to accept tax payments in the form of offsets, that is, by canceling out mutual debts between the government and taxpayers and other nonmonetary forms of payment, such as barter. Because some regional and local tax inspectors still owed consider- able allegiance to subnational bosses, regions enjoyed a greater collection rate on shared taxes. Re- gional and local tax offices may have tried harder to collect revenues accruing to subnational budgets, or cash-strapped businesses might have been allowed to clear tax arrears to subnational governments before paying on federal taxes. Own-Source Revenues During the economic transition period, the importance of own-source revenue collections increased over time (see table 3.3).1 The relative importance of the various subnational own-source revenues is explored in table 3.5. More than 80 percent of subnational own-source revenues came from three categories: property and natural resource taxes and a local levy on housing and militia maintenance. While the increased magnitude of these own-source revenues is not disputed, it is often questioned whether these increases in subnational own-source revenues have translated into real increases in rev- enue autonomy. To achieve revenue autonomy, subnational governments must enjoy a substantial de- gree of discretion over either the tax base or rate. While in many cases the revenues from a tax were assigned to the regional or local level of government, the rates and bases of these own-source revenues were often determined or limited by federal law. The degree to which regional and local governments exerted control over their own-source revenues is outlined in table 3.6. As can be seen, regional governments were able to set the rate of the enterprise property tax, albeit within federal limits, which accounted for almost all property tax collections. In a similar manner, the Housing Maintenance Tax fell clearly within the purview of local governments. By contrast, table 3.6 shows that subnational governments had no control over either the tax base or rate for 1. Unlike the Russian Federation tax system, under international practices shared revenues are never consid- ered own-source revenues. Instead, the regional portions of shared revenue sources are considered transfers because subnational governments hold little influence over either the bases or the rates of these taxes. Revenue Assignments and Tax Administration Issues 41 Table 3.4. Revenue Distribution between Government Levels, 1992-97 (percent) Level of government 1992 1993 1994 1995 1996 1997 Total tax revenues 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 60.3 45.6 48.3 48.4 49.8 42.5 Consolidated subnational budget 39.7 54.4 51.7 51.6 50.2 57.5 Regional budgets 19.4 28.3 26.4 26.2 23.7 Local budgets 20.3 26.1 25.3 25.4 26.5 Profit tax 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 41.3 32.5 35.1 35.6 36.6 32.4 Consolidated subnational budget 58.7 67.5 64.9 64.4 63.4 67.6 Regional budgets 30.1 39.7 38.9 39.0 38.4 Local budgets 28.6 27.7 26.1 25.4 25.0 VAT 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 75.1 64.4 67.2 75.4 74.8 68.5 Consolidated subnational budget 24.9 35.6 32.8 24.6 25.2 31.5 Regional budgets 15.1 22.5 21.1 14.0 15.7 Local budgets 9.8 13.1 11.7 10.6 9.5 Personal income tax 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 0.0 0.0 0.6 9.0 9.2 2.3 Consolidated subnational budget 100.0 100.0 99.4 91.0 90.8 97.7 Regional budgets 22.8 23.2 25.8 24.2 24.6 Local budgets 77.2 76.8 73.6 66.8 66.3 Excise taxes 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 47.9 50.5 60.1 72.8 86.8 80.2 Consolidated subnational budget 52.1 49.5 39.9 27.2 13.2 19.8 Regional budgets 41.9 40.2 32.0 21.3 10.3 Local budgets 10.2 9.3 7.9 5.8 2.9 Property taxes 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 0.0 0.0 0.0 5.0 0.8 1.2 Consolidated subnational budget 100.0 100.0 100.0 95.0 99.2 98.8 Regional budgets 52.4 45.5 41.8 42.3 44.1 Local budgets 47.6 54.5 58.2 52.8 55.1 Natural resources tax 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 0.0 29.1 15.3 24.4 25.1 19.7 Consolidated subnational budget 100.0 70.9 84.7 75.6 74.9 80.3 Regional budgets 63.9 40.6 44.9 36.4 39.0 Local budgets 36.1 30.3 39.7 39.2 35.9 Other taxes 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 91.8 77.8 70.7 61.3 43.1 42.9 Consolidated subnational budget 8.2 22.2 29.3 38.7 56.9 57.1 Regional budgets 2.3 8.9 13.1 17.3 16.6 Local budgets 5.9 13.3 16.2 21.4 40.2 (table continues onfollowing page) 42 Jorge Martinez-Vazquez and Ja meson Boex Table 3.4 continued Level of government 1992 1993 1994 1995 1996 1997 Nontax revenues 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 59.2 - 38.0 70.9 75.1 66.4 Consolidated subnational budget 40.8 - 62.0 29.1 24.9 33.6 Regional budgets 33.6 - 35.7 22.2 18.8 - Local budgets 7.1 - 26.3 7.0 6.1 - Intergovernmental transfers 100.0 100.0 100.0 100.0 100.0 100.0 Federal budget 6.6 - 3.5 1.7 0.3 0.2 Consolidated subnational budget 93.4 - 96.5 98.3 99.7 99.8 Regional budgets 33.1 - 54.4 42.6 46.4 - Local budgets 60.3 - 42.1 55.7 53.2 - Not available. Note: Data may not sum to total because of rounding. Source: Freinkman, Titov, and Treisman (1998); Ministry of Finance data. Table 3.5. Consolidated Subnational Own-Source Revenue Collections, 1997 Own-source tax revenues Trillions of rubles Percentage of total Property taxes 46.9 39.0 Individual property tax 0.4 0.4 Enterprise property tax 46.4 38.6 Other property taxes 0.1 0.1 Natural resource collections 28.6 23.8 Mineral resource payments 11.9 9.9 Tax on depletion of mineral resources 6.9 5.7 Land tax 8.4 7.0 Other natural resource collections 1.4 1.2 Other taxes 44.8 37.2 Tax on housing and militia maintenance 23.6 19.6 Wage fund tax 9.1 7.6 Other income taxes 0.4 0.3 Special sales tax 0.9 0.7 Other sales taxes 6.6 5.5 Other taxes 4.1 3.4 Total subnational own-source tax revenues 120.3 100.0 Regional nontax revenues 21.0 17.5 Shared revenue collections 208.7 173.4 Note: Data may not sum to total because of rounding. Source: Authors' calculations based on data from the Ministry of Finance. revenues from natural resource taxes. Except during 1993-96, in which subnational governments were allowed to introduce their own taxes, all taxes were federally legislated, distancing regions from actively participating in the development of the tax system. When the information contained in tables 3.5 and 3.6 is considered jointly, one can see that regional and local governments had some degree of revenue autonomy, particularly with regard to enterprise, property, Revenuie Assignments and Tax Administration Issues 43 and the housing-maintenance taxes. However, federally imposed limits on the tax rate placed an upper bound on the ability of subnational governments to be more responsive to their fiscal needs and wants. Greater revenue autonomy could be accorded by allowing regional and local governments increased input in determining the rates for federally legislated taxes, greater freedom in legislating their own revenue sources, and more self-determination in administrating own-source revenues. Regional autonomy in designing tax bases is less desirable because it would create greater compliance costs for taxpayers operating in more than one jurisdiction. How Has the System Performed? The trends in revenue assignments impacted the performance of the revenue system in a number of ways. First, compared to pretransition levels, subnational revenues increased as a share of gross domestic product (GDP). Second, increased subnational revenue autonomy resulted in improvements in the verti- cal fiscal balance. Third, greater freedom at the subnational level combined with increases in regional economic disparities resulted in greater horizontal fiscal imbalances. Fourth, distinctive patterns emerged in the distribution of public resources between regional and local govemments. Revenues as a Share of GDP The share of overall subnational revenues in GDP changed over the transition period, albeit not drastically. This share was 13.8 percent of GDP in 1992, peaked at 18.2 percent in 1994, and stood at 16.6 percent of GDP in 1997 (table 3.7). The share of federal revenues in GDP remained lower than for subnational governments after 1992, between 12 and 14 percent through 1997. The combined share of revenues from both levels of governments, excluding extrabudgetary funds, was 32.6 percent in 1992 and 28.6 percent in 1997. The Western press generally reports that poor revenue collections are a main cause of Russia's eco- nomic problems. Even though Russia's tax-to-GDP ratio is below the average for countries of the Organisation for Economic Co-operation and Development (OECD), an often missed story is that Russia's share of taxes in GDP, excluding extrabudgetary funds, exceeds that of several OECD countries, includ- ing Japan, Switzerland, and the United States.2 Because GDP decreased considerably over the transition, table 3.8 was created to provide a more complete picture of the evolution of real tax revenues for subnational governments. In constant 1995 rubles, per capita revenues declined steadily during the early transition period. The trend improved slightly in 1997, when real per capita revenue collections increased by almost 10 percent, but real rev- enues per capita were still considerably lower in 1997 than in 1994. The indices used in table 3.8, normalized at 100 for 1994, show that of the major taxes considered, the EPT had the worst performance as a revenue source during the transition period. These data reflect not only the poor performance of the general economy and profits in particular, but they also reflect changes in regulations. By allowing more deductions and eliminating excess-wage taxes, the federal government narrowed the base of the EPT. Tax incentives and holidays provided by regional govern- ments to enterprises may have also affected the EPT figures. Although no stellar performers existed among the different revenue sources, the VAT and excises performed relatively better than the others, which may explain the continued strong opposition of subnational governments to reform revenue- sharing arrangements that would allot them a larger share of direct income taxes, EPT and PIT, and a smaller share, or no share, of the VAT. 2. The Russian Federation's tax-to-GDP ratio, excluding nontax revenues, equaled 22.0 percent for 1997. Ac- cording to OECD (1997b), the unweighted average of tax revenue to GDP ratios, excluding social security, for all OECD countries in 1993 was 28.5 percent. The Russian Federation's tax ratio exceeds that of Japan (19.3 percent), Spain (21.7 percent), Switzerland (20.8 percent), Turkey (18.9 percent), and the United States (21.0 percent). Table 3.6. Summary of Legislative Control over Subnational Own-Source Revenues, 1997 Legislative level Legislative level Budget(s) to Tax determining tax base determining tax rate which tax revenues go Tax on enterprise property Federal Regional (Federal limit) Federal/Regional Tax on property of physical persons Federal Federal a Local Land tax Federal Federal b Federal/Regional/Local Tax on use of mineral resources Federal Federal Federal/Regional/Local Tax on depletion of mineral resources Federal Federal Federal/Regional Transport tax (incl. road users, vehicle Federal/Regional Federal/Regional Regional owners, and acquisition tax) Housing and militia maintenance taxes Federal Local (Federal limit) Local Educational charges Federal Regional (Federal limit) Regional/Local Advertising tax Federal Local (Federal limit) Local Resort fees Local Local Local Exchange transaction fees Federal Federal Federal/Regional Tax on resale of automobiles and computers Federal Local (Federal lirnit) Local Racing and gambling charges Local Local Local Tax on industrial construction in national resort areas Local Local Local a. Regional and local governments can lower the rate and add exemptions. b. Regional govemments set concrete minimum rate for particular kind of land, and local governments can raise the rate. Source: Based on information from the Ministry of Finance. Revenue Assignments and Tax Administration Issues 45 Table 3.7. Revenue Distribution between Government Levels, 1992-97 (percentage of GDP) Level of government 1992 1993 1994 1995 1996 1997 Federal budget 18.8 13.1 13.6 14.2 14.6 12.0 Consolidated region budget 13.8 16.6 18.2 14.8 15.5 16.6 Regional budgets 6.8 8.4 9.7 7.5 8.1 Local budgets 7.0 8.2 8.5 7.3 8.5 Total 32.6 29.7 31.8 29.1 30.0 28.6 - Not available. Note: Transfers to other levels of government may result in double counting. Data may not sum to total because of rounding. Source: Freinkman Titov, and Treisman (1998); Ministry of Finance data. Vertical Fiscal Imbalance Vertical balance exists when a broad correspondence can be found between the expenditure responsi- bilities assigned to each level of government and the fiscal resources available to them to carry out those duties. In the Russian Federation, the lack of revenue autonomy at the subnational level was the central cause of some vertical imbalance. There are several other reasons for vertical imbalance, includ- ing the following: * The perceived need for the centralized administration of the most significant taxes * The assignment of the most elastic sources of revenues to the federal government, even though subnational governments officials were assigned responsibility for public services with a more elastic demand with respect to income * The federal government's fear of tax competition or mismanagement among regional and local governments * A reflection of the dominant political power of the federal government. Revenue sharing and transfers were typically designed to redress this vertical imbalance. However, both political accountability and economic efficiency require that subnational governments retain their own-source revenues (Boadway, Roberts, and Shah 1994). Although subnational governments increas- ingly relied on own-source revenues to fund their budgets, their control over these revenue sources was often marginal. In many cases, the bases and rates of taxes were determined by the federal government. Conventionally, vertical fiscal imbalances have been measured in two different ways. The first is to look at the surplus or deficit position of each consolidated level of government before borrowing but after all revenue sharing and transfers have been implemented. This measure tends to be biased to- ward the central or federal government level because in many countries, either by law or by practice, subnational governments operate more conservatively than the national government. In most coun- tries, the central government is allowed to borrow freely to cover current operations. In contrast, local governments in many countries can only borrow for capital investment purposes, if at all. In addition, this measure of vertical imbalance is only meaningful under the assumption that deficits are not the result of mismanagement or waste. During the transition period, the Russian Federation accrued large, though declining, budget deficits, while the consolidated subnational government budget moved from a small surplus to a small deficit. By this deficit measure, therefore, there has been a lack of vertical fiscal balance in Russia during the transi- tion. However, these data demand careful interpretation. Because regional and local governments could not borrow significant funds, especially early in the transition, the early surpluses and minor deficits of subnational governments may reflect the institutional constraints and the budgetary uncertainty and not the state of fiscal health. In the early transition period, regional and local governments may have kept a cushion of resources in extrabudgetary accounts to cover unpostponable expenses, especially because Table 3.8. Composition of Revenuesfor Subnational Governments in Real Terms, 1992-97 In real terms: per capita 1995 rubles In real terms: index (1994 = 100) Revenue category 1992 1993 1994 1995 1996 1997 1992 1993 1994 1995 1996 1997 Own-source and shared revenues of which 1,692.9 1,805.9 1,345.5 1,277.5 1,182.5 1,296.9 125.8 134.2 100.0 94.9 87.9 96.4 EPT 684.3 849.1 576.0 511.4 297.5 270.3 118.8 147.4 100.0 88.8 51.7 46.9 PIT 320.4 329.6 318.6 224.8 239.5 290.2 100.6 103.5 100.0 70.6 75.2 91.1 VAT 369.7 302.0 194.0 164.5 193.9 210.1 190.6 155.7 100.0 84.8 100.0 108.3 Excise tax 82.4 66.0 54.8 44.6 38.3 49.5 150.4 120.4 100.0 81.4 69.9 90.3 Other and nontax revenues 32.3 19.0 237.1 113.1 102.4 141.6 13.6 8.0 100.0 47.7 43.2 59.7 Transfers (excludes loans) 106.0 340.2 468.0 191.8 211.0 ]93.6 22.7 72.7 100.0 41.0 45.1 41.4 Total revenues 1,831.2 2,165.1 2,050.6 1,583.9 1,497.1 1,632.9 89.3 105.6 100.0 77.2 73.0 79.6 Source: Authors' calculations based on data from the Ministry of Finance. Revenue Assignments and Tax Administration Issues 47 unexpected revenue shortfalls were particularly frequent due to economic instability, poor revenue fore- casts, and sequestering of intergovernmental transfers. Increased subnational borrowing to finance bud- get deficits, even after institutional and market constraints were removed, suggests that vertical imbal- ance continued beyond the early transition period. Another way to measure vertical fiscal imbalance involves the examination of subnational govern- ment expenditures that are financed with sources of revenues under the control of regional and local governments. The coefficient for vertical imbalance is calculated as one minus-the share of the subnational government expenditures that are financed from sources of revenues not controlled by the subnational government. Because federally controlled revenues are inexactly defined, different versions of the coeffi- cient are typically computed (Hunter 1977). Table 3.9 shows the computation of three vertical fiscal imbalance indexes that correspond to different definitions of federally controlled sources for 1997. The first definition considers all revenue sources assigned to regional and local governments, excluding shared revenues and transfers, as own-source revenues. The second definition also considers revenues from tax sharing as own-source revenues, while the third definition adds untied equalizing grants to own-source revenues and shared taxes as subnationally controlled revenues. By construction, the coefficients of vertical imbalance take values between zero and one, with values closer to zero indicating a larger vertical fiscal imbalance. The small values of the coefficient for the first definition (when shared revenues and transfers are not considered part of subnational own-source rev- enues) indicate a significant degree of vertical imbalance. A proper perspective is given when the coeffi- cients of vertical imbalance for Russia are compared with coefficients for a selected number of transition and Western European countries in the 1990s (table 3.10). The Russian Federation achieved a relatively high score, suggesting a degree of vertical balance quite similar to that of Germany. However, the com- parison in table 3.10 does not reveal the degree of discretion that subnational governments had over their own revenue sources. Subnational governments in Russia had limited freedom in setting the rates or defining the base of regional and local taxes. Horizontal Disparities in Regional Revenue Collections Like for per capita expenditures, disparities in per capita revenue collections among regions increased steadily during the transition. As shown in table 3.11, the mean coefficient of variation for revenue collections across regions, induding the federal share, increased by 124 percent from 1993 to 1997. The highest per capita re- gional revenue in 1993 was 30 times higher than the lowest. By 1997 this difference had increased substantially. Disparities were created by the increasing concentration of economic activity in a small number of regions. There is, of course, a strong correlation between economic activity and tax bases and revenues. The five highest collection regions represented 40.9 percent of all revenue collections, including the fed- eral share, in 1997, up from 33.7 percent in 1993 (table 3.12). The 20 highest collection regions represented 70.9 percent of all revenue collections in 1997. When federal shares were not considered, as can be seen in table 3.13, the degree of concentration decreased only slightly. The uneven distribution of natural resources in the Russian Federation contributed to the marked horizontal imbalance in per capita revenue collections. Subnational revenues from taxes, fees, and charges on natural resources, including subsoil charges for use of mineral deposits and for prospect- ing, represented on average only 7.7 percent of regional tax revenues in 1997. However, the uneven- ness in the distribution of natural resources means that these revenues are highly concentrated in a small number of regions. While in 1997 the five richest regions, Khanty-Mansi, Yamalo-Nenets, Tyumen, Tatarstan, and Yakutia (Sakha), accounted for only 5.5 percent of the population, they collected 52.7 percent of all regional revenues from taxes, fees, and charges on natural resources. Two factors multiply the impact of differences in natural resource endowments on regional revenue disparities. First, the relative importance of natural resource taxes and fees in the regional budgets, on average, grew between 1995 and 1997. Second, the high correlation between regional collections from 48 Jorge Martinez-Vazquez and Jameson Boex Table 3.9. Measures of Vertical Fiscal Imbalancefor the Subnational Sector, 1997 Coefficient Measure Coefficient of vertical fiscal imbalance # 1 0.270 Coefficient of vertical fiscal imbalance # 2 0.841 Coefficient of vertical fiscal imbalance # 3 0.916 Note: Coefficients of vertical fiscal imbalance were calculated, following Hunter (1977): coefficient #1 = 1 -(revenue sharing + ututied (equalizing) and other transfers\ total expenditures Coefficient #2 = 1 - untied (equalizing) and other transfers) total expenditures ( other transfers Coeffcient #3 = 1 - (total expendituresJ Source: Authors' calculations based on data supplied by the Ministry of Finance. Table 3.10. Measures of Vertical Fiscal Imbalancefor the Subnational Sector: An International Comparison Country Coefficient #1 Coefficient #2 Russia, 0.270 0.841 Austriab 0.186 0.717 Belgium 0.156 0.512 Denmark 0.172 0.611 Estonia, 0.088 0.686 Finlandd 0.191 0.654 Germanyb 0.244 0.794 Latvia, 0.069 0.491 Lithuania, 0.111 0.869 Netherlands 0.288 0.345 Note: See table 3.9 for definitions of the coefficients. Unless otherwise mentioned, 1991 data used. a. Based on 1997 fiscal data. b. Subnational sector includes Lander as well as local governments. c. Based on 1994 fiscal data. d. Coefficients calculated using 1990 local expenditure data. Source: Calculated from data supplied by the ministries of finance of the Russian Federation, Estonia, Latvia, and Lithuania; IMF (1994); OECD (1997b). Table 3.11. Measures of Regional Fiscal Disparities: Per Capita Revenue Collec- tions (including Federal Share), 1993-97 (thousands of rubles) Coefficient Year Mean of variation Minimum Maximum 1993 251.9 0.68 31.2 943.5 1994 722.0 0.85 24.4 3,878.0 1995 1,950.9 1.12 158.7 15,681.4 1996 2,732.0 1.33 268.9 24,862.2 1997 3,723.3 1.52 251.5 41,940.4 Source: Authors' calculations based on data from the Ministry of Finance. Revenue Assignments and Tax Administration Issues 49 Table 3.12. Concentration of Total Revenue Collections (including Federal Share) among Regions, 1993-97 (percentage share of total revenues collected by selected regions) Revenue collection regions 1993 1994 1995 1996 1997 5 highest colection regions 33.7 32.0 34.5 37.6 40.9 10 highest collection regions 47.5 45.8 48.0 50.9 54.9 20 highest collection regions 64.9 64.6 67.2 68.9 70.9 Coefficient of variation, per capita revenue collections 0.68 0.85 1.12 1.33 1.52 Source: Authors' calculations based on data from the Ministry of Finance. Table 3.13. Concentration of Regional Revenue Collections among Regions, 1992-97 (percentage share of regional revenues collected by selected regions) Revenue collection regions 1992 1993 1994 1995 1996 1997 5 highest collection regions 27.0 28.0 30.2 30.4 31.6 34.1 10 highest collection regions 42.2 43.1 44.1 45.2 46.5 49.4 20 highest collection regions 62.7 62.7 64.1 64.5 65.6 67.3 Coefficient of variation, per capita revenue collections 0.89 0.67 0.86 0.86 1.14 1.51 Source: Authors' calculations based on data from the Ministry of Finance. natural resources and gross regional product (GRP) increased the regional revenue disparities between resource rich and poor regions (table 3.14). Note also that the correlation between natural resource collec- tions and total regional collections is high, 0.96 in 1997, and had increased throughout the transition period. These correlations are a reflection of the fact that businesses involved in natural resource extrac- tion also paid taxes on related economic activities, such as processing and transportation. Another factor that may have contributed to the existing disparities among regional governments is the manner in which shared taxes are apportioned among subnational governments. The impact of rev- enue apportionment mechanisms is hard to quantify without detailed taxpayer information, but because enterprises paid all their EPT in the region where they were registered, rather than where they conducted economic activity, collected revenues were likely concentrated in fewer regions. Proper apportionment mechanisms would establish a correspondence between the location of the economic activity, such as production, and the apportionment of the shared revenue, such as the EPT. The lack of proper apportionment of the EPT has worked to the benefit of regional and national economic centers, especially the city of Moscow. The derivation-based apportionment of VAT probably also contributed to the concen- tration of regional revenues; regional revenue disparities would have been reduced if VAT had been shared with the regions on a per capita basis, as is done in Germany, or on an estimated consumption expenditure basis, as is done with the maritime provinces in Canada. The apportionment of PIT by work location rather than by place of residence also contributed, at the margin, to the concentration of tax revenue in regions that neighbor an industrial economic base. Distribution of Revenues between Regional and Local Governments The distribution of revenues between the regions and local governments varied significantly by type of tax (table 3.4). The most centralized revenue sources at the regional-local level were the VAT and excises, while the most decentralized were property taxes and the PIT. Less wealthy regions, which were more dependent on federal transfers, generally exhibited a higher degree of decentralization. Since 1996, re- gional governments retained a higher proportion of PIT and compensated local governments with rela- tively higher transfers (Lavrov and others 1996). 50 Jorge Martinez-Vazqoez and Jameson Boex Table 3.14. Analysis of Regional NRT Collections, 1994-97 Variable 1994 1995 1996 1997 Average NRT (percentage of regional collections) 2.60 5.23 6.55 7.68 Coefficient of variation, per capita NRT collections 3.03 2.35 2.85 3.40 Correlation per capita NRT collections with Per capita GRP, 0.81 0.88 0.95 0.95 Per capita regional collections 0.80 0.85 0.94 0.96 NRT Natural resource tax. a. Data on 1996 GRP were used for 1997. Source: Authors' calculations based on data from the Ministry of Finance. However, revenue assignments within the regions continued to have very little significance given the practice of minimum expenditure budgeting. Because regional governments have discretion to change revenue-sharing rules and transfer levels, the composition of funding sources matters much less than the overall level of financing. The latter is often determined prior to any discussion on revenue assignment in negotiations between the regional and local governments. The Role of Perverse Incentives and Soft Budget Constraints A major shortcoming of intergovernmental fiscal relations in Russia is the lack of incentives to subnational governments to increase their own revenue collections and to rationalize their expenditures. During the early transition period, 1992-93, the bargained approach to expenditure levels, the regulated revenue- sharing rates, and the negotiated transfers failed to provide subnational governments with hard budget constraints, motivation to raise their own revenues, and incentives to rationalize expenditures. The re- forms of 1994, which have been in place through the end of the decade, took significant steps toward the imposition of a hard budget constraint, but ultimately fell short of this goal. Both regional and local governments still had a powerful incentive to understate and hide their revenues in extrabudgetary funds. Although the importance of extrabudgetary funds may be on the decline at the subnational level, no good measure of their importance is available (Freinkman, Titov, and Treisman 1998; Kirkow 1996). Several fiscal practices can be identified that caused perverse fiscal incentives in Russia. First, subnational governments responded to incentives, resulting from the reliance on budget expenditure norms to measure regional expenditure needs, to maintain existing infrastructure capacity rather than to rationalize their expenditures. As a result, regional and local governments failed to exploit efficiency gains by improving the mix of inputs, and they continued to have an incentive to hide funds from higher- level authorities in extrabudgetary funds. Second, poor design of intergovernmental transfers reduced the incentives for subnational govern- ments to collect revenues. For example, equalization transfers were provided to subnational govern- ments for which own-source and shared revenues were not large enough to fund minimum expendi- ture budgets. By using current or base year revenue collections as a means for evaluating fiscal capacity, this gap-filling approach ensured that any reduction in fiscal effort was offset, at least in part, by an increase in transfers. Perverse fiscal incentives were even more prominent among local govemments, which in most re- gions negotiated their budget allocations in much the same way as was done in the former Soviet Union. Local governments routinely saw a large percentage of additional revenues clawed back by regional governments. As a result, they were still more concerned with hiding fiscal resources through all sorts of extrabudgetary funds and bargaining for assistance from the upper-level government than with increas- ing tax collections. Freinkman, Titov, and Treisman (1998) argued that soft budget constraints caused increased local government spending on subsidies to housing and communal services, transportation, and agriculture and less on health and education. Rezvenue Assignments anzd Tax Administration Issues 51 The Role of Tax Administration A major difficulty that has plagued the implementation of the system of intergovernmental fiscal relations in Russia has been the inability of the federal government to exert effective control over the regional branches of the STS. The lack of a functioning treasury system during the transition also impeded the smooth functioning of revenue assignments. Administrative Structure of the State Tax Service Since the beginning of the transition, the Russian Federation struggled with regional noncompliance with federal tax legislation. For example, the revenue assignments provided in the Law on the Basic Principles of the Tax System of the Russian Federation of December 1991 were never implemented be- cause of regional opposition; regional and local offices of the federal STS took orders from the regional authorities rather than their federal supervisors. There are historical reasons for these problems with the tax administration system. Under the former Soviet Union, the system of tax administration consisted of a network of tax divisions of local and re- gional financial offices with subordination to local, republican, and federal authorities. With indepen- dence, the Russian Federation moved quickly to create a national tax administration system solely under federal authority to avoid a repeat of the Soviet collapse. Though, after the establishment of the Russian Federation, the STS was legally made a federal agency and renamed the Ministry of Taxation, in many ways it remained a confederation of local and regional offices with personnel responding to both federal and subnational authorities. Many regional and local tax administration officials still respected and took informal directives from subnational authorities in return for housing, office space, and other logistical supports. The arrangements affected regional tax revenue collections and remittances to the federal government. In substance, the inverted system of collections had not changed from the Soviet system. In certain cases, local and regional offices still remitted collections to the federal level, creating the opportunity for withholding payments. This problem is especially poignant for Tatarstan and Bashkortostan because these regions had a so-called single-channel arrangement. Of the pending tax administration reform issues, the creation of an agency that responds exclusively to federal authorities may be the most important. However, there have been no serious attempts to intro- duce policies that could address this subnational allegiance problem, such as providing all compensation for tax administrators from the federal budget or rotating personnel geographically. Many other structural issues remain to be addressed in tax administration in the Russian Federation. Are regional and local government tax administrations necessary under the draft Tax Code? Should the Ministry of Taxation and the Tax Police be integrated as in many other countries? Should the Ministry of Taxation collect the social security contributions to extrabudgetary funds? What is the role of the federal government with respect to property taxes? Until these questions can be properly answered, the Ministry of Taxation could assess and collect new regional and local taxes, with the possible exception of the property tax, while the government uses its limited resources to improve tax administration at the fed- eral level and to avoid creating new tax administrations at the regional and local levels of government. However, because the Ministry's primary mission is to raise revenues for the federal budget, it might neglect the assessment and collection of solely regional and local taxes. Consequently, it would be more effective to have local and regional tax administrations administer their own taxes in the longer term. The Ministry of Taxation also needs to play a significant role in fostering the development of transparent and fair intergovernmental relations between all government levels. The distrust between the federal- and oblast-level governments is exacerbated by the inability of the Ministry of Taxation to collect the budgeted revenues for each government level. This distrust is evidenced by the requirement that taxpayers write separate EPT checks for the federal and oblast governments. This simple but significant revenue-sharing responsibility clearly should be transferred from the taxpayer to the Ministry of Taxation. 52 Jorge Martinez-Vazquez and Jameson Boex The Treasury System Regional mistrust of the federal tax administration extends to remittance of taxes. Regional governments doubt the ability and willingness of the federal authorities to resend funds to the regions once the money gets into federal budget accounts. During the transition, Russia lacked a developed treasury system that would allow the easy transfer of funds throughout the national territory As a result, the system of inter- governmental fiscal relations was designed to minimize the occurrence of financial flows and counterflows, whereby some funds would be sent to Moscow (such as, collected federal revenues) while other funds would flow from Moscow to the regions for some other reason (such as, equalization transfers). A mere accounting issue in many other countries, in Russia the existence of counterflows significantly impacted the design of intergovernmental fiscal relations. Pending Reforms in Revenue Assignments An extensive reform agenda pending in revenue assignments pertains to both federal-regional and re- gional-local relations. Some issues at the federal-regional level include revenue sharing and apportion- ment problems, own-source revenues, extent of revenue autonomy, and vertical balance. Several ongoing reform blueprints, including those in the draft Tax Code and the Concept of Reform of Intergovernmental Fiscal Relations in the Russian Federation in 1999-2001, missed opportunities for squarely addressing these issues. The Tax Code, if implemented, would marginally address revenue autonomy. The Concept of Reform, other than recommending the use of a three-year sharing rate, re- mains silent on revenue assignment issues. At the regional-local level, the need for reform is even more dramatic because many regional governments still rely on bargaining budget levels and regulating taxes and transfers with local governments. Revenue Sharing and Apportionment Issues The VAT should not be shared with subnational government based on the location where the tax is collected. A mismatch frequently arises between the VAT payment made in a region and the actual value added in that region, making it unsuited for direct revenue sharing. Based on this argument, the VAT should be assigned 100 percent at the federal level,3 but eliminating revenue sharing from the VAT will be politically difficult or even impossible. It was tried in 1991-92 under the Law of Basic Principles of the Tax System, but regions considered VAT revenues as their own and did not comply with the law. As a result, a more appropriate alternative might be sharing the VAT using a different apportionment mechanism. For example, the VAT could be shared among the regions on a per capita basis, or founded on other simple formulas (Igudin 1998). An additional revenue-sharing problem involves the progressive PIT as currently used. It should not be assigned 100 percent at the subnational level, because in contrast with the purpose of progressive income taxation, this prevents the redistribution of resources across the national territory. By keeping the progressive PIT as a federal tax, with lower rates, and granting subnational governments the right to introduce a piggyback proportional income tax with minimum and maximum rates, regional governments would gain some spending autonomy using the same tax base as the federal PIT. A third problem with shared revenues regards the apportionment of EPT revenues. An apportion- ment formula that allots EPT revenues based on the average of payroll and property that enterprises have in different regions already exists in the tax laws, and the draft Tax Code respects this formula. 3. The arrangement between the Canadian federal government and the province of Quebec may be the only example of a workable dual VAT between the national and subnational levels (Bird and Gendron 1998; McLure 1998). This dual system requires a high level of tax administration performance that does not exist in the Russian Federation at this time. Revenue Assignments and Tax Administratioin Issues 53 However, the existing formula is used only on rare occasions. Instead, businesses pay all their taxes in the region where they are registered. The obvious solution to this problem is to require multiregional enterprises to file tax returns in each of the regions where they operate; this is a viable option only if the overwhelming filing requirements are simplified. In addition, the formula, even when implemented, would hurt regions that buy more than they produce. The equation could be adjusted by introducing a sales component, as is done in the United States, but this would further add to the formula's data re- quirements. As a result of the noted apportionment problems, the EPT should not be considered a good basis for revenue sharing. Unless an individual has multiple income sources and must file a tax return, PIT collectors accrue to the place-of-work jurisdiction and not to the place of residence. However, individual taxpayers probably consume and demand public services more in their place of residence than in their place of work. In other nations, regional income taxes are often based on the place of residence. Revenue Assignment Issues An important outstanding issue regarding own-source revenues is whether taxes on natural resources should be assigned to the regions, as is currently the case, or whether they should be reassigned to the federal level (McLure 1994; McClure, Wallich, and Litvack 1996; Mikesell 1998). The literature suggests that revenues from natural resource exploitation should generally flow to the federal budget because natural resources are often considered a possession of the whole nation. Also, accrual of natural resource taxes at the regional level could result in gross subnational inequities. Yet, fees that compensate regions for the cost of exploiting natural resources should flow to the regions. The final arrangement for the sharing of these taxes will require a national consensus and a balance between objectives involving eco- nomic efficiency and horizontal balance, on the one hand, and the rightful compensation to subnational governments for environmental damage and the exploitation of natural endowments on the other. The Tax Code reassigns the enterprise asset, land, and individual property taxes (tables 3.5 and 3.6) as one property tax at the regional level. The allocation is inconsistent with international practice, where the tax is often levied by local governments. However, in the Russian Federation, property tax assignment at the regional level may be justified for administrative reasons. Requiring regional governments to share the tax with local governments and giving local governments the ability to add to the rate of property tax, within given limits, may prove a reasonable compromise. However, until local governments can be as- sured that extra revenues will not be clawed back by the regional governments, local governments will see little advantage in introducing a shared property tax. The lack of real estate markets, land ownership, and the multiple institutional constraints are powerful obstacles for the development of an ad valorem real estate tax in the Russian Federation. At this early stage of development of the real estate property tax in Russia, the emphasis should be put on successfully finalizing the pilot work in the cities of Novgorod and Tver and extracting useful lessons for the introduction of this tax in other regions. Revenue Autonomy Issues A critical concern in the assignment of revenues is that subnational governments in Russia continue to lack adequate revenue autonomy. Revenue autonomy is the key element of a functioning system of inter- governmental fiscal relations. The draft Tax Code provides a greater degree of revenue autonomy to the regions with the introduction of the regional sales tax, which in theory would be an adequate instrument to fulfill the goals of revenue sufficiency and autonomy at the subnational level. However, the high costs of administering and complying with a regional retail sales tax are compounded by the existing VAT. A move to a nationwide retail sales tax at a rate of 5 or 10 percent and a reduction of the VAT rate to 10 percent is neither optimal nor feasible in terms of revenues. Available estimates indicate that a combination of a reduction in the tax rate for the VAT and regional sales tax with a rate of 10 percent would yield only 60 percent of the revenues now produced by the VAT. The VAT is generally a more 54 Jorge Martinez-Vazquez and Jameson Boex efficient (economically neutral) tax than a retail sales tax. The problem with the sales tax is that it is difficult to avoid the taxation of intermediate production without opening serious avenues for evasion. One of the important reasons for this difference is that the VAT is easier to collect because it is levied at different levels of production. If tax is not paid at some level, it can still be fully recovered at the next level. In contrast, a retail sales tax is only levied on final sales or at the last stage of production where it is most difficult for tax inspectors to enforce collections. In addition, international experience indicates that a retail sales tax with a rate of 10 percent could lead to considerable evasion. A second avenue of increased revenue autonomy in the draft Tax Code is the proposed regional prop- erty tax. As noted earlier, the real estate property tax has great potential, but will not be fully functional for some time. Meaningful and quicker added revenue autonomy at the subnational level could come from a piggyback-style personal income tax. This tax would use the same tax base as the national income tax and give regional governments the discretion to vary a flat rate up to a federally determined maximum. Even though a piggyback proposal has been on the table since the beginning of the transition, and even though it has gained acceptance among analytical and academic circles and is a proposed part of the 1999 budget, regions resist its introduction because they currently receive all the revenues collected from the PIT. Vertical Balance and Revenue Assignments For a decentralized system to be efficient, subnational governments must be given sufficient revenue autonomy. If taxpayers of a subnational jurisdiction find that the available resources are inadequate to finance needed government services, they have the option to change their own taxes. Thus, revenue autonomy is the key, not only to more efficient expenditures, but also to reaching vertical balance. Revenue Assignments to Local Governments Many regional governments continue to organize their relations with local governments as in the past. For these regions, revenue assignments are an afterthought in financing a negotiated minimum expendi- ture budget. On a more positive note, an increasing number of regions are imitating the reforms at the federal-regional level. The piecemeal approach to revenue assignments at the regional-local level, such as the one followed by the 1997 Law of Financial Foundations of Local Self-government, is clearly ineffective. The entire process that governs the relationships between the two levels of government must be reformed, not only addressing revenue assignments but also expenditure assignments, transfers, and the budget process itself. Federal intentions to spearhead comprehensive reform between regional and local governments raise important constitutional issues regarding the extent of federalism in the Russia. The steps that will be needed to reform regional-local relations will be similar to those for completing the reform of federal- regional relations. This is true whether the federal authorities find the ground to impose these reforms, as in a Subnational Budget Code, or the regions are left on their own to carry out these reforms. It is also clear that the regional-local reform process would benefit profoundly from rational, stable, and transpar- ent relations between the federal government and the regions. 4 The System of Transfers During the transition, intergovernmental transfers in Russia provided a relatively small share of subnational revenues; in 1997, the sum of all federal transfers to subnational governments amounted to only 14 percent of consolidated subnational revenues. However, transfers were not an insignificant fund- ing source for many regions. For the average region, transfers accounted for 20 percent of budget re- sources; for 10 regions, transfers accounted for 40 percent of available resources. Intergovernrmental transfers in the Russian system can be divided in two general categories: equaliza- tion transfers, specifically provided to minimize differences in fiscal capacities and needs across regions, and other transfers, including subventions and mutual settlements, which are mainly unbudgeted, nontransparent allocations made during budget execution. Equalization transfers were drastically re- formed in 1994 when the system of negotiated subventions was replaced with a formula-driven equaliza- tion approach. However, the negotiated and nontransparent nature of mutual settlements provided re- gions with soft budget constraints and perverse collection incentives. The negotiation of transfers at the regional-local level presented identical limitation and motivation problems among local governments as experienced at the regional level. Evolution of the Transfer System Three distinct periods identify the system of intergovernmental transfers from the federal government to the regions. These phases parallel those of revenue assignments. Transfers in the Early Transition Years, 1992-93 During 1992-93 most transfers were negotiated subventions to those regions for which own-revenues and shared revenues were not large enough to fund the minimum expenditure budget that was negoti- ated, as a first step, with the Ministry of Finance. The entire budgetary process, and by association the determination of these lump-sum subventions, lacked transparency and was subject to political manipu- lation and pressure. Despite these shortcomings, these transfers had a positive impact on equalization efforts. The same methodology of negotiated transfers was followed at the regional-local level. Transfer Reforms of 1994 In 1994 the Russian government introduced a new system of formula-driven equalization transfers that was used in earnest in 1994 and 1995; between 1996 and 1998, the formula's results were subjected to political negotiation. Alongside the formula-driven equalization transfers, the federal government continued to employ a number of other types of transfers, which were mostly of an ad hoc and nontransparent nature. The Fund for Financial Support of the Regions (FFSR), as the new system of equalization transfers was called, had two equalization windows in its formula. The first formula window, called "Regions in Need of Financial Assistance," attempted to equalize the availability of revenues across regions. The second window, called "Regions in Need of Additional Financial Assistance," was designed to provide additional funding to regions with unmet expenditure needs. The overall level of available money for the FFSR was determined every year in the federal budget. The funding rule for the FFSR has been subject to change virtually on an annual basis since the fund's 55 56 Jorge Martinez-Vazquiez and Jameson Boex introduction in 1994. Until 1994, the funding for the FFSR was set at 22 percent of the federal share of VAT collections, while funding was increased to 27 percent of federal VAT collections in 1995. For 1996 and 1997, the funding rule was changed to 15 percent of all federal tax collections, excluding import duties and the 10 percent federal share of the PIT. The 1998 budget lowered FFSR funding to 14 percent of federal collections with the same exclusions. Table 4.1 outlines the first-window formula of the FFSR. Although the basic calculations remained un- changed during the transition years, some of the coefficients were altered. For 1997, the first mathematical step used revenues per capita adjusted from a base year (the most recent year for which data were avail- able) to equalize revenues across the regions. Revenue data from an earlier year were used to reduce disin- centives to revenue mobilization. To determine FFSR transfers, the Russian Federation was divided into three groups (two groups of regions in the northern territories and one for the rest of the Federation) mainly to capture differences in the cost of living.! Within each of these three groups, regions for which adjusted per capita revenue collections were below 92 percent of the group average were entitled to capacity-equalizing transfers. Steps two and three of the first-window calculations determined the size of the transfer by figur- ing the available funds to those regions with a positive claim as determined in the first step. The formula for the second window of the FFSR (table 4.1), in its 1997 version, was based on expendi- ture data for the base year of 1991. Until 1996, the base year for measuring expenditure needs was 1993; in 1997, the previous year was considered more representative of actual expenditure needs. To approximate the 1997 expenditure-levels needs for all regions, base year expenditure data were adjusted for changes in legislation for the years in between. Revenue capacity for the second window was determined the same way as for the first window. Each transfer claim was computed in the second window formula as the difference between the sum of adjusted revenues, and the transfer from the first window, if any, and the estimated expenditure needs. The remaining window steps calculated the funding amounts for those regions with positive expenditure gaps. Significant weaknesses of the FFSR system included the following: * After 1995, the transparency and objectiveness of the FFSR approach was compromised when the formula results were subjected to political negotiations. * The equalizing capacity of the system was limited because the funds available for equalization were a relatively small share of the subnational budget. * Continued changes in the formulas for both windows turned many regions against the system, as the alterations did not hold them harmless against losses of funding. * A widespread perception was held that the mechanism did not achieve the goal of equalizing resources between regions. * The use of actual revenue collections and actual expenditure per capita data from a base year did not properly measure fiscal capacity and expenditure needs. For instance, the assumption that expendi- tures for 1991 reflected the regional spending needs of 1997 was unrealistic, especially in light of the sweeping changes in the economic structure of Russia during the transition period. * The division of regions into geographical groups may have prevented equalization across groups and did not adequately address differences in the cost of living across regions. * The formula for the second window was based on the "gap filling" philosophy of the subvention system that was used during 1992-93 and previously in the Soviet Union. * The gap-filling approach and the annual adjustment process introduced perverse incentives, as regions that increased revenue collections had a legitimate concern that their FFSR transfer would be reduced. In addition to the FFSR monies, several other types of transfers were used from 1994 through 1998, none based on formulas or rules, and in some years, they allocated more funds than did the FFSR (Freinkman, Titov, and Treisman 1998; IEPT 1998; Morozov 1998; Titov 1997). The most important of these other forms of transfers were "mutual settlements." Technically, mutual settlements were transfers 1. In 1996 and before, the Federation was divided into 11 groups of regions to allow for cost-of-living differences. The System of Transfers 57 Table 4.1. Fundfor Financial Support of the Regions First window: regions in need offinancial assistance (variables arefrom base year adjustedfor legislation changes) Step 1: . Average revenue Per capita \ Claim by =* per capita in the *0.92 - reenues in * Populaion of group of regions region r [Average expenditure per capita ] Step 2: (For those regions with a positive claim in step 1) Share of region i Claim by region i in the pool offunds =S, Total claimsfrom for window I all regions Step 3: Actual equalization Pool offunds grant to region i - s * availablefor under window I window I Second window: regions in need of additionalfinancial assistance (variables arefrom base year adjustedfor legislation changes) Step 4: (lai7 ib?,y urrent expenditures Rer en es Equalization Claim by Curretexpenditure Reeue grntfor region j regionj= in regionj in region + in window I Step 5: (For those regions with a positive claim in step 1) Shared region j Claim by region j in the pool offunds = S,,, = Total claimsfrom for window 11 all regions Step 6: Actual eqoialization Pool offunds grant to region j - S availablefor under window II "' window II Source: Martinez-Vazquez and Boex (1997). arranged after the annual budget law had been adopted. However, in actuality mutual settlements formed an umbrella category that included transfers given in lieu of financing regional investment programs in the federal budget, to compensate regional governments for the mandated delivery of federal programs, and as emergency aid (Morozov 1998). In the case of many federal mandates, current budget legislation did not require that these expenditures be compensated through mutual settlements. Instead, these ex- penditures were included into the estimate of the expenditure needs of the regions, so equalization transfers supposedly included funding for these federal mandates. However, in reality many federal mandates did not get compensated either through mutual settlements or through equalization transfers. 58 Jorge Martinez-Vazquez and Jameson Boex Many different forms of negotiated and discretionary funds were included in the mutual settlement umbrella. After 1994, a key component of mutual settlements was ex post, negotiated payments to subnational governments for housing maintenance. Though these transfers were determined during the budget execution process, all but a handful of regions were recipients of mutual settlement funds. Other transfers used in the Russian system of intergovernmental fiscal relations included grants to Moscow for performing the role of capital city, Sochi, the northern territories for the delivery of goods, and "closed" cities that were part of the military-industrial complex and subject to maximum secu- rity.2 Budget loans issued from the Ministry of Finance to regional governments often became trans- fers when they were written off after the close of the fiscal year. Some regions also received federal support to support particular sectors or enterprises, such as the coal industry or agriculture. As in the case of mutual settlements, many of these other transfers were assigned as needed and lacked trans- parency (Morozov 1998). Transfer Reforms of 1998 The general dissatisfaction with the FFSR mechanism, which was used since 1994, led to its gradual demise. For 1999, the Russian government introduced a new simplified equalization formula for the FFSR (table 4.2). The crux of the new formula consisted of index of budget expenditures, or expenditure needs, to "normalize" or adjust per capita revenues. The new formula assigns equalizing transfers to regions for which the normalized per capita revenue falls below a determined threshold. The Current System of Transfers Three elements of the current system of transfers are considered in this section: equalization transfers; other transfers, including mutual settlements; and regional-local transfers. Because of the high frequency with which budget loans are forgiven, they are sometimes considered transfers, but they will be consid- ered as part of deficit finance in chapter 5. Equalization Transfers The 1998 formula for the FFSR was an improvement over the past formula; it was simpler and got closer to the ideal of using measurements of fiscal capacity and expenditure needs rather than actual revenues and expenditures for a base year as the standards for equalization (Boex and Martinez-Vazquez 1998a). Like the original FFSR mechanism, the new formula had two components: the determination of funding and the equalization formula. The overall level of funding for the FFSR was still determined in the an- nual budget as a share of federal tax revenues, excluding import duties. The basis of the new equalization instrument was the measure of revenue collections (or fiscal capac- ity in step 2) and a measure of budget expenditures in step 3. In step 4, the fiscal capacity of each region was normalized by the index of expenditure needs. Each region that has below average, normalized, per capita revenues could potentially receive an equalization transfer. However, if sufficient funds were un- available to equalize all below-average regions, the equalization threshold was gradually lowered until the neediest regions could be pulled up to the threshold level (steps 5 and 6). The specific measures of fiscal capacity and fiscal need were crucial to the successful performance of the new FFSR transfer instrument. The updated equalization formula used per capita revenues as a proxy for fiscal capacity; in 1999, per capita revenues were based on 1997 data. Annual adjustments to the measure will be made until per capita revenues can be substituted with a more appropriate measure of regional fiscal capacity. For expenditure needs, differences in the cost of living across groups of regions 2. Closed city budgets were handled directly by the Ministry of Finance. Residents are reputed to have en- joyed a privileged standard of living (Campbell 1996). The System of Transfers 59 Table 4.2. Formulafor Allocating FFSR Transfers, 1999 Step 1: Determine the amount of funding for the FFSR Step 2: Determine per capita revenue (a measure of fiscal capacity): RPC -j Step 3: Determine index of budget expenditures (a measure of expenditure need): K. =zH / min, , 1l, Step 4: Determine normalized per capita revenue: RPC RP; = KRc Step 5: Assign equalization transfers (t) for regions for which RP < RP: t, = ( RP- - RP,) K; N;1~ Step 6: If necessary, only equalize to threshold RP.: X RPij Ni Ki i=] E Ni Ki ,=1 Note: In Step 2, R, is adjusted revenue collections (a measure of fiscal capacity), and Nj is population; in Step 3, Hq, is the per capita expenditure norm for region i and expenditure category j. When the amount of transfers assigned in Step 5 exceeds the size of the FFSR, the equalization threshold is lowered to RP,, which is determined in an iterative process that simultaneously determines the number of regions eligible for equalization funds (m). Source: Boex and Martinez-Vazquez (1998a). were assessed for 1999. The Ministry of Finance expects to develop a comprehensive set of norms to estimate differences in expenditure needs across regions. The selected measures of fiscal capacity and need presented two main areas of concern with the new FFSR formula. With regard to the measure of fiscal capacity, using actual collections data for 1997 for the 1999 budget continued to send the wrong message to the regions regarding their levels of tax effort. When up-to-date data for tax bases or the composition of GRP become available, it will be possible to improve on the measure of fiscal capacity. Another concern is the proposed use of phased-in budget norms during 1999 for housing and utilities, health care, education, social welfare, public transportation, culture, and public administration. By 2000, all regional expenditure needs will be solely determined on the basis of norms. Numerous attempts have been made in the past years to estimate different indexes of need. Lavrov (1998) discussed these indexes, including those developed by the Ministry of Finance, the Ministry of Economy, the Central Research Institute of Economics, and the State Committee for Construction. He found that there are very signifi- cant correlations among most of them. Whether the plans are to go back to using "physical norms" (as those used in the Soviet era) or more broadly defined per capita norms is not clear. While the use of cost or funding standards in the formulation of the budget presents no problem, the explicit use of legislative norms raises major concerns. BUDGETING FROM NoRMs RESULTS IN SUBoPTiMAL FISCAL MANAGEMENT. A conspicuous feature of budget preparation in Russia is the philosophy that budgets should be based on a perspective of needs rather than from a perspective of feasible public services given the current revenue resources. This perspective leads to unrealistic budgets and poor fiscal performance. 60 Jorge Martinez-Vazquez a7nd Jameson Boex All budget practitioners use some form of norms, or expenditure benchmarks, to compile a budget. The process becomes problematic when norms for different functional expenditure categories are estab- lished by law before funding implications are considered. This causes taxpayers to expect a certain level of public service provision even though sufficient funds may be unavailable to deliver all programs. In contrast, de facto norms can be used to assist budget practitioners in successfully allocating funds within the relevant budget constraint without raising such expectations. Also, in a decentralized federalist system the role of the federal government in providing for equaliza- tion between different regions is often less extensive than in the Soviet tradition. A certain degree of variation in regional expenditures should be considered beneficial as it allows regional governments to adjust spending patterns to region-specific concerns and to respond to regional demands for public ser- vices. In contrast, budgetary norms convert the process into an input-oriented activity, making budget performance and the level and quality of services more difficult to assess. PHYSICAL NoRMs PROVIDE MISDIRECTED FISCAL MANAGEMENT INCENTIVES. Budgetary norms based on physi- cal measurements provide incentives that are incompatible with the efficient allocation of budgetary resources. For example, physical norms based on infrastructure capacity, such as the number of schools, number of hospitals, and so forth, give subnational governments an incentive to generate excess capacity or preserve obsolete capacity. In addition, the use of physical expenditure norms in the assignment of transfers may actually cause greater inequality between regions. Greater disparities could occur if wealthy or politically fa- vored regions enlarged their physical capacity, for example, more hospitals or universities than poorer or less-favored regions with similar public service needs. Under a system of physical expenditure norms, the wealthier regions would be considered to have greater need and thus receive a greater share of the equalization fund. It should be noted that such problems can be avoided by using per capita monetary norms instead of physical norms. If expenditure need were evaluated by nonphysical measurements, public officials would have an incentive to close rather than maintain schools and clinics with excess capacity and to allocate the available funds to improve the quality of service at the remaining locations. NoRMs ARE NONTRANSPARENT AND ADMINISTRATIVELY COSTLY. A large number of norms cause complex, nontransparent, and costly budget administration. The practice of setting expenditure norms in isolation of revenue availability is likely to result in public disenchantment, as taxpayers feel cheated by unful- filled promises and disengage from the political process. If revenues fall short of the announced norms, the expenditure norms must be reduced through negotiations to achieve budgetary balance, and over several years, the measure of expenditure needs may become so convoluted that the budget process will be reduced to a de facto system of politically negotiated transfers. Other expenditure-need measures could be used more effectively in the allocation of equalization transfers. One measure that received increased attention from the Russian government in the late 1990s was the Representative Expenditure System, a measure of expenditure needs developed by the U.S. Ad- visory Committee on Intergovernmental Relations (1990). Although this approach is theoretically sound, its success depends on budgetary discretion and revenue autonomy by subnational governments at lev- els inconsistent with those in place during the transition. Mutual Settlements and Other Transfers Despite the continued reform of the FFSR mechanism, the Russian government continued to rely on mutual settlements and other ad hoc transfers. However, mutual settlements were the dominant form of nonequalizing transfers, accounting for approximately one-fourth of all transfers, including the FFSR, and three-fourths of nonequalization transfers. Mutual settlements were employed before the transition to compensate the budgets of lower-level governments for legislated tax changes or the imposition of The System of Transfers 61 new expenditure mandates. During the transition, mutual settlements became mostly a one-way stream of transfers from the federal to regional governments, while the scope of mandate settlements grew to include emergency aid, block grants, and settlements to substitute for the financing of regional invest- ment programs (Morozov 1998). A final group of transfers consists of block grants and subsidies. Subventions were earmarked grants for capital expenditures or current expenditures and allocated by the State Duma after negotiations with the government and the regions. These grants accounted for approximately one-fourth of all nonequalization transfers but only one-tenth of all transfers. Two main issues arise with the allocation of nonequalization transfers. First, these transfers fail to provide a stable budgetary environment as none of them are allocated based on objective, predetermined, and transparent criteria. Subventions are allocated through bargaining and budget negotiations; mutual settlements most often are not budgeted at all. Second, in combination with the availability of budget loans and other means of financing for current expenditures, mutual settlements provide a soft budget constraint for regional governments. Regional-Local Transfers To a large extent regional-local relations during the transition propagated the system practiced during the Soviet regime. Often, minimum expenditure budgets were negotiated between the regions and local governments. The revenue gap between local expenditure needs and own- and shared-revenue sources was addressed using negotiated gap-filling transfers. The continued reliance on this outdated approach preserved the major incentive problems that were prevalent under the previous system. Some regions, including Sverdlovsk and Penza, introduced standard rates for revenue sharing within the regions (Igudin 1998). When revenues were shared at a standard rate, the local allocations remained relatively low to avoid wealthier districts and municipalities from attaining large fiscal surpluses, leav- ing an increased role for transfers at the regional-local level. The analysis of one specific region may provide some additional insights into the issues surrounding intraregional relations. Recently, Bahl and others (1998) provided a comprehensive review of intergov- ernmental transfers in Leningrad oblast. Leningrad oblast used three types of grants for its local govern- ments: earmarked subventions to cover price subsidies for medicines and to partially compensate com- munications companies for mandated rate reductions for pensioners and others; subsidies, which are paid to local governments essentially as gap-filling transfers to cover the general shortfall between the minimum expenditure budget and other revenue sources; and mutual settlements, ex post unbudgeted transfers. Mutual settlements paid within the Leningrad oblast accounted for 45 percent of all regional transfers, much more than between federal-regional entities. The gap-filling and earmarked subsidies accounted for 45 and 10 percent, respectively, of all regional transfers (Bahl and others 1998). How Has the Transfer System Performed? The data on equalization transfers appear to support the use of transfers as an effective equalization tool between the regions of the Russian Federation. However, the transfer system had significant failings, including limited funding, increased expenditure disparities among regions, and disincen- tives for tax efforts. Decreased Transfer Amounts during the Transition At their peak in 1994, on average transfers represented 35.8 percent of funding for regional governments. As shown in table 4.3, by 1997 the significance of transfers and other settlements had declined to only 20.6 per- cent. The intergovernmental transfers as a subnational source of funding was largely replaced by "other sources of financing," mainly consisting of budget loans and different forms of borrowing. The most significant 62 Jorge Martinez-Vazquez and Jameson Boex Table 4.3. Funding Sourcesfor the Regions, 1992-97 (average funding by source as percentage of total expenditures) Variable 1992 1993 1994 1995 1996 1997 Own, shared, and other regional revenues 87.2 75.7 66.5 73.6 68.5 67.9 FFSR transfersa 0.0 0.0 9.4 17.3 15.0 16.4 Other transfers and settlements 16.3 30.1 26.4 5.6 7.6 4.2 Other sources of financingb -3.5 -5.8 -2.3 3.6 9.0 11.5 a. The FFSR was introduced in 1994. In earlier years equalization was achieved through variable sharing rates and other equalizing transfers. b. Determined as expenditures minus revenues and transfers. Consists mainly of budget loans and borrowing. Source: Freinkman, Titov, and Treisman (1998); Ministry of Finance data. decrease in transfers took place in 1995; while mutual settlements and other (nonequalization) transfers on average represented 26.4 percent of all regional funds in 1994, this percentage dropped to 5.6 percent in 1995. The relative role of equalization transfers in subnational budgets remained fairly constant between 1995 and 1997. In 1997, for an average region, FFSR transfers made up 16.4 percent of subnational public funds. The level of federal transfers to the regions expressed as a share of gross domestic product (GDP) provides another useful perspective on the relatively small significance of these funds (table 4.4). In 1997, total federal transfers, excluding budget loans, represented 1.92 percent of GDP. When budget credits and loans from the federal government to the regions were considered, this figure increased to 2.67 percent for 1997. These 1997 figures are representative of transition years 1992-93 and 1995-97. In 1994, the introduction of the FFSR and a sharp increase in mutual settlements pushed total federal transfers and loans to the regions to 3.84 percent of GDP. In real terms, per capita transfers have vacillated during the transition. In real terms, federal transfers to the regions, excluding budget loans, were reduced by half between 1994 and 1997 (table 4.5). The FFSR, by design, was an equalizing mechanism. Yet, the equalizing capacity of the FFSR has been placed into question since its introduction in 1994. However, the equalizing role of the FFSR cannot be consid- ered in a vacuum. Did other transfers have an equalizing effect on regional budgets? If not, to what extent did these transfers offset the equalization obtainable through the FFSR? It is important, therefore, to consider the equalizing effect not only of the FFSR, but of the entire system of transfers, and ultimately of the entire system of intergovernnental relations. Have Transfers Been Equalizing? Whether the FFSR has actually been equalizing has been one of the most analyzed issues in Russian fiscal federalism, and, to some extent, the answer to this question depends on how one chooses to answer it. The influence of economic factors on the allocation of transfers is analyzed in tables 4.6 and 4.7. Table 4.6 presents Table 4.4. Federal Transfers to the Regions, 1992-97 (percentage of GDP) Transfers 1992 1993 1994 1995 1996 1997 Total federal transfers and loans 1.67 2.57 3.84 2.18 2.80 2.67 Total federal transfers 1.58 2.53 3.82 2.13 2.58 1.92 Subventions 0.75 0.66 0.43 0.12 0.12 0.13 FFSR 0.00 0.00 0.37 1.25 1.18 1.27 Mutual settlements 0.83 1.85 2.93 0.70 0.14 0.43 Other transfers 0.00 0.02 0.09 0.06 0.14 0.09 Budget credits and loans 0.09 0.04 0.02 0.05 0.22 0.75 Note: Data may not sum to total because of rounding. Source: Ministry of Finance data. Table 4.5. Transfer Compositionfor Subnational Governments in Real Terms, 1992-97 In real terms: per capita 1995 ruble In real terms: index (1994 = 100) Category 1992 1993 1994 1995 1996 1997 1992 1993 1994 1995 1996 1997 Transfers (excludes budget loans) 106.0 340.2 468.0 191.8 211.0 193.6 22.7 72.7 100.0 41.0 45.1 41.4 FFSR 0.0 0.0 95.6 128.0 109.1 131.5 0.0 0.0 100.0 133.9 114.1 137.5 Mutual settlements - 255.7 307.2 47.1 85.2 44.6 - 83.3 100.0 15.3 27.7 14.5 Total revenues (excludes loans and borrowing) 1,831.2 2,165.1 2,050.6 1,583.9 1,497.1 1,632.9 89.3 105.6 100.0 77.2 73.0 79.6 - Not available. Source: Ministry of Finance data. 64 Jorge Martinez-Vazquiez anid Jamesoni Boex Table 4.6. Regression Analysis: Per Capita FFSR Transfers, 1994-97 Independent variable 1994 1995 1996 1997 Intercept -1,743.0 -4,219.4 -3,450.8 -5,000.5 Per capita GRP (thousands) -13.0 -18.6 -7.6 -10.3 Cost of living index 6.8 16.6 13.1 18.9 Percentage of the population younger than working age 26.3 72.0 65.3 93.8 Percentage of the population older than working age 37.3 69.8 50.7 72.1 Percentage of the population in poverty -1.4 -4.5 0.3 -1.8 Tax effort -8.6 -1.0 -2.3 -3.0 R2 0.68 0.81 0.76 0.74 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Source: Authors' calculations based on data from the Ministry of Finance. Table 4.7. Regression Analysis: Per Capita Other Transfers (excluding Buidget Loans), 1994-97 Independent variable 1994 1995 1996 1997 Intercept -2,569.6 -1,509.4 -1,306.9 -1,498.2 Per capita GRP -55.2 -6.7 -0.6 -6.1 Cost of living index 12.3 6.5 6.1 0.8 Percentage of the population younger than working age 37.3 16.3 15.0 -29.4 Percentage of the population older than working age 29.1 24.2 24.5 -35.4 Percentage of the population in poverty -1.1 0.9 -1.3 -0.8 Tax effort 14.6 4.9 3.0 1.5 R2 0.79 0.63 0.59 0.36 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Source: Authors' calculations based on data from the Ministry of Finance. regression analysis of per capita FFSR transfers by per capita gross regional product (GRP) and several mea- sures of need: cost of living, percentage of the population under and over working age, and percentage living in poverty (see table 2.10 for sample descriptive statistics).3 For all years the regression coefficients for per capita GRP were negative and statistically significant for 1995-97, demonstrating that wealthier regions received fewer FFSR transfers. The positive and statistically significant regression coefficients for cost of living and young and old population also support the notion that FFSR transfers worked to equalize fiscal needs. However, the percentage of the regional population living in poverty did not affect FFSR allocations in any of the years under consideration. The regression analysis of per capita, non-FFSR transfers for 1994-97 (table 4.7) tells a similar, but not as compelling, story of equalization. The category "other transfers" was defined to include subventions, mutual settlements, and some other types of minor transfers, but excluded budget loans. All the regression coeffi- cients for per capita GRP were negative but were only statistically significant for two out of the four years studied. Stewart (1996) noted that both FFSR and other transfers can be interpreted as equalizing in relation to expenditure needs. Regions with higher expenditure needs, as measured by the cost of living and percentage 3. The dependent variables in tables 4.6 and 4.7 do not take into account the fact that transfers to different regions may actually provide different levels of fiscal support due to regional price differences. When all else is equal, a federal transfer to a region with a relatively lower price level has a more equalizing effect. Therefore, the regressions were also estimated using real per capita transfers levels, adjusted for the regional cost of living, as dependent variables. The regression results and conclusions were unaffected by this adjustment. The System of Transfers 65 of the population under and over working age, were also positive but not always significant from 1994 through 1996. However, in 1997 these transfers became disequalizing in terms of the needy population; the regression coefficients for the young and old population were negative and statistically significant. This suggests that in 1997 regions with a disproportionately small, economically active population received fewer non-FFSR trans- fers. Based on the data presented, one can conclude that FFSR transfers were equalizing between 1994 and 1997. The incidence analysis in table 4.6 further suggests that economic factors were important determinants in FFSR allocations. This was not the case for the other transfers analyzed in table 4.7. The explanatory power of the economic variables included in the regression has steadily declined over time; by 1997, economic vari- ables accounted for only roughly one-third of the variation in these transfers. According to the regression analyses shown in tables 4.8 through 4.11, transfers had a limited impact on reducing fiscal disparities. These tables considered five dependent variables in per capita terms-total collec- tions in the regions, including the federal portion of shared taxes; regional revenues before transfers of any kind; regional revenues after FFSR transfers; regional revenues after all transfers; and regional expenditures. Table 4.8. Regression Analysis: Allocation of (Per Capita) Regional Resources, 1994 Total Regional After After all Regional Independent variable collections revenue FFSR transfers expenditures Intercept -932.9 -1,746.2 -3,489.2 -6,058.9 -6,147.0 Per capita GRP 182.2 138.8 125.8 70.6 60.4 Cost of living index -0.4 2.4 9.2 21.6 22.7 Percentage of the population younger than working age 11.4 30.4 56.7 94.0 91.0 Percentage of the population older than working age -4.7 15.1 52.4 81.5 84.9 Percentage of the population in poverty 0.3 -0.7 -2.1 -3.1 -3.2 Tax effort 44.2 29.3 20.7 35.3 35.3 R 2 0.98 0.89 0.94 0.90 0.89 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Source: Authors' calculations based on data from the Ministry of Finance. Table 4.9. Regression Analysis: Allocation of (Per Capita) Regional Resources, 1995 Total Regional After After all Regional Independent variable collections revenue FFSR transfers expenditures Intercept -4,352.6 120.5 -4099.0 -5,608.3 -6,690.1 Per capita GRP 268.1 100.6 82.0 72.3 61.1 Cost of living index -1.6 5.55 22.1 28.7 33.2 Percentage of the population younger than working age 51.6 16.9 55.1 71.4 82.2 Percentage of the population older than working age 39.2 -21.9 47.9 72.1 68.0 Percentage of the population in poverty 8.5 -5.6 -10.1 -9.3 -7.9 Tax effort 88.8 32.2 31.1 36.0 66.5 R 2 0.98 0.91 0.91 0.90 0.88 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Source: Authors' calculations based on data from the Ministry of Finance. 66 Jorge Martinez-Vazquez and Jameson Boex Table 4.10. Regression Analysis: Allocation of (Per Capita) Regional Resources, 1996 Total Regional After After all Regional Independent variable collections revenue FFSR transfers expenditures Intercept -4,305.9 -878.6 -4,329.4 -5,636.3 -13,435 Per capita GRP 254.3 127.5 119.9 119.3 115.7 Cost of living index -4.4 8.3 21.4 27.5 48.8 Percentage of the population younger than working age 44.1 -4.4 61.0 76.0 202.8 Percentage of the population older than working age 38.9 -2.2 48.4 72.9 203.6 Percentage of the population in poverty 4.6 -5.9 -5.67 -6.9 -11.7 Tax effort 113.3 24.1 21.8 24.8 48.1 R 2 0.96 0.92 0.88 0.89 0.92 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Source: Authors' calculations based on data from the Ministry of Finance. Table 4.11. Regression Analysis: Allocation of (Per Capita) Regional Resources, 1997 Total Regional After After all Regional Independent variable collections revenue FFSR transfers expenditures Intercept -9,363.2 -4,187.1 -9,192.0 -7,693.9 -16,465 Per capita GRP 348.0 210.6 200.3 194.2 200.2 Cost of living index -4.9 3.57 22.4 23.2 48.2 Percentage of the population younger than working age 106.5 63.0 156.8 127.4 264.8 Percentage of the population older than working age 135.4 51.7 123.8 88.3 240.8 Percentage of the population in poverty 17.0 -1.0 -2.8 -3.6 -4.22 Tax effort 115.2 27.5 30.2 31.7 43.7 R 2 0.97 0.96 0.93 0.94 0.94 Note: Italics indicate statistical significance at the 5 percent level. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Source: Authors' calculations based on data from the Ministry of Finance. With a few exceptions, the transfers reduced the role of fiscal capacity, measured by per capita GRP, in the allocation of subnational resources. In addition, transfers were allocated to regions with higher expenditure needs (as proxied by cost of living and percent old and young population). Ultimately, the equalizing impact of transfers was weakened considerably in later transition years and, as the regres- sion results for per capita regional expenditures show for differences in regional fiscal capacity, created substantial disparities in regional expenditure levels. Have Transfers Discouraged Tax Effort by Regional Governments? One of the most serious criticisms of the system of transfers in force during 1994-97 was that its formulas gave regions the perverse incentive to lower their tax effort. This issue is explored in table 4.12. The dependent variable, tax effort, is defined as the level of regional tax collections expressed as a percentage of GRP. The same set of regression estimates, but with tax effort defined in terms of The System of Transfers 67 overall collections, including the federal share, is presented in table 4.13. The results are similar to those obtained when only regional revenues are incorporated in the measure of tax effort. The regression coefficient for FFSR transfers, defined as a percentage of total regional expenditures, was negative for all four years and, except for 1997, was statistically significant. The negative impact of FFSR transfers on tax effort is not extremely pronounced, but neither is the impact immaterial. For example, in 1996 for every 10 percent increase in FFSR transfers as a share of the regional budget, the tax effort ratio decreased by one-half percent. The empirical evidence indeed appears to suggest that FFSR transfers discouraged tax effort during the transition. However, shortcomings in the tax effort definition allows for alternative explanations. Perhaps GRP is a poor measure of regional fiscal capacity because it does not account for the economic composition of a region. For example, the measure of GRP may overstate the fiscal capacity of regions that rely heavily on agricultural production. If these agricultural regions have limited ability to raise revenues, then the measure of tax effort would understate their actual level of effort; as a result, the regression parameters would reveal a negative relationship between tax effort and FFSR transfers. Also, the measure of fiscal effort may also fail to take into account higher expenditure needs for certain types of regions. If regions for which GRP overstates their fiscal capacity systematically have greater legitimate expenditure needs, such as for maintenance of infrastructure, the allocation of FFSR funds would be greater to regions with a lower level of measured tax effort. Discouragement of tax effort was more pronounced at the local level where negotiated minimum expenditure budgets, regulated sharing rates, and negotiated transfers often prevailed throughout Table 4.12. Explaining Tax Effortfor Regional Collections, 1994-97 Independent variable 1994 1995 1996 1997 Intercept 6.40 8.35 11.36 6.65 Per capita GRP -0.46 -0.06 -0.02 0.05 Cost of living index 0.06 0.03 0.02 0.02 Business taxes 0.08 0.06 0.06 0.07 Personal income tax -0.12 -0.12 -0.09 0.00 Sales taxes 0.14 0.07 0.01 0.12 FFSR transfer (percentage of expenditures) -0.14 -0.04 -0.05 -0.03 R 2 0.48 0.45 0.34 0.41 Note: Italics indicate statistical significance at the 5 percent level. Expressed as percentage of total regional revenues. Source: Authors' calculations based on data from the Ministry of Finance. Table 4.13. Explaining Tax Effortfor Overall Collections (including Federal Share), 1994-97 Independent variable 1994 1995 1996 1997 Intercept 28.5 38.3 47.0 36.4 Per capita GRP -0.64 -0.06 -0.02 0.09 Regional share of collections -0.25 -0.31 -0.40 -0.42 Cost of living index 0.08 0.04 0.02 0.03 Business taxes 0.09 -0.01 -0.01 0.09 Personal income tax -0.17 -0.10 -0.10 0.01 Sales taxes 0.17 0.04 -0.07 0.23 FFSR transfer (percentage of expenditures) -0.21 -0.10 -0.09 -0.02 R 2 0.57 0.73 0.66 0.51 Note: Italics indicate statistical significance at the 5 percent level. Expressed as percentage of total regional revenues. Source: Authors' calculations based on data from the Ministry of Finance. 68 Jorge Martinez-Vazquez and Jameson Boex the transition. Zhuravskaya (1998) studied the budgets of 35 large cities in 29 regions from 1992 through 1997 and found that for every ruble that local government own revenues increased in a given year, approximately 90 kopeks were clawed back by reductions in transfers or tax sharing by the regional governments. Differences in Budgeted and Actual Transfers One important source of dissatisfaction among the regions with the FFSR was the considerable time lag between the crediting of taxes to the federal budget and the final remittance of funds to the regions from the FFSR. Indeed, delays were only part of the problems in administering the FFSR. The record on actual imple- mentation of budgeted transfers was poor. As table 4.14 shows, the budgeted amounts and the actual amounts received in the execution of the budget deviated in both directions. These variations for 1995, 1996, and 1997 defy easy explanation; anecdotal evidence suggests that they may have resulted because of political factors. However, the extensive use of budgetary loans may have also contributed to the disparities. Pending Policy Agenda for Transfer Reform The overhaul of the entire system of transfers is crucial to the successful reform of intergovernmental relations in the Russian Federation. Without addressing the problems with the transfer system as a whole, subnational governments will continue to face fiscal incentives that reward irresponsible fiscal behavior. The reform agenda with regard to transfers should contain reform proposals in three broad policy areas. First, improvements to the new FFSR equalization mechanism should be considered. Second, sweeping reforms are needed to replace the current system of nontransparent and ad hoc mutual settlements, forgiven loans, and special subventions, with a system of budgeted, objective, and trans- parent transfers that target federal government objectives other than equalization. Third, regional- local transfers must be redesigned in the context of wider reforms in revenue assignments and the assignment of expenditure responsibilities. Reforming the System of Equalization Transfers The introduction of the formula-driven equalization instrument in 1994 and its further improvement in 1998 were two major steps in the direction of an objective, transparent, and effective equalization mecha- nism. Nonetheless, a number of issues prevented the FFSR from achieving equalization among regions. THE FUNDING RULE. The determination of appropriate FFSR funding level is clearly a political choice; a balance between the objective of equalization and spending on other government programs must be achieved. Lawmakers must also realize that the redistribution of funds to poorer regions inevitably re- quires taxing wealthier regions, a process which may slow overall economic growth. They also must consider the tradeoff between stability and flexibility when they determine a funding rule for FFSR. The current approach of deciding the level of funds in each year's budget provided the federal government with desirable flexibility but sacrificed certainty for regional governments. Many countries fix the overall funding as a percent of central government revenues for a period of three years, which can only be altered in the case of a stipulated emergency situation. This approach may prove to be a workable compromise for the Russian Federation. Even with an efficient FFSR mechanism, the method of its implementation must be considered. Equalization can be achieved either through vertical arrangements using federal sources or through a horizontal or "fraternal" approach. In the latter case, wealthier regions would contribute some of their own revenues to an equalization fund. The "fraternal" approach, common in Scandinavian and some Central European countries, and also partially adopted in the Baltic states, is unlikely to be accepted in Russia. The System of Transfers 69 Table 4.14. Executed Share of FFSR in Each Region, 1995-97 (percentage of budgeted amount) Region 1995 1996 1997 Adygeya, Republic of 96.1 115.3 178.0 Aginsk Buryat AR 100.4 102.5 110.0 Altay Krai 105.9 98.1 112.0 Altay, Republic of 111.0 137.6 172.0 Amur Region 100.5 77.8 116.0 Arkhangel'sk Region 103.0 136.2 74.0 Astrakhan Region 106.0 106.1 106.0 Bashkortostan, Republic of Donor Donor Donor Belgorod Region 80.9 136.4 80.0 Bryansk Region 97.9 134.1 84.0 Buryatia, Republic of 101.5 77.4 143.0 Chelyabinsk Region 72.4 121.4 81.0 Chita Region 97.7 87.7 103.0 Chukotka AR 111.7 84.1 128.0 Chuvashia, Republic of 107.0 101.6 122.0 Dagestan, Republic of 133.5 87.8 111.0 Evenk AR 94.7 129.0 153.0 Ingushetia, Republic of 101.0 107.6 216.0 Irkutsk Region 86.2 105.1 73.0 Ivanovo Region 101.7 82.7 113.0 Jewish AR 156.2 100.7 105.0 Kabardin-Balkar, Republic of 99.0 105.7 138.0 Kaliningrad Region 94.9 73.3 123.0 Kalmykia, Republic of 98.0 98.2 120.0 Kaluga Region 94.2 120.5 96.0 Kamchatka Region 94.5 39.5a 74.0 Karachevo-Cherkess, Republic of 94.9 90.0 132.0 Karelia, Republic of 50.0 103.7 115.0 Kemerovo Region 95.3 136.6 83.0 Khabarovsk Krai 93.2 123.9 101.0 Khakassia, Republic of 99.4 91.8 100.0 Khanty-Mansi AR Donor Donor Donor Kirov Region 92.2 96.8 113.0 Komi, Republic of 87.6 157.2 111.0 Komi-Permyak AR 101.4 89.1 102.0 Koryak AR 103.3 147.4 95.0 Kostroma Region 93.1 132.0 87.0 Krasnodar Krai 93.8 89.2 116.0 Krasnoyarsk Krai Donor Donor Donor Kurgan Region 91.6 105.5 110.0 Kursk Region 94.6 105.5 96.0 Leningrad Region 104.5 73.6 114.0 Lipetsk Region Donor Donor Donor Magadan Region 101.0 100.6 108.0 Mari El, Republic of 102.2 105.0 109.0 Mordovia, Republic of 101.6 98.9 121.0 Moscow City Donor Donor Donor Moscow Region 74.1 50.1 b 69.0 (table continues onfollowing page) 70 Jorge Martinez-Vazquez and Jameson Boex Table 4.14 continued Region 1995 1996 1997 Murmansk Region 90.0 25.8 C 67.0 Nenets AR 94.3 90.7 142.0 Nizhny Novgorod Region Donor 114.2 94.0 North Ossetia, Republic of 95.2 112.2 99.0 Novgorod Region 107.7 98.3 105.0 Novosibirsk Region 103.1 112.4 141.0 Omsk Region 94.4 214.3 74.0 Orenburg Region 97.1 118.4 102.0 Orlov Region 102.1 91.6 149.0 Penza Region 95.4 126.6 106.0 Perm Region 81.0 26.4d 66.0 Primorsky Krai 90.3 19.0e 68.0 Pskov Region 96.3 113.8 94.0 Rostov Region 107.3 88.6 116.0 Ryazan Region 70.1 121.1 80.0 Sakha (Yakutia), Republic of 89.8 56.2 68.0 Sakhalin Region 98.9 93.3 106.0 Samara Region Donor Donor Donor Saratov Region 95.0 88.8 104.0 Smolensk Region 92.9 101.6 119.0 St. Petersburg City Donor 79.2 94.0 Stavropol Krai 109.8 114.0 100.0 Sverdlovsk Region Donor Donor Donor Taimyr AR 98.1 103.3 124.0 Tambov Region 126.8 82.3 110.0 Tatarstan, Republic of 0.0 Donor 121.0 Tomsk Region 98.1 112.8 104.0 Tula Region 97.8 116.8 114.0 Tuva Republic 105.6 118.4 178.0 Tver Region 93.9 109.1 112.0 Tyumen Region 89.6 46.7 107.0 Udmurtia Republic 108.0 102.7 38.0 Ulyanovsk Region 94.7 117.9 89.0 Ust-Ordinsky Buryat AR 93.3 1.9Y 111.0 Vladimir Region 92.0 104.6 114.0 Volgograd Region 83.5 151.2 83.0 Vologda Region 90.3 122.8 13.0 Voronezh Region 101.2 104.1 113.0 Yamalo-Nenets AR Donor Donor Donor Yaroslavl Region Donor 25.1 111.0 AR Autonomous region. Note: Instead of 0.0 percent, received (a) 39.5 percent; (b) 50.1 percent; (c) 25.8 percent; (d) 26.4 percent; (e) 19.0 percent; (f) 1.9 percent from its 1995 share. Source: Computed by the authors based on data from the Ministry of Finance. IMPROVING THE EQUALIZATION FORMULA. Improvements of the equalization formula can be suggested in three directions: * The measure of fiscal capacity used in the equalization calculation should be improved to avoid the incentive problems raised by using past actual collection data. A number of fiscal capacity measures The System of Transfers 71 that would improve the equalization process are discussed in Boex and Martinez-Vazquez (1997a) and computed for the regions of the Russian Federation in Boex and Martinez-Vazquez (1997b). * The measure of regional expenditure needs should be reconsidered. Alternative measures of ex- penditure needs are per capita monetary expenditure norms, indexes of fiscal needs based on broad social and economic measures of need, and the Representative Expenditure System. The use of monetary norms and expenditure needs indexes as measures of fiscal need is discussed by Boex and Martinez-Vazquez (1998a). A preliminary approach and initial estimates for a Representative Expenditure System for the Russian Federation were developed by Rafuse (1996). * A more dubious improvement to the equalization formula involves the introduction of stimuli to increase the tax effort by regional governments. The reforms of 1998 did not incorporate any such inducements for elevated tax collections. Instead, each FFSR-recipient region was required to imple- ment a program, subject to approval by regional legislative acts and the Ministry of Finance, to increase own-source revenue collections. Although a generally accepted principle in fiscal federal- ism states that transfer systems should neither encourage nor discourage tax effort, inspiring tax effort may be temporarily justified in a country, such as the Russian Federation, with substandard subnational collections. Introducing a stimulus to tax effort in the equalization formula is not com- plicated, although poorly directed incentives to wealthy regions could create further inequality among regions. Stimulating tax effort through incentives should prove more effective than main- taining separate revenue-raising agreements with the regions. REGIONS. The 1998 Budget of the Russian Federation mandated the distribution of Rb 1.2 billion among "depressed regions." The 1998 budget neither proposed any scheme or formula for the distribution of these funds nor identified "depressed regions." A few other countries besides the Russian Federation set aside a special pool of transfer funds for disadvantaged regions. For example, Canada's Territorial Formula Financing provides special aid to disadvantaged regions such as the Yukon, Northwest Territo- ries, and Nunavut Territories. This option may be quite suited to Russia given the large fiscal and eco- nomic disparities between the regions and the presence of a core of regions that is dismally poor. The 1999 introduction of the new FFSR equalization formula already focused much greater attention on the neediest regions than did the previous formula. The implementation of an equalization program for economically depressed regions would call for criteria to objectively identify depressed regions and require an adapted formula to distribute the available funds among the qualifying regions (Martinez- Vazquez, Boex, and Bahl 1998). HOLDING REGIONS HARMLESS. Reputedly, one main reason for the Duma's dissatisfaction with the FFSR allocations for the 1998 budget was that some regions would not receive the same allocation as they had enjoyed in the past. By holding regions partially harmless for reform measures, any new mechanism will likely become more politically acceptable. The disadvantage of holding regions harmless is that it sacri- fices funds that would be otherwise available for equalization. Nevertheless, some transition measures should be considered if they result in political acceptance for an improved equalization mechanism. Introduction of Categorical Transfers While policymakers paid a great deal of attention to FFSR reform, the reform of the remainder of the transfer system has received insufficient attention. The myriad of transfers that characterize the transi- tion period, including mutual settlements, forgiven loans, and other forms of transfers, provide an unstable and nontransparent source of regional funding and perverse fiscal incentives. The use of mutual settlements should be discontinued in favor of a system of formula-based, transparent, and objective categorical transfers. The public finance literature suggests that, in addition to equalization, intergovernmental transfers can be used to pursue several types of legitimate federal policy objectives. To achieve these multiple 72 Jorge Martinez-Vazquez and Jameson Boex goals, the Russian government should design a system of transfers using separate transfer programs for each separate objective. The pursuit of multiple goals with one instrument seldom performs as desired; the use of equalization grants from the FFSR for supporting part of the 1997 regional health insurance funds exemplifies this confusion of objectives. The design of an effective transfer system was further constrained by the limited ability of the federal authorities to monitor the behavior of regional govern- ments and the lack of enforcement mechanisms. Notwithstanding those problems, the federal govern- ment could introduce the following: * Explicit categorical grants, earmarked for programs of national policy significance, such as those in education, health, or the environment * Direct transfers to compensate subnational governments for federally mandated programs * A set of matching grants to provide incentives to subnational governments to increase expendi- tures in areas with externalities, such as transportation networks, or those with national signifi- cance, such as education and health. One feature of Russia's system of transfers has been the lack of focus on capital investment needs. None of the transfers were provided with the explicit objective of promoting investment in infrastructure by subnational governments. Capital investments are needed to rehabilitate and expand Russia's social and physical infrastructure and to offset the significant differences in capital infrastructure among the regions. The Concept of Reform of Intergovernmental Fiscal Relations called for the creation of a new transfer fund, Investment Support of Regional Development, to address capital infrastructure at the subnational level. The proposed arrangement is, properly speaking, not a separate fund, but rather pro- vides a methodological integration of federal, regional, and sectoral investment programs already in place. The main idea of this approach is to create an integrated investment plan for the regions under the coordination of the Ministry of Economy, which would provide matching-capital investment grants to the regions. This may prove to be the right approach to the needs for capital infrastructure at the subnational level, presuming the process does not become overly bureaucratic. Morozov (1998) presented proposals for improving the Investment Support of Regional Development fund, including the cancelling or freez- ing of the majority of the investment programs earmarked by the federal government, the reduction in their numbers, and the implementation of a smaller number of programs as closed-end matching grants. The Concept of Reform also calls for the creation of the Regional Finance Development Fund with the objective of providing incentives to subnational governments to reform their public finances and to en- courage revenue raising and expenditure rationalization. This would be accomplished through the granting of loans with conditionality for reform by the regions. This program could also prove to be a useful approach to encourage the implementation of a Subnational Budget Code. Reforming the System of Regional-Local Transfers In many regions the entire system of regional-local relations during the transition was simply carried over from the Soviet system. As this system was based on a minimum expenditure budget, regulated sharing rates, and gap-filling transfers, the policies lack any incentives for local governments to collect any own-source revenues. Although expressly forbidden in the Law on the Financial Foundations of Local Self-government of September 1997, many regions continued to claw back increases in local collec- tions by reducing transfers to localities. The heedless use of gap-filling transfers at the regional-local level will perpetuate both poor fiscal incentives and suboptimal fiscal behavior at the local level. What is truly needed at the regional and local government level is a reform of the entire budgeting process that governs the relationships between the two levels of governments, including the allocation of transfers. Such an overhaul cannot be simply mandated and will take time to take hold. However, several immediate improvements can be made within the regional-local framework to reduce the incentive prob- lems associated with the current transfer system, including the introduction of objective and incentive- free measures of expenditure needs and fiscal capacity for the determination of regional and local The System of Transfers 73 expenditure needs and intergovernmental transfers (Bahl and others 1998). In addition, if an adequate treasury system is put in place, it will be possible to downplay the operational concerns over financial counterflows in the design of the system of regional-local fiscal relations, allowing regions to focus on securing both vertical balance and horizontal balance in the allocation of regional resources. Projects that provide technical assistance to regional governments in the reform of intraregional fiscal relations in certain pilot regions have the ability to demonstrate to other, more hesitant regions that re- form can benefit both the regional and local govemments. The federal government could showcase posi- tive reform experiences by holding conferences and seminars and by providing documentation describ- ing success stories of regional reform. Constitutional issues and regional animosity toward federal interference in regional-local relations may surface with more direct federal involvement in comprehen- sive subnational reforms. Consequently, the federal authorities may wish to limit their role to educating regions about the severity of disincentives, developing and providing legislative support for a Subnational Budget Code, and promoting the voluntary use of formula-based transfer systems. The federal govem- ment should also set a good example regarding the responsible use of transfers. 5 Subnational Borrowing, Debt, and Financial Controls Reform of the legal framework for subnational borrowing and debt would form the capstone of a solid basis for intergovernmental fiscal relations in the Russian Federation. Responding to increasing budget- ary pressure and a permissive regulatory framework, subnational governments over the transition pe- riod increasingly resorted to borrowing and other types of deficit financing to maintain expenditure levels. By 1997-98, federal authorities grew increasingly concerned about the dangers to national eco- nomic stability and the lack of responsible fiscal management arising from the permissive approach to subnational borrowing. As a result, the Russian government has started to take measures to limit the borrowing power of subnational governments. Evolution of the Legal Framework for Borrowing Under the Soviet Union, subnational budget resources were fine-tuned to provide sufficient resources to finance a negotiated minimum expenditure budget. As a result, subnational governments did not have deficits, and they did not borrow, except from the Ministry of Finance for bridge financing purposes. Regions were to repay loans before the end of the fiscal year, and as a rule they did. During the early transition period, this traditional borrowing system continued to function. Two pieces of legislation were introduced that provided a much more permissive borrowing frame- work: The Law on the Foundation of Budgetary Rights of 1993 and the Law on the General Principles of Organization of Local Self-government in the Russian Federation of 1995. These regulations im- posed no explicit limits on subnational borrowing, allowing them to borrow for all purposes, includ- ing the financing of current expenditures. These rights were immediately put to use by many subnational governments. Since 1996, concerns about the lack of responsible fiscal management at the subnational level and its impact on national economic stability caused the Russian government to reverse its course. Legislation and decrees tried to reign in the subnational governments with regard to loan requests. The Law on the Securities Market of 1996 required federal authorization and imposed registration disclosure require- ments for bonds issued by subnational governments. The Law on the Financial Foundations of Local Govermnent in the Russian Federation of September 1997 restricted the amount that local governments could borrow to 15 percent of budget expenditures. This law included several other important restric- tions: Bonds could be issued only for investment purposes, not for current expenditures; municipally owned banks were banned from borrowing to avoid the moral hazard of joint borrowing and lending functions; and local bonds were not guaranteed by the federal government. However, the Law on the Financial Foundations applied only to local governments and not to regional governments. The Budget Code approved in 1998 limited the budget deficits of regional governments to 5 per- cent of budget revenues, before transfers and loans, and the budget deficits of local governments to 3 percent of budget revenues before transfers and loans. By requiring that expenditures for yearly debt service not exceed 15 percent of budget expenditures, the Budget Code limited the overall level of debt for both regional and local governments. Failure to comply with these limits could result in a budget audit or in the assumption of control over budget execution by upper-level government. The Budget Code further restricted subnational borrowing to that for capital investment and granted ex- ternal borrowing privileges only to the federal government. 75 76 Jorge Martinez-Vazquez and Jameson Boex A series of presidential decrees in 1997 and 1998 authorized a small number of subnational govern- ments to issue Eurobonds. These decrees imposed borrowing limits similar to those in the 1998 Budget Code; they required registration with the Ministry of Finance and imposed disclosure rules for subnational finances. Freinkman, Titov, and Treisman (1998) reported that by the end of 1997, 12 regional govern- ments had issued Eurobonds and many more were preparing to do so. Analysis of Subnational Deficits and Debts The balance of the trends in expenditures, revenues, and transfers at the subnational level between 1993 and 1999 was toward sharp increases in subnational budget deficits. Although budgetary balances can be as- sessed in several ways, each led to the same conclusion: Subnational governments were accumulating debt. Subnational Deficits and Debt Levels During the transition period, the consolidated subnational government budget went from a surplus, after transfers and budget loans, of 1.44 percent of gross domestic product (GDP) in 1992 to a deficit, after transfers and budget loans, of 0.80 percent of GDP in 1997 (table 5.1). If budget loans between govern- ments were considered a means of deficit finance, as opposed to a type of transfer, the consolidated subnational budget deficit evolved from a surplus of 1.35 percent of GDP in 1992 to a deficit of 1.55 percent of GDP by 1997. The true size of the deficit for subnational governments is understated in table 5.1. Not all types of subnational debt, including nonconventional means of financing such as arrears or the use of extrabudgetary funds, were reported in executed budget accounts. Liabilities for loan guarantees to municipal enterprises or suppliers were also undisclosed in the executed budget deficit. When both re- ported and unreported debts of subnational governments are considered, excluding loans from the fed- eral government, the total accumulated amount of subnational debt is estimated to have reached ap- proximately 6 percent of GDP by early 1998. Of the subnational debt, most was accumulated at the regional level; the municipal share of the consolidated subnational debt was estimated to be around 20 percent (Freinkman, Titov, and Treisman 1998). At 6 percent of GDP, the level of subnational government indebtedness is not considered high by international standards (table 5.2). Instead, concern with subnational borrowing in Russia is caused by the characteristics of the debt, high interest costs and short maturities; the purpose of the debt, often used to finance current expenditures; and the deteriorating fiscal situation and the growing inability of subnational governments to service this debt. Also, subnational debt levels increased extremely rapidly, 70 percent in 1997 (Freinkman, Titov, and Treisman 1998). But, these rapid percentage increases seem to be as much the result of the increase in borrowing as the initial low level of indebtedness. When ex- pressed in real terms, the increased reliance on deficit finance by subnational governments actually lev- eled off during the transition. Subnational Deficit Financing Subnational governments found multiple ways to finance their budget deficits, including both tradi- tional and less conventional methods of financing. Many of the nontraditional means do not show up in budget accounts. BUDGET LOANS. Traditionally, the Ministry of Finance lent funds to regional governments mainly to address short-term liquidity problems. The level of federal budget loans provided to subnational gov- ernments increased over the transition period; funds initially allocated as loans increasingly were writ- ten off as transfers by the federal government when the fiscal position of regional governments dete- riorated. As a result, budget loans were commonly counted as a form of transfer. In fact, budget loans provided subnational governments with a soft budget constraint, negatively affecting subnational Subnational Borrowing, Debt, and Financial Controls 77 Table 5.1. Consolidated Budget Execution, 1992-97 (percentage of GDP) Government level 1992 1993 1994 1995 1996 1997 Federal government Revenues 18.80 13.13 13.60 14.24 14.56 12.00 Expenditures 39.65 23.77 23.40 18.45 21.79 18.95 Balance, after transfers -20.85 -10.64 -9.80 -4.21 -7.23 -6.95 Balance, before transfers -19.44 -8.07 -6.18 -2.11 -4.45 -4.28 Consolidated regional government Revenues 13.78 16.59 18.23 14.83 15.45 16.62 Expenditures 12.34 16.06 18.24 15.15 15.99 17.42 Balance, after transfers 1.44 0.53 -0.01 -0.32 -0.54 -0.80 Balance, before transfers 0.04 -2.04 -3.63 -2.42 -3.32 -3.47 GDP, trillions of ruibles 19.00 171.50 610.70 1,630.10 2,200.00 2,602.30 Source: Freinkman, Titov, and Treisman (1998); Ministry of Finance data. governments' incentives to limit spending or increase revenues. Because loans were often forgiven, they were usually not counted toward the size of the subnational debt. Excessive provision of such loans may have contributed to the deterioration of the federal financial position that led to the collapse of the ruble in August 1998. COMMERCIAL DEBT. By early 1998, the reported commercial debt of subnational governments, not in- cluding budget arrears, guarantees, and debts to the federal government, was equivalent to 1.5 percent of GDP (Freinkman, Titov, and Treisman 1998). Commercial bank loans were the predominant source of commercial deficit finance, especially as the use of veksels (promissory notes) was prohibited in early 1997. Approximately one-third of all the commercial debt of subnational governments was held as loans. The most worrisome aspect of the market for commercial loans to subnational governments was that these involved nontransparent transactions that were subject to abuses and corruption. These concerns are further aggravated by the fact that certain regions actually owned the commercial banks from which they were borrowing. The use of bonds to finance subnational deficits grew more important during the transition pe- riod. Since their introduction in 1994, markets for municipal and regional bonds evolved from scratch, largely mirroring the growth of the market for federal bonds and government securities. The pioneer in the domestic market was St. Petersburg in 1994, followed by Moscow and Tatarstan. The use of bonds as a means of finance increased rapidly; by the end of 1997 more than one-quarter of the out- standing subnational commercial debt was held as bonds. In total, some 70 regions have issued Table 5.2. Relative Size of Subuzational Debt, Comparison of Selected Countries Subnational debt (percentage Subnational debt Country of subnational expenditures) (percentage of GDP) Australia (1996) 57 11 Germany (1996) 91 21 Switzerland (1995) 100 25 United States (1995) 65 13 Russia (estimated 1997) 40 6 Source: Freinkman, Titov, and Treisman (1998); computed by authors based on International Monetary Fund data. 78 Jorge Martinez-Vazquez and Jameson Boex domestic bonds, and a dozen or so have issued Eurobonds. All bond issues were supposed to be registered with the Ministry of Finance. Due to the lenient regulatory requirements regarding their issue and circulation, veksels were used extensively until 1996. Veksels are simply bills of exchange or promissory notes issued by regional gov- ernments instead of money in return for the provision of goods or services. Their use as a means of finance was banned in early 1997 by the Law on Bills of Exchange and Promissory Notes. However, many subnational governments continued to issue promissory notes or switched to alternative instruments that performed a similar function; for example, commercial entities, including public companies under control of subnational governments, issued veksels to finance subnational operations. BUDGET ARREARS AND OTHER UNREPORTED METHODS OF DEFICIT FINANCE. Budget deficits were calculated on a cash basis. As a result, the budget accounts do not reflect the accumulation of accrued but unpaid expen- ditures, such as arrears in wages to employees or payments to suppliers. Budget arrears of subnational governments became an increasing source of deficit financing midway through the transition. They in- creased from 1 percent of GDP in 1994 to 2.6 percent of GDP in 1997 (Freinkman, Titov, and Treisman 1998). In addition to arrears, alternative debt instruments, including some forms of external loans, borrowing through extrabudgetary funds, and commercial financial instruments, were not reflected in the budget. Why Do Regions Resort to Deficit Financing? With few exceptions, Russian regions were subject to sharp decreases in available real resources dur- ing the transition period. Their choices were to either cut expenditures or incur a deficit. Although most regions did both, some regions have been more prone to running deficits and accumulating debt. Have some regions borrowed more because they were poorer, because they have had higher fiscal needs, or simply because they had greater access to financing (see table 5.3)? The dependent variable is "other sources of finance," which incorporates both budget loans from federal authorities and from any other reported source. The results of the regression analyses reveal a clear pattern. In 1996 and 1997 regions with greater fiscal need, as measured by the shares of the population above and below working age, and regions with higher levels of cost of living, consistently relied more on budget loans and borrowing than other regions. The negative coefficient for per capita gross regional product (GRP) indicates that poorer regions borrowed more; however, these regression coefficients were not statistically significant. The increase in the explanatory power of the regressions (from 0.20 in 1994 to 0.60 in 1997) may indicate that expenditure needs became more binding determinants of regional bud- get deficits and the use of debt financing. Table 5.3. Regression Analysis: Per Capita Other Sources of Financing, 1994-97 Independent variable 1994 1995 1996 1997 Intercept -88.2 -1,081.8 -7,798.3 -8,541.6 Per capita GRP -10.1 -14.2 -3.5 7.7 Cost of living index 1.1 4.6 21.3 24.8 Percentage of the population younger than working age -2.9 10.7 12.7 136.0 Percentage of the population older than working age 3.4 -4.1 13.1 151.4 Percentage of the population in poverty -0.1 1.4 -4.8 0.7 Tax effort -0.1 30.5 23.3 0.6 R 2 0.20 0.35 0.62 0.60 Note: Analysis based on expenditures minus revenues and transfers. "Other sources of financing" mostly consists of budget loans and borrowing. Tax effort is defined as total collections as a percentage of gross regional product (GRP). Italics indicate statistical significance at the 5 percent level. Source: Authors' calculations based on data from the Ministry of Finance. Subnational Borrowing, Debt, and Financial Controls 79 Other Factors that Explain Subnational Deficit Growth Several other explanations could account for the continued growth of subnational deficits. Most of them are related in one way or another to fiscal needs. OFF-LOADING BY THE FEDERAL GOVERNMENT. The trend of increasing subnational govemment deficits during the transition stands in sharp contrast with that of the federal deficit, which fell from 20.85 percent of GDP in 1992 to 6.95 percent of GDP in 1997 (table 5.1). From 1992 through 1997, federal expenditures declined from approximately 40 percent of GDP to 20 percent of GDP. However, many federal expendi- tures were off-loaded in the early years of transition (1992-93) when, in the aggregate, subnational gov- ernments ran surpluses. The shifting of expenditure responsibilities continued after 1994 largely in the form of unfunded mandates. LACK OF FISCAL DISCIPLINE. Regional governments have been blamed for increasingly resorting to bud- get loans and external borrowing to finance expenditures that should have been reduced or cut years ago. Continued high levels of subnational government spending on the national economy sector, which consists largely of subsidies on housing and public utilities, are the most egregious examples of misdi- rected spending. Subnational governments also have been accused of not rationalizing their expendi- tures; they have not pursued more cost-effective options of service delivery, such as closing facilities with excess capacity or contracting with the private sector to deliver some services. The lack of adaptation may be a reflection of perverse incentives transmitted through the budget process; for example, subnational governments face an incentive to keep public facilities open to maximize budgetary allocations. No MOTIVATION TO INCREASE REVENUE COLLECTIONS. Poorly designed revenue-sharing arrangements and gap-filling transfer schemes, especially at the regional-local level, contributed to a lack of fiscal effort. Low tax effort at the local level required higher intergovernmental transfers and potentially resulted in higher regional government deficits. LIMrTED REVENUE AUTONOMY AND VERTICAL FISCAL IMBALANCES. The ability of regions to act upon the grow- ing deficit was restricted by the lack of significant own-source revenues. Subnational dependence on federal money created vertical imbalances. Vertical fiscal balance exists when the expenditure responsi- bilities assigned to each level of government correspond with the fiscal resources available at their discre- tion to carry out those responsibilities. While both revenue sharing and transfers can be used to redress vertical imbalances, both political accountability and economic efficiency require that regional and local governments have discretion over some of their sources of revenues, such as taxes and fees that they can expand or contract according to their needs. Yet, subnational governments in the Russian Federation were not given sufficient latitude to collect their own revenues. Further increases in fiscal autonomy could allow regional and local governments to respond to increasing budget pressures by increasing revenue collections, thus giving them an alternative to increased deficit finance and expenditure cuts. Has Excessive Subnational Borrowing Contributed to Macroeconomic Instability? Irresponsible borrowing behavior by subnational governments could affect the economic stability of the Russian Federation, especially if the federal government became the lender of last resort. Subnational government borrowing also crowds out domestic loan seekers, which may impair economic growth, especially if loans are used by subnational governments to finance current expenditures. And yet, at least on the surface, it has not been the borrowing by subnational government that caused the financial crisis and ruble devaluation in August 1998, but federal government borrowing combined with the rapid accumulation of short-term debt at extremely high interest rates. However, the question remains whether the federal government was forced to shoulder the subnational deficits through budget 80 Jorge Martinez-Vazquiez and Jameson Boex loans. Was the need to indirectly finance regional debts the cause of excessive borrowing by the federal government? In retrospect, the federal government neglected opportunities to eliminate its deficit. For example, in 1997 the federal deficit could have been significantly reduced if the federal government had kept the funds transferred to subnational governments (table 5.1). Naturally, the federal deficit could also have been eliminated by an equal-sized reduction in other areas of federal spending. Perhaps the federal government's attempts to transfer its deficit to subnational governments through off-loading of respon- sibilities and expenditure mandates were only partially successful. In the end, the federal government still had to borrow to finance part of the subnational government expenditures. Pending Policy Agenda for the Reform of Subnational Borrowing Several basic approaches could bring discipline and responsibility to subnational government borrowing. Financial markets could self-enforce prudent borrowing behavior by subnational governments. In a mar- ket system, lenders would charge higher borrowing costs to excessive or irresponsible borrowers and would ultimately refuse funds to irresponsible governments. This approach only works effectively if there are well-developed capital markets and institutions, including disclosure of information requirements, ratings agencies, and bankruptcy laws. The relatively simple self-enforcing properties of the market ap- proach make it a more attractive and ultimately a superior approach. However, the absence of sufficiently developed markets and financial institutions caused the capital market not to be a viable alternative. A second approach relies on federal government legislation to impose limits on borrowing for subnational governments and on federal government agencies enforcing these limits. Countries with less well-developed capital markets and institutions often rely on this method for subnational borrow- ing. Success resolution of the subnational debt problem further requires instituting bankruptcy and financial emergency controls and addressing the imbalances that underlie the fiscal problems of subnational governments. Imposing Legislative Limits on Borrowing Before implementing a legislative approach to subnational government borrowing, policymakers must consider, first, whether appropriate limits have been set on borrowing by regional and local governments and, second, whether appropriate monitoring and enforcement mechanisms have been put in place. The limits to the overall level of debt for regional and local governments imposed by the Budget Code appear adequate. Limiting debt service expenditures to no more than 15 percent of annual expenditures is prudent and falls in line with international practices. Other aspects of the limits to subnational borrowing in the 1998 Budget Code are more problematic and should be reconsidered. For example, allowing subnational governments to finance all kinds of ex- penditures with loans, in particular current expenditures, constitutes poor financial management. Bor- rowing should be limited to capital investment expenditures. The economic rationale for borrowing in the case of capital expenditures is that future generations of taxpayers should contribute to pay the cost of infrastructure that benefits them. Clearly, this rationale disappears when borrowed funds are used to finance current expenditures. In that case future generations of taxpayers would be held responsible for the consumption of the current generation. An additional aspect of subnational borrowing that should be considered is international financing. The overriding objectives of national macroeconomic stability call for the close federal scrutiny and ap- proval of any foreign borrowing by subnational governments. However, the outright prohibition of all forms of international borrowing may deprive responsible subnational governments of cheaper and more readily available funds for worthwhile and well-planned infrastructure investment projects. The effectiveness of any limits on subnational government borrowing will depend on the quality and effectiveness of monitoring and enforcement of regulations by the federal authorities. The only enforcement tool incorporated in the 1998 Budget Code is the requirement that all bonds issued by Subnational Borrowing, Debt, and Financial Controls 81 regional and local governments be registered with the Ministry of Finance. This mandate falls dramatically short of what will be needed to monitor and enforce legislated borrowing limits. In 1999, the federal gov- ernment announced plans to create a monitoring agency for subnational governments within the Ministry of Finance, but provided no details on how this program will operate or how limits will be enforced. Be- cause federal authorities in the 1990s failed to monitor and control subnational government activities, this is an issue that will require immediate attention and significant federal government resources. Subnational Bankruptcy and Financial Emergency Controls Part of a comprehensive strategy to counteract irresponsible fiscal management at the subnational level in Russia should be formed by the implementation of regional or municipal bankruptcy procedures in case regional or municipal governments are unable to pay their debts as they come due. Throughout the international community, municipal bankruptcy or insolvency laws provide the debtor government with some degree of protection from its creditors so the insolvent subnational government can resolve its debts through negotiated debt restructuring, reorganization, and refinancing. When a region or munici- pality seeks bankruptcy protection, its financial administration may be placed under court supervision to ensure implementation of an adequate debt adjustment plan. A subnational government should be considered insolvent when it is unable to pay its debts as they become due. Failure to pay wages and salaries, to make pension payments, to pay contractors, or to meet debt-service payments of principal or interest are evidence of government insolvency. Under this defini- tion, a large number of regional and local governments in the Russian Federation would qualify for bankruptcy protection. At the same time, a region that is merely experiencing temporary financial diffi- culty should not be granted a bankruptcy settlement. The intention of a subnational bankruptcy procedure would be to compel subnational governments to engage in responsible fiscal behavior. For those governments that are unable to remain solvent, bank- ruptcy provides time to correct the situation that caused the insolvency and to allow the government unit to restore its financial health. Unlike bankruptcy legislation for businesses and individuals, procedures for subnational governments should not provide for liquidation of the government's assets to satisfy the claims of creditors. Such a dissolution would leave the citizenry without a local or regional government. Instead, reorganization of regional or municipal debts could be accomplished either by extending debt maturities, reducing the amount of principal or interest, or refinancing the debt by issuing a new loan. During the reorganization, bondholders are typically considered superior creditors, so debt service will ordinarily flow on schedule; however, other contractual obligations may not be met. For instance, wages and salaries may have to be reduced from their previously contracted levels for the regional or local government to regain its financial health. Regional or municipal bankruptcy is a drastic measure that should only be used in the most serious cases of fiscal mismanagement. An alternative to bankruptcy includes employment of emergency finan- cial management controls. In this case, a superior-level government intervenes with extraordinary con- trols to constrain the financial and economic damage caused by mismanagement of a regional or local government. For example, in the United States, 20 states have laws for provision of state assistance and control for municipalities in financial distress. Although each state uses a different approach, the general objectives of such programs are as follows: * To preserve and protect the fiscal solvency of local governmental entities * To assist local governments in providing essential services without interruption and in meeting their financial obligations * To assist local governmental entities by improving local financial management procedures. The determination of a financial emergency for U.S. states is made based upon a set of objective criteria; for instance, when a subnational government fails to pay its obligations or when it has insuffi- cient reserves to cover a budget deficit for two successive years. When a financial emergency arises, the 82 Jorge Martinez-Vazquez and Jameson Boex municipality or region is brought under close control of the superior-level government by establishing a board to oversee the local operations and to provide technical assistance. The subnational govern- ment and the oversight board jointly develop a plan that provides payment for all debt obligations, pension payments, and all other mandatory payments while reducing expenditures by eliminating low-priority budget items. Compared with the regional or municipal bankruptcy option, the state emergency boards involve far greater loss of local control and focus greater effort on reforming and educating local officials about fiscally responsible practices. Nonetheless, emergency boards are empowered to change local priorities and fiscal decisions originating from the local political process and consequently may cause greater dis- comfort for local officials than would a bankruptcy. Addressing Vertical and Horizontal Fiscal Imbalances Subnational deficits in the Russian Federation will not disappear without policies that address the existence of vertical and horizontal fiscal imbalances. Funding pressures are likely to continue at all levels of government. The balance of expenditure needs and available resources at different levels of government must be reexamined to provide some basis for reorienting expenditures toward priority goals and increasing the efficiency of public expenditures at all levels of government. Although during the transition, substantial progress was made in correcting vertical fiscal imbalance, there is a contin- ued need for expanding the revenue autonomy of subnational governments. Giving subnational gov- ernments the capacity to impose taxes, if so desired, is the most effective, if not the only, way to address these vertical imbalances. Because regions with greater expenditure needs are more likely to use deficit financing, significant horizontal imbalances likely exist, and the transfer system has failed to correct them. As a result, impor- tant subnational programs such as health and education may go unfunded. A focus on vertical and hori- zontal imbalances will provide the necessary perspective of the "big picture" that has been missing in much of the reform process during the transition. 6 Implementing Reform This chapter focuses on the process through which reforms in intergovernmental fiscal relations should be implemented. It addresses key issues that the Russian government must consider in the formulation and implementation of the reforms: (a) piecemeal versus unified strategies for reform, (b) asymmetric versus uniform treatment of subnationa] governments, (c) compliance with and enforcement of federal legislation among subnational governments, (d) the improvement of communications between federal and regional governments, and (e) the appropriate form of federalism for the country (in particular, deciding how much the federal government should intervene and regulate regional-local relations). Piecemeal Versus Unified Reform During the transition, the Russian Federation lacked a coherent and unified strategy for reform of inter- governmental fiscal relations. A unified strategy would have required stating, balancing, and imple- menting the key objectives underpinning the decentralization policy. To reach a comprehensive reform strategy, the government would have to address the extent and nature of federalism that it seeks to achieve, the desirable degree of equalization in the distribution of fiscal resources across regions, the efficient use of budgetary resources, the creation of incentives for revenue mobilization, and the preservation of fiscal discipline and responsibility among subnational governments. The difficulties in reaching a national consensus on the general objectives of fiscal decentralization policy may help explain the piecemeal nature of the reform process. The process of defining an overall strategy for reform, which requires the weighting of the different objectives, will likely continue to be difficult and slow. The views of the president's administration and the Ministry of Finance on decentrali- zation issues differ considerably from those held at the State Duma and especially at the Federation Council, where diversity of interests and values is significant among the "have" and "have not" regions. The Russian government's current strategy for reform of intergovernmental relations is outlined in the Concept of Reform of Intergovernmental Fiscal Relations drafted in 1998 by the Working Group on Intergovernmental Fiscal Relations. Even though this document was revised numerous times, with every version adopting a more comprehensive approach than its predecessor, the 1998 Concept of Reform still falls short of delivering a unified reform strategy. The Concept of Reform addresses independent policy issues and recommendations without clarifying the connections between them. Furthermore, it does not quantify the fiscal impact of proposed reforms or spell out their effects on vertical and horizontal bal- ances in intergovernmental fiscal relations. The piecemeal nature of the reform strategy has given rise to a fragmented, incomplete, and often contradictory or inconsistent legal framework for intergovernmental fiscal relations. In addition, some regions have simply refused to accept certain legal provisions or negotiated with the federal authorities to be exempted from legislation. The most conspicuous feature of the transition's legal framework for intergovernmental fiscal relation has been its ambiguity and the government's tendency to postpone addressing fundamental questions. Tackling the wide array of problems in intergovernmental fiscal relations all at once amounts to an impossible task. Rather, the government needs to consider strengthening the Concept of Reform of Inter- governmental Fiscal Relations in light of a comprehensive and unified strategy by systematically review- ing all the existing laws and drafting amendments to ensure that legislation conforms to the overall strategy and objectives. 83 84 Jorge Martinez-Vazquez and Jameson Boex To improve the chances of creating successful reforms, representatives from all three levels of govern- ment should be directly involved in the definition and implementation of the plan. Reforms imposed in a top-down process are certain to be received with suspicion and hostility by subnational governments. This is true anywhere in the world, but especially in the Russian Federation. Regional and local govern- ments should be particularly instrumental in the improvements of fiscal relations at the regional-local level, resulting in the development of a Subnational Budget Code. In this arena, the federal government should play more of a coordinating role rather than relying on its executive power. Although the Federa- tion Council represents regional interests, it cannot adequately represent and articulate the views and interests of local governments. A new institutional vehicle might have to be considered, which could benefit from both regional and local government input in the formulation and implementation of com- prehensive fiscal relations reform. Uniform Versus Asymmetric Federalism and Addressing Fiscal Disparities Asymmetric federalism, the nonuniform treatment and standing of regions, may have been unavoidable given the historical precedents inherited from the Soviet Union and perpetuated by the need to accom- modate ethnic regions' demands. Asymmetric federalism practices may also have been justified by the need to deal with regions in unusual circumstances and with significantly differing regional abilities to implement policies and programs, but they undermined national solidarity and imposed a system that lacks "unity of interest" (Igudin 1998). During much of the transition period, the federal government was pressured into granting new privi- leges to regions or expanding existing ones. However, there has been a growing awareness at the federal level of the problems associated with the use of bilateral treaties. As a result, the 1998 federal budget called for the reassessment of all bilateral treaties to ensure that they were congruent with the provisions in the annual budget. No evidence exists that any evaluation actually took place. Similar plans for the reconsideration of all bilateral treaties were proposed at different stages of 1999 budget development. Even though the demands for autonomy from ethnic regions will continue to pose problems in the future, the most difficult long-term challenge to the reform of fiscal federalism in Russia will likely come from the growing fiscal disparities among the regions. Even though large amounts of resources were expended by the Soviet regime to equalize fiscal disparities among the regions and to keep afloat unviable regional economies in remote areas of the country, economic and fiscal differences among regions were significant at the time of independence and grew significantly during the transition period. The lack of market development outside a few geographical areas, labor immobility, and the absence of adequate and universally available social services led to a situation in which economic and fiscal disparities can only be redressed by nonmarket means. The federal government's failure to equalize regional economies created increasing political tensions (Polishchuk 1996). Federal neglect is not without reason. The growing inequality in the distribution of tax bases made the federal government more vulnerable to political pressures from richer regions, which increasingly have opposed the idea of paying higher federal taxes to redistribute to the poorer regions. At the same time, the plight of the weak regions deteriorated because their troubles have been increasingly perceived at the federal level as self-inflicted, a result of poor fiscal and economic management. The growing economic and fiscal disparities between regions led to a pronounced divergence of inter- est among the regions and to a lack of consensus in national policies. This has left the federal government in the crucible of either engaging in more redistribution among the regions or achieving faster, but geo- graphically uneven, economic growth. In the first paradigm, the federal government would tax resources away from those regions that were performing better economically. Under the second paradigm, equal- ization, if it could be achieved at all, would come from migration and other market adjustment mecha- nisms. The government has not chosen explicitly between these two reform models, but the direction in federal policies seems better represented by the second paradigm. Federal policies also appear to move in the direction of further reductions in the funds available for equalization. In addition, the federal government is proposing a "dual" regime in which those regions 1, ^ !.-'..- .Reform 85 that are most dependent on federal transfers would be forced to institute reforms of regional budgeting practices. Under such a regime, regions that receive 50 percent of more of their budgets from Funds for Financial Support of Regions (FFSR) would need to completely relinquish their budget powers to the federal authorities (IEPT 1998).1 These policies may be the most viable politically, but it remains to be seen whether they can improve the ability of poor regions to provide adequate basic services, including health and education. In addition, these federal proposals do not distinguish between regions with poor fiscal management and a poor economic base. Though these two phenomena are highly correlated, policymakers should carefully differentiate between them to improve the fiscal management of regions that are both poorly managed and underprivileged. The question needs to be raised whether the pro- posed policies will create a permanent underclass among regions that will depend on federal support and controls, providing a basis for further division among the regions. Although some asymmetry in intergovernmental fiscal relations must be accepted as unavoidable in the Russian Federation, future reforms should approach federal-regional relations in a uniform way when possible and prevent expan- sion of the asymmetry already contained in the system. Regional Compliance with Federal Laws Fiscal federalism in Russia has been complicated by the open hostility between the federal government and some of the regions. This problem was more severe in the early years of the transition but had not disappeared by the end of the 1990s. Contentious relations in combination with the inability to monitor subnational governments seriously inhibited the ability of federal authorities to enforce federal laws throughout the national territory. While some particularly defiant regions did not remit revenues from federal taxes to the federal government, noncompliance took many other forms. In recent years there have been more open attempts by the federal government to enforce federal laws. The Law on the 1996 Federal Budget contained a number of measures to punish those regions breaching federal legislation on issues of taxation and budgetary matters. These disciplinary actions in- cluded the suspension of funding for all federal programs; termination of transfers, loans and credits; termination of export license issuance; suspension of centralized product delivery from government sources; and the retrenchment of credit and cash supplies to the regions by the Central Bank of Russia. Although it is hard to tell how many, if any, of these steps have been implemented, the Ministry of Fi- nance did de facto transfer additional expenditure responsibilities for federal programs to those regions where unauthorized shares of revenue collections were held. In addition, the federal government opted to "bribe" regional governments with transfers to ensure compliance with federal policies (Memorandum of the Government 1998).During the first half of 1998, in accordance with a presidential decree of May 5, 1997, the Ministry of Finance signed agreements to clear local wage arrears with most of the regions that had received budget loans. The federal government forgave the budget loans for regions that agreed to implement a set of federal policies that they had previously refused to accept: discontinuation of tax offsets, elimination of wage arrears, adjustment of subnational government pay scales to the standards of the federal government, and implementation of cost recovery for housing and public utilities using federal standards. To enforce the agreements, the federal government threatened to reduce the amount of equalization transfers for any noncompliant region by an amount equal to the scheduled loan repayment. The 1999 budget draft offered a similar enforcement mechanism. It contained provisions that call for the withdrawal of equalization transfers and any other grants or loans to regions that fail to comply with a specific set of federal policies. However, some of the federal requirements, such as the elimination of all nonmonetary forms of payments in regional budgets, are unlikely to be fulfilled, because more than one- half of all tax receipts by regional governments are noncash payments. Making pronouncements that later cannot be enforced weakened the standing of federal authorities in the regions. 1. Adgeya, Dagestan, Ingushefia, Kabardino-Balkar Republic, Karachayevo-Cherkess, Marii-El, and Tuva Republic. 86 Jorge Martinez-Vazquez and Jameson Boex How the August 1998 ruble crisis affected relations between the federal and regional governments re- mains unclear. Many regions openly defied federal authorities and legislation in a number of areas in the weeks following the crisis. Gaining the ability to enforce federal laws in the regions will be crucial for a working system of intergovernmental fiscal relations. To successfully produce compliance, the 1993 Consti- tution may need to be amended, and the federal government may have to prosecute regional violations of federal laws in court. On a day-to-day level, better compliance will require strengthening and reforming existing institutions, such as the Ministry of Taxation, or the introduction of new ones, such as a treasury agency for the execution of the federal budget. When, and if, the federal authorities get real control of the collected federal tax revenues, federal efforts to enforce compliance by all regions will be solidified. Federal authorities also need to treat regional governments as partners; cooperation and consensus generally lead to much better regional compliance with federal laws than do conflict and forced imposition. Communications between Federal and Regional Governments An important feature of Russian fiscal federalism has been the lack of communication and exchange between the federal and regional authorities. International practices differ greatly with respect to fiscal communications, ranging from the German practice of cooperative federalism, whereby federal and subnational fiscal strategies and targets are coordinated; to the Premiers' Conferences in Australia, Canada, and New Zealand, which provide for an informal exchange of views; to the complete lack of coordination between the fiscal strategies of different levels of government in the United States. The nature of Russia's federalism and the history of intergovernmental fiscal relations during the tran- sition suggest that many of the current conflicts and the lack of smoothness in intergovernmental opera- tions could be alleviated by increased communications between the federal and subnational levels of gov- ernment. Improved communications could be accomplished by replicating some of the features of Australian or Canadian approaches. The Ministry of Finance could host a formal annual conference on intergovern- mental fiscal relations to which finance directors of all regions would be invited. Without binding powers, this conference could allow for open exchange of information and policy priorities between the Federal Ministry of Finance and its regional counterparts. As an alternative to the conference of regional finance directors, or in addition to it, another meeting could involve the prime minister or president and the governors of the regions. On a more permanent basis, the federal government could establish an advisory council on intergovernmental relations comprising representatives from all three levels of government, similar to the U.S. Advisory Commission on Intergovernmental Relations. Such a nonpartisan advisory body could prove particularly useful in identifying programs that contain unfunded federal mandates and in making policy recommendations to either eliminate, fund, or modify these programs. During the transition, the Ministry of Finance actually held some annual conferences with the regions to discuss the rules for the distribution of FFSR money. This annual conference could be formalized as part of the budget process and its scope expanded to cover not only equalization-fund issues, but also medium-term plans for revenue sharing and expenditure assignments, borrowing, and budgetary strat- egies from all the regions. In addition, the conference could provide a forum where regions could share experiences on regional-local relations issues. Determining the Form of Fiscal Federalism Significant differences exist between the de jure fiscal federalism and the system of fiscal federalism that is actually in place in the Russian Federation. This schism has left several important questions unan- swered that could greatly affect the performance and efficiency of intergovernmental fiscal relations. How much autonomy should local governments have in relation to regional governments? Should the federal government directly address some policy issues with local governments, thereby entirely bypass- ing regional governments? How much control can the federal government exert over regional govern- ments in mandating certain regional government policies toward local governments? These are Implementing Reform 87 fundamental questions concerning the nature of federalism in Russia. No universal model for a federalist structure of government can be applied; federalist countries around the world have answered those ques- tions in many different ways, depending upon their history and political needs. In Russia, the 1993 Constitution proclaimed the independent right of local governments to manage their affairs separately from the sovereignty of the state, which includes both the federal and regional governments. In reality, local governments have been subject to regional dictates in funding and expendi- ture assignment matters, if not to the outright control of regional governments. Regional governments, for the most part, were satisfied with these transition arrangements; at the very least, regional authorities saw the dictates as a necessary evil in an unstable economic environment. However, federal and local officials became dissatisfied with regional governments' handling of fiscal relations with local authorities. The appropriateness of federal government intervention in the fiscal relations between regional and local governments remains unclear. However, Presidential Decree No. 685 in 1996 mandated regional gov- ernments to share a minimum of 5 percent of tax revenues with local governments. The law had little substantial effect. The Law on Financial Foundations of Local Self-government in the Russian Federation of September 1997 was passed by the State Duma and the Federation Council after several revisions demanded by the Federation Council. It proclaimed a prohibition on unfunded mandates and prohibited the clawing back of local revenues. It also established minimum, average sharing rates for shared taxes with local gov- ernments. Overall, the Law on Financial Foundations provided few actual constraints on the way in which regional governments dealt with local governments. However, regional opposition was based on the fear that regions will eventually lose control over local governments. The significance of the Law on Financial Foundations is that it attempts to dismantle the traditional hierarchical structure in which local govern- ments respond exclusively to regional authorities and that it seeks to establish a system that separates, or at least regulates, the fiscal relationships between state authorities and local self-governments. Given the historical context and the lack of institutional development in many regions, a policy that provides broad federal guidelines regarding subnational fiscal parameters but still leaves choice for the region may be best suited to Russia as it continues its economic transition. In this context, federal legisla- tion could provide local governments with much more revenue autonomy and budgetary certainty than they have enjoyed until now. The blueprint for intergovernmental fiscal reform should include legisla- tion that imposes actual constraints on regional government behavior in relation to local governments. The enactment of a Subnational Budget Code may be the first step in accomplishing a workable fiscal federalism in the Russian Federation. 7 Postscript: Recent Reforms in Intergovernmental Relations in the Russian Federation The manuscript for Russia's Transition to a New Federalism was completed in June 1999, six months before one of the most significant political changes in Russia since the beginning of the transition: President Boris Yeltsin announced his resignation on December 31,1999, and appointed his prime minister, Vladimir Putin, as acting president. Putin has openly stressed the role of federal-regional relations as an important theme in Russian public policy and has begun to implement a directed agenda of tax reforms and reforms of intergovernmental relations. Many assessments of these policy reforms, especially by the international news media, have concluded that Putin's reforms constitute a new phase of power recentralization in Russia and have expressed concerns about the possible undemocratic tendencies of the reforms.' Following his election to the presidency in March 2000, Putin indeed took a number of steps to strengthen the federal position in relation to the regional governments. In May 2000 he issued a decree that divided Russia into seven groups of regions, each with a presidential envoy to monitor regional legislation and ensure that regional administrations are adhering to federal laws. In addition, Putin has pursued increased federal control over regional administrations by pursuing legislation that gives the Russian president the right to suspend regional legislation that conflicts with federal law and to dismiss regional governors if their actions violate federal statutes. In addition, tax reforms proposed by the current administration are expected to result in a sharp decline in revenue autonomy for regional and local governments. Careful analysis of the prevailing trends in intergovernmental relations in Russia during the past decade as presented in the preceding chapters provides the necessary context for an evaluation of the reforms pursued by President Putin thus far. A key observation in this regard is that the trends of federal strength- ening and the recentralization of fiscal resources began in 1997-98, well before Putin's rise to power, as a result of the federal authorities' growing concerns about regional noncompliance with federal laws. As discussed in chapter 1, the federal government had reason to be concerned. Throughout the early stages of the transition, certain regions habitually ignored federal legislation, forcing the powerless federal adminis- tration to negotiate deals with these renegade regions. These compromises resulted in an asymmetric feder- alist approach that created tensions between the federal and regional govermnents and among regions. The fiscal policy reforms that took place during the final years of the 1990s, as well as Putin's reforms in the months following his election, constitute an important phase in Russian decentralization policy. Putin's reform agenda could be seen as a continuation, albeit a more coherent continuation, of ongoing policy efforts to strengthen the federal government's constitutionally defined position in the federal struc- ture. The extent to which President Putin intends to reassert the role of federal institutions and whether future reforms will limit subnational self-governance are unclear at this time. We felt that this manuscript would be incomplete if it did not address Putin's reforms in the context of the decentralization trends that took place in the first decade of the transition. We expect to update the book when current fiscal data for the Russian regions become available and include a more complete discussion of the ongoing reforms. 1. An article in the New York Times (June 1, 2000, p. All) suggested that Putin's reforms were moving "much too far toward making Mr. Putin... not a democratic president, but a Latin American style strongman." An article on Putin's reform plans in the Financial Times (July 20, 2000, p. 2) quoted a regional governor who suggested that the reform set "a bad precedent, which has nothing in common with a democratic society." 89 90 Jorge Martine--Vazquiez and Jameson Boex Reform of the Vertical Power Structure The political dynamic between the federal government and the regions during the beginning of the tran- sition was defined to a large extent by the disintegration of the Soviet Union. The ensuing power vacuum enabled ethnic regions to exert pressure with separatist demands on divided federal executive and legis- lative powers, which were fighting each other for hegemony. In stark contrast to the central control dur- ing the Soviet era, during the early transition regions were able to take advantage of the fact that they had de facto control over key elements of government administration, including the tax administration and internal security forces. This left the federal government without tools to deal with noncompliant regions and enforce federal legislation. As discussed in previous chapters, this political reality forced the federal government to cut deals with dissenting regions to prevent secession. Increased concerns about its lack of effectiveness and credibility caused the federal government to take concrete steps to strengthen its position in the vertical power structure after 1998. Arguably, the first piece of major legislation that turned the tide in favor of the federal government was the Law on the Principles for the Demarcation of Jurisdictions and Powers, enacted in June 1999. This law delineated the powers of federal bodies and regional governments and clearly restated the supremacy of the federal constitution, legislation, and decrees over regional constitutions, legislation, and decrees. During the first half of 2000, additional reforms were introduced to reassert federal primacy over inter- governmental relations. The reform agenda set out by Putin to bring political order to the Russian federal system comprises a number of administrative reforms, including greater direct scrutiny of regional policies by the federal government as well as political and legislative reforms that provide the federal government with legal avenues of recourse when regions disregard and violate federal legislation. As mentioned earlier, one of Putin's first reforms of the vertical power structure was a presidential decree that established seven groups of regions, each headed by a presidentially appointed envoy to monitor regional compliance with federal legislation. These envoys replaced 80 presidential regional representatives who had been appointed by a 1997 presidential decree and had largely turned into inef- fective figureheads. Political reactions to the decree varied from fear by regional governments concerned with losing regional powers, to skepticism and concern that the decree would merely impose an addi- tional layer of ineffective federal bureaucracy. Although it is too soon to pass final judgment on the effec- tiveness of the new administrative controls, in concert with the legislative measures discussed later, they appear to have contributed to increased regional compliance with federal laws. Perhaps the most controversial set of political reforms proposed by Putin, and adopted by the State Duma and the Federation Council with some modifications, were a number of reform measures aimed at reducing the power of the regional governors. These reforms can be seen as logical and justifiable when considered in the context of Russia's recent history, during which some regional governments systemati- cally and blatantly failed to comply with federal legislation and regulations in order to advance their own political agendas. This situation was exacerbated by the fact that regional governors had parliamen- tary immunity, as their position automatically gave them a seat in the Federation Council, Russia's upper house of parliament. To redress this issue, the Bill on Amending the Law on the Procedure of Forming the Federation Council proposed a change in the makeup of the Federation Council, replacing the governors with regional representatives elected by the regional legislatures. Despite initial overwhelming opposi- tion from regional governors in the Federation Council, the law was eventually approved by both houses of parliament, thereby clearing a legal obstacle for the federal justice system to deal effectively with corrupt regional governors.2 In a compromise with the Federation Council, these reforms will not take effect until 2002, when the council's current legislative term expires. 2. The State Duma approved the bill on June 23,2000, but the Federation Council initially voted it down. How- ever, after the State Duma overturned the council's veto on a related bill, the Federation Council voted by a wide margin to approve the bill on July 26, 2000. Postscript: Recent Reforms in Intergovernmental Relations in the Ruissian Federation 91 A second step in Putin's political reform agenda was the Bill on Amending the Law on the Principles of Organizing the Legislative and Executive Bodies of Russian Territories, which proposed giving the Russian president the power to remove governors or suspend regional legislatures when they fail to comply with federal laws. While the bill was slightly modified by the State Duma to assure a process of judicial review, the core intent of the law was left intact.3 Whereas the federal government had previously been essentially toothless in its attempts to rein in regions that ignored federal laws, the final version of the law gave the Russian federal government an effective legal remedy to deal with recalcitrant regions. In addition, proposed modifications of the Law on General Principles of Local Self-Government would give the president the right to dismiss local gov- ernment officials for failure to comply with federal law, but the dismissal of local officials could only occur after judicial review and only if regions failed to bring localities into compliance with federal regu- lations first. These reforms have been touted as bordering on undemocratic in some news reports. In all fairness, the law as originally intended would have gone too far by giving the president a free hand to remove subnational chief executives or suspend legislatures. However, by requiring a process of judicial review, the current set of reforms do not appear to have diminished regional or local powers so long as the regions and local governments act within the bounds of the Russian constitution and federal legisla- tion. Whether the president uses his new powers with restraint and equanimity and whether the courts can truly prevent any abuse of these new presidential powers remains to be seen. Expenditure Assignments and Budgetary Responsibilities As throughout most of the transition, the assignment of expenditure responsibilities between the differ- ent levels of government has remained stable in recent years. As concluded in chapter 2, the de facto assignment of expenditure responsibilities generally follows the principles of sound fiscal decentraliza- tion. Budgetary reforms since 1998 have focused on rationalizing and clarifying the legislated division of expenditure responsibilities between the different levels of government. For instance, the 1999 Law on the Principles for the Demarcation of Jurisdictions and Powers clearly established the supremacy of fed- eral legislation and decrees over areas of joint federal-regional responsibility. In addition, Article 85 of the Budget Code was modified in 2000 to require that joint expenditure responsibilities must be separated in the annual budget law for every type of activity. A number of further changes to the Budget Code aimed to improve the budget process by clarifying sanctions for violating federal budget laws and imposing additional limits and regulations for subnational government borrowing. Two main characteristics of the assignment of expenditure responsibilities are of ongoing concern. First, expenditure assignments remain too centralized as the federal government continues to rely on budget norms and unfunded mandates. Second, assignment of expenditure responsibilities at the regional- local level remains unclear. Whereas previously intraregional relations were largely ignored in the decen- tralization process, the Federal Ministry of Finance now recognizes the importance of regional-local rela- tions to the system of fiscal federalism and is actively providing support to regions that are reforming fiscal relations at the regional-local level. The federal government is also taking concrete steps to resolve the difficult issue of unfunded man- dates. First, the budget proposal for 2001 includes Rb 71 billion earmarked for transfers to the regions as compensation for federally mandated programs. For those federal mandates that the federal government is unable or unwilling to fund, changes to Article 83 of the Budget Code, which will take effect for the 2001 budget year, require that the annual budget law includes the list of mandates that will be stricken because of a lack of funding. For mandates that are partially funded, the budget law will need to specifi- cally identify which parts of the mandates would not be financed by the federal government. Of course, whether these policies will suffice to end the practice of unfunded mandates remains to be seen. The 3. The Federation Council vetoed the bill on June 30,2000. The State Duma overturned the veto on July 19,2000. 92 Jorge Martinez-Vazqez and Jameson Boex success or failure of these provisions will ultimately depend on their actual implementation and on the federal legislative and executive branches' adherence to these policies. Revenue Assignments and Tax Reforms A beginning was made with the reform of the Russian tax system in 1998 as part of the emergency mea- sures taken in response to the economic crisis of August 1998. The first part of the new Tax Code, which contains the general principles of taxation in the Russian Federation, was passed by the legislature in 1999. Subsequently, however, the government's tax reform efforts effectively stalled as the more impor- tant, second part of the Tax Code, which contained legislation covering individual taxes, became bogged down in the legislative process because of a lack of political consensus on the nature of the reforms within the executive and legislative branches. In the months following his election to the presidency, Putin used an unprecedented level of direct influence to achieve consensus within the executive branch regarding the details of the government's tax reform package and to guide the tax reforms through parliamentary consideration. Important tax re- forms were contained in four chapters of the second part of the Tax Code, as well as interim tax legisla- tion, which received parliamentary approval in July 2000.4 Putin has pursued tax reforms aimed at sim- plifying and rationalizing the tax structure and lowering the overall burden of taxation. Public sector resource centralization was another goal of these reforms. Note that both the draft Tax Code and the trend of revenue centralization were set in motion in 1998, well before Putin's ascent to power. While a complete analysis of the recent Tax Code reforms and other interim tax legislation falls be- yond the scope of this chapter, two components of the tax reform are having a substantial impact on intergovernmental fiscal relations in the Russian Federation and warrant mention. First, subnational turn- over taxes, such as the Housing Maintenance and Road Users taxes, which provided a substantial level of own-source revenues for local and regional governments, are being substantially reduced and will even- tually be eliminated. Second, the PIT, the revenues of which flow largely to the subnational governments, will be overhauled in 2001. The structure of the tax will be substantially simplified in 2001, from a pro- gressive rate schedule with a 30 percent top rate to a flat rate of 13 percent. This rate change is expected to result in a corresponding loss of revenues for subnational governments. Other recent tax reforms affect the fiscal position of subnational governments as well. As part of the overall rate reduction of the EPT in 1999 from 35 to 30 percent, the regional EPT rate was reduced from 22 to 19 percent. Despite earlier confusion and policy statements to the contrary, regions will continue to be able to grant exemptions over the regional share of the EPT, effectively allowing them full control over the regional share of the tax. A number of other changes will affect the corporate income tax as well. In particular, starting January 1, 2001, municipalities will have the option of introducing a local "piggy- back" corporate income tax up to a maximum of 5 percent. Many, but not all, of the recently enacted tax reforms make good fiscal sense. Overall, the reforms tend to reduce the complexity of the tax system and lower the marginal tax rates, thereby reducing incentives for tax avoidance and evasion. However, some reforms will actually complicate rather than simplify the system, for example, the proposed municipal piggyback tax on the EPT will substantially complicate the administration of the profit tax. While generally in line with sound fiscal practices, these tax reforms and changes in the assignments of tax revenues are causing a substantial loss of shared revenues for regional and local governments and have added very little to the tax autonomy of these governments. In contrast, federal revenue sources, such as excise taxes, have been substantially increased as part of ongoing tax reforms. In this sense, the recent tax reforms do not facilitate, and in fact may jeopardize, many of the potential benefits of a decentralized fiscal system. 4. The Tax Code chapters on the personal income tax, the value added tax, excise taxes, and social security taxes were adopted by the legislature and signed into law by President Putin in August 2000. The remaining chapters of the Tax Code include those addressing corporate income and property taxes. Postscript: Recent Reforms in Intergovernmental Relations in the Russian Federation 93 A second set of fiscal policy reforms in recent years also has been aimed at reallocating tax revenues away from subnational governments and toward the federal government. Table 7.1 summarizes revenue sharing arrangements from 1999-2001. Whereas revenue sharing arrangements had been relatively stable from 1994 through 1997, several substantial changes were made in revenue sharing arrangements before and after the economic crisis of August 1998. Changes in the assignment of revenues, especially the assign- ment of PIT collections, reflect a tug-of-war over resources between the federal and regional governments. Despite the long period of stability in revenue sharing arrangements from 1994 through 1997, the sharing of PIT revenues has been subject to annual changes in recent years. Although regional govern- ments had received 100 percent of PIT collections since the beginning of the transition (with the excep- tion of a 3 percent federal surcharge in 1995-96), in 1998 the federal government increased the federal share of PIT collections to 40 percent. In 1999 PIT collections were once again fully reassigned to the regions while the 3 percent federal surcharge was reintroduced, resulting in an effective regional sharing rate of 90 percent. In another policy reversal, the budget law for 2000 eliminated the surcharge and im- posed a new sharing regime altogether, assigning 84 percent of PIT collections to the regional govern- ments while allocating the remaining 16 percent of PIT revenues to the federal government. According to the federal government's budget proposal for 2001, this sharing arrangement will be maintained as the new, flat rate income tax structure is introduced. Therefore, subnational government revenues will con- tinue their decline in 2001, not only due to the continued use of the lower sharing rate, but also due to the new, lower rate structure of the PIT. In addition to the centralization of PIT revenues, the gradual reassignment of VAT collection is con- tributing to an important increase in the centralization of public sector resources. From 1994 until the first quarter of 1999, the VAT was shared between the federal and regional governments on a 75/25 percent basis. During this period, the regional share of the VAT contributed between 15 to 20 percent of subnational own-source and shared revenue collections. However, beginning in the second quarter of 1999, the re- gional governments' share of VAT collections was reduced to 15 percent. The budget proposal for 2001 envisions assigning 100 percent of the VAT to the federal level. While this move would eliminate the current problems with the apportionment of VAT collections between regions, the centralization of VAT revenues would cause a substantial decline in regional revenue availability. The cumulative impact of the post-1998 tax reforms and the reassignment of revenue sources on the vertical fiscal balance is reflected by the changing distribution of resources across the different levels of government over time. During the early years of the transition the federal share of overall tax revenues steadily declined. While in 1992 approximately 60 percent of overall tax collections were assigned to Table 7.1. Legislated Sharing Rates of Major Taxes, 1999-2001 (percent) 1999 Ql 1999 Q2-Q4 2000 2001 (proposed) Tax Federal Regional Federal Regional Federal Regional Federal Regional VAT' 75 25 85 15 85 15 100 0 EPTb 37-34 63-66 37-29 63-71 37-29 63-71 37-29 63-71 piTc 10 90 10 90 16 84 16 84 Excises on alcohol 50 50 50 50 50 50 50 50 Energy excises 100 0 100 0 100 0 100 0 Excises on domestic production 0 100 0 100 0 100 0 100 Q Quarter. a. VAT on imports, precious metals, and stones is assigned 100 percent to the federal level. b. In 1999 Q2, the federal EPT rate was reduced to 11 percent, and regional EPT rates were lowered to 19 percent for most businesses and 27 percent for banks and insurance companies. c. In 1999, a 3 percent federal PIT surcharge was levied. Source: Budget legislation (various years). 94 Jorge Martinez-Vazquez and Jinneson Boex the federal level, by 1997 the federal share had declined to 42.5 percent (table 3.4). A sharp reversal in this trend began in 1998; since its low point in 1997, the federal share of consolidated tax revenues has steadily increased over time. The federal government's share of overall tax collections is once again expected to reach 60 percent by 2001 (table 7.2). Thus, after a period of decentralization of fiscal re- sources early in the transition, in recent years the revenue balance in Russia has shifted substantially in favor of the federal government.5 Changes in Intergovernmental Transfers Since 1998, reforms of the system of intergovernmental transfers have taken place on three fronts. First, substantial improvements have been made in the assignment of equalization transfers. Second, progress has been made in allocating earmarked transfers to pay federal mandates. Third, the federal government is prcmoting the introduction of objective, formula-based transfers at the regional-local level. These re- forms are consistent with the recommendations made in chapter 4. Arguably, the greatest progress in the realm of transfers has been made in improving the methodol- ogy for allocating the FFSR. The methodology for distributing the FFSR used until 1999 had many fea- tures of the Soviet-era, gap-filling approach. It provided regions with incentives to lower tax collections and maintain obsolete or excessive infrastructure in order to maximize intergovernmental transfers. However, the new methodology introduced in 2000 uses a modified representative revenue system to measure regional fiscal capacity, instead of revenue collections, and client-based measures of regional fiscal need, instead of physical budget norms. These improvements have resulted in a transparent and objective methodology that has removed the perverse incentives present in the methodology used until 1999. Federal authorities are also advocating the use of similar, formula-based equalization transfers at the regional-local level. While the first steps have been taken in the introduction of earmarked transfers, substantial work is still needed. The federal govemment has recently put in place a number of measures to end the practice of unfunded federal mandates, thereby restoring a certain measure of vertical fiscal balance to federal- regional relations in the Russian Federation. The extraordinary increase in revenue centralization envi- sioned in the 2001 budget proposal has allowed the federal government to set aside Rb 71 billion for ear- marked transfers to compensate regions for the cost of these federal programs. These transfers would lead to a better correspondence between regional expenditure responsibilities and fiscal resources available to the regions. However, as of this writing, the allocation formula for compensating regions for federal man- Table 7.2. Federal and Suibnational Shares of Consolidated Tax Revenues, 1998-2001 (percent) Category 1998 1999 2000 (budget) 2001 (estimate)a 2001 (proposed)a Federal share 43.4 50.8 52.3 55.4 59.7 Subnational share 56.6 49.2 47.7 44.6 40.3 Total 100.0 100.0 100.0 100.0 100.0 Note: The figures include revenues from the Mineral Resource Replacement Fund. a. Fiscal 2001 amounts reflect the estimate under 2000 legislation and the government's budget proposal. Source: Authors' calculations based on data from the Ministry of Finance. 5. The determination of the "right" level of revenue centralization is a policy decision that balances competing policy objectives, including macroeconomic stability and administrative efficiency, which often require centralized control over fiscal policy tools, and the desire to provide subnational governments with own-source and shared revenues to fund the decentralized provision of public goods. Postscript: Recent Reforms in Intergovernmental Relations in thie Russial Federatioln 95 dates has not yet been determined, causing substantial fiscal uncertainty as regions prepare for the 2001 budget. Thus, the effects of the proposed earmarked transfers cannot be determined at this time. Regardless of the distribution of the funds among the regions, the introduction of earmarked transfers would substantially improve the vertical fiscal balance among the different levels of government (table 7.3). In aggregate, the proposed level of intergovernmental transfers in the 2001 budget would cause public sector resources (tax revenues plus transfers) to be allocated in correspondence with expenditure responsibilities, which have historically been divided equally between the federal and subnational gov- ernments. However, despite these improvements in intergovernmental transfers, the current reforms ultimately fail to address the most important cause of vertical fiscal imbalance: the lack of tax autonomy for regional and local governments. Concluding Remarks The presence of this chapter highlights the fact that Russia's transition to a new federalism is an ongoing and rapidly changing process. While the opening years of the transition witnessed a substantial decen- tralization of government revenues, the reforms that have characterized Russia's decentralization poli- cies since 1997-98 emphasize the themes of federal strengthening, rationalization of intergovernmental fiscal relations, and increased revenue centralization. Despite the increased emphasis placed on federal- regional relations in Russia since Yeltsin's resignation, both by the Russian government as well as by the news media, the current reforms in intergovernmental relations should clearly be considered the con- tinuation of trends that were put in place well before Putin's rise to power. The ongoing policy reforms will undoubtedly leave the federal government politically and fiscally stronger at the expense of regional and local governments. However, the decentralization of fiscal and political power in the early years of the transition may have been more a side effect of the disintegration of the Soviet Union and the transition process than a path toward a sustainable federal framework. The trend in recent years toward greater federal control should not be surprising. Current trends are in line with pretransitional levels of fiscal decentralization in Russia, and the proposed reforms are not out of step with federalist arrangements in many other countries. Some centralization of fiscal resources in Russian may be defended in light of the prominence that should be given to the objective of macroeco- nomic stability, and perhaps also as a result of the continued presence of centrifugal forces in the Russian Federation. By all appearances, Putin is seeking to reassert control over federal institutions that Yeltsin first gave away and over which he ultimately lost control. A benign view of President Putin's policies maintains that he is seeking to return order and symmetry to an unruly and asymmetric federal system and impose structure on the sometimes chaotic transition process. A not-so-benign view put forward by some of those most critical of the reforms is that these may be the first steps toward a much more Table 7.3. Budget Proposal 2001: Federal and Regional Shares of Consolidated Revenues (rubles billions) Category Federal Subnational Tax revenue before grants, 1,193.5 881.9 (as a percentage of total) (57.5) (42.5) Total grants -208.7 208.7 Of which FFSR -98.8 98.8 Of which compensation for mandates -71.1 71.1 Total revenues after grants 984.8 1,090.6 (as a percentage of total) (47.5) (52.5) a. Revenues include those from the Road Fund. Source: Authors' calculations based on data from the Ministry of Finance. 96 Jorge Martinez-Vazquez and Jameson Boex centralized political system that practically eliminates subnational self-governance and brings back the traditional role of authoritarian control from Moscow. One aspect of the reform is certain: the declining importance of decentralized own-source and shared revenue sources presents a substantial threat to subnational revenue autonomy and the benefits associated with fiscal decentralization in Russia. Putin has pushed the pendulum of fiscal federalism further in the direction of centralization, and only time will tell how far the pendulum will swing. References Bahl, Roy, Galina Kourliandskaia, John Mikesell, Sally Wallace, Natalia Golovanova, Dmitry Shiskin, Andre Timofeev, Alexander Derugin, Yelena Nikolayenko, Inna Verbina, and Natatsha Minkova. 1998. "Intergov- ernmental Fiscal Relations in Leningrad Region." Georgia State University Consortium Russian Fiscal Re- form Project, Moscow. Berkowitz, Daniel, and David N. DeJong. 1998. "Russia's Internal Borders." University of Pittsburgh, Pittsburgh, Pennsylvania. Bird, Richard M., and Pierre Pascal Gendron. 1998. "Dual VATs and Cross-border Trade: Two Problems, One Solution." International Tax and Public Finance 5:429-42. Bird, Richard M., Robert D. Ebel, and Christine I. Wallich, eds. 1995. Decentralization of the Socialist State: Intergov- ernmental Finance and Transition Economies. Washington, D.C.: World Bank. Boadway, R., S. Roberts, and A. Shah. 1994. "The Reform of Fiscal Systems in Developing and Emerging Econo- mies: A Federalism Perspective." Policy Research Working Paper WPS 726. World Bank, Washington, D.C. Boex, Jameson, and Jorge Martinez-Vazquez. 1997a. "An Analysis of Alternative Measures of Fiscal Capacity for Regions of the Russian Federation." Georgia State University, School of Policy Studies, Atlanta, Georgia. _ - 1997b. "Fiscal Capacity: An Overview of Concepts and Measurement Issues and Their Applicability to the Russian Federation." Georgia State University, School of Policy Studies, Atlanta, Georgia. _ 1998a. "Reforming the Fund for the Financial Support of the Regions: An Analysis of the Proposed Equal- ization Mechanism." Georgia State University Consortium Russian Fiscal Reform Project, Moscow. _ - 1998b. "A Review of Solutions to Unfunded Mandates and Their Potential Application to the Russian Federation." Georgia State University Consortium Russian Fiscal Reform Project, Moscow. Campbell, Adrian. 1996. "City Government in Russia." In John Gibson and Philip Hanson, eds., Transformation from Below: Local Power and the Political Economy of Post-communist Transition. Cheltenham, U.K.: Edward Elgar. Freinkman, Lev, and Michael Haney. 1997. "What Affects the Propensity to Subsidize: Determinants of Budget Subsidies and Transfers Financed by the Russian Regional Governments in 1992-1995." Policy Research Work- ing Paper no. 1818. World Bank, Washington, D.C. Freinkman, Lev, and Stepan Titov. 1994. "The Transformation of the Regional Fiscal System in Russia: The Case of Yaroslavl." Report no. IDP-143. World Bank, Washington, D.C. Freinkman, Lev, Stepan Titov, and Daniel Treisman. 1998. "Russian Federation: Preempting a Potential Debt Crisis: Fiscal Performance in Russia's Regions." World Bank, Washington, D.C. Hunter, J. S. H. 1977. Federalism and Fiscal Balance: A Comparative Study. Canberra, Australia: Australian National University Press and Centre for Research on Federal Financial Relations. Igudin, A. G. 1998. "Analysis and Assessment of the Existing System of Intergovernmental Fiscal Relations in the Russian Federation." Ministry of Finance of the Russian Federation, Research Institute of Finance, Moscow. IEPT (Institute of Economic Problems in the Transition Period). 1998. "Basic Issues of the Concept of Develop- ment of Fiscal Relations between the Federal and Subnational Governments." Moscow. IMF (International Monetary Fund). 1994. Government Finance Statistics Yearbook. Washington, D.C. Kelly, Janet M. 1995. "Lessons from the States on Unfunded Mandates." National Civic Review 84(2): 133-39. Kirkow, Peter. 1996. "Distributional Coalitions, Budgetary Problems and Fiscal Federalism in Russia." Commu- nist Economies and Economic Transformation 8(3): 277-98. Lavrov, Alexei. 1995. "Russian Federation: First Steps, First Results." Moscow: Segodnya 104. _ . 1998. "Intergovernmental Fiscal Relations in Russia: An Agenda for Reforms." Presidential Administra- tion, Moscow. 97 98 References Lavrov, A. M., 0. V. Kuznetsova, J. W. David, and E. E. Skatershchikova. 1996. "The Budgetary System of the Subjects of the Russian Federation." World Bank, Moscow. Le Houerou, Philippe, and Michal Rutkowski. 1996. "Federal Transfers in Russia: Their Impact on Regional Rev- enues and Incomes." Comparative Economic Studies 38(2/3): 21-44. Martinez-Vazquez, Jorge. 1993. "Financing Social Expenditures in the Russian Federation: The Role of Oblast and Rayon Governments." Unpublished report prepared for the World Bank. Georgia State University, Atlanta. . 1994. "Expenditures and Expenditure Assignments." In C. I. Wallich, ed., Russia and tlhe Challenge of Fiscal Federalism. Washington, D.C.: World Bank. Martinez-Vazquez, Jorge, and Jameson Boex. 1997. "A Methodological Note on the Reform of Equalization Trans- fers in the Russian Federation." Report funded by the World Bank and the U.S. Agency for International Development. Georgia State University, Atlanta. Martinez-Vazquez, Jorge, Jameson Boex, and Roy Bahl. 1998. "A Methodology for Implementing Transfers to Depressed Regions in the Russian Federation." Georgia State University Consortium, Russian Fiscal Reform Project, Moscow. McLure, Charles E., Jr. 1994. "The Sharing of Taxes on Natural Resources and the Future of the Russian Federa- tion." In C. I. Wallich, ed., Russia and the Challenge of Fiscal Federalism. Washington, D.C.: World Bank. . 1998. "Federal and Oblast Taxation of Sales in Russia: Issues, Solutions, and International Experience." Georgia State University Consortium, Russian Fiscal Reform Project, Moscow. McLure, Charles E., Jr., Christine I. Wallich, and Jennie I. Litvack. 1996. "Special Issues in Russian Federal Fi- nance: Ethnic Separatism and Natural Resources." In Richard M. Bird, Robert D. Ebel, and Christine I Wallich, eds., Decentralization of the Socialist State: Intergovernmental Finances in Transition Economies. Washington D.C.: World Bank. Mikesell, John. 1998. "The Distribution of Natural Resource Taxes in the Russian Federation." Georgia State University Consortium, Russian Fiscal Reform Project, Moscow. Morozov, Alexander. 1998. "Intergovernmental Transfers Reform in the Russian Federation." Fiscal Policy Note. World Bank, Washington, D.C. OECD (Organisation for Economic Co-operation and Development). 1997a. Economic Surveys: Russian Federation. Paris. . 1997b. OECD Revenue Statistics 1965-1996. Paris. Ordeshook, Peter 0. 1995. "Reexamining Russia: Institutions and Incentives." Journal of Democracy (April): 46-61. Polishchuk, Leonid. 1996. "Russian Federalism: Economic Reform and Political Behavior." Social Science Work- ing Paper no. 972. California Institute of Technology, Division of Humanities and Social Sciences, Pasadena, California. Rafuse, Robert W. 1996. "Fiscal Federalism and Fiscal Disparities among the Regions of the Russian Federation: Some Preliminary Estimates." Intergovernmental Fiscal Reform Project. Barents Group LLC, Moscow, Rus- sian Federation. Solnick, Steven L. 1995. "Federal Bargaining in Russia." East European Constitutional Review 4(4): 52-58. Stewart, Kitty. 1996. "Are Intergovernmental Finances in Russia Equalizing?" European University Institute, Department of Economics, Florence, Italy. Sutherland, Douglas, and Philip Hanson. 1996. "Structural Change in the Economies of Russia's Regions." Eu- rope-Asia Studies 48(3): 367-92. Teague, Elizabeth. 1996. "Russia and the Regions: The Uses of Ambiguity." In John Gibson and Philip Hanson, eds., Transformationfrom Below: Local Power and the Political Economy of Post-communist Transition. Cheltenham U.K.: Edward Elgar. Titov, Stepan. 1997. "Different Aspects of Developments in Russia's Regional Fiscal System in Transition." Insti- tute for the Economy in Transition Project, Moscow. Treisman, Daniel. 1996. "The Politics of Intergovernmental Transfers in Post-Soviet Russia." The British Journal of Political Science 26(3): 299-335. References 99 .1998a. Fiscal Redistribution in a Fragile Federation: Moscow and the Regions in 1994. The British Journal of Political Science 28(1):185-222. . 1998b. Deciphering Russia's Federal Finance: Fiscal Appeasement in 1995 and 1996. Europe-Asia Studies 50(5): 893-906. United States Advisory Commission on Intergovernmental Relations. 1990. Representative Expenditures: Address- ing the Neglected Dimension of Fiscal Capacity. Information Report M 174. Washington, D.C. . 1996. The Role of Federal Mandates in Intergovernmental Relations: A Preliminary ACIR Report. Washington, D.C. Wallich, Christine I., ed. 1994. Russia and the Challenge of Fiscal Federalism. Washington D.C.: World Bank. Working Center for Economic Reform. Various years. Russian Economic Trends. London: Whurr Publishers. World Bank. 1996. Fiscal Management in the Russian Federation. Washington D.C. Zhuravskaya, Ekatherina. 1998. "Incentives to Provide Local Public Goods: Fiscal Federalism Russian Style." Harvard University, Cambridge, Massachusetts. THE WORLD BANK WBI Learning Resources are designed for use in WB1 courses and seminars. They discuss 1818 H Street, N.W. 7 Washington, D.C. 20433, U.S.A. issues in economic development policy and lessons from experience in a way that can be Telephone: 202 477 1234 u Facsimile: 202 477 6391 understood by persons without extensive Internet: www.worldbank.org background knowledge or technical expertise. E-mail: feedback@worldbank.org They will be of particular interest to readers concerned with public policy. 8.2 0-821 3-4840-X