Tax Systems and Tax Reforms in South and East Asia (Routledge International Studies in Money and Banking)

Tax Systems and Tax Reforms in South and East Asia

An international group of contributors provide a comprehensive analysis of tax systems and tax reforms in South and East Asian countries. Utilizing a broad range of case studies, the book discusses the determinants of the current structure of tax systems and examines the prospects for reform. The analysis is supported by an analytical overview of the tax systems of six countries. Tax Systems and Tax Reforms in South and East Asia identifies and analyses the common aspects of the countries’ taxation, the broad issues they face when devising tax policy and the need to build an efficient tax administration. Particular attention is devoted to the control of tax evasion and the role of tax administration. The hotly debated topic of fiscal federalism in relation to current tax systems is assessed with particular regard to China and India. The volume also explores the extent to which political factors, in addition to economic and demographic trends, may also represent a crucial challenge for the development of a welfare state in the region. The existing interactions between the economic structure, the corporate tax system and the attraction of FDI inflows are comprehensively analyzed and critical aspects of the design of a tax incentive policy are stressed: namely the risks of tax competition and the need of a comprehensive cost benefit analysis of FDI inducing measures. Luigi Bernardi is a Professor of Public Finance at the University of Pavia, Italy. Angela Fraschini is a Professor of Public Finance at the University of Eastern Piedmont, Italy. Parthasarathi Shome is an Adviser to the Union Finance Minister, Ministry of Finance, India.

Routledge international studies in money and banking

1 Private Banking in Europe Lynn Bicker 2 Bank Deregulation and Monetary Order George Selgin 3 Money in Islam A study in Islamic political economy Masudul Alam Choudhury 4 The Future of European Financial Centres Kirsten Bindemann 5 Payment Systems in Global Perspective Maxwell J. Fry, Isaak Kilato, Sandra Roger, Krzysztof Senderowicz, David Sheppard, Francisco Solis and John Trundle 6 What is Money? John Smithin 7 Finance A characteristics approach Edited by David Blake

8 Organisational Change and Retail Finance An ethnographic perspective Richard Harper, Dave Randall and Mark Rouncefield 9 The History of the Bundesbank Lessons for the European Central Bank Jakob de Haan 10 The Euro A challenge and opportunity for financial markets Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF) Edited by Michael Artis, Axel Weber and Elizabeth Hennessy 11 Central Banking in Eastern Europe Edited by Nigel Healey and Barry Harrison 12 Money, Credit and Prices Stability Paul Dalziel

13 Monetary Policy, Capital Flows and Exchange Rates Essays in memory of Maxwell Fry Edited by William Allen and David Dickinson

20 Central Banking Systems Compared The ECB, the pre-euro Bundesbank and the Federal Reserve System Emmanuel Apel

14 Adapting to Financial Globalisation Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF) Edited by Morten Balling, Eduard H. Hochreiter and Elizabeth Hennessy

21 A History of Monetary Unions John Chown

15 Monetary Macroeconomics A new approach Alvaro Cencini

23 Islamic Economics and Finance A glossary, 2nd edition Muhammad Akram Khan

16 Monetary Stability in Europe Stefan Collignon

24 Financial Market Risk Measurement and analysis Cornelis A. Los

17 Technology and Finance Challenges for financial markets, business strategies and policy makers Published on behalf of Société Universitaire Européenne de Recherches Financières (SUERF) Edited by Morten Balling, Frank Lierman, and Andrew Mullineux 18 Monetary Unions Theory, history, public choice Edited by Forrest H. Capie and Geoffrey E. Wood 19 HRM and Occupational Health and Safety Carol Boyd

22 Dollarization Lessons from Europe and the Americas Edited by Louis-Philippe Rochon and Mario Seccareccia

25 Financial Geography A banker’s view Risto Laulajainen 26 Money Doctors The experience of international financial advising 1850–2000 Edited by Marc Flandreau 27 Exchange Rate Dynamics A new open economy macroeconomics perspective Edited by Jean-Oliver Hairault and Thepthida Sopraseuth 28 Fixing Financial Crises in the Twenty-first Century Edited by Andrew G. Haldane

29 Monetary Policy and Unemployment The U.S., Euro-area and Japan Edited by Willi Semmler 30 Exchange Rates, Capital Flows and Policy Edited by Peter Sinclair, Rebecca Driver and Christoph Thoenissen 31 Great Architects of International Finance The Bretton Woods era Anthony M. Endres 32 The Means to Prosperity Fiscal policy reconsidered Edited by Per Gunnar Berglund and Matias Vernengo

33 Competition and Profitability in European Financial Services Strategic, systemic and policy issues Edited by Morten Balling, Frank Lierman and Andy Mullineux 34 Tax Systems and Tax Reforms in South and East Asia Edited by Luigi Bernardi, Angela Fraschini and Parthasarathi Shome

Tax Systems and Tax Reforms in South and East Asia

Edited by Luigi Bernardi, Angela Fraschini and Parthasarathi Shome with a foreword by Vito Tanzi

First published 2006 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business

This edition published in the Taylor & Francis e-Library, 2006. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2006 Selection and editorial matter, Luigi Bernardi, Angela Fraschini and Parthasarathi Shome; individual chapters, the contributors. All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested

ISBN 0-203-08824-7 Master e-book ISBN

ISBN10: 0-415-38959-3 (hbk) ISBN10: 0-203-08824-7 (ebk) ISBN13: 978-0-415-38959-4 (hbk) ISBN13: 978-0-203-08824-1 (ebk)

Contents

List of figures List of tables List of contributors Foreword: tax systems and tax reforms in South and East Asia by Vito Tanzi Preface Acknowledgments

ix x xiii xv xxi xxviii

PART I

A general picture of tax systems and tax reforms in South and East Asia 1 Overview of the tax systems and main tax policy issues

1 3

LUIGI BERNARDI, LAURA FUMAGALLI AND LUCA GANDULLIA

2 The control of tax evasion and the role of tax administration

35

PARTHASARATHI SHOME

3 Fiscal federalism in the big developing countries: China and India

63

ANGELA FRASCHINI

4 Democracy and welfare without welfare state

92

MATTEO CACCIATORE, PAOLA PROFETA AND SIMONA SCABROSETTI

5 Features and effects of corporate taxation on FDI FRANCESCO D’AMURI AND ANNA MARENZI

109

viii Contents PART II

Country studies of tax systems and tax reforms in South and East Asia 6 China

135 137

DOMENICO D’AMICO

7 India

164

LUIGI BERNARDI AND ANGELA FRASCHINI

8 Japan

192

LUIGI PASCALI

9 Malaysia

218

GIANPAOLO FANARA

10 South Korea

234

LIDIA CERIANI

11 Thailand

254

MARCO BARTOLICH

Index

275

Figures

2.1 2.2 2.3 4.1 4.2 5.1 5.2 6.1 8.1 8.2 8.3 8.4 8.5 8.6

8.7 8.8

Tax and expenditure: policy and administration Tax and expenditure: gaps in administration and control Income tax: Online Tax Accounting System (OLTAS) Democracy in South and East Asian countries: China, Korea and India Democracy in South and East Asian countries: Japan, Thailand and Malaysia GDP per capita, PPP, 1975–2003 Gross FDI inflows as percentage of GDP 1975–2003 The distribution of tax revenue in China Rates of economic growth in Japan, 1990–2005 (%) Rates of growth of the main components of GDP in Japan, 1990–2005 (%) Rate of unemployment in Japan, 1990–2005 (%) A comparison between the fiscal revenues in G7 countries and Korea, 2000 The movement of income distribution in terms of Gini coefficient in Japan, 1951–2000 Redistributive effects of income tax (percentage difference of the Gini coefficient before and after taxes) in Japan, 1951–2000 Revenue, expenditure and bond issuance in Japan, 1990–2003 General government gross financial liabilities, 1990–2005. A comparison between some OECD countries

46 47 55 95 95 111 111 144 194 194 195 196 206

206 213 213

Tables

1.1 1.2 1.3 1.4 1.5 1.6 1.7

3.1 3.2 3.3 3.4 3.5 5.1 5.2 5.3 5.4 6.1 6.2 6.3

Some economic and social indicators in selected South and East Asian countries, 2003 US$ per capita incomes and taxes/GDP % in selected South and East Asian and other countries, early 2000s Structure and development of fiscal revenue in selected South and East Asian countries, 1992–2002 (% of GDP) Regression results for total fiscal pressure (TFP) growth – selected countries, 1991–2002 Implicit tax rates in selected South and East Asian countries, year 2000 (approximately) Effectiveness of value added tax – selected countries and 1997 data Central government percentage share on general government main tax revenues – selected federal countries Government layers by countries Comparison of selected items Intergovernmental transfers, 1997–2001 Number of local bodies at different tiers Revenue and expenditure of local bodies (rural and urban) in % of total The main features of the corporate tax systems Summary of corporate tax incentives in China and India Summary of corporate tax incentives in Malaysia and Thailand Summary of corporate tax incentives in South Korea and Japan General government expenditure in China, 1997–2002 (% of GDP) Shares of central and local governments in selected budget expenditures in China, 2001 Structure and developments of consolidated general government revenue in China, 1991–2002 (% of GDP)

8 10 11 13 14 24

28 65 67 71 75 76 119 123 124 125 140 141 142

Tables xi 6.4 6.5 6.6 7.1 7.2 7.3 7.4 7.5 7.6 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8 9.9 10.1

10.2 10.3 10.4

PIT rates on employment income in China, 2002 PIT rates on household business income in China, 2002 Business tax rates in China, 2003 Indian yearly percent rates of economic growth Percent shares of Indian GDP Structure and developments of consolidated general government revenue in India, 1989–2002 (% of GDP) Structure and development of consolidated central government revenue in India, 1989–2002 (% of GDP) Structure and developments of state government revenue in India, 1989–2002 (% of GDP) Tax rates, assessment year 2004–5, India (Finance Bill) Structure and development of fiscal revenue in Japan, 1965–2002 (% of GDP) Progressive tax rates on personal income in Japan Tax rates on corporate income in Japan A comparison between the Gini coefficient before tax and after tax among some OECD countries Dispersion of deduction rate by income class in Japan (%) Total tax wedge on labor in 2003 in G7 countries and Korea (marginal rate in %) Marginal effective tax wedge in manufacturing in G7 countries in 1998 (%) Expansionary fiscal packages since 1992 in Japan (trillion YEN) Malaysian GDP by sector of origin, 1960–95 (% of GDP) Fiscal revenue and public expenditures in Malaysia, 2002 (% of GDP) Structure and development of government revenue in Malaysia (% of GDP) Comparison with the fiscal revenue structure of some selected countries, 2001 (% of GDP) Different forms of relief for the determination of taxable income in Malaysia Income tax rates for resident individuals in Malaysia The rate of the RPGT in Malaysia Current rates of sales taxes in Malaysia Structure and developments of taxation by function and by implicit tax rates in Malaysia, 1992–2002 Structure and developments of consolidated general government revenue in South Korea, 1975–2001 (% of GDP) Personal income tax rates in South Korea Inheritance and gift tax rates in South Korea Corporation tax rates in South Korea

146 147 153 167 168 170 171 172 176 199 201 202 208 209 210 211 212 219 221 222 223 225 225 227 227 230

237 242 242 243

xii Tables 10.5 10.6 11.1 11.2 11.3 11.4 11.5 11.6

Average effective tax rates on capital, labor and consumption in South Korea Distributional effects of taxes and social security contributions in South Korea, 1998 (%) Structure and development of fiscal revenues in Thailand, 1992–2002 (% of GDP) Structure of fiscal revenues in selected countries, 2001 (% of GDP) PIT rates in Thailand CIT rates in Thailand Tax rates of the customs duties undergoing reform in Thailand Structure and development of taxation by function and by implicit tax rates in Thailand, 1992–2002

246 247 257 260 262 263 265 268

Contributors

Marco Bartolich is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia. Strada Nuova 65, 27100 Pavia, Italy. Luigi Bernardi is Professor of Public Finance, University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Matteo Cacciatore is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Lidia Ceriani is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Domenico D’Amico is at the Mediterranean University of Reggio Calabria; Università degli Studi Mediterranea di Reggio Calabria, Via Zecca 4, 89125 Reggio Calabria, Italy. Francesco D’Amuri is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Gianpaolo Fanara is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Angela Fraschini is Professor of Public Finance at the University of Eastern Piedmont; Dipartimento di Politiche Pubbliche e Scelte Collettive, Università del Piemonte Orientale “A. Avogadro.” Via Cavour 84, 15100 Alessandria, Italy. Laura Fumagalli is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy.

xiv

Contributors

Luca Gandullia is Associate Professor of Public Finance at the University of Genoa; Dipartimento di Scienze Economiche e Finanziarie-DISEFIN, Università di Genova. Largo Zecca 8–14, 16124 Genova, Italy. Anna Marenzi is Associate Professor of Public Finance at the University of Insubria; Dipartimento di Economia, Università dell’Insubria, Via Ravasi 2, 21100 Varese, Italy. Luigi Pascali is a PhD student at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Paola Profeta is Professor of Public Economics and Lecturer of Economics at the Università Bocconi; Instituto di Economia Politica, Via Gobbi 5, 20136, Milano. Simona Scabrosetti is a Research Assistant at the University of Pavia; Dipartimento di Economia Pubblica e Territoriale, Università di Pavia, Strada Nuova 65, 27100 Pavia, Italy. Parthasarathi Shome is Adviser to Union Finance Minister; Government of India, Ministry of Finance, North Block, New Delhi, 110001, India. Vito Tanzi is a Senior Consultant at Inter-American development Bank, 5912 Walhonding Road, Bethesda Md 20816, USA.

Foreword: tax systems and tax reforms in South and East Asia Vito Tanzi

This book contains descriptions and analyses of the tax systems of six Asian countries. These are among the most important countries of the region. The six contain three colossi (China, India, and Japan) and three less large, but still important countries (Malaysia, South Korea, and Thailand). With the exception of Japan, that has been undergoing some economic difficulties in recent years, the countries are growing very fast so that as a group they constitute the fastest growing region in the world today. Japan, of course, had its period of extraordinary growth when, within a little more than a generation, it went from being a relatively poor country much damaged by World War II to becoming the second largest and most powerful economy in the world. Today, in spite of its difficulties, it is still the second largest economy. What strikes a reader who reads these chapters is the fact that some of the concerns that have guided the tax policy of Western countries have been absent, or have not played much of a role in the tax systems of these countries. In them the clear and almost single-minded objective has been the pursuit of growth. Equity has definitely taken a back seat. The concept of ability to pay, that has been important in guiding tax policy in the West, has played almost no role. It is interesting to read that during the American occupation of Japan, after World War II, general MacArthur invited Professor Carl S. Shoup, Professor of Public Finance at Columbia University and, at that time, perhaps the most famous public finance economist in the world, to come to Japan and design for the Japanese a good tax system. Shoup spent some time in Japan preparing his recommendations. Of course, he brought with him the intellectual baggage that was common at that time among American economists. This baggage gave a prominent role to the principle of ability to pay and to equity and to the view that a global income tax, imposed with highly progressive rates, satisfied best that principle. This was the view made popular by Henry Simons in Personal Income Taxation, his 1938 book. A global income tax required that all the earnings of a person, regardless of source, be put into a kind of virtual basket and then taxed as a whole with progressive rates. It was thus conceptually different

xvi

Foreword

from the kind of taxation of income proposed, for example, by De Viti De Marco that preferred to tax different sources differently. The global income tax acquired the status of the fairest of all taxes in the United States in the 1950s and 1960s and American economists tried to push this idea to other countries. The Japanese politely listened to Professor Shoup and introduced a tax system that reflected the Americans’ recommendations. Then, as soon as the Americans left, politely and quietly they started changing that system to make it better reflect Japanese traditions and preferences that were different from the Americans’. In a society with a low Gini coefficient and low per capita income, as was the Japanese society at that time, equity did not seem to be as important an objective as growth. The Japanese, then, and other South East Asian countries, later, were driven by the objective of stimulating growth. This drive is reflected clearly in the various chapters. Another aspect that immediately strikes the reader of these chapters is the relatively low level of taxation in these countries. All of these countries have ratios of taxes to gross domestic product (GDP) that are less than 30 percent. In three of them (Thailand, India, and China), the ratio is less than 20 percent. To Europeans that cannot finance all their public spending with tax levels that are often well above 40 percent of GDP, it must be a great puzzle to understand how these Asian countries can manage with so little taxation. The answer is, again, tied to the earlier point. These countries have not been seduced by the often inefficient and confused redistributive role of the state as European countries have. The main function of the government in these countries has remained largely an allocation function. This has made it possible to keep low the ratio of taxes to GDP. These ratios rarely grow high because of the allocation role of the state. However, pressures have been building up so that several of these countries have been developing problems common to several European countries: growing or high fiscal deficits (India, Malaysia, Japan) and substantial public debts (Japan, India, Thailand, Malaysia). In time, the tax levels of these countries will need to increase, but it is unlikely that they would increase to the European levels. The financial crisis that, in 1997–8, hit several of the countries of South and East Asia, and especially Thailand, Korea, and to a lesser extent, Malaysia showed the cost to the citizens of these countries of not having a well developed and government-financed safety net for individuals who, from one day to the next, found themselves without a job and without an income to support their families. Many tragic stories came out of these countries in that period. The 1997–8 financial crisis indicated that the traditional, non-government sponsored safety net that had existed in these countries failed in a situation when the equivalent of what could be defined a “perfect storm” hit them. That crisis was interpreted, especially by Western observers, as a strong warning to the governments of these

Foreword xvii countries that they needed to play a larger role in protecting their citizens against economic risks. That role would of course require much higher and, perhaps, different taxes. There is little if any evidence that taxes have gone up after the crisis. Maybe the governments have reached the conclusion that “perfect storms” are rare events and that higher taxes may retard growth. Therefore, in their calculations they may have concluded that it may be a better course of action to continue with low tax levels even if this means that, should another crisis come, they would be as unprepared to face it as the last time. After all, these countries recovered rather quickly from that crisis and they have resumed significant growth rate. Changing this policy mix might reduce their growth rate, an outcome that they would find undesirable. The apparent, relative indifference toward equity concerns is also evident from particular features of the tax systems. For example, while the European countries have preferred to have a lower rate for more essential products in the value added tax, Thailand, Japan and Korea have introduced value added taxes with single rates on relatively broad bases. Again, the equity concern has been ignored. Only China has seen it convenient to have preferential rates for some goods, but its basic rate is relatively high at 17 percent and in China, the VAT is a relatively recent addition. Furthermore, its income distribution is becoming a major concern for Chinese policymakers. They may thus see the merit of using the tax system to reflect equity concerns. These countries’ lack of concern, at least as expressed through their tax systems, toward equity may also be seen in the predominance of indirect taxes. In practically all of these countries, indirect taxes, including import duties, seem to play a bigger role than they do in OECD countries. However, these indirect taxes cover still a broad range of different and not always modern taxes and they need to be streamlined to facilitate their administration. As already mentioned, the tax systems of these countries seem to be oriented towards promoting growth. The counterpart of this is the limited use of income taxes and especially of personal income taxes. Even in Japan, where income taxes are important, we find that for the most part, they are levied with rates that subject most taxpayers to largely proportional and low-rate taxes. Most of the taxpayers face only the first rate in the progressive rate structure. Furthermore, returns to savings (interest incomes, dividends, etc.) are taxed at relatively low rates to prevent discrimination against saving. This introduces a schedular character to these systems. The absence, or the almost absence of significant social security systems means that the tax wedge on dependent labor is low. Thus, there are no or lesser disincentives to work as compared with European countries. Workers do not face the push to remain unemployed or to go into

xviii Foreword informal or underground activities in order to avoid high taxes. Perhaps because of this, the employment rate in these countries is very high and the unemployment rate is very low. Even at the peak of the financial crisis of 1997–8, the unemployment rates remained lower than they are in Europe. The absence of taxes on labor income and the dependence on taxes on consumption also make the tax systems relatively neutral across generations. Retired people continue paying taxes levied on their consumption. Reading these chapters, one gets the distinct impression that the tax systems of these countries have been chasing the structural changes in their economies and, to a much lesser extent, the macroeconomic policy developments. Especially the formers have been moving at a rapid pace reflecting the fast rates of growth of these countries. Some of the literature on tax structure changes that was popular some decades ago stressed the fact that tax systems are much influenced by the structure of the economy and that they adjust to that structure. Thus, as the structure changes, so must the tax systems. But this normally happens with a lag. When the economies change at the phenomenal rate experienced by these countries in recent years, it is obvious that the process of adjustment of the tax system may appear too slow and need to be speeded up. But changing tax systems is never easy. It is a difficult and time-consuming process. This means that, inevitably, the systems look a bit out of phase from the economy and a bit out of date. The reader who is familiar with the European tax systems, systems that had a much longer period to adjust to the (slow changing) structure of their economies, will be struck by the differences of these systems from those of most OECD countries. It is realistic to expect that the six countries will continue to reform their tax systems and to bring them more closely in line with those of OECD countries. These changes are likely to: (i) raise somewhat the tax level, because of the existing fiscal deficits and because of the need for more public spending; (ii) redress the direct– indirect taxes imbalances; (iii) eliminate many antiquated taxes; and (iv) reduce the recourse to features that may have been common in the past, but that are no longer so at the present time. How far the systems will change will, in part, be determined by the role that these countries’ policymakers will wish to play in the economy. It is unlikely, given their traditions, that they will want to adopt the European social system that requires high levels of taxation. Two particular features that merit some comments are tax incentives and earmarking. Tax incentives have been relied on by all the countries in the sample to pursue particular objectives and especially to push for more investment and more growth. Western tax economists have been generally skeptical about the role of these incentives, preferring lower general tax rates over reduced rates for specific purposes. However, this campaign against tax incentives generally had little success, especially in Asia.

Foreword xix Nobody can answer with certainty whether these incentives had the hoped for positive effect. Those who favor tax incentives for investment can point to the high investment rates of these countries and the high growth rates as an indication that the incentives had the desired effect. Those who oppose them can argue that these results may not be linked to the incentives. They would have occurred without the incentives. This debate has been going on for decades and this book will certainly not settle it. However, it will provide information on the extent of these incentives and on the implicit belief on the part of the policymakers of these countries that incentives can change the behavior of economic agents. For example, Korea has gone as far as using tax incentives to stimulate the use of credit cards on the part of citizens. Less general, but still common in these countries is the use of earmarking. Earmarking is shown to be particularly common in Japan and Korea. This practice is also frowned upon by economists. They argue that a unified budget with an effective political process that chooses how public resources will be used is more likely to lead to optimal results. Earmarked resources by definition escape this political filter and can lead to distortions and abuses. Also in a dynamic sense, over the long run, they can provide too many or too little resources to the activities to which the earmarking is directed, because the outcome depends on the sensitivity of the earmarked taxes with respect to economic growth. However, this is not a first-best world so that many decisions are made within a process that faces many constraints and many imperfections. When political cycles, lobbies, or corrupt practices lead to the appropriation of resources for uses that have a lower social value, it may be justified to rely on earmarking to protect specific areas or sectors from the possibility that they will be bypassed by the political process. Obviously, some of these countries have considered it worthwhile to earmark resources for particular uses that they wanted to isolate from the political process. A last issue that receives some particular attention in the chapters is fiscal federalism. This is especially the case in the two countries with the largest population and territory (China and India). This is a big issue in many countries. Fiscal decentralization reduces the degrees of freedom that national policymakers face in their tax policies and tax reforms. In 1994, China made a major tax reform that had the main objective of restoring some balance between the revenue that went to the national government and the revenue that went to the sub-national governments. The central or national government was running the risk of being cut out from much of the (shrinking) revenue. The ratio of taxes to GDP in China reached a very low point in 1996 and started rising after that year, in part as the result of the 1994 reform. But equally important is the fact that a much higher share of the higher revenue started going to the national government as a consequence of the reform. This, however, may have created difficulties for the provincial and local governments that saw

xx

Foreword

their resources shrink. They have been relying on fees and other “nontax,” but still obligatory, payments by farmers and enterprises to provide them with needed revenue. These “non-tax” revenues have been growing rapidly. This has created pressures to replace them with regular taxes. At some point, this change will take place and, when it does, it will push up, perhaps substantially, the ratio of taxes to GDP in China. India also had great difficulties in dealing with the problems created by fiscal federalism. Here, too, a large share of total taxes goes to the subnational governments. Constitutional limitations have made it difficult for the national government to modernize and rationalize the country’s tax system. Ancient and inefficient taxes (such as the octroi) have continued to exist and to fragment the economy while modern taxes, such as the value added tax, could not be introduced because of the legal impediments and opposition from the states that did not want to lose this tax autonomy. The result has been a system of indirect taxation that could only be defined as a nightmare. At the same time, the structure of the economy, as well as the political and administrative obstacles, prevented the greater use of the personal income tax that continued to be collected from only a small minority of potential taxpayers. As the economy grows, the number of potentially taxable individuals will increase. A reform introduced in April 2005 is supposed to bring improvements to the system of indirect taxation. India, like China, is undergoing a period of fast growth that is creating a dual economy, with part of the population in a modern and increasingly efficient sector, while a larger part remains in the much less efficient agricultural sector. One of the future challenges for both countries will be to create tax systems that accommodate this duality and that do it within a context of fiscal federalism. Time will tell how successful they will be. I hope that the reader will benefit from reading this book.

Preface

In the first century BC, Lucius Annaeus Seneca, the distinguished Roman philosopher and scientist, wrote that the Roman Empire was suffering large money losses because of the goods imported from India and “Sera” (⫽China). The flow continued until the Renaissance, along the silk ways and spice routes. Since the late Middle Ages, Venice and other European City-States have raised custom duties on South and East Asian imports. The rates were astutely low on raw materials and high on manufactured products. The industrial revolution brought back European fears about Asian competition. These fears are nowadays rising again, as a consequence of the impressive catching up phase entered by most Asian countries and especially by China and India. Indeed, a critical aspect of this catching up phenomenon is that it is largely export-led. Producers, workers and consumers living in Western European countries, where the average yearly income per capita is about US$22,000, appear to be willing to restore custom duties and to fight a commercial war against countries whose per capita income is one twentieth or less. Apart from its economic validity, we cannot help finding this attitude ethically unsuitable. But among the branches of an increasingly hot debate, there is an increasing willingness to uncover the “inevitable trick” of Asian countries’ competitive edge. Does industrial production maintain security and quality standards? What about the systems of social protections? Last but not least, do firms, workers and consumers pay any tax, and if so, how much? How efficiently and equitably? Trying to give an answer to the last questions is the main aim of this book, which follows the Routledge companions of 2004 (EU 15 Countries, edited by L. Bernardi and P. Profeta) and 2005 (New EU Members, edited by L. Bernardi, M. Chandler and L. Gandullia). Both books were prepared under the supervision of V. Tanzi and run out in a few months. But the new aim is not one easy to reach. We selected a sample of representative countries, made up of two big developing countries (China and India), two transition “tiger” economies (Malaysia and Thailand) and two industrialized countries (Japan and South Korea). Only two countries in the sample (Japan and South Korea) are considered by OECD in revenue

xxii

Preface

statistics and analyses. Databases for the other countries are lacking or critically unsatisfactory: the most common source, IMF’s Government Statistics, does not present recent data and data only refer to central government. National sources differ in standards and itemization. With this book we have thus produced a consistent dataset on the main taxes prevailing in the sampled countries. In order to do so, we have critically evaluated and compared different sources, especially as to the opaque giants of China and India. Moreover, we tried to build some more elaborate if approximate indicators of fiscal pressure, like ITR in particular. We present the main trends of taxations in selected countries, at least from the early 1990s up to the early 2000s, and comment on them. We give a userfriendly picture of the basic features of the main taxes in each country. We summarize the most recent tax reforms, those under way and those deemed as necessary to go on the remaining critical points. However, our aim is not just to inform the interested reader. We want to equip him with a series of critical analyses on the main tax policy issues which arise across or inside the selected countries. To begin with, in his Foreword, Tanzi first notices that the main concern that has guided the tax policy of these countries has been the pursuit of growth. Equity has taken a back seat. This has had a series of consequences. Taxation has a relatively low level because the function of the government has remained largely an allocation function. The relative indifference toward equity concerns also played an evident role in shaping particular features of the tax systems. Where introduced, VAT is single rate, except in China and, now, in India. Indirect taxes predominate and the use of income taxes, especially personal income tax, is generally limited. The absence, or the almost absence, of significant social security systems means that the tax wedge on dependent labor is low. Thus, there are no or lesser disincentives to work compared with European countries. However, pressures (1997–8 financial crisis and its social consequences; budget deficits; high public debts) have been building up so that in time the tax levels of these countries will need to increase. Furthermore, Tanzi points out that the tax systems of these countries have been chasing the structural changes in their economies. But this normally happens with a lag. Inevitably, the systems look a bit out of phase from the economy and a bit out of date. It is realistic to expect that these countries will continue to reform their tax systems and to bring them more closely in line with those of OECD countries, but it is unlikely that they will reach European levels of taxation. Tanzi concludes by focusing on three particular features. First, tax incentives, which have been relied upon especially to push for more investment. Tax economists have been generally skeptical about the role of these incentives preferring lower general tax rates. However, this campaign has had little success, especially in Asia. Nobody can answer with certainty whether these incentives have had the hoped-for positive effect. Second, the use of earmarking. This practice is also frowned upon by economists. However, when

Preface xxiii political cycles, lobbies or corrupt practices lead to the appropriation of resources for uses that have a lower social value, it may be justified to rely on earmarking. Finally, the troubles of fiscal federalism, especially in China and in India. Tanzi reminds us that these two largest countries are undergoing a period of fast growth that is creating a dual economy. One of the future challenges for both countries will be to create tax systems that accommodate this duality and that do so within a context of fiscal federalism. An overview of taxation and the main tax policy issues are presented in the chapter by Bernardi, Fumagalli and Gandullia (Chapter 1), mostly in line with Tanzi’s intuitions. The authors point out that South and East Asia is a fast developing and economically integrated area, but made up of not so homogeneous (developing, transition and industrialized) countryclusters, lacking in any supra-national authority. Total fiscal pressure is somewhat low when compared with that of other worldwide countries where per-capita income is not so different. There is, however, the evidence of a (smooth) Wagner’s law, which supports the forecast of a future increase in tax levels. According to common experience of developing and transition economies, indirect taxes prevail over direct ones. Thus a low tax wedge on labor improves efficiency, by inducing both supply and demand of work. The price paid is that of a heavy burden on consumption, which tends to lessen equity and increase welfare losses. According to the authors, any next attempt to go forward towards a uniform analysis of South and East Asia countries’ tax policy issues would however be misdirected. They believe it would be far better to consider tax policies issues concerning the whole area separately from those more specific to each cluster of similar countries. In this vein the authors notice that intraregional economic integration poses severe challenges to the Asian area. Three tax policy issues seem most problematic: the emerging feature of intra-country free trade agreements, the following revenue consequences of reduction in foreign trade taxes and the increasing tax competition for FDI. Country clusters’ tax policy issues are different from each other. As to the industrialized Japan and South Korea, different levels of comprehensiveness characterize the basis of the personal income tax, whose burden is poorly distributed. The two countries are, on the contrary, facing the same problem of aging population, requiring them to raise social contributions, and, eventually, VAT. Malaysia’s direct taxes look higher than in Thailand, another transition country, but just by accident. Thailand has adopted VAT, while Malaysia has not yet changed its old-fashioned sales tax. Both the countries are engaged in enhancing revenue by improving tax administration. In both developing countries China and India, income tax is still not large and tends to be poorly redistributive. VAT is well established in China, while it is just arriving in India, after a long awaited but challenging reform, especially to intra-layer tax relationships among different levels of government. Taxing power is now more centralized in

xxiv Preface China, but this needs correction to avoid a lack of accountability on the part of provinces. The role of tax administration in maximizing revenue generation and minimizing tax evasion cannot be over-emphasized, especially to developing and/or transition countries like most South and East Asian countries. This topic is dealt with in Shome’s chapter (Chapter 2). The chapter begins with a general discussion of the problem of tax evasion. In the following sections, Shome analyzes how to keep tax evasion in check. The tax administration must: (i) incorporate genuine threat of penalty but ensure due process; (ii) computerize as many administrative processes as possible; and (iii) not remain aloof from tax policy but assist in every way possible to help design a simple tax structure and its commensurate tax law. Emphasis is on actual experiences drawn from South and East Asia. A lot of conclusions follow. There are various sources and causes of tax evasion. However, globalization seems to have led to a growth in taxevasive behavior. To combat tax evasion, tax administrations should have adequate resources. Incentive schemes should be provided for tax officers. In fact, a review of Asian countries reveals an increase in administrative action and follow-up in the administration of international taxation. As to the specific tools to be placed in force, Shome stresses that tax administrations are increasingly requiring third-party information returns in which a third party is asked to submit returns on transactions with others. Furthermore, large taxpayer units are being used in many countries to reduce tax evasion by requiring all taxes to be paid through one window by large taxpayers. Given the growth in cross-border transactions, tax administrations are also increasingly training officers and applying arm’s length rules in the setting of transfer prices. Joint audits by central and local governments are being carried out to facilitate matching of information across a wide geographic area. In an increasingly complex world faced by the tax administration, computerization is an essential step to take. It is, therefore, crucial to design tax policy that is implementable. Shome concludes by observing that innovations in tax design are appearing to reduce evasion. But he reminds us that taxes should not be devised merely to accommodate tax administration if the consequence follows of raising inefficiencies in resource allocation and inequity among taxpayers. As already pointed out by Tanzi in his foreword, fiscal federalism is a hotly debated aspect of current tax systems. The topic is analyzed in depth in the chapter by Fraschini (Chapter 3). Broadly speaking, in South and East Asian countries a highly centralized government prevails, although recently some trends are moving towards a greater degree of decentralization. Also the two giants China and India, which cannot rely on a merely centralized government, have experienced a greater or lesser degree of fiscal unionism. As to China the local government system provides four levels: provincial, city, county and township. Intergovernmental fiscal relations were revamped by the 1994 reform which established a new tax

Preface xxv sharing system. The local financial revenue mainly derives from local taxes, shared taxes and non-tax revenue. As to India, the federal system is quite complex. The center–states relations are envisaged in the Constitution also for financial aspects: two 1992 constitutional amendments made India one of the most politically decentralized countries among developing ones. However, the implementation of the decentralization program is still lagging. Trying to evaluate the Chinese and Indian systems, Fraschini observes that both countries have not viewed fiscal decentralization as a comprehensive system. Until now India seems to have considered decentralization mainly in terms of the local election system, without the transfer to local authorities of all functions provided for by decentralization. With regard to China, the 1994 fiscal reform gave local governments more control over the administration of local taxes but no significant degree of tax autonomy and no substantial expenditure assignments. Only India set up a different system of local bodies in rural and urban areas with different expenditure responsibilities and financing powers. On the contrary, China has a unitary fiscal system. In India it is necessary to redesign the transfer system to improve accountability, incentives and equity, whereas in China the fiscal revenue sharing schemes limit intergovernmental budget transfers. Finally, the rule of hard budget constraint in China is faced by all levels of government, while in India sub-national governments face soft budget constraint. Low fiscal pressure is possible in South and East Asian countries especially because they have a “light” welfare state, characterized by low public spending on welfare. In their chapter (Chapter 4), Cacciatore, Profeta and Scabrosetti note that enterprises and families have traditionally played a major welfare role that has partially compensated for the low public spending. However, these forms of enterprise and family welfare are currently being challenged by economic conditions: the financial crisis, the falling fertility and aging process, as well as by some common trends, such as urbanization, family nuclearization and the rise of female employment (which implies a reduced ability of women to care for their parents or children). As a consequence, public welfare spending is expected to increase in South and East Asian countries. Some countries have already introduced important reforms in the last decade to strengthen their social welfare systems (Japan, Korea), while others have them on their agenda (China). In this chapter, the authors explore whether, in addition to economic and demographic trends, political factors may also represent a crucial challenge for the development of a welfare state in South and East Asian countries. Many of these countries show a trend towards increased democratization and participation of civil society, which raise demands for government to assume more responsibility for the unemployed, sick, poor and the elderly. This chapter contributes to the analysis of the relation between the democratization process and the emergence of a welfare state in South and East Asian countries. The authors argue that, although it

xxvi Preface seems not to be essential to have democracy in order to start reforms aimed at raising the welfare programs (e.g. in China), the interrelation between economic and political liberalization may be important to characterize the outcome of the final stage of the reform process, with the best performances arising when political and economic development go hand in hand. In South and East Asian countries, the existing interactions between the economic structure, the corporate tax system and the attraction of FDI inflows are of paramount importance. They are analyzed in depth in D’Amuri’s and Marenzi’s chapter (Chapter 5). The authors begin by remarking that with respect to FDI taxation, countries design their tax policy aiming at particular economic goals that differ depending on their economic structure. Developing countries, such as China and India, aim to attract FDI in order to boost economic growth. China has managed to attract much more foreign investment thanks to a resolute policy of trade liberalization. India has performed more cautiously in this field. Transition countries such as Malaysia and Thailand are interested in attracting those FDI that can generate positive spillovers in strategic sectors, technology and knowledge transference. Finally, industrialized countries such as Korea and Japan offer fewer incentives to foreign investors, strictly targeting IT and R&D enterprises. D’Amuri and Marenzi go on with a broad analysis of the economies of the countries included in the sample, especially by considering the openness of their markets, the level of industrialization and their capacity to attract FDI. Then the authors afford a detailed review of the main ideas in the literature about the controversial effectiveness of the different systems of corporate income taxation and tax incentives on economic development and capital inflows. On these grounds, they move to analyze the corporate tax structures of the selected countries, by stating the main features of the basic tax regimes and going through the jungle of the various and complex systems of tax incentives. Finally they bring the topics together, by evidencing how the different stages of economic development of the countries are reflected in their respective corporate tax systems. Moreover some critical aspects of the design of a tax incentive policy, namely the risks of tax competition and the need a comprehensive cost–benefit analysis of FDI inducing measures, are stressed. So much for the matter. To sum up and conclude, South and East Asian tax systems are not very different from the corresponding ones of, respectively, developing, transition and industrialized countries, according to the cluster to which each of them pertains. Of course, they have some particular features which deserve some specific attention. Almost all countries made a choice in favor of a low fiscal pressure, a “light” welfare state and small or unaccomplished level of fiscal decentralizations. However, these choices and trends are changing somewhat, as a consequence of the increase of political democracy, which is underway. Inevitably this takes

Preface xxvii place with some lags. Obviously research as wide and deep as performed here suggests many improvements of selected countries’ tax systems. Tax structures need to be modernized, in accord with OECD standards, to reach higher levels of equity, neutrality, efficiency and simplification. Many tax policy issues that are country specific must be solved. Tax administration and the fight against evasion must be strengthened, in order to reduce the share of the black economy. Fiscal federalism should be deepened, especially in giant countries. Intra-area tax relationships have to find some coordinating supra-national authority, to avoid harmful tax competition practices and a “race to the bottom” aimed at attracting FDI. These are our evaluations and our policy suggestions. However, we have to alert the reader that the “trick,” if any, of South and East Asian countries’ competitive edge does not seem to have its roots only in tax systems. A low fiscal pressure may well have contributed to a higher growth. However, no tax wedges have also played a relevant role. Luigi Bernardi, Angela Fraschini and Parthasarathi Shome

Acknowledgments

This book is the result of the research on “Globalizzazione, dinamiche demografiche e futuro dei sistemi fiscali,” fostered by the Italian Ministry of Education, University and Scientific Research, “Scientific research programs of national interests – financial years 2004.” Operating Unit, “Development and issues of tax systems within main areas of world economy,” directed by L. Bernardi: Department of Public and Environmental Economics, University of Pavia (Italy). Additional financing from the Fondazione Cassa di Risparmio delle Provincie Lombarde and from the Provincia di Pavia is gratefully acknowledged. The Editors are grateful to S. Scabrosetti, who, in addition to her own contribution, provided excellent research assistance during the whole project. They also thank A. Badhuri, from the University of Pavia, G. Barbieri from Ref-Milan, G. Crespi-Reghizzi, from the University of Pavia, the IBFD of Amsterdam, L. Le Masurier from Oxford University, R. Puglisi from MIT and LSE, and Y. Y. Yjng from Shanghai’s University for information, comments, suggestions and help in a number of areas.

Part I

A general picture of tax systems and tax reforms in South and East Asia

1

Overview of the tax systems and main tax policy issues Luigi Bernardi, Laura Fumagalli and Luca Gandullia

Introduction, contents and main conclusions When compared with other areas of the world’s economy, the case of South and East Asia is somewhat particular. The region is not just fast growing, but also highly integrated, as in North America or in Western Europe, for instance. The participant countries are, however, barely homogeneous, like in South America or, to a lesser degree, in Eastern Europe. There is a lack of a supra-national authority able to provide coordinating policies for single countries and to harmonize their institutions. This particular feature is fraught with consequences for most tax policy issues. The total fiscal pressure of South and East Asian countries looks somewhat low when compared with that of countries with a similar per-capita income, pertaining to other economic world areas. The main explaining factors can essentially be found, first, in the absence, or in a very small level, of social contributions, and, second, in a still widespread infant stage of personal income tax (PIT). However, a smooth Wagner law is confirmed by the data, so that fiscal pressure is destined somewhat to increase as growth continues. According to a common experience of developing and transition countries, indirect taxes prevail over direct ones. The exceptions are, of course, Japan (but not South Korea) and, more surprisingly, Malaysia. Corporation tax revenue usually stays higher than personal income tax, despite the flood of incentives allowed for corporations. On the contrary, PIT is still in its primary stages everywhere except in Japan. VAT is well established in China, Japan and South Korea, where it prevails on excise duties, and has just been introduced in April 2005 in India. Custom duties, entirely on imports, are still present in India, China and Thailand. A relevant consequence of such a prevailing tax structure is the particular ranking of the implicit tax rates. It is only in Japan and Korea that labor income is more heavily taxed than capital and consumption, while the opposite happens in Malaysia and Thailand. The same may be said for China and India. A low tax wedge on labor (due to the limited role played by PIT and the absence or the irrelevance of social contributions)

4 Luigi Bernardi et al. improves the efficiency, by inducing both supply and demand of labor. Generally speaking, the heavy burden on consumption lessens the equity, as the taxes affect the prices in a regressive way. In terms of welfare (consumer’s surplus) the excess burden increases. The previous data and information make it clear that any uniform analysis of the South and East Asian countries’ tax policy issues would be quite fruitless. It is far better to consider the following tax policy issues separately: first, those which arise inside the whole area and, second, those more specific to each country. The latter may be organized according to the clusters of some countries which emerge owing to their economic and social characteristics, and also in their tax systems (India and China; Malaysia and Thailand; Japan and South Korea). Intra-regional economic integration poses severe challenges to the tax structure in the Asian area. As trade barriers come down and capital mobility increases, the challenges for South and East Asian countries become particularly acute, because of their tax administration capabilities and their dependence on foreign trade taxes that are relatively more limited. Three tax policy areas seem more problematic: the building of intra-countries’ agreements on reducing trade tariffs, the revenue consequences of trade reform with reduction in foreign trade taxes and the increasing tax competition for direct foreign investment. Around the world there has been a substantial growth in common markets, customs union and free trade areas. Also in the Asian area there are two blocs of countries where economic cooperation and integration has been strengthened during the past few years. The first bloc is represented by the Association of Southeast Asian Nations (ASEAN) that recently implemented the ASEAN Free Trade Area (AFTA), making significant progress in trade and investment liberalization by the lowering of tariffs in intra-regional trade. At present a second bloc is represented by the South Asian Free Trade Area (SAFTA), comprising of India and six other countries, where tariffs on internal trade are going to be eliminated. SAFTA is an agreement between the seven South Asian countries that form the South Asian Association for Regional Cooperation (SAARC). SAFTA will come into effect in 2006. As is well-known, developing countries and emerging markets still rely on trade taxes. For different reasons, trade taxes on imports have often been introduced to protect domestic production and those on exports reflect in part the export of primary products over which the country has some monopolistic power. Standard economic theory suggests that taxes on international trade have a major distorting effect and that efficiency gains deriving from their reduction far outweigh the loss of such revenue sources. The reduction of taxes on import trade often runs into serious political opposition, due to the pressure of domestic producers. Moreover, one of the most important constraints to trade reform in such countries is the conflict between tariff reforms and macro-stabilization goals. The

Overview of tax systems and tax policies 5 adverse revenue impact of tariff reductions could be redressed in the short run by reducing existing exemptions, by removing highly restrictive non-tariff barriers and by relying on the expected import volume growth. But in the long run it will require the implementation of compensatory revenue measures. It is in the area of tax competition where the growing economic integration and capital movements between Asian countries pose relevant challenges to the existing national tax structures. As non-tax barriers decline, investment decisions and location of investment become more tax sensitive. Within free trade areas firms can supply different national markets from a single location. The relevant issue here is the temptation for such countries to broaden the scope of tax incentives to attract and compete for foreign direct investment (FDI). The main argument in favor of these national policies is that FDI can contribute to increase the productivity of the domestic economy, but the effectiveness of tax incentives is highly questionable. Almost all South and East Asian countries have made and still make extensive use of tax (but also non-tax) incentives to compete for promoting domestic investment and especially to attract FDI. However, in the next years the use of tax incentives for the promotion of FDI will require a coordinated multilateral approach, at least on a regional basis. Agreements on a regional basis – for instance between the member countries of the ACFTA (ASEAN with China) and those belonging to SAFTA – could create different kinds and varying degrees of cooperation. Countries, for instance, could agree on a limited set of tax incentives, conditioned to certain criteria. With regards to intra-countries clusters’ tax policy issues, it is primarily necessary to note that in Japan and South Korea, a striking difference emerges in terms of their PIT structure. Looking at the features of PIT, we can point out an important dissimilarity that attests the existence of two alternative theoretical visions in addressing the problem of personal taxation. The original (1940s) structure of the first Japanese personal income tax system was based on an idea of comprehensive taxation, but this was soon transformed into a system founded on a notion of expenditure income. On the contrary, South Korea still maintains a “global” income taxation that aggregates most personal income (inclusive of many forms of capital revenues), taxing it at progressive rates. Therefore, in the debate between the two aspects of equality, Japan and South Korea have taken opposite directions. Whereas the reforms implemented in South Korea in the 1990s put increasing effort in extending the tax base, Japan sets up a complex set of allowances that made the base smaller and smaller. A second issue that has to be discussed when comparing South Korea and Japan is the way in which the two countries are handling the problem of a troublesome rise of the expense needed for pensions. The growing imbalance due to an aging population constitutes a worrying threat for

6 Luigi Bernardi et al. the pension system. Both Japan and South Korea opted for a sharp increase in the share of social security contributions, but, whereas in Japan the population structure shows the typical features of a developed country (inverted pyramid type), in South Korea the demographical transition is still at an earlier stage so that the pensions’ system imbalance looks less pressing. It is clear, however, that a VAT increase could be recommended in both countries, provided that VAT hits in a non-distortionary way, especially for the aged, whose pensions in these countries are exempt from income tax. As to the cluster made up by Malaysia and Thailand, at first glance Malaysia presents direct taxes that seem far higher than in Thailand and nearly in line with OECD standards. In fact the value of corporate income taxation accounts for a comparatively high value. Nevertheless, this value is inappropriate for evaluating the real impact of the average fiscal pressure on corporations in Malaysia, since it also includes a peculiar tax levied on petroleum companies whose tax rate is much higher than the “standard” one. In both the two countries PIT has a narrow weight. It relies on a progressive schedule with many brackets and a widely spread set of tax rates. However, the scarce pervasiveness of the tax, along with the massive use of personal reliefs and personal tax rebates makes the pursuit of horizontal equity almost infeasible. With regards to indirect taxation a peculiar feature in Malaysia is, first, the complete absence of any kind of value added tax. In fact the most important heading of indirect tax is an ad valorem single stage tax, imposed at the import and manufacturing levels. The main problem that arises is the existence of cascading non-neutral price effects. An attention to low income earners in the shaping of consumption taxes can be found in both countries. After the Asian financial crisis, Malaysia and Thailand also had to find a way to recover their revenues. The main field in which both countries are increasing their efforts is in the strengthening of revenue collection. In order to achieve this goal, one of the fundamental pillars in the agenda of the two governments is the rationalization of administrative procedures and the improvement of the efficiency of tax administration. Finally, we must consider China and India. Let us first look at direct taxation. In India, personal income tax is imposed by the Union Government. It may look supply friendly: the tax brackets are few and rate graduation is not steep. Taxable income is very similar to global income according to Haig-Simons definition. However, many personal exemptions narrow the tax base, so limiting the effects of the aggregation of incomes. The Chinese system of personal income taxation presents opposite features. The Chinese PIT has a “pure” schedular structure with many different types of income taxed at different rates, no aggregation of alternative sources of earnings and no personal deductions. As a consequence, in China like in India, albeit due to different factors, there is both a loss

Overview of tax systems and tax policies 7 on the ground of progressivity, and a fall in the total tax burden. With regards to corporate taxation, the main field for tax planning in China is the tax environment created by some special incentives granted to foreign enterprises, in particular, a generous tax holiday which adds to an investment-encouraging tax regime of amortization. While useful to attract foreign investors, tax holidays in time may be harmful because some “race to the bottom phenomena” can ensue, as well as a rise in the relative tax burden on home taxpayers. In the field of indirect taxation the two countries feature important differences. In China the main indirect tax is the value added tax, which recently has widely been updated. In India until 2004 VAT was absent. From 1 April 2005 VAT has been finally introduced. This reform is expected to be welfare improving, since VAT will substitute for a huge, complex and distortionary amount of sales tax and excise duties. In practice, however, such a large reform will have to get over some difficult challenges, such as the taxation of services, the adequacy of the tax administration and concerning the complex structure of intergovernmental tax relationships. The system of public finance is now, after the reforms of the 1990s, more centralized in China, and most taxes are charged by the central government which transfers part of the revenue to the inferior layers. The system can be powerful to per-equate fiscal capacities among provinces, while weakening their fiscal responsibility. Hence now it is under reform, to avoid harmful fiscal imbalances of the lower layers and in order to increase their budget transparency and fiscal effort.

How much are South and East Asia economies and tax systems uniform? Like, unlike or cluster economies? To begin with, look at the top rows of Table 1.1. Our selected sample of South and East Asian countries seems to be made up of dissimilar countries, in terms of their geo-demographical features. China is 9,597 thousand km2 large, Malaysia just 48. India 3,287, South Korea not more than 99. China and India are populated by more than one billion people; Malaysia, South Korea and Thailand stay under 40 million. Therefore, at once, the population to area ratio looks uneven. A historical and cultural perspective confirms this first glance. China and Japan have their roots in long-lasting unitary empires, although they reached their present state in very different ways. India was a British dominion until 1947. South Korea was dominated by Japan from 1919 until 1945. At that time an independent Republic was formed in the whole South Korean peninsula. Subsequently, South Korea emerged as a separate and Western-liking state in 1953, after the war with the communist North region. Thailand has been an independent kingdom since 1932; Malaysia only ceased to be a part of

4.3 33.1 127

556.2 520 2,900 5.6 4.0 ⫺6.2

3,287 1,058 322

India

5.3 11.1 9

4,190.7 32,859 28,000 2.0 ⫺0.3 ⫺9.2

372 127.6 342

Japan

3.8 10.9 59

101.0 4,042 9,000 4.2 1.7 ⫺5.2

329 25 76

Malaysia

Notes Some data may differ from that given by the country chapters because of different sources and/or reference year.

Sources: United Nations, IMF, FACT-CIA for per capita income PPP corrected and country chapters.

4.3 14.2 94

1,372 1,062 5,000 9.1 1.2 ⫺2.7

GDP (US$ billion) Per capita GDP (US$) Per capita GDP (US$ PPP corrected) Yearly rate of growth (%) Yearly rate of inflation (%) Central government deficit/GDP (%)

Unemployment (%) Index of human poverty (%) Ranking of human development

9,597 1,284 134

Area (1,000 km2) Population (millions) Pop./area km2

China

Table 1.1 Some economic and social indicators in selected South and East Asian countries, 2003

3.4 11.0 28

515.3 10,641 17,700 2.5 3.3 2.8

99 48 481

South Korea

2.0 12.9 76

130.7 2,037 7,400 5.0 1.4 0.8

513 60 125

Thailand

Overview of tax systems and tax policies 9 the British Empire in 1957. Races, religions, languages, social aptitudes and institutions are fairly diverse. Notwithstanding all this, especially during recent decades, albeit beginning at different starting times, these countries shared a common way of fast catching up growth that was quite sustainable in terms of inflationary pressures and (with just Japan’s and Malaysia’s exceptions) budget balance. The employment record of any country is near that of other countries (see Table 1.1). Current trends in industrial production, inflation and real rates of exchange are broadly speaking very similar. The evolving economic environment unavoidably caused huge changes in production structures and social aptitudes. A process like this strengthened both the economic and the financial integration of the whole area (REF 2004) and created strong domino effects along cyclical moves. Intra-area commercial exchanges of total exports’ shares reach from 33 percent (China towards Japan and other Asian countries) up to more than 47 percent (Japan towards China and other Asian countries). The whole region shares a common financial conflicting position inside world rates of the exchange arena, especially with respect to the US$. Rates of exchange re-evaluations, as requested by EU, US and other commercial and competitive countries, would have the twin effect of losing competitiveness but increasing internal financial wealth and vice versa. Now go back, to Table 1.1, but instead look at the central and bottom rows. Some countries are still in the first stages of development with a very low level of per capita income, not over US$520 yearly in India. Some others are very mature and rich. In Japan per capita income reaches nearly US$33,000 yearly. The other countries lie in between, but not in an evenly dispersed order. China is close to India. South Korea stays well above the other countries. The two “tigers,” Malaysia and Thailand, have their place between the two poor and the two rich countries. A lot of indicators confirm this ranking: per capita income PPP corrected (albeit with some peculiarities), the index of human poverty and the degree of human development (see Table 1.1). To sum up, the final picture shows a structure of clusters. Two fast developing yet still poor countries (China and India); two transition countries (Malaysia and Thailand); and finally, two more or less mature industrialized countries (Japan and South Korea). When compared with other areas of the world’s economy, the case of South and East Asia therefore is somewhat particular. The region is highly integrated, from an economic viewpoint, as happens, for example, in North America or Western Europe. However, the participant countries are not entirely homogeneous, like in South America or, to a lesser degree, as in Eastern Europe. There is no supra-national authority which serves to coordinate single countries’ policies and to harmonize their institutions. Subsequently, we see the relevance for tax policy issues of the topics which we have briefly discussed.

10

Luigi Bernardi et al.

A general overview of countries’ tax systems and their development since the early 1990s The total fiscal pressure of South and East Asian countries looks somewhat low when compared with that of countries with a similar per-capita income (Table 1.2), pertaining to other economic world areas. This is more or less true for India and China as developing countries which stay only just below some Latin American or Central Asian low fiscal pressure countries. However, they are very far from CIS or Eastern Europe countries. About the same may be said for the two transition countries, Malaysia and Thailand. By considering now the pair of industrialized countries, we may notice that South Korea’s fiscal pressure stays below the figure of Western European countries by an average of about GDP eight points. Japan is far under the Western European standard, while it is very near to that of the United States. The main explanatory factors of the South and East Asian countries’ low fiscal pressure essentially can be found, first in the absence, or in a very small level, of social contributions, and, second, in a still widespread infant stage of personal income tax. We go back to these features subsequently. Before doing this, we notice, from Table 1.3 (as to its time-series evidence), and from Table 1.2 (along a cross-countries path) that Wagner’s law might not be confirmed. From the early 1990s to a decade Table 1.2 US$ per capita incomes and taxes/GDP % in selected South and East Asian and other countries, early 2000s Per capita income

Taxes/GDP

India Costa Rica Ukraine China Kazhakistan Belarus

520 384 640 1,062 1,230 860

14.6 17.5 34.2 16.8 19.6 44.3

Thailand Russia Colombia Malaysia Poland Chile

2,037 1,730 2,290 4,042 4,100 4,630

16.6 37.0 16.2 17.8 40.4 19.3

10,641 12,127 12,103 32,859 25,299 35,676

27.4 35.9 33.9 27.3 45.9 26.4

South Korea Greece Portugal Japan Finland United States

Source: Country chapters; Bernardi (2005) and related sources; OECD 2004; Datastream on IMF data.

■

About 2002

14.6 8.1 6.5 4.0 1.3

2.2 – 2.7 21.3 8.7 4.5 3.3 1.0 30.0

13.1 7.8 9.1

2.4 1.1 1.2 12.3 –

8.8 3.3 0.6 15.3 – – – – 15.3

7.4 7.9 –

n.a. n.a. –

n.a. – 0.7 20.3 – – – – 20.3

9.8 n.a. n.a. 9.8 n.a.

13.5 2.6 3.0

5.1 – 2.7 18.1 1.0 0.7 0.3 – 19.1

6.2 3.5 2.7 9.2 3.9

15.9 1.1 0.2

3.9 3.0 – 16.8 0.2 0.1 0.1 – 17.2

5.6 1.8 2.9 11.2 3.9

10.1 6.7 –

0.9 2.3 1.3 16.8 – – – – 16.8

4.2 1.0 2.9 11.3 5.9

6.3 8.3 –

8.8 1.8 0.7 14.6 – – – – 14.6

3.2 1.4 1.8 10.7 –

Source: Own elaborations on IMF, OECD, World Bank and national data. See country chapters for details.

Direct taxes, of which: 3.0 Personal income 0.0 Corporation income 2.7 Indirect Taxes, of which 8.5 VAT-general consumption 2.6 Specific excises and sales taxes 2.6 Custom duties 0.8 Other taxes 0.9 Total taxes revenue 12.4 Social contributions – Employers – Employees – Self-employed – Total fiscal revenue 12.4 Administrative levels Central government 3.2 Local government 9.2 Social Security – 10.1 6.9 10.3

2.1 – 2.8 17.0 10.3 5.1 4.1 1.2 27.3

9.0 5.5 3.5 5.2 2.4

n.a. n.a. –

1.2 1.0 1.1 17.9 – – – – 17.9

11.8 2.9 7.3 5.0 2.7

16.4 5.5 5.5

5.2 – 4.2 22.4 5.0 2.8 2.2 – 27.4

7.3 3.7 3.6 10.9 4.9

13.9 2.1 0.6

4.4 1.8 – 16.0 0.6 0.3 0.3 – 16.6

5.5 1.9 2.9 10.5 3.0

China India Japan Malaysia South Korea Thailand China India Japan Malaysia South Korea Thailand

About 1992

Table 1.3 Structure and development of fiscal revenue in selected South and East Asian countries, 1992–2002 (% of GDP)

12

Luigi Bernardi et al.

after, a significant increase in total fiscal pressure may be observed just for China and South Korea. The remaining countries stayed in the same position, as consequence of the policies adopted to recover economic growth after the 1997–8 slowdowns (Malaysia and Thailand) or to overcome the long-lasting after-bubble stagnation (Japan). Furthermore, as the per capita income doubles or grows (from India to China, to Thailand, to Malaysia, to South Korea and to Japan), fiscal pressure’s increase is far smoother. Therefore, a more formal test seems interesting. According to standard literature (e.g. Musgrave 1969; Burgess and Stern 1993; Tanzi 1994), the test to be performed assumes as its maintained hypothesis the dependence of total fiscal pressure (TFP) from real per capita income (RPCI), the share of agriculture (usually untaxed or taxed very slightly) on GDP (AGR), the openness of the economy (OPE: imports are still a relevant tax handle, see the following sections), and the debt/GDP ratio (DEBT), as a measure of push on taxes to fulfill the long-running sustainability of the country’s public finance. According to the data of our countries’ panels, the regression results are depicted in Table 1.4. The unconstrained (log) specification shows that total covariance performs well and all coefficients but the one for DEBT are fairly significant. By dropping DEBT, results on the remaining variables and the overall statistical performance of the whole estimated equation hardly change at all. The sign of AGR is however the opposite to that expected. This uncomfortable result may be explained as a consequence of a decreasing and/or flat fiscal pressure in more than one of our countries during the 1990s, when the share of agriculture went down almost everywhere. Finally, by constraining the model to just RPCI and OPE as explaining variables, the latter still performs well, as does the overall equation. Two observations deserve, however, some further attention. First, the figure of roughly 0.3 of the elasticity of TFP to RPCI is not very high and seems more determined by the cross-countries’ than by the time-series shares of the panel. Second, OPE does not seem, as has been maintained, to represent the tax handle for custom duties but instead the “quality” of economic development. Not surprisingly therefore, it seems to be correlated with RPCI, as it might be suggested by the high value of the standard error of the estimated coefficient. To conclude, a smooth Wagner’s law seems rooted in the data and is able to predict a certain but smooth increase of countries’ TFP as economic growth continues. According to a common experience of developing (Burgess and Stern 1993) and transition (Bernardi 2005) countries, indirect taxes prevail (Table 1.3) over direct ones in our sample of South and East Asian countries. The exception is, of course, Japan (but not South Korea) and, more surprisingly, Malaysia,1 where direct taxes dominate, as occurs more often in industrialized countries. Corporation tax revenue usually stays higher than personal income tax, despite the flood of incentives allowed to corporations, especially to attract FDI. On this topic we have to underline the

Values

0.1989 0.3115 0.1157 0.0425 0.64 23.12

Variables

RPCI AGR OPE DEBT R2 Log likelihood

16.92 6.61 5.58 0.95

T-stat 0.2035 0.3498 0.1219 0.63 22.65

R2 Log likelihood

Values

RPCI AGR OPE

Variables 18.98 14.35 6.2

T-stat

R2 Log likelihood

RPCI OPE

Variables

Table 1.4 Regression results for total fiscal pressure (TFP) growth – selected countries, 1991–2002

0.57 25.26

0.3112 0.1112

Values

19.80 3.00

T-stat

14

Luigi Bernardi et al.

effects for Asian countries (except Japan and South Korea) to be free from OECD surveillance from harmful tax competition. These countries are then particularly attractive for FDI. China is an extreme case. A specific and “ring fenced” corporation tax regime is established just for foreign companies and a generous tax holiday is allowed just to them, but not to the national ones. The control of transfer prices both for internal and international transactions is, however, increasingly performed, but just by national tax administrations, as it is shown in Chapter 2. Hence the institution of a World Tax Organization, as repeatedly and authoritatively suggested by V. Tanzi (e.g. 1999), seems called for. The other great personal tax, i.e. PIT, on the contrary, is still at an infant stage almost everywhere but in Japan and has not significantly increased since the early 1990s. A relevant consequence of such a prevailing tax structure is the ranking of implicit tax rates (Table 1.5), which looks quite different from that which characterizes industrialized countries, especially in Western Europe (Bernardi 2004). It is important to note that while in Japan and in South Korea labor income is more heavily taxed than capital and consumption, as in almost all industrialized countries, the reverse happens in Malaysia and Thailand. Herein consumption is taxed two- or threefold labor income. The same should happen in China and India, albeit we lack the data needed for fine ITR estimations. One unavoidable conclusion arises. A low tax wedge on labor (due to the early stage of PIT and the absence or mild relevance of social contributions) improves efficiency, by enhancing both labor’s supply and demand. The heavy burden on consumption, however, lessens equity, as taxes are passed on prices and result in being regressive. In terms of welfare (i.e. consumer surplus) (e.g. Steve 1976) excess burden is raised. Only a question of country clusters or also one of tax system clusters? The previous discussion makes it clear that any further uniform analysis of South and East Asian countries’ tax policy issues would be quite futile. It is

Table 1.5 Implicit tax rates in selected South and East Asian countries, year 2000 (approximately)

Labor income Consumption Capital

China

India

Japan

Malaysia

South Korea

Thailand

n.a. n.a. n.a.

n.a. n.a. n.a.

28.3 6.0 n.a.

6.3 11.1 24.1

27.3 13.8 n.a.

6.2 16.6 7.1

Source: OECD for Japan and South Korea, own calculations for the other countries. See country chapters for details. Notes Capital means National Accounts gross operating surplus.

Overview of tax systems and tax policies 15 far better to consider separately the tax policy issues which arise inside the whole area and those more specific of each country’s cluster. But do the country clusters that we discovered on economic and social characteristics (India and China; Malaysia and Thailand; Japan and South Korea) correspond also to tax system clusters? The answer seems to be affirmative, albeit with some exceptions, as follows (see Table 1.3). Total fiscal pressure China and India share the lower figures among the whole set of selected countries, although China is very near to Thailand. South Korea and Japan have a very similar value, about ten points over the remaining countries. Malaysia and Thailand do not differ by more than about one GDP point. Broad tax headings China and India have the lower and similar ratio between direct and indirect taxes. Social contributions are completely lacking. Japan and South Korea share a similar figure for direct taxes, while Japan stays in a higher position for social contributions and South Korea (where social contributions are already present and growing) for indirect taxes. Also in Malaysia and in Thailand there is an opposing ratio between direct and indirect taxes. This depends on the tax handle opened to Malaysia by the profits of the petroleum companies. Social contributions do not exist in Malaysia, but in Thailand they are very light. Single taxes In China and in India corporation tax remains higher than personal income tax. To a lesser degree, this is also true for both Malaysia and Thailand, but not for South Korea and Japan. In China VAT prevails on excise duties; in India excise duties have commonly allowed for deduction on taxes paid on intermediate and capital goods bought. This feature is shared also by South Korea and Japan and by Malaysia too, while a different pattern is observed only in Thailand. Custom duties reach a higher level in India and in China, are on the whole lower in Malaysia and in Thailand, and are totally absent in South Korea and Japan.

Intra-zone tax policy issues Intra-regional economic integration raises severe challenges to the tax structure in the Asian area. As trade barriers come down and capital mobility increases, the challenges for South and East Asian countries become particularly acute, because of their relatively poor tax administration

16

Luigi Bernardi et al.

capabilities and their high dependence on foreign trade taxes. Three tax policy areas seem most problematic: first, the building of intra-country agreements on reducing trade tariffs; second, the revenue consequences of trade reform because of reduction in foreign trade taxes and third, the increasing tax competition for foreign direct investment. Around the world there has been a substantial growth in common markets, customs’ unions and free trade areas. Also, in the Asian area there are two blocs of countries where the economic cooperation and integration has been strengthened over recent years. The first bloc is represented by the Association of Southeast Asian Nations (ASEAN) that recently implemented the ASEAN Free Trade Area (AFTA), making significant progress in trade and investment liberalization by the lowering of tariffs in intra-regional trade. ASEAN is developing enhanced cooperation with other countries in the area, namely with China, South Korea and Japan (so-called “ASEAN plus Three cooperation”). The intention is to create a larger East Asia Free Trade Area (EAFTA). Between China and Southeast Asian countries in particular, a new free trade agreement has recently been reached. This is likely to encourage competition among countries and improve economic efficiency in the region (Cordenillo 2005). The agreement recently put into practice creates the largest free trade area (ACFTA). Tariffs on almost all inter-ASEAN–China trade in goods will be gradually eliminated. The new free trade area between ASEAN and China has caused a domino effect, with Japan, South Korea and even India moving faster to sign similar agreements with Southeast Asian countries. At present a second bloc is represented by the South Asian Free Trade Area (SAFTA), comprising India and another six countries, where tariffs on internal trade are going to be eliminated (Choon 2002). SAFTA was decided upon amongst the seven South Asian countries that form the South Asian Association for Regional Cooperation (SAARC). SAFTA will come into effect in 2006, by reducing tariffs for intra-regional trade and replacing the earlier South Asia Preferential Trade Agreement (SAPTA). It may lead to a fully fledged South Asia Economic Union. The new free trade area provides for free movement of goods in the region through the elimination of tariffs, other duties that include border charges and fees, and nontariff restrictions on the movement of goods. On the one hand these new free trade areas produce economic benefits to the countries involved in terms of increased bilateral trade, greater economic efficiency and increased bilateral FDI. On the other hand, progress in economic integrations poses new challenges: intensified competition in domestic markets, loss of tariff revenues and the possibility of temporary unemployment. Commonly understood as a difference (Heady 2002) from developed countries, developing countries and emerging markets still rely on custom duties and less on internal sources. The difference in revenue patterns is mainly explained in terms of administrative convenience, as for a develop-

Overview of tax systems and tax policies 17 ing country it is relatively easy to observe and value goods as they cross international frontiers. For different reasons, trade taxes on imports have often been introduced to protect domestic production and those on exports reflect in part the export of primary products over which the country has some monopolistic power. Standard economic theory suggests that taxes on international trade have a major distortionary effect and that efficiency gains deriving from their reduction far outweigh the loss of such revenue sources. As a general principle, it is well known that it is optimal for a small open economy to raise any revenue it needs by setting all tariffs to zero and relying entirely on destination-based taxes on consumption (Dixit 1985). Keen and Ligthart (2002) formally show that any tariff reduction that increases production efficiency, coupled with a consumption tax reform which leaves consumer prices unchanged, increases both welfare and public revenue. This provides a rationale for those strategies (often recommended by the World Bank and the IMF) of sequential tariff reforms with strengthening of domestic consumption taxation, often in the form of a value added tax. The reduction of taxes on import trade often runs into serious political opposition, due to the pressure of domestic producers. Moreover, one of the most important constraints to trade reform in such countries is the conflict between tariff reforms and macro-stabilization goals (Mitra 1992). Concerns about revenue losses may be exacerbated by the short-term expenditure pressures that can arise, due to increases in social outlays for displaced workers (for instance, IMF 2005). A review of revenue statistics confirms that the decline in tariff revenue in developing countries and emerging markets around the world is underway. According to Zee (2004), tariff revenue in non-OECD countries accounted for 22 percent of total tax revenue in the first half of 1990s and for 17.4 percent in the second half. The comparison with the OECD countries’ figures (respectively 1.9 percent and 1.4 percent) suggests that the process will continue in the future. The challenges vary with the degree of economic development. In the selected South and East Asian countries, excluding Japan and South Korea (where, as in the other OECD countries, trade taxes are so small that the OECD Revenue Statistics does not compute them on average fiscal pressure), the role of trade taxes is almost stable, accounting for about 2 percent of GDP on average and for 10–14 percent of the total tax revenue in countries like China and Thailand. In the transition economies of Southeast Asia (Cambodia, Lao, Myanmar and Vietnam) the role of indirect taxes and particularly of trade taxes is predominant; last years’ custom duties accounted for figures between 11 and 35 percent of the total tax revenue; it has been estimated that the participation in the new free trade area will cause revenue losses ranging from 14 to 60 percent of the total customs revenue (Tongzon et al. 2004). For a number of countries the adverse revenue impact of tariff reductions could be redressed in the short run by reducing existing

18

Luigi Bernardi et al.

exemptions, by removing highly restrictive non-tariff barriers and by relying on the expected import volume growth. But in the long run it will require the implementation of compensatory revenue measures, like (Tanzi and Zee 2000): a broadening of tax bases; a review of existing exemptions and preferential regimes in the field of indirect taxation (turnover taxes, VAT, import duties) in order to verify their justification and effectiveness; as well as compensating for tariff reductions on excisable imports by an increase in their excise rates and increasing the rates of the general consumption taxes (VAT, sales taxes, turnover taxes). More generally, as noted by Zee (2004), this process will induce policymakers in developing countries and emerging markets to reform and modernize their tax systems fundamentally by virtue of necessity rather than choice. As a general principle, attempts to reform the structure of protection trade taxes cannot ignore the role of the domestic indirect tax structure since the latter can – and in many countries does – affect protection (Mitra 1992). In many cases the World Bank has made reference to the need to “harmonize” the domestic tax treatment of imports and domestic production, given that an asymmetric treatment of domestic and imported goods under the domestic indirect tax system implies that domestic indirect taxes may add to the protective effect of trade taxes. A central issue in tax policy design is whether – as recommended by the World Bank – such countries should be encouraged to move away from trade-based taxes and existing commodity taxes towards more broad consumption taxes. There are conceptual merits in using domestic consumption taxes – mainly general sales taxes like VAT, but also excises on particular goods – to offset any revenue loss from tariff reduction (IMF 2005). A strategy of matching a reduction in the tariff rate on some final consumption good with an increase in the corresponding domestic tax on consumption on that same good will leave, for a small open economy, the price faced by consumers unchanged. However, the government’s total tax revenue will increase, since these revenues are now collected on all consumption, domestically produced as well as imported goods. There are also practical merits in this strategy. Like trade taxes, also a considerable part of excise and VAT revenues are collected at the border (many developing countries collect more than half of their VAT revenues from imports) using the same administrative organization. Anyway, the revenue consequences of trade reform and tariff reduction pose serious challenges. In a recent work based on a panel of 125 countries over 20 years, Baunsgaard and Keen (2005) find that low-income countries are typically able to recover only a small percentage of lost trade tax revenue, even over a longer term. The presence of VAT does not, in itself, appear to enhance the ability to recover revenue. Excises also play an important role in the transition from trade taxes to domestic consumption taxes, since excisable goods are often a large part of the import base. In those countries with high recovery, there has also been a strengthening of income tax

Overview of tax systems and tax policies 19 revenues, suggesting that the burden of adjustment has not been borne solely by shifting to taxes on consumption. The arguments in favor of using VAT rather than trade taxes and other commodity taxes are well known (Heady 2002). Because the tax base is much larger (it also includes services), tax rates can be lower and as a consequence, the distortion effects are lower; through the destination principle it does not distort relative prices in international trade. Finally, the self-enforcing mechanism means that compliance is generally higher. There is, however, an important structural feature of a developing country that acts against the desirability of VAT: the existence of a large informal sector that escapes VAT. While a radial (across the board) uniform reduction in trade taxes reduces distortions in production, a revenue-neutral radial increase in VAT increases the inter-sectoral distortions between formal and informal sectors. As a result, such a reform can reduce welfare (Emran and Stiglitz 2005). Also Hines (2004) concludes that increasing consumption taxes definitely fosters the expansion of the hidden economy. Currently, the large majority of South and East Asian countries apply VAT, with standard tax rates ranging from 5 percent in Singapore and Japan to 17 percent in China (ITD 2005). On average the standard VAT rate is lower in Asian countries than elsewhere (ITD 2005).2 Asian countries have had mixed success with their VAT systems (Adhikari 2002): in some cases the implementation of the VAT led to an increase in indirect tax revenues compared with previous turnover taxes; in other instances the success of the new tax has been lower. Generally, Asian countries – like other transition or developing countries – face difficulties in VAT administration owing to its complexity. On the one hand the adoption of VAT is often seen as an opportunity for the overall modernization of tax administration (ITD 2005; see also Chapter 2). On the other hand the administration of VAT is more difficult in these countries where underground or shadow economies are considerably larger in comparison with developed countries (Bird 2005). The risk is that an increase in tax may foster the expansion of the underground economy, especially when the labor intensity of production in the informal sector is greater than in the formal sector (Hines 2004). VAT performance in such countries is still much below its potential, mainly as a consequence of the existing exemptions and of the large informal sector. VAT systems are often limited to a small number of economic sectors or goods and services. Exemptions and preferential tax treatment should be minimized as they create distortions and difficulties in the administration and compliance of the tax.3 A third general tax policy area, where the growing economic integration and capital movements between Asian countries poses relevant challenges to the existing national tax structures, is in the area of tax competition. As non-tax barriers decline, investment decisions and location of investment become more tax sensitive. Within free trade areas

20

Luigi Bernardi et al.

firms can supply different national markets from a single location. The relevant issue here is the temptation for such countries to broaden the scope of tax incentives to attract and compete for foreign direct investment (FDI). The main argument in favor of these national policies is that FDI can contribute to increase the productivity of the domestic economy, but the effectiveness of tax incentives is highly questionable. Moreover, as noted by Tanzi and Zee (2000), a tax system that is riddled with such incentives will inevitably provide fertile grounds for rent-seeking activities. Recent evidence (OECD 2001) suggests that tax incentives can have effects on the location of investment, especially between locations that are similar in other respects. However, it is also recognized that neighboring countries within a region could compete against each other in offering tax incentives in a way that provides benefits to the investor without increasing the total amount of FDI allocated to the region. Almost all South and East Asian countries have made, and still make, extensive use of tax (but also non-tax) incentives to promote domestic investment and especially to attract FDI (see Chapter 5) in forms such as tax holidays, accelerated depreciation or investment tax credits (Easson and Zolt 2003). As reported by Asher and Rajan (1999) there is evidence of tax competition between certain Asian countries for FDI from EU countries, the US and Japan. There is also evidence that some South and East Asian countries have emulated or responded to the tax incentives provided by the leading country in the region, that is Singapore. Similar patterns of tax competition between neighboring transitional or developing countries are very frequent; for instance they can be found between some new EU member states, with Poland and the Czech Republic responding to the investment incentives provided by Hungary (see Gandullia 2005). The 1997 East Asian crisis put additional pressure on these countries to attract foreign investment (Adhikari 2002); recently Thailand has expanded the scope of its tax (and non-tax) incentives for FDI and Indonesia has reintroduced the income tax incentives after having repealed them in the 1983 tax reform. At present, international agreements between Asian and Western countries are unlikely to be successful in this area. While there has been a trend towards transparency in tax structures, preferential treatment in Asian countries is often negotiated on a case-by-case basis and agreements are not made public. Hence, enforcement of any type of international rule becomes very difficult (Asher and Rajan 1999). However, in the coming years, the use of tax incentives for the promotion of FDI will require a coordinated multilateral approach, at least on a regional basis. Agreements on a regional basis – for instance between member countries of the ACFTA and member countries of the SAFTA – could create a different kind and intensity of cooperation. Countries, for instance, could agree on a limited set of tax incentives, conditioned to certain criteria (Easson and Zolt 2003).

Overview of tax systems and tax policies 21

Intra-cluster tax policy issues Japan and South Korea Among the countries in the area, Japan and South Korea are those that have reached the highest level of development. In spite of the aforementioned degree of homogeneity in their economic and productive structure, the forces that shape their tax systems are broadly speaking similar but also somewhat different. A simple comparison of the composition of their tax burden shows some differences in the tax mix. In fact, while in Japan the primary source of revenues is direct taxation (although the relative share of direct taxation on income is well below the level of other OECD countries), in South Korea the main role is still played by indirect taxes, in particular specific excises, sales and general consumption taxes. Custom duties are totally absent in both countries because these countries have been participating for a long time in free international trade. Looking at the features of PIT we can point out an important dissimilarity that attests to the existence of two alternative theoretical visions in addressing the problem of personal taxation. The first Japanese personal income tax system, designed in the late 1940s, was based on an idea of comprehensive taxation, but this original structure was soon transformed into a system founded on a notion of expenditure income in which saving was practically exempt. According to this view (the “schedular view,” following a definition by Musgrave 1969), incomes from different sources are taxed at separate rates so that the ratio of the total individual fiscal burden to the total personal income depends on the income composition. On the contrary, South Korea adopted a “global” (Musgrave 1969) income taxation that aggregates most personal income (inclusive of many forms of capital revenues) by taxing it at progressive rates. The choice in favor of a comprehensive income was due to the well known aim of pursuing horizontal and vertical equity, by taxing individuals taking into account not just wages or expenditure, but all the possible incomes they receive in order to broaden the tax base as much as possible. Also in their policies of the last decade concerning tax equity, Japan and South Korea have taken opposite directions. Whereas the reforms implemented in South Korea in the 1990s put increasing effort in extending the tax base, Japan has set up a complex set of allowances that have made the base smaller and smaller. In principle, such deductions could have made a more equitable system by adopting a multidimensional concept of ability to pay4 and by also taking into consideration extra economic variables. In practice, empirical analysis has shown (e.g. Dalsgaard and Kawagoe 2000) that the allowances have favored only the very low and the very high income earners without giving any relief to those in between. Leaving aside the distortionary effects in the taxpayers’ behavior caused by a complex ensemble of deductions, let us take a glance at the

22

Luigi Bernardi et al.

main consequences that these measures had in terms of fairness. Since, with a progressive tax schedule, the tax values of deductions increase as long as the tax rate rises, higher income groups receive a larger tax relief as a result of these kind of allowances. It follows that a disproportionate use of exemptions could have a regressive consequence instead of being a helpful instrument to pursue vertical equity. A simple way to cope with the problems concerning equity while broadening the tax base could be that of strengthening the effective taxation of the self-employed. In fact, in Japan the self employed are allowed to deduct necessary expenses (including private consumption) from the taxable income and to split their personal business income by paying salaries to family members. When introduced, the relief on employment income had the precise goal of filling the gap between the self-employed and wage earners, by also permitting the latter to subtract part of their income from taxation. For example, we find that the theoretical roots for the spouse deduction exist in terms of giving comparable opportunities for income splitting to all taxpayers. Although complex, this is just one aspect of the question. From a revenue raising perspective, the main difficulty that threatens Japan is a sharp decline in the tax base. This is due to two different factors. The broad scope for tax planning permitted for the self-employed on the one hand encourages both evasion and avoidance, on the other hand it points out the necessity to lower the effective tax rate for the wage earners to avoid horizontal inequities. We have just shown how an extensive system of tax exemptions limits the equalizing power of Japanese taxation by removing a big share of the taxpayers’ income from the tax roll. So, why does the Japanese government not opt for a more comprehensive taxation? The first reason regards labor supply elasticities. Although the narrowing of the tax base for PIT does not permit the government to lower tax rates (thus limiting the tax wedge on salaries), the tax-induced distortions to the Japanese labor market are likely to be low, at least for primary earners (see Tachibanaki 1997 and Tyrväinen 1995). The second reason deals with the way in which Japanese salaries are determined. A growing literature defines the Japanese labor market as an “organizational-oriented system” instead of a “market-oriented” one. This means that salaries are established at a firm level5 and present a high degree of heterogeneity. Hence, it is not surprising that, in such an individualistic environment, taxation tries to preserve net wage specificity by rejecting the idea of a comprehensive PIT. The recent development of South Korean PIT shows a fairly different pattern. The aim of improving the redistributing consequences of personal taxation without strengthening the progressivity of the tax schedule led the South Korean government to broaden the tax base while keeping the tax rates at very low levels. This strategy allowed a partial counterbalance to the regressive outcomes of indirect taxation that is still the South Korean main source of tax revenue.

Overview of tax systems and tax policies 23 A second issue that has to be discussed when comparing South Korea and Japan is the way in which the two countries are handling the problem of a troublesome rise of the expense of pensions. The growing imbalance caused by an aging population constitutes a worrying threat for the pension system. Both Japan and South Korea opted for a sharp increase in the share of social security contributions. Looking once more at Table 1.3 we can observe that from about 1992 to about 2002 social contributions rose in Japan from 8.7 to 10.3 percent of the GDP, while in South Korea throughout the same decade they reached about 5 percent from a starting point of 1 percent. The striking gap of the starting level partly explains the large spread between the two increases, but not all of it. In fact, whereas in South Korea the development of the pension system was strongly supported by policymakers as an important source of investment capital for financing development-oriented projects (the South Korean pension system is basically fully founded), in Japan, the government seems lenient in proposing an increase in contribution rates or a cut in retirement benefits.6 An increase in Japanese social contributions would raise problems in terms of intergenerational equity, since it would be hitting the young generation without affecting (or even favoring) the elderly population. Nor could it help to alleviate the intergenerational unfairness by removing the tax allowances for private pension earnings (that are nowadays exempt), since the cost would be borne principally by the youngest cohorts.7 In general, it is important to bear in mind that in Japan lump sum payment to employees at the point of retirement and pension incomes8 receive a preferential treatment compared with wage incomes. Would it be fair to abolish these privileges? Is the strengthening of taxation on pension benefits a possible way to fill the intergenerational financial gap? A recent proposal (Dalsgaard and Kawagoe 2000) suggests the use of VAT to collect the missing revenues. The main argument is that such a tax, besides being neutral with respect to saving behavior, could be an adequate instrument to restore a kind of intergenerational equity. VAT is a consumption tax so it could be considered as a tax whose burden especially hits individuals with a low saving rate. Hence, in a lifetime perspective, this could be the case for elderly people given that old age is typically the age of consumption9 and the VAT would be, in effect, a proportional tax on lifetime income or intergenerational income. Furthermore the scope of Japanese PIT is weakened by the aforementioned disproportionate set of exemptions; on the contrary, Japanese value added tax is more effective since it is broadly based and extremely difficult to avoid. Table 1.6 shows the effectiveness of VAT systems in some selected worldwide countries. Looking at the data we gain an idea of the coverage of this kind of taxation in terms of the effective tax base. It is easy to grasp the efficacy of the value added tax in the countries that are considered here. In both Japan and South Korea the effective VAT rate (calculated as

24 Luigi Bernardi et al. Table 1.6 Effectiveness of value added tax – selected countries and 1997 data

Japan Germany France Italy United Kingdom Czech Republic Denmark Finland Greece Hungary South Korea Mexico The Netherlands Norway Spain Turkey OECD average

Value added tax revenues (% of GDP)

Standard rate ( %)

Effective VAT rate (%)

Effective VAT rate (% of standard rate)

1.8 6.6 7.9 5.7 6.9 7.1 9.8 8.2 7.5 7.9 4.3 3.1 7.0 8.8 5.8 6.5 6.7

5.0 16.0 20.6 20.0 17.5 22.0 25.0 22.0 18.0 25.0 10.0 15.0 17.5 23.0 16.0 15.0 17.7

3.1 11.5 14.7 9.8 10.8 12.6 22.2 18.1 11.6 14.4 8.5 4.7 13.2 21.7 9.9 9.9 12.5

89 77 71 52 62 57 89 82 64 58 85 32 75 94 62 66 73

Source: OECD (2004).

a percentage of the standard rate) is very close to the standard “legal” rate, thus demonstrating that this tax could be a useful instrument in the fight against avoidance.10 Our analysis has shown how different theoretical visions have shaped Japanese and South Korean fiscal systems in particular directions. These are still reflected by the pillars of the reforms that are being implemented in the two countries. South Korea is continuing in the process of broadening its tax base while increasing the share of direct taxation on the total fiscal revenues.11 In Japan, the main result of the reforms that were implemented during the 1990s was that of a dramatic compression of the tax base. This raises both a problem in collecting revenues and a concern over inequality, since the redistributive impact of the Japanese tax system has recently been lost. The reforms should also take into account what is stated by Musgrave (1969), i.e. that the “non-global approach which involves progressive rates is an incongruity.” The simplest option to improve equity while increasing revenues would be to reduce personal allowances in personal income taxation. This requires either an effort towards a more comprehensive system, or a consolidation of a separate taxation of labor and capital income. Without any doubt, the Japanese government has chosen the second way. Inside such a dichotomist scheme the main concern regards the need to ensure tax neutrality between alternative forms of investment. Accordingly, the latest Japanese tax

Overview of tax systems and tax policies 25 reform was towards a general alignment in the tax rate on different types of capital12 and on alternative forms of savings. Some people agree that this harmonization should encompass every kind of pension earning and retirement benefit. Malaysia and Thailand Malaysia and Thailand have experienced an economic escalation before the sharp fall in the late 1990s and finally reached a sustained growth in the more recent past. Despite such a dynamic performance the “catching up” phase is not yet concluded and the two countries still have to cope with the difficulties of the transition towards full development. In comparing the Malayan and Thai fiscal burden, it may be interesting to draw a rough link with Musgrave’s (1969) theory of tax structure development. In fact, both countries show some typical features of the so-called “early period,” but Malaysia presents a use of fiscal direct levy that seems to be more in line with that of the OECD countries.13 However, if we look at disaggregated data while taking into account the institutional context, we notice that the picture is somewhat more complex. The initial information that we obtain from Table 1.3 is the share of revenues from CIT, which is fairly large in the case of Malaysia. In fact, the value of corporate income taxation accounts form much more than half of the yield that comes from direct taxation. Nevertheless, these data are just a proxy (upward biased) for evaluating the real impact of average fiscal pressure on all corporations, since they also include a specific tax levied on petroleum companies whose rate is much higher than the “standard” one (38 percent instead of 28 percent). Hence, interpreting the CIT share in Malaysian tax revenue as a proof of the existence of a more developed fiscal system could lead to methodological misunderstandings and to inaccurate interpretations. The petroleum tax can be interpreted as a tax on non-competitive profits that in developing countries is a much used revenue raising tool since it has a well established tax base and is harder to evade.14 However such a tax presents some ambiguities given that it does not hit the firm according to its ability to pay, but determines its tax liability by conforming to the specific role that the corporation plays in the market. In its turn, in Thailand a distinction is made among corporations. It is based on their dimension, since the smaller firms (but just the national ones) are taxed at lower rates by means of a progressive schedule.15 With regards to personal income taxation there are several analogies between the two countries. The first observation concerns the importance of PIT whose weight is consistently narrow and rather stable. Then, there are important similarities in the structure of personal taxation which has a progressive schedule with many brackets and a widely spread set of tax rates.16 In principle, such a configuration could be powerful in a per-equating perspective, but the scarce pervasiveness of this kind

26 Luigi Bernardi et al. of taxation, along with the massive use of personal reliefs and personal tax rebates, makes the pursuit of horizontal equity a goal that is far from being attained. In spite of the likeness between the two, Malaysia differs from Thailand in the sense that it applies a specific tax rate (equal to the highest rate for resident taxpayers) on the chargeable income of foreign taxpayers. In addition, Malaysian citizens who do not reside in Malaysia are not entitled to any form of personal exemption. This is not a common feature in personal taxation where, beside the principle of residence, the specific characteristics that are relevant for defining the personal tax liability do not refer to the country of residence of the taxpayer, but relate to their ability to pay. In this vein it could be seen as relevant to bring up a proposal of Mahatir (the Malaysian Prime Minister) who suggested putting a tax on rich countries into practice, in order to support the poorer countries. In effect the reforms that have been implemented in Malaysia in the last decades are rooted in a peculiar mix of nationalism and openness to the international capital market that has, as a consequence, a corporate taxation that is only apparently onerous and a personal taxation whose scope is highly limited. By analyzing the composition of indirect taxation, a peculiar feature particular to Malaysia is that of the complete absence of any kind of value added tax. In fact the most important form of indirect tax is an ad valorem single stage tax, imposed at import and manufacturing levels. The main problem that arises due to the lack of a “well behaved” VAT is the existence of cascading non-neutral price effects.17 Second, there is the old and widespread (albeit debatable) consensus on the role played by a value added tax in creating a useful information externality that can have a compliance-enhancing outcome on other taxes.18 The nature of indirect taxation is quite important for a country like Malaysia. In fact, in countries that are yet to be fully developed, which have an economic and productive structure that is still rooted on traditional and informal sectors, personal taxation has a fairly limited coverage. In addition, the lack of adequate administrative skills does not allow direct taxation to be completely effective, thus weakening real progressivity. On the contrary, the greater ease in the collection of consumption taxes makes this kind of fiscal instrument a useful tool both in a revenue raising perspective and (sometimes) also in the fight against avoidance. Since, in the case of sales taxes, underreporting is more difficult, there are several incentives for taxpayers towards a correct declaration of their status given that poorer individuals are associated with a smaller tax burden from indirect taxation. There is some proof that, in both countries, low income earners have been considered in the shaping of consumption taxes. In Malaysia a uniform tax rate (10 percent) is applied to most products, but some goods such as primary commodities, basic foodstuff and basic building materials are completely exempt, while in Thailand, the operators earning less than

Overview of tax systems and tax policies 27 bath 600,000 per year do not have to pay VAT at all. Strictly speaking, this personal exemption is not completely in line with a restrictive definition of indirect taxation. Musgrave (1969) reminds us that amongst the characteristics that attempt to draw a separation line between direct and indirect taxation, a much used criterion states that “(indirect taxes) are assessed on objects rather than on individuals and therefore not adaptable to the individuals’ special position and his taxable capacity.” In fact, the prevailing literature about taxation considers indirect tax as a useful but regressive instrument for collecting revenue, given that the demand for basic commodities presents a very low elasticity to prices and cannot be shifted towards alternative tax-free goods. However, we have shown that, where fiscal systems are still embryonic and direct taxation has a limited scope, indirect taxation may help governments to achieve some objectives on the ground of equity.19 However, an excessive reliance on indirect taxation could lead to a big loss in terms of efficiency; the more the indirect taxation adopts a progressive schedule, the more it differs from an “optimal taxation structure” whose goal is the minimization of the “excess burden.”20 After the Asian financial crisis which strongly hit both countries (although with several important differences in the depth and in the timing of the crash), Malaysia and Thailand had to find a way to build the roots for a quick recovery. With regards to taxation, the main field in which both are putting increasing efforts is the strengthening of revenue collection. In fact, a crucial issue in order to keep the situation of public finance under control is to increase taxpayers’ compliance so that it could be possible to finance public expenditure without creating distorsive consequences on the economy. In order to achieve this goal, one of the fundamental pillars in the agenda of the two governments is the rationalization of administrative procedures and the improvement in the efficiency of tax administration. For Thailand, this objective means, first of all, the creation of a more effective collaboration among local and central governments to reach a better coordination and take advantage of an increasing fiscal decentralization. For Malaysia it implies a simplification in the complex set of incentives that have been introduced both for economic and for political reasons. China and India Let us now look first at direct taxation. In India, personal income taxation is completely levied at the level of Union Government (see Table 1.7). In principle, Indian income taxation shows some important features which the supply-siders’ views on taxation (for a discussion: Stiglitz 2000) look on favorably. In fact, Indian personal income taxation has few tax brackets with a low degree of progression. In addition, the notion of taxable income that has been adopted in the Indian fiscal system is very similar to

28

Luigi Bernardi et al.

Haig-Simons definition. This is, however, not entirely problem-free. An important difficulty concerns the presence of many personal exemptions that narrow the tax base, so limiting the effects of income agglomeration. The point is analogous to what has already been discussed in the case of Japan, thus it is not necessary to repeat the whole debate. Anyway, it could be interesting to reiterate that the dimensions of India, together with its greater administrative weakness, makes the problem even worse and the role of income taxation even more limited. Chinese personal income taxation presents opposite features. The Chinese PIT has a “pure” schedular structure with many different types of income taxed at different tax rates,21 no aggregation of alternative sources of earnings and no personal deductions. As a consequence, there is both a loss on the ground of progressivity, and a fall in the total tax burden, since taxpayers can choose to transform activities that are heavily hit by taxation into others that are less taxed or even tax free.22 The main field for tax planning in China, however, is the tax environment created by some special incentives that have been recently granted to foreign enterprises. In China, specific systems of corporate income taxation are in force for foreign corporations, and have been recently updated. The CIT imposed on Chinese enterprises has a fairly simple structure that applies a 33 percent tax rate23 on the gross income after subtracting allowable deductions. Foreign companies receive a generous system of tax holidays (see Chapters 5 and 6). In fact, for companies that meet some requirements (such as firms operating in some designated industries) the possibility of enjoying a very favorable fiscal treatment throughout long periods of time exists. For example, a ten-year tax

Table 1.7 Central government percentage share on general government main tax revenues – selected federal countries Country

Period

Domestic indirect taxes

Income taxes

Argentina Australia Austria Belgium Brazil Canada Germany India Mexico Switzerland United States

1975–98 1975–2002 1975–2002 1975–2002 1977–97 1975–2002 1975–2002 1975–97 1975–2002 1975–2002 1975–2002

81.72 64.49 70.99 80.72 47.15 32.25 63.71 50.71 99.00 72.83 14.52

74.58 100.00 63.44 82.03 97.64 64.84 40.63 100.00 100.00 22.62 82.24

Source: Own elaboration on IMF and OECD data. Notes The time periods stop at the last available data.

Overview of tax systems and tax policies 29 holiday entails a total exemption from taxation for the first five years and a big tax cut (50 percent reduction) for the following five years. These fiscal stimuli add to an already investment-encouraging tax regime that provides for the depreciation and the amortization of tangible assets. While useful in the short run since it has the consequence of attracting foreign investors, the use of tax holidays may be harmful in the long run because some “race to the bottom phenomena” can ensue, thus creating a continuous reduction in tax revenues. A parallel trend that emerges from several empirical studies is that of a rise in the relative tax burden on home taxpayers. This fact could have bad consequences both with regards to horizontal equity and with regards to the growth of Chinese domestic industry. Indirect taxation is perhaps the most interesting analytical field because, on these grounds, many reforms have been recently elaborated. In this respect the two countries feature some important differences in the composition of their fiscal revenues. While in China the main indirect tax is value added tax that is charged on a broad set of goods including power, heating, gas and services, in India until 2004 VAT was absent. From 1 April 2005 VAT has finally been introduced in India. Obviously, there are still not enough data to evaluate the effect of the reform on the fiscal system as a whole. However, some comments will be made at the end of this section (see also Chapter 7). It is generally agreed that an introduction in India of a value-added type taxation should be a welfare improving reform. In fact, in India, a huge share of fiscal revenues24 comes from sales tax and excise duties that, as we have already noted in the previous pages, have distortionary effects on relative prices. These kind of distortions could be partially eliminated by a uniform tax on consumption (like VAT). According to the somewhat forced view of Jha (2001) and Jha and Mittal (1990),25 taxing consumption at a uniform rate could have specific positive effects in the case in which consumer utility functions are weakly separable between consumption and leisure,26 so that this tax “would approximate a lump sum tax” (Jha 2001). Actually, the principle could be valid for India only in the case of a properly harmonized state and central VAT capable of substituting the complex set of sales taxes that are nowadays levied by provincial governments. So far, so good. But what could be the result of such a reform in terms of revenue sharing among national and sub-national governments? Before answering the question we will briefly present some basic data about intergovernmental fiscal relations. Table 1.7 shows the national government tax collection share by type of levy for a number of selected federal countries. The picture drawn by Indian data is quite clear. Although all direct taxation is collected by the central government in line with the redistributive function of income taxation, the share of revenues from indirect taxation that is collected by the center falls dramatically.27 Among the central government’s revenues,

30

Luigi Bernardi et al.

however, are all the production excise taxes that naturally pertain to that level because they have to be uniform across all the country. We can infer from this datum that nearly all the percentage to the provincial government is composed by the whole set of internal sales taxes. It follows that the new VAT system must be implemented on a dual-layer level in order to substitute sales taxes as a main source of local financing. On the contrary, after the thorough reform of the middle 1990s, in China the system of public finance is more centralized and is based on a “pyramidal” structure in which most taxes (inclusive of VAT) are charged by the central government which, successively, shares part of the tax revenues to the inferior governmental levels by means of a broad set of transfers. In principle the use of transfers from the center could be powerful in per-equating local fiscal capacity, but could generate an excess of spending behavior in the provinces while weakening fiscal responsibility. Now however, it is crucial to redefine the system of transfers, given that provincial governments have a small fiscal autonomy. Hence, to avoid harmful fiscal imbalances at lower government levels and in order to increase herein budget transparency and fiscal effort,28 many rule-based mechanisms have been introduced.29 It is almost mandatory to conclude now that in India most of the current debate about taxation concerns the value added tax introduced in 2005 (see also Chapter 7). The reform, whose features are not completely clear yet, introduces a two-rate tax (4 and 12.5 percent) and it will cover more then 500 different goods. The Indian VAT will present some characteristics that we have described while analyzing the Thai tax system. In fact traders with turnovers of less than rupees 500,000 will be completely exempt from the tax30 so that it is worthwhile to think of this measure as a progressivity-enhancing tool. Nevertheless, the main argument in favor of VAT states that such a change helps to improve transparency, may contribute to limit tax evasion31 and will simplify the system, while correcting distortionary effects on prices. An important question is whether to charge an indirect tax on services. At the moment in India services constitute 53.3 percent of GDP. Thus, whereas incomes from services are taxed with incomes, services themselves face very few indirect taxes. According to Jha (2001) “this is inefficient as well as inequitable because it discriminates between providers of goods and services, inefficient because it has the potential of creating several distortions thus increasing nonlabor costs.” Certainly, the transition from a multifaceted system of indirect taxes to a homogeneous value added tax will not be easy since central sales taxes will continue to run alongside VAT at least until 1 April 2006. In China the debates and the changes concerning VAT are also relevant. First, the tax base has been broadened, the number of tax rates have been reduced and the whole system has been simplified. Second, in three north-eastern provinces, the Chinese government is about to carry out a new project which has the principal aim of replacing the current produc-

Overview of tax systems and tax policies 31 tion-type VAT with a new consumption-type VAT. Such a change is in line with the rapid growth of Chinese industry that is becoming more and more capital intensive. In fact consumption-type VAT does not discourage investment in capital because the expense for capital goods can be deducted from the VAT tax base.

Notes 1 The case of Malaysia depends on the heavy taxation which hits native petroleum companies. See below in this chapter. 2 Many countries in the area apply a single rate (Bangladesh, Cambodia, Japan, Korea, Nepal, Philippines, Singapore, Thailand). Multi-rate VATs are present in China, Indonesia, Pakistan and Sri Lanka (ITD 2005). The choice depends mainly on balancing tax administration arguments – favoring a single rate structure – against the availability of other instruments better targeted to achieve distributional objectives – the absence of which tends to favor multi-rate structures. 3 The main challenge that the existing VAT systems face is in the context of trade liberalization in the area. As commonly understood with respect to international trade, the standard approach is to levy the tax on domestic consumption through the destination principle, which assures production efficiency even in the presence of differentials in national tax rates. Around the world recent trends toward regional integration, the development of the Internet, the growth in trade involving services and intangibles and the trend toward the reduction of traditional custom formalities complicate the implementation of the destination principle. In the less developed Asian countries further difficulties arise when the zero-rating has to be applied to exports which requires the appropriate refunding of excess VAT input credits to exporters (Adhikari 2002; Choon 2002). As trade barriers come down, methods to redress this situations will be increasingly critical. 4 This could explain, for example, employment income deductions. 5 The Japanese system is based on the so-called “keyretsu:” sort of “families” of firms that define the wage patterns autonomously. As a consequence, wages are a result of a complex set of personal variables that are only partially related to the bargaining contracts. 6 It is useful to note here that the Japanese pension system is financed essentially by the pay-as-you-go system. 7 However, taxing this form of saving could be acceptable to pursue the neutrality among different types of capital investment subject to taxation (albeit the tax rates for capital income are low). 8 The situation in South Korea is not far from the Japanese one. Although most capital income is added to wage income and taxed at progressive rates, income and timber income are taxed separately at low tax rates. 9 Note that available data do not seem to confirm this theoretical hypothesis. 10 In particular in Korea, where the VAT share, with regards to the total amount of indirect taxation, is less than the share of specific excise and sales taxes, a substitution of some consumption taxes with a less distortive value added tax could be a welfare-improving measure. 11 In fact, the effort in pursuing the full concept of the ability to pay at the PIT level is counterbalanced (and strongly weakened) by the huge share of indirect taxation which form a more distortive and less equitable source of revenue. 12 The tax treatment of interest capital gains from stock and dividends on listed stocks have been unified at the rate of 20 percent (see Chapter 8).

32

Luigi Bernardi et al.

13 In the area that we are analyzing only Japan shows a bigger share of tax revenues from direct taxation. 14 It is a widely accepted fiscal instrument since it allows the government to finance public expenditure while limiting the creation of monopolistic power in the market. 15 Although the ordinary tax rate for CIT is 30 percent, smaller companies are taxed at a lower rate (20 percent). See Chapter 11. 16 From 2 to 29 percent in Malaysia and from 5 to 37 percent in Thailand. 17 In Malaysia indirect taxes are production taxes that can induce distorting changes in relative prices. 18 In Malaysia, an alternative spill-over of information comes from the obligation for manufactures to be licensed under the Sales Act 1972. This requirement has the further effect of allowing a lesser taxation on the smaller (and poorer) manufactures, which are permitted to buy tax-free inputs. 19 See Musgrave (1969); on the contrary Sah (1983) states that the extent of the redistribution possible through this kind of fiscal instrument is quite limited. 20 As it is well known, in order to limit the deadweight losses, optimal taxation calls for a tax rate to vary inversely with the compensatory elasticity of demand. It easy to understand that such a structure differs from a progressive one. 21 Whether to apply a progressive schedule or a proportional one depends on the single source of income. 22 Actually this possibility is open also to Indian taxpayers due to the justdiscussed existence of many exemptions. 23 The tax rate applied to foreign enterprises is substantially similar (30 percent plus a local surcharge of 3 percent). 24 The 8.8 percent of GDP, well above the share of the whole income taxation. 25 Such a view assumes a Walrasian equilibrium theoretical framework that cannot be easily applied to a country like India. See also Chapter 7 with regards to the famous Ahmad’s and Stern’s simulation of optimal taxation in India during the 1980s. 26 This requirement implies that taxation on goods does not have any implication in the labor–leisure choice and hence does not affect the supply of labor. 27 The share of indirect taxes levied by the central government is about 50 percent. 28 Jha et al. (1999) showed that the higher the share of central financing of state government expenditures, the lower is their tax effort. 29 More broadly on fiscal federalism in China and in India, see Chapter 3. 30 Actually this threshold is different in some states where the exemption applies to traders with an annual turnover of up to one million rupees. 31 Developing countries have a genuine faith in the conventional wisdom according to which VAT, by providing input credit from the manufacturing stage to the retail stage, should reduce underreporting. This may happen because whoever does not provide correct declarations incurs a loss by getting lesser input credit. Countries, like the EU members, where VAT has been in force for decades, do not all confirm this feature of the “VAT’s chain” (see Table 1.6, especially for the Mediterranean countries, where corruption and inefficiency of tax administration is high and tax-payers’ compliance is low). In fact, this chain breaks down at the link between the retailer and the final consumer, since the latter is not interested in crediting paid VAT. Collusive evasion may arise, and then feeds back along the chain so that it undermines the same income tax returns. Note that from the supply side, VAT’s basis is broadly the sum of wages and profits (less investment in the EU model).

Overview of tax systems and tax policies 33

References Adhikari, R. (2002) “Tax policy and administration in Asian countries: a review of key issues and options,” IBFD Bulletin, 46–59. Asher, M. G. and Rajan, R. S. (1999) “Globalization and tax systems: implications for developing countries with particular reference to Southeast Asia,” Discussion Paper 23, Adelaide University, Australia: Centre for International Economic Studies. Baunsgaard, T. and Keen, M. (2005) “Tax revenue and trade liberalization,” mimeo, Washington, DC: IMF. Bernardi, L. (2004) “Rationale and open issues for more radical tax reforms,” in Bernardi, L. and Profeta, P. (eds) Tax Systems and Tax Reforms in Europe, London: Routledge. —— (2005) “Main tax policy issues,” in Bernardi, L., Chandler, M. and Gandullia, L. (eds) Tax Systems and Tax Reforms in New EU Members, London: Routledge. Bird, R. M. (2005) “Value-added taxes in developing and transitional countries: lessons and questions,” Paper prepared for the first global international tax dialogue conference on VAT, Rome: 15–16 March. Burgess, R. and Stern, N. (1993) “Taxation and development,” Journal of Economic Literature, 2: 762–830. Choon, C. (2002) “Major issues and challenges in fiscal restructuring in Asia,” IBFD Bulletin, 146–54. Cordenillo, R. (2005) “The economic benefit to ASEAN of the ASEAN-China free trade area (ACFTA),” Bureau for Economic Integration: Studies Unite. Dalsgaard, T. and Kawagoe, M. (2000) “The tax system in Japan: a need for comprehensive reform,” OCSE Economics Department Working Papers 231, Paris: OECD. Dixit, A. (1985) “Tax policy in open countries,” in Auerbach, A. and Feldstein, M. (eds) Handbook on Public Economics, Amsterdam: North Holland. Easson, A. and Zolt, E. M. (2003) “Tax incentives,” Paper prepared for World Bank course on practical issues of tax policy in developing countries: 28 April–1 May. Emran, M. S. and Stiglitz, J. E. (2005) “On selective indirect tax reform in developing countries,” Journal of Public Economics, 89: 599–623. Gandullia, L. (2005) “An overview of taxation,” in Bernardi, L., Chandler, M. and L. Gandullia (eds) Tax Systems and Tax Reforms in New EU Members, London: Routledge. Heady, C. (2002) “Tax policy in developing countries: what can be learned from OECD experience?,” Paper prepared for the seminar Taxing perspectives: a democratic approach to public finance in developing countries, University of Sussex: the Institute of Development Studies, 28–29 October. Hines, J. R. Jr. (2004) “Might fundamental tax reform increase criminal activity?,” Economica, 71: 483–92. International Monetary Fund (2005) “Dealing with the revenue consequences of trade reform,” Washington, DC: IMF. ITD (2005) “The value added tax: experiences and issues,” Background Paper prepared for the International Tax Dialogue Conference on the VAT, Rome: 15–16 March. Jha, R. (2001) “The challenge of fiscal reform in India,” ASARC Working Papers, Australian National University: Australia South Asia Research Centre.

34

Luigi Bernardi et al.

Jha, R and Mittal, S. (1990) “Savings, investment and marginal effective tax rates in India,” International Journal of Development Banking, 8: 3–12. Jha R., Mohanty, M., Chatterjee, S. and Chitkara, P. (1999) “Tax efficiency in selected Indian states,” Empirical Economics, 24: 641–54. Keen, M. and Ligthart, J. E. (2002) “Coordinating tariff reduction and domestic tax reform,” Journal of International Economics, 56: 489–507. Mitra, P. (1992) “The coordinated reform of tariffs and domestic taxes,” The World Bank Research Observer, 7: 2. Musgrave, R. A. (1969) Fiscal Systems, New Haven: Yale University Press. OECD (2001) “Corporate tax incentives for foreign direct investment,” OECD Tax Policy Studies 4, Paris: OECD. —— (2004) Revenue Statistics 1965–2004, Paris: OECD. REF (2004) Congiuntura Ref, XI, 6: December. Sah, R. K. (1983) “How much redistribution is possible through commodity taxes?,” Journal of Public Economics, 20: 89–101. Steve, S. (1976) Lezioni di scienza delle finanze, 6th edn, Padua: CEDAM. Stiglitz, J. E. (2000) Economics of the Public Sector, 3rd edn, New York: W.W. Norton. Tachibanaki, T. (1997) Public Policies and the Japanese Economy: Savings, Investment, Unemployment, Inequality, London: Macmillan. Tanzi, V. (1994) “Taxation in developing countries,” in Bernardi, L. and Owens, J. (eds) Tax Systems in North Africa and European Countries, Deventer: Kluwer. —— (1999) “Is there a need for a World Tax Organization?,” in Razin, A. and Sadka, E. (eds) The Economics of Globalization: Policy Perspective from Public Economics, Cambridge: Cambridge University Press. Tanzi, V. and Zee, H. (2000) “Tax policy for emerging markets: developing countries,” National Tax Journal, 53: 299–322. Tongzon, J. L., Khan, H. and Doanh, L. (2004) “Options for managing revenue losses and other adjustment costs of CLMV participation in AFTA,” ASEAN: October. Tyrväinen, T. (1995) “Wage determination in the long run real wage resistance and unemployment: multivariate analysis of cointegrating relations in 10 OECD countries,” Discussion Paper, Helsinki: Bank of Finland. Zee, H. (2004) “World trends in tax policy: an economic perspective,” Intertax, 32: 352–64.

2

The control of tax evasion and the role of tax administration Parthasarathi Shome

Introduction, contents and main conclusions1 The role of tax administration in maximizing revenue generation and minimizing tax evasion cannot be over-emphasized. These remain challenging tasks at every stage of development of a tax administration. This is because it is not just a matter of maximization or minimization but, rather, one of optimization. Thus, there has to be a genuine threat and actual carrying out of audit, scrutiny, investigation, penalty and punishment for an errant taxpayer while, at the same time, his compliance costs must be minimal, and he should be guaranteed an avenue for redressal of genuine grievances for payment of tax, with robust systems of adjudication, appeal and final decision. And, to minimize lengthy adjudication processes, there should be practical avenues such as advance rulings or tribunals that would cut back on undue delay and facilitate speedy administration decisions. Computerization has become a mainstay in efficient administration that can successfully curb tax evasion. For example, for income tax, forms should be downloadable electronically. Electronic filing should be made possible. If there is tax deduction at source (TDS), the mechanisms – including filing at banks – and transfer of data and revenue to further points in the system should be computerized. For indirect taxes, for example the value added tax (VAT), there has to be a computerized method for cross-checking invoices that is precise and incisive, and certainly not cumbersome, for both domestic and international transactions. An increasingly recognized element in controlling tax evasion is to reduce, as much as possible, the likelihood of interface between taxpayer and tax administrator. Computer-assisted processes have helped in this objective. Thus, selection of audit or scrutiny cases should be primarily random and computer assisted except in the case of criminal suspicion. The outcome of a fully developed tax administration is, therefore, one that is essentially invisible and that ensures due process of law to the taxpayer, yet fully enables the efficient collection of revenue for the exchequer.

36

Parthasarathi Shome

The role of appropriate tax design in assisting efficient tax administration is equally important. A complex tax structure or tax law is difficult to administer and, therefore, easy to evade. For example, if VAT has three rates but no exemptions, then it should not create too many problems. A taxpayer would just need to subtract total taxes paid by him on his purchases – irrespective of tax rates – from all taxes collected by him on his sales, and transfer the difference to the administration. But as soon as one introduces exemptions for individual goods or services, as most countries do, the VAT structure assumes complexity since it breaks the VAT chain. VAT credit cannot be taken against exempted items. So the administration would have to devise a mechanism to keep track of any excessive credit that may be taken by a tax evader. Similar administrative problems emanating from faulty income tax design arise all the time. It is, therefore, imperative for the tax structure to be comfortably administrable and the tax law to be easily interpretable. In sum, in order to keep tax evasion in check, the tax administration must: (i) incorporate genuine threat of penalty but ensure due process; in order to do this, of course, the tax administration should be adequately financed and structured; (ii) computerize as many administrative processes as possible to minimize the interface between taxpayer and tax official; and (iii) not remain aloof from tax policy but assist in every way possible to help design, in reflection of its field experience, a simple tax structure and its commensurate tax law. In the following sections, these three major tax administration elements in the control of tax evasion and the generation of revenue are examined individually. Emphasis is on actual experiences drawn from South and East Asia. In particular, since tax administration reform has taken up speed in India, the lessons from this process are universal and are cited where found to be useful. However, first, in the next section, a discussion of the problem of tax evasion – in juxtaposition to its variants, tax avoidance and corruption – is attempted. The final section concludes. However, before embarking on that journey, we summarize below the main conclusions from the analysis that follows. •

•

There are various sources and causes of tax evasion. An inequitable or highly burdensome tax structure will not be tolerated. However, globalization seems to have led to a growth in tax-evasive behavior, reflecting income sources from many countries and increased possibilities to hide income from any individual country tax administration. Ineffective tax administration will exacerbate tax evasion under these circumstances. Tax evasion has been estimated in many ways. An indirect way is to estimate the underground economy since tax evasion directly generates it. A more direct way is to estimate tax potential through a country’s input–output matrix and comparing it with actual revenue

Control of tax evasion and tax administration 37

•

•

•

•

•

•

collection. It could be concluded that in many developing countries, half the income tax may be evaded and one-third the VAT or consumption taxes may be evaded. To combat tax evasion, tax administrations should have adequate resources. In most countries, however, they form part of general government administration and must, therefore, compete for scarce resources. Under the circumstances, the literature has discussed the impact of incentive schemes for tax officers. Typically, the cost of collection is about 1 percent of revenue collected. The more inefficient a tax administration is, the less effective would be the use of a dollar in revenue collection. A review of Asian countries reveals an increase in administrative action and follow-up in the administration of international taxation. However, tokenism is being avoided and focus is shifting to critical areas of intervention through the development of robust practices in operations, investigation and judicial processes. Thus tax administrations are requiring third-party information returns in which a third party is asked to submit returns on transactions with others, if such transactions are over a certain threshold. The objective here is to cross-check such information with large taxpayer income tax returns. Tax deduction at source is also a third-party device to facilitate tax collection and is increasingly used for many sources of income. Large taxpayer units are being used in many countries to reduce tax evasion by requiring all taxes to be paid through one window by large taxpayers. This facilitates cross-checking across all taxes while providing a single window facility to the taxpayer. The small taxpayer is being assisted with the establishment of help centers external to the tax office to reduce his inherent fear of tax officers. The potential taxpayer is being required to file returns even with zero taxable income as long as he is captured under specified expenditure criteria. Given the growth in cross-border transactions and the quantum increases in international capital flows involving transfer pricing, tax administrations are increasingly training officers and applying arm’s length rules in the setting of transfer prices by multi-national companies in both international transactions and domestic transactions among multiple subsidiaries. Joint audits by central and local governments are being carried out to facilitate matching of information across a wide geographical area. Such audits focus on transfer pricing behavior and use a wide variety of indicators for selection and examination. Audits can be of various types: random audits, audits of evasion-prone sectors, in particular, emerging service sector activities, and of high-income individuals identified through information and intelligence. The latter remains limited in scope and is reserved usually for cases of repeat offenders.

38

•

•

•

Parthasarathi Shome A review of Asian countries reveals an increase in administrative action and follow-up in the administration of international taxation. In an increasingly complex world faced by the tax administration, computerization is of the essence. Experience reveals that computerized systems for tax collection may be successfully put in place; but it remains a challenge to make such systems user-friendly. There are likely to be quite a few initial glitches, but these can be resolved with time and a full understanding of the underlying needs of the tax administration to implement the tax structure. It is, therefore, crucial to design tax policy that is implementable. At the same time, innovations in tax design are appearing to buttress tax administration to reduce evasion and improve revenue productivity. Two recent examples are a financial transactions tax and a fringe benefits tax. They have been introduced in Latin America and the Asia-Pacific region. While they have been rather controversial, there is no gainsaying the fact that they are turning out to improve information flows to the tax administration and to assist in additional revenue mobilization. Nevertheless, taxes should not be devised merely to accommodate tax administration. For, if they increase inefficiencies in resource allocation and inequity among taxpayers, they will affect economic growth adversely in the medium term and, consequently, be deleterious to revenue productivity.

Tax evasion – its forms and effects Tax administration is closely linked with the three concepts of tax avoidance, tax evasion and corruption. They are sometimes used interchangeably; however, there are essential differences among the three, mainly reflecting the degree of malpractice involved. Tax avoidance is not illegal in the sense that it usually results from the creativity and planning of tax accountants and lawyers within the gambit of existing tax laws in aiding taxpayers to minimize their tax. Tax avoidance is obviously facilitated by complex tax statutes in which opportunities exist to interpret the tax law in the taxpayer’s favor when it was not so intended by the drafters of the law. Therefore, tax avoidance as a matter is mostly addressed in the context of tax policy reform that is the focus in the last section. Tax evasion, on the other hand, is illegal though, if detected, it normally leads to civil rather than criminal penalties. It involves flouting of the tax law. It generates an underground economy in the form of income that is not reported or accounted for in tax returns. As a result, a subterranean economy begins to function in parallel to the organized economy with generally deleterious effects on equity among taxable persons, on the efficiency of resource allocation, and on the stability of revenue collection and the macro-economy (Tanzi and Shome 1993).

Control of tax evasion and tax administration 39 Some authors whose views have been summarized by Brooks (2001) have, however, claimed that the underground economy minimizes the government’s reach over the private sector and, thus, assists in unfettered economic growth. This argument essentially ignores the fact that it leads to distorted growth inasmuch as one part of the economy grows faster at the cost of the other part, and that the latter could have been more efficient and productive in an even playing field. Corruption goes one step further. It has been defined as the “abuse of power for private gain” (World Bank 1997). Its ramifications are grave since it burrows into the very foundations of society and the institutional framework of government. Unlike tax avoidance and tax evasion, corruption involves collusion and, often, criminal connotations. Thus, bribery is a form of corruption since it involves at least two parties (Asher 2001). In this sense, Transparency International’s (2001) Corruption Perception Index (CPI), in vogue for citation in cross-country studies, is essentially misconceived. It appears to recognize the demand and acceptance of bribes rather than the offer and supply of bribes, thereby missing the quid pro quo nature of corruption. For example, in international contracts, assigning the perception of corruption to one party leads, expectedly, to far higher CPIs in developing countries than in developed countries. A similar fallacy appears when the illegal outflow of money from developing countries is considered corrupt but not its acceptance, banking and use elsewhere. Both arms of corruption are pernicious. Without the adroit cooperation of both providers and recipients of such resources, the web of international money laundering could hardly have spread so far. The role of tax administration in curbing corruption is through its investigative wings but it has to be supported by the crime control wings – such as intelligence, enforcement and internal security – of government. This chapter does not attempt to go further into the criminal aspects that are centered around the issue of corruption. Sources and causes of tax evasion Tax evasion is the outcome of individual taxpayer behavior and social norms on the supply side, and shortcomings in tax administration on the demand side. To begin with, the tax structure has to be perceived as equitable across various groups of taxpayers. Also, if marginal rates are too high, taxpayers are likely to find ways to evade tax. Across the world, these aspects are well recognized today, and the outcome of international tax harmonization has been a lowering in the rate structure (to improve incentives) and attempts at broadening the tax base (to reduce inequities). Nevertheless, tax evasion tends to arise also from other factors. Individual behavior in certain sectors is known to be highly tax-evasive. These include professional services including the usually highly paid medical

40

Parthasarathi Shome

profession, and the construction industry, to name a few. They tend to receive fees and make payments in cash, fuelling the underground economy. Even in developed countries such as the United States, medical practitioners, in particular, comprise a difficult sector to tax. Indeed, a number of Western authors have dubbed the individual income tax a voluntary tax. Difficult-to-tax taxable persons may even include large corporations that typically contract out – through work contracts – to smaller units. With globalization, this phenomenon has assumed challenging dimensions. As capital becomes increasingly fungible, sources of income and expenditure are spread across the globe and difficult to track or even identify since they often leave little trace in any one national tax jurisdiction. Increasingly, therefore, the globalized economies of East Asia are demonstrating their conservatism in international taxation rules and practices (Pricewaterhouse Coopers 2003). Social norms have been changing in many economies. Since the 1990s, as large public corporations have been privatized in Latin American and South Asian countries, the small service sector – one that is traditionally difficult to tax – has grown. With the emergence of Eastern Europe and the former Soviet Union (FSU), and their entrance into the international market economy, the taxpayer roll has multiplied but revenue intake is not reflected commensurately. In traditional economies, if eminent persons do not pay their fair share to the exchequer, that becomes an example to be emulated by all and sundry. An assessment that the use of revenue is inefficient or ineffective or, worse, suffers from unacceptable leakages, provides grounds for society to evade tax. Ineffective administration can exacerbate tax evasion. For example, extremely punitive penalties encourage taxpayers to appeal tax demands easily since it gives them time until final determination of the tax. By that time, the real – as opposed to nominal – value of the revenue in the hands of government is likely to diminish. Bureaucratic lethargy in applying the law – perhaps in reflection of social norms – increases evasion and hurts revenue. Slow development of information technology (IT) due to its complexity, or its ineffective use by the administration, signals to taxpayers that tax evasion carries low risk. Indeed, technological advances have, in particular instances, translated into tax evaders’ computer capabilities running well ahead of the administrator’s, exacerbating the problem. Estimating tax evasion The difference between actual revenue collection and an estimation of potential collection is the tax gap. It is an estimate of total leakage, comprising the effects of tax avoidance, tax evasion and corruption. Typically, however, the tax gap is referred to as estimated tax evasion. Estimating tax evasion can be indirect or direct. The indirect method

Control of tax evasion and tax administration 41 links tax evasion to the underground economy since the latter could be thought of as the size of economic activity that would be taxed if reported in tax returns. The estimate of the underground economy is essentially indirect in that it is attempted through estimating the demand for currency in the economy, or estimating discrepancies between national expenditure and income statistics, or linking growth in electricity consumption to the growth in economic activity because of their observed close relationship. However, one school of thought is that the tax rate – usually the income tax in this context – to be applied to an estimated underground economy to arrive at an estimate of tax evasion would have to be perspicacious. Otherwise the estimated tax gap may turn out to be too large for corrective policies that are practical and implementable. Authors have discussed the issue of optimal auditing in a context of audit probability and penalty (Mookherjee and Png 1989; Chander and Wilde 1998). In a recent cross-country survey on the size of the underground economy, Schneider and Enste (2000) have grouped countries into regions and shown that, in a few, the underground economy could be as large as three-quarters of the reported economy though, in the representative Asian, Latin American and FSU countries, it ranges between one-third and half. Among OECD countries, southern Europe was at one-fifth and others at one-tenth. However, as argued above in the context of the CPI, these analyses tend to reflect some double counting in developing countries and undercounting for others, if one considers how the underground economy is generated in a globalized world economy. Nevertheless, there is no gainsaying the fact that, even if distributional adjustments are made, tax evasion, derived through indirect methods, is likely to yield significant estimates. A more direct method of estimating the tax gap, in particular for VAT, is to estimate potential VAT revenue from the country’s input–output matrix and to compare it with VAT revenue collection. This method was developed for Mexico by Aguirre and Shome (1988) and later applied to other Latin American countries (Shome 1995). These methods have revealed that as much as one-third of potential VAT collection and half of income tax collection may remain uncollected. Another concept of VAT evasion has emerged in the discussions and practices of international tax experts. This is VAT productivity. Thus if a country has a general VAT rate of x percent, it should ideally be able to collect half x percent of GDP in VAT revenue. If it collects less than onethird x percent of GDP in VAT revenue, VAT productivity is low. The VAT structure possibly suffers from many exemptions and breaks in the VAT chain, and the VAT administration is unlikely to be combating tax evasion successfully (Shome 2002).

42

Parthasarathi Shome

Growth in tax evasion There is a prevalent view that tax evasion has grown in recent years and is growing. Many factors account for it. The role of tax administration in the control of tax evasion is, therefore, not a static one but one that is dynamic and remains ever challenging. Among the significant reasons is that, first, in the last two decades the volume of internationally mobile global capital has increased by leaps and bounds and its flows keep increasing. Its dimensions – both direct investment and financial capital in the latter’s many diverse forms – also transmute and grow. Opportunities for tax evasion abound and increase before even well developed and astute tax administrations can understand or confront them. A later section will draw upon the experience of Asian countries in addressing tax evasion in international taxation. Second, the services sector has been growing phenomenally in many countries, in particular in the Asian region, as agriculture loses its comparative advantage, manufacturing faces stiff international competition under new World Trade Organization rules and swift economic growth shifts the comparative balance towards the specialized services sector. The services sector is often represented by micro-units whether they are business processing offices (BPOs) covering a host of service-oriented activities, highly skilled medical practices catering to a primarily foreign clientele, or highclass restaurants serving the nouveau riche. These new types of economic agents are notoriously difficult to tax. And, if they have emerged as colossal public sector units – producing deficits but having nevertheless been a source of guaranteed revenue for government – and have then disappeared, then the loss to revenue collection could be significant. The challenge for the tax administration here is to develop precision instruments that are able to aim directly at newly emerging revenue sources. Such instruments comprise third-party information sources, compulsory return filing, and other methods, that will be addressed further in this chapter. Third, a mix of sociological and psychological hypotheses of tax evasion has received some attention. Among the sociological are factors such as the premise that the younger generation are less concerned about social or national responsibilities, necessarily the outcome of increased global mobility, the arrival and advancement of the information super-highway, and a common global cultural denominator represented by Music Television (MTV), fast foods and the fashion ramp. The psychological hypotheses take off from there, in that the new generation is perceived to possess utility functions that are more individualistic – the go-getter – that is explicable in the context of a much faster moving world in which there is cut-throat competition and where they have to run to be at the same spot, their goal increasingly represented by the famous metaphor, “famous for fifteen minutes.” While fascinating to the economist and posing a challenge as to how to incorporate such factors in economic analysis (Roth and Scholz

Control of tax evasion and tax administration 43 1989; Slemrod 1992; Cullis and Lewis 1997), for the tax administrator it adds another dimension to his work in the control of tax evasion. Effects of tax evasion Despite erring on the side of brevity, there needs to be some mention of the negative effects of tax evasion on the economic environment. Its most important adverse effect is perhaps on equity. A wage-earning factory worker pays tax. A restaurant worker whose income is the same but who receives part of his income in tips does not reveal it for tax purposes. Thus, one blue-collar worker gains at the expense of the other. This is horizontal inequity. A barrister charges his fees, in part, in non-pecuniary terms, for example, a foreign trip for his family, all expenses paid. This is not reported on his income tax return. A salaried employee in the organized corporate sector earns the same reported income as the barrister. Their incomes appear to be the same for tax purposes. This is vertical inequity since the barrister’s actual taxable income is higher (it should have included the value of his foreign trip). In sum, for both forms of inequity, the higher-taxed person pays for the lower-taxed person since, had there been no tax evasion, the tax rates would have been lower under the premise of revenue neutrality. Second, tax evasion distorts economic efficiency. In sectors that are less subject to the administrator’s scrutiny, there will be more investment. That may be one reason why certain service sector activities – for example, the construction industry – have grown so phenomenally as companies move across national barriers in a globalized world economy. Similarly, the unorganized sector may evade taxes much more easily than the organized sector. It is no wonder that, in the VAT where there is usually a threshold level below which taxpayers are not expected to keep detailed accounts, allowing them to pay a small percentage of their turnover as tax (termed compounding), there is a clustering of registrants just below this threshold. Small taxpayers have remained very difficult to tax and maintain a constant presence in the list of administrative concerns (Shome 2004). Based on a theoretical model, Sanchez and Sobel (1993) conclude that only incomes below some threshold level should be audited with a probability that is positive and less than unity. Third, as both inequity and inefficiency lead to lower revenue intake for government, its functional capacity, efficiency and effectiveness suffer because of tax evasion. Capacity suffers due to lower availability of resources. Efficiency declines since important functions may have to be given less priority than others. And effectiveness declines as compliant taxpayers realize that government is unable or unwilling to take corrective action and, therefore, feel increasingly comfortable in joining the rest in the act of tax evasion.

44

Parthasarathi Shome

The result could very well be an increase in tax rates, or the imposition of distortive taxes, thereby initiating a vicious cycle of inequity and inefficiency. Clearly, the escalation in property tax rates or the imposition of entry taxes by many municipalities, turnover taxes in addition to the sales tax at the level of states or provinces, or inventive surcharges and cesses at the level of the central government are cases in point. Fourth, tax evasion being under-reporting of income, implies underestimation of GDP and all its commensurate macro-economic ramifications. Since the denominator is under-estimated, the ratios of tax to GDP, the fiscal deficit to GDP, and public debt to GDP are all over estimated. The perceived higher tax/GDP ratio leads to false comfort, but exaggerated deflationary action may be taken to rein back an exaggerated fiscal deficit or public debt ratio. Policy could thus turn pro-cyclical, leading to suppression of the organized economy and exacerbating a vicious cycle of less revenue generation and higher tax rates. If the underground economy absorbs labor, then the adverse impact of unemployment – by focusing primarily on the organized sector – will also be over-emphasized. Inflation estimates would be similarly distorted unless care is taken to include all segments of the market economy, both organized and underground.

Cost and design of tax administration Having traversed the causes and effects of tax evasion, we now look at particular aspects in the formation and functioning of an effective tax administration. The first ingredient is whether tax officials are remunerated appropriately in order to check the incidence of tax evasion and corruption. This is inherently linked to the cost of tax administration and, therefore, to the cost of revenue collection. Authors have discussed these matters in stylized frameworks, for example, the issue of increasing inspectors’ salaries (and increase in the opportunity cost of firing them for poor job performance) by Vasin and Panova (1998), or giving bonuses to tax officers in relation to their individual tax collections (Bardhan 1997). Cost of tax collection Over and above incentive schemes that could raise revenue collection, a more direct preoccupation is the matter of cost of revenue collection. Overall, the cost of collection seems to vary a good deal across countries. In a recent cross-country study, Highfield (2001) demonstrates that, in 1998–9, the cost of collection as a percentage of revenue has been 0.72 in Thailand, 0.84 in Australia, 1 in the Philippines, 1.04 in Singapore, 1.1 in Hong Kong, 1.21 in Japan and 1.37 for the UK (Inland Revenue), though, for customs and VAT, it is much lower. Indian data indicate that the cost of collection as a percentage of customs and excise is about 0.86 while that of the income tax is near 1.

Control of tax evasion and tax administration 45 In general, the cost of collecting income tax, as compared with indirect taxes, is higher reflecting the relative complexity of an Income Tax Act, a larger number of income taxpayers, far more points at which the income tax has to be collected, and the wider sources of revenue, both domestic and spread internationally. What is certain is that resources for tax administration should be adequate. More important perhaps, they need to be efficiently spent. This is because tax administration is usually placed well within the framework for overall government administration expenditure allocations and does not enjoy any special dispensation. Therefore, it has to compete for scarce resources. Attempts to set up tax administrations as autonomous agencies external to the ministry of finance have had, at best, mixed success across the world. A framework for administration resources An essential element in the control of tax evasion is the adequacy of administration resources. A theoretically correct tax policy and expenditure structure may be set up. But, without sufficient tax administration resources and mechanisms or expenditure management systems, the tax law or expenditure allocations by themselves would not yield the national income or economic growth envisaged (Shome 2002). That framework is delineated below. The objective of the framework is to demonstrate that there is a binding constraint that tax administration may pose on tax collection over that predicted by the structure of taxes. The same type of constraint applies also to the expenditure side between expenditure policy and expenditure control. Thus Figure 2.1 refers to a wider fiscal context, incorporating both tax and expenditure sides. For simplicity, we shall use tp, ta, ep and ec to denote the tax policy, tax administration, expenditure policy and expenditure control variable, respectively. Conventionally, if tp is the tax rate and Y1 ⫺ Y0 is income, then tp (Y1 ⫺ Y0) ⫽ TP is the tax revenue that is collected. This is demonstrated in the top left quadrant of Figure 2.1. TP can then be consumed or invested by government in its expenditure programs EP, leading to income in the next period of Y2 ⫺ Y1, as sketched in the top right quadrant of Figure 2.1. The expenditure–income relationship is: TP ⫽ EP ⫽ ep (Y2 ⫺ Y1) where ep is the expenditure/income ratio. So far, we have related tax policy and expenditure policy in the top two quadrants of Figure 2.1.

46 Parthasarathi Shome Tax and expenditure policy and resources

TP⫽EP

tp Y0

ta

ep A

B

Y1

Y2

Income variable

ec

C C

D

Tax administration and expenditure control

Figure 2.1 Tax and expenditure: policy and administration (source: see text).

There has been no role specified for tax administration or government expenditure control in constraining revenue generation or its expenditure by government. This influence is depicted in the bottom quadrants of Figure 2.1 and is explained below. The achievement of Y2 subsumes a given state of tax administration and expenditure control which is usually not recognized in policy circles to the extent that it should be. The underlying tax administration coefficient is ta and the underlying expenditure control coefficient is ec. To make the concept of correspondence clearer, diagrammatically tp and ta are shown to be equal. Similarly, ep and ec are equal. To put it another way, ta and ec are given an equal role in the generation of the economy’s income flows. They would simply lie underneath or veiled as it were, just as the oil to grease the wheel of revenue generation and concomitant expenditure. Problems would arise, however, if the underlying assumptions were not correct. This is elaborated further. The role of tax administration could be specified as: TA ⫽ ta (Y1 ⫺ Y0) where TA would be the imputed resources – reflecting the efficiency with which its available finances are administered and utilized – that tax administration would require to generate revenue TP, given income (Y1 ⫺ Y0). This is shown in the bottom left quadrant of Figure 2.1 as the distance AB.

Control of tax evasion and tax administration 47 The expenditure control side, ec, follows directly. Essentially, its role is added on: EC ⫽ ec (Y2 ⫺ Y1) This is depicted in the bottom right quadrant of Figure 2.1 as the distance BD. Obviously, the overall imputed resources needed for government’s taxexpenditure policies as designed and projected, and the resultant sequence of incomes to take effect, are depicted by the distance AB ⫹ BD. Equivalently, the minimum resource requirement translates to C ⫽ C. This may not represent reality, however, if administrators are simply not able to target the universe of potential taxpayers. Thus, administration has to be enhanced, or greater administrative resources made available to administrators. In that case, a higher underlying requirement K ⫽ K would apply, to generate the same income flows (Figure 2.2). K is determined as follows. Working backwards within Figure 2.2, it is easy to see how administration practices that are not able to be addressed or cover the complete tax structure, or inadequacies in expenditure management and control, would constrain the result depicted in Figure 2.1. The constraint can be broken up into tax administration and expenditure control constraints. Tax and expenditure policy and resources

TP⫽EP

ep

tp Y0

ta

ta*

A Y1

B B1 C

K

Y12

D

D1

ec1

Y2

Income variable

ec ec* C

K

Tax administration and expenditure control

Figure 2.2 Tax and expenditure: gaps in administration and control (source: see text).

48 Parthasarathi Shome Say ta is inadequate, resulting in a need for additional administrative resources. Then ta* ⬎ ta would be needed. Similarly, deficient expenditure management implies ec* ⬎ ec. Only with these additional resources, AB1 ⫹ B1D1 ⬎ AB ⫹ BD, will the same income, Y2 ⫺ Y1, result. Otherwise, with any combination of resources represented by coefficients lower than ta* or ec* (such as say ta and ec1), a lower income than Y2 ⫺ Y1 (such as Y21 ⫺ Y1) will result. (Note that ta ⬍ ta* but ec1 is set to equal ec*). Once this happens, future income flows would be similarly adversely affected and, ultimately, economic growth would suffer. The essential lesson from this simple exposition can be useful. The gaps between tax policy and tax administration (and between expenditure policy and expenditure control), have to be bridged if government’s objectives in formulating its tax–expenditure policy mix are to materialize. The difference can take the form of lack of required resources for full implementation, or the inefficient use of allocated resources or, simply, an application of the law that does not completely reflect its original intent and purpose. Next we look further into the role of modern tax administration methods in the control of tax evasion and at specifically designed tax policy measures for the control of tax evasion. Design of modern tax administration controls The basic design of a tax administration is well known and has been widely discussed, whether along a tax administration’s functional lines (Bird and Casanegra 1992; Silvani and Baer 1997), or along economic lines or as a firm maximizing its objective (Bagchi et al. 1995; Tanzi and Pellechio 1995). It is not the purpose here to traverse those same grounds. Instead, as indicated in the introductory section, we intend to take up three specific matters in the development of modern tax administrative practices that represent a movement away from tokenism and towards incisive facilitation of the primary operations – assessment, audit, scrutiny, investigation and judicial – of tax administration. First, what are some of the incisive instruments that have been developed in terms of taxpayer information flow that could be usefully utilized by the administration to expand the taxpayer base and facilitate additional revenue mobilization? Such instruments include, among others, additional information (third party) returns for large transactions; tax deduction at source (TDS) or withholding; large taxpayer units (LTUs); and compulsory filing of returns to enable non-tax information matching for those who have zero taxable income but fulfill certain expenditure criteria. Second, with the spread of international businesses across national boundaries, how do tax administrations monitor arm’s length rules in the setting of transfer prices? Various Asian countries are dealing with it to different intensities, but it is clear that they are all tightening their rules and

Control of tax evasion and tax administration 49 implementing them more strictly. These two aspects will be examined in this section. Third is the issue of computerization of the tax administration. This has become a mainstay of a modern tax administration. But the design of computerization itself is a challenging task as will be explained in the next section. The ultimate success of a tax administration in collecting revenue rests squarely on a well-designed mainframe into which all information flows are channeled and based on which all cross-checks are possible. Tax administration precision instruments As modern tax administration has moved to financial – as opposed to physical – control methods, its instruments have become more precise. They are oriented towards cross-checking rather than merely unilinear checks of consignments leaving the factory in the case of domestic production (excises) and consumption (VAT) taxes, or single-file examination of an income taxpayer, where the emphasis is increasingly on third-party information. In the case of an annual information return (AIR), a taxable person (A) is asked to submit returns on transactions with others (B) over a certain threshold. This allows the administration to cross-check whether the others (B) have filed returns or, if they have filed, to verify income. For example, in 2005, the Indian income tax authorities have enacted AIR requiring a number of such transactions with respect to a person in one year to be reported to the income tax administration: banks on cash deposits of INR1 million or above in any savings account; credit card companies on payments (INR0.2 million); mutual fund managers on investments (INR0.2 million); companies or institutions issuing bonds or debentures on investments (INR0.5 million), or shares through public issue (INR0.1 million); registrars or sub-registrars of properties on purchase or sale of immovable property (INR0.3 million); and the Reserve Bank of India (central bank) on investment in RBI bonds (INR0.5 million). This is just to illustrate the long arm of the tax administration that is increasingly intended for use in cross-checking and detecting tax evasion. The challenge remains, of course, in effective utilization of such information for the intended objective, in particular, because the compliance cost for third parties in meeting AIR requirements is unlikely to be insignificant and cannot be ignored. Among the more traditional instruments involving two parties is tax deduction at source (TDS), which has borne the test of time. It is no longer an issue of its use but, rather, to how many sources of income it is applied. TDS for salary earners is most common in Asian countries. India utilizes TDS for a much wider set of income sources above designated thresholds, for example, rent, interest from banks, securities (though not

50

Parthasarathi Shome

dividends since they are not taxed in the hands of the shareholder) and inter-corporate interest payments, payments to contractors and subcontractors, insurance commission to corporations and firms (not individuals), commission and brokerage (other than insurance), payments to non-residents, payments on account of repurchase of Unit Trust of India and other mutual fund units, winnings from lotteries, crossword puzzles and horse races, and commission on the sale of lottery tickets. But effective administration of TDS requires the use of computerization, a matter that will be taken up in the next section. Another widely tested instrument is the large taxpayer unit (LTU) through which taxpayers whose tax contributions are above a threshold are required to file. More than 50 countries use this window (Baer 2002). Its advantage from the administration’s point of view is that information exchange across taxes is a built-in feature of the framework. From the taxpayer’s point of view, the availability of an exclusive window for tax payment and the likelihood of speedy processing for dispute resolution are important attractions. LTUs have proved to have assisted in facilitating revenue generation and lowering compliance costs, and have become a common feature in Asian countries. In her annual 2005–6 fiscal budget, India likewise announced her intention to set up LTUs. On the other hand, 20–30 percent of total potential revenue may be represented by small taxpayers. Thus the best resources of tax administration cannot be focused on large taxpayers alone and small taxpayers cannot be ignored (Shome 2004). Incisive carrot-and-stick instruments are needed for garnering cooperation from the latter. Their fears of the tax administration may be reduced through the establishment of help centers in cooperation with private chambers of industry and commerce. At the same time, small potential taxpayers may be required to file returns even if they self-assess zero taxable income. Such an instrument requires persons to file returns irrespective of taxable income as long as they meet certain expenditure criteria. India again has utilized such a scheme since 1997–8, in particular with the objective of requiring non-taxpayers to file. Today it is known as the 1 ⫻ 6 scheme since, if an individual meets any one of six criteria, he is required to file. These include ownership of property, automobile, credit card, undertaking of foreign trip, annual electricity bill above INR50,000, and membership of a club with an entrance fee over INR25,000. Again, even as the taxpayer register and information flows increase remarkably as a result, the utilization of such data to achieve the originally intended purpose remains a major challenge. For indirect taxes, effective cross-checking forms the crux of modern tax administration based on financial control methods. The most applied method for VAT is random cross-checking of invoices. Of course computerization is of paramount importance with appropriate systems applications for scrutiny selection based on predetermined indicators. They

Control of tax evasion and tax administration 51 usually include input/output ratio, credit availment, export/output ratio, sector-specific characteristics and others. But in evasion-prone environments – where multiple account books are likely to be maintained by taxpayers – such cross-checking may not be sufficient since it is not likely to reveal serious evasion (Mukhopadhyay 2005). Here, second-generation computer models that limit the role of cross-checks by the tax administration and roll over the responsibility of credit availment flow within the VAT chain are being contemplated. Such a method would automatically check availment of input tax credit claimed by a taxpayer (buyer) unless the credit appears on the scroll submitted by the depositor (seller) in the bank where the VAT is deposited. In sum, the role of computer-assisted systems has become imperative in effective tax administration as physical forms have been essentially replaced by financial forms of control. Tax enforcement in a globalized world economy If the previous section dealt with the means to improve the tax administration of domestic taxation, an increasingly important feature that is discussed in this section is one that deals with emerging administration challenges for international taxation. One important effect of globalization is on taxation. On the one hand, as cross-border related-party transactions overpower domestic transactions, a typical tax director of a multinational company (MNC) tends to claim that geographically split tax attributions, that will completely satisfy a tax administration, have become difficult to achieve. On the other hand, tax authorities that call the splitting “transfer pricing” maintain that extreme caution is required regarding MNC tax avoidance and evasion because of greater opportunities to hide income across the globe. They have stepped up field audits, visiting business premises and factory sites, and imparting a more visible presence to enforcement activities. Such audits seem to have facilitated the control of both tax avoidance and tax evasion. Pricewaterhouse Coopers (2003) has carried out an extensive survey of the East Asian region. Some country experiences are summarized here. In 2002, China conducted more than two million “tax reviews” that resulted in more than 800,000 audits and a recovery of more than RMB39 billion. The focus on transfer pricing practices of MNCs even for intra-China transactions – so much so that these have essentially come to be called “transfer pricing audits” (TPA) – has tended to restore revenue that earlier suffered from cross-border tax avoidance activities. The selection of cases for TPA begins with desk-top analyses based on: low profits for extended periods pari passu with company expansion, sizeable drop in profits after expiration of tax holiday, profit consistently lower than industry average, and inexplicably fluctuating profits. Import–export transactions are, of course, examined, but intangible transactions such as licensing, financing and provision of services are increasingly scrutinized.

52 Parthasarathi Shome The audits require cooperation between national and local tax authorities. Cooperation enables the latter to establish their own databases for audit targets while being supported by the former on transfer pricing adjustments. “National joint audits” (NJA) featuring simultaneous audits of Chinese subsidiaries of MNCs are, therefore, made possible. From just coastal areas, audits now extend into the far reaches of the interior. The joint audit approach obliges MNC subsidiaries to improve coordination in management and accounting, in particular, transfer pricing. Transfer pricing adjustments can be imposed by the administration within three years (in extraordinary cases, even ten years), resulting in additional taxes payable on the adjusted income. After the TPA, a follow-up review could also be undertaken by the administration. Controlling individual income tax evasion has been attempted through general investigations, informers – with rewards up to RMB100,000 for appropriate information – and tax recovery operations on a targeted industry basis. In 2002, for example, executives in the high-technology sector were scrutinized with respect to declarations on offshore stockoption benefits. Such audits tend to affect goodwill of industry adversely. They also lead to financial consequences for industry since the resultant increased tax burden – gross-up tax rates applied to net-of-tax employment contracts – is often borne by the employer, in particular where the employee leaves China. Hong Kong follows the territorial principle of taxation, taxing only Hong Kong sources of income. For companies incorporated in Hong Kong, the territorial principle attempts to tax profits arising in or derived from Hong Kong. Commensurately perhaps, tax evasion is dealt with strictly. Using field audits and investigation to control tax avoidance and evasion, pecuniary punishment – sometimes 100 percent of tax evaded – and prison sentence are both levied. For audit, criteria used include: whether deductions claimed reflect commercial basis, to what extent a Hong Kong company’s expenses are incurred for a related offshore company (hence not deductible), what portion of the profits are attributable to Hong Kong, and intra-company pricing policies for services provided internally. In Korea, transfer pricing has occupied center stage in recent years. For administrative support, the tax administration initiated a project to train 1,000 tax officers in transfer pricing during 2002–7. The Law for Coordination of International Tax Affairs (LCITA) applies to cross-border transactions since 1996, following OECD guidelines on transfer pricing. Its main objective is to ensure arm’s length application in price determination. A company has to submit a statement on transfer pricing choices made in overseas inter-company transactions. Additional explanations may be asked for, with replies expected within 60 days, with an extension of another 60 days. Failure to comply could result in a penalty of up to W30 million and is

Control of tax evasion and tax administration 53 likely to lead to a TPA. TPAs are based on a close watch over MNCs and their polices, investments and practices in Korea. They increasingly attempt to cover management service fees that comprise fees for services provided from a central office to subsidiaries. Comparable prices are difficult to find and their benefits not easy to evaluate, thereby carrying the potential for abuse. As a result, an incisive TPA could yield huge amounts through transfer pricing adjustments. Malaysia uses its investigation and intelligence branches for controlling tax evasion. Officer training has been received from both the United States and Japan. While a field audit unit, established in 2001, usually carries out audits with advance notice – with a target of auditing a company every five years – unannounced visits for tax investigation are also conducted. Visits have resulted in MNCs facing additional taxes and penalties. Malaysia’s transfer pricing rules follow OECD guidelines. They apply both to MNCs and domestic companies and include cross-border and domestic intra-group transactions affecting profits, such as transfer of tangible goods and assets, and transfer and use of intangibles and services. They require documentation to justify that transfer prices are set at arm’s length. Based on a self-assessment system, penalties escalate over the stages of voluntary disclosure, non-disclosure and repeated offence. Singapore has typically conducted tax investigation of local businesses. Though random audits – of which transfer pricing examination forms a part – to verify tax returns are carried out, tax investigation per se remains focused on those suspected to be repeat tax evaders. Voluntary disclosure of evasion reduces publicity, investigation and penalty. In fact, for unwitting tax evaders, the penalty was lowered in 2000 to 10 percent of underpayment, from a 100 percent maximum earlier. This could be one explanation for the low number of investigation cases. But if evasion is determined, conviction could lead to severe pecuniary and non-pecuniary consequences, since the courts do not shy away from imposing maximum penalties. Punishment escalates – for incorrect return to fraud – to 400 percent of amount evaded with a ceiling of S$50,000 and imprisonment of up to five years. And case details may be put on the authorities’ website. To combat tax evasion, Bangladesh has set up a Central Intelligence Cell (CIC) in its revenue administration. The CIC can investigate income tax files, check financial accounts, gather evidence and seize suspicious files. It has focused on monitoring large taxpayers, detecting several cases of tax evasion involving significant sums. A recent case of matching credit card expenses with the income tax returns detected huge tax evasion. Examination of tax files of clearing and forwarding agents found that, on average, they had concealed half their income derived from commissions. This resulted in a tax recovery of BDT20 million from this sector in the city of Chittagong alone in 2003 (Kibria 2004). It is pertinent to mention, at least in passing, the experience of the developed economies since, in a manner, the Asian economies emulate

54

Parthasarathi Shome

their experiences. France, Germany and Belgium aggressively investigate and prosecute nationals and residents investing in foreign investment vehicles – life insurance and annuities – to evade tax. France took Pan Euro Life – a subsidiary of Ohio-based Nationwide Global Holdings – to court in 2001, claiming it had been established to evade tax (Zagaris 2001). Given that its chairman was a former head of the European Commission and Prime Minister of Luxembourg, it revealed determination and resolve on the part of the tax administration. The United States has also played a role in enforcement efforts in international taxation. In 2001, third-party information was stipulated through qualified intermediary (QI) regulations. QI required banks and financial institutions abroad to institute mechanisms to identify and inform on benefiting investors and account holders who should be subject to withholding. Extensive raids have also been carried out on promoters of tax evasion schemes using offshore banks and trusts. In sum, recent experience in Asia and the developed economies demonstrates an increasing awareness of tax effects of trans-national economic activity and follow-up action through joint audits of subsidiaries within national boundaries, fine-tooth comb examination to ensure arm’s length application of transfer pricing in both domestic companies and MNCs and, if warranted, use of investigation, raids and pecuniary and non-pecuniary punishment.

Computerization and automation It is needless to emphasize at this point, after describing the complexities that a modern tax administration encounters, how important the role of appropriate computerization is in its overall management and in the control of tax avoidance and evasion. Successful administration is essentially based on the assimilation of a comprehensive information pool, combined with intelligent, result-oriented data mining. Physical controls including raids for search and seizure can be used, but they are mainly reserved for extreme circumstances. In this section, the issue of computerization is addressed. It is not the objective to provide a checklist for computerization needs as tends to appear in many an advisory report of tax and computer experts. Rather, what is attempted is to provide an essence of the type of problems faced in developing a computerized system, for example, for cross-checking information on tax collection and crediting of taxpayer accounts for income tax or VAT. Let us take income tax. India has recently computerized various processes in income tax that provide a good illustration of lessons that could be learnt in computerizing a tax department. This case study is described below. One aspect of computerization of income tax processing is the On-Line Tax Accounting System (OLTAS) that was initiated in 2004 on the basis of

Control of tax evasion and tax administration 55 a single hard-copy system. Income tax payment can be made through one return form in banks. The form appears to be quite simple and welldesigned. It has two “major heads” (individual income tax and corporate income tax) and four “minor heads”(self-assessment, TDS, advance tax and regular tax). It also has a block for “assessment year.” Let us take the case of TDS and track its processing. This is illustrated in Figure 2.3. An employer who has withheld tax at source from his employees goes to any collecting bank branch and deposits the money. The bank keeps the form and gives him a receipt, which is the portion below a perforated separation on the form. At the end of the day, a bank entry operator enters the details into the bank’s self-developed software that processes the form. A supervisor tallies the entrance and approves it. The bank passes the money through a nodal branch (clearing house) that is designated to receive all inflows in one day (T ⫹ 1). Within three days (T ⫹ 3), the receipts are transferred to a central link cell. Here the monies are, in effect, separated from the information flows. The receipts are sent for deposit in the central bank (RBI) that simply credits the government’s accounts. The data with the details on the return form, as entered by the banks, are sent electronically to a data depository and processing entity (NSDL)

Collecting Bank Branch

Nodal Bank Branch

Money and Data to Central Link Cell

Money deposited in central bank (RBI) Data to National Data processed and Information sent to National Depository (NSDL) – Computer Center Tax Information (NCC) Network (TIN)

Connected to Assessing Officers (AOs) online in 60 cities

Single scroll and hard copy of return form sent to Zonal Accounts Office (ZAO) Principal Chief Controller of Accounts (PCCA) who is outside Income Tax Department

Data sent to 36 Regional Computer Centers (RCCs) Should copy of scroll and return forms go to RCCs in the future? For past return forms, would RCCs scan from ZAO?

Current framework of OLTAS Possible modification to OLTAS

Figure 2.3 Income tax: Online Tax Accounting System (OLTAS) (source: see text).

56

Parthasarathi Shome

which is the pivot of the Tax Information Network (TIN). NSDL processes and transmits the data to a National Computer Center (NCC). From the NCC, the data are fed into 36 Regional Computer Centers (RCCs). The RCCs are connected to Assessing Officers (AOs) in 60 cities so far (to be extended to 510 cities when the project is completed). Ideally an AO will have ready online access to the information when he makes an assessment or undertakes an audit. The only hard copy of the return form is sent by the nodal bank branch to the Zonal Accounts Office (ZAO) where it rests with the Principal Chief Controller of Accounts. In sum, the monies are sent to the RBI; the electronic data are accessible by the AO at the RCCs after some routing; and the single hard copy of the return form rests with the ZAO. This results in three final destinations for the money, electronic data and hard copy. The structure may be expected to function well except where human error in the previous stages of entering the data frustrates information matching. Initial experience has unveiled such practical lacunae. First, in India, the term “income tax” covers both individual and corporate income tax in the Income Tax Act. Initially, therefore, the form was often filled out wrongly so that collection showed a disproportionate individual income tax collection over corporate tax collection. The form was accordingly modified by separately identifying the corporate income tax. Second, the fact that data entry is left to bank entry operators who may not be familiar with the Income Tax Act led to errors of non-matching. For example, TDS in 2005–6 must necessarily relate to assessment year 2006–7. But TDS entries have been found to be entered with 2005–6 as the assessment year (mistakenly interpreted to be the current financial year). Even employers who turn in the return form sometimes commit this error. Third, no hard copy of the return form remains with the TDS deductor/depositor (who only has a receipt). This tends to leave little trace of the bank-attested details of his deposit payments with him that he could usefully produce if summoned by an AO. Fourth, the AO cannot easily access a hard copy on his own (for making corrections with respect to a return) since the only hard copy lies at the ZAO where the keepers of government accounts could rightfully be expected to be cautious about parting with the sole hard copy information. Thus the system’s initial creases would need to be ironed out. Would the requirement of a second hard copy, that would go from the nodal bank branch to the RCCs be a retrograde step for modern computerization techniques, or should that be perceived as an evil necessity for crosschecking electronic information that arrive at RCCs from the NCC? Further, should bank entry operators be trained to familiarize them with the basic relationships underlying the return form, or would it be more efficient and user-friendly for the income tax administration to make available (or insist upon) its own software to all banks that would cut out builtin inconsistencies?

Control of tax evasion and tax administration 57 Alternately, should the prevailing system itself be tightened up rather than trying to modify it? For example, by insisting that bank data entry be done at the front end (rather than end-of-the-day entry) under the assumption that those who physically come to the banks to deposit the checks are more familiar with the return form so that their presence would minimize errors in data entry? But this may not be possible since data entry is carried out only after a bank’s clearing house has cleared a check. On the other side, if forms are not correctly completed by depositors, should banks be instructed to refuse them? These are considerations that would need to be weighed appropriately before arriving at a comprehensive solution to the teething problems. Another lacuna is with respect to the deductee in this system. The deductor deducts from his employee (the deductee) and deposits TDS in a bank as explained. Today the deductee receives, from the deductor, a certificate of deduction specifying the amount deducted. It is proposed to do away with this requirement (de-materialization of TDS certificates) under a fully functioning OLTAS. However, until the remaining lacunae are fully plugged, the deductee has been temporarily protected with continued issuance of TDS certificates to him. In sum, the minutiae of computerization can be challenging and are certainly interesting; there is no option but to address issues as they emerge since its ultimate purpose is to facilitate the process from both tax administration and taxpayer points of view. They are unlikely to be resolved in the short term. A medium-term framework is its essence. The medium-term nature of the solutions, in turn, poses a related problem. The advancements in information technology are so overwhelming, and hardware and software changes are so rapid, that project proposals are likely to become dated faster than due process of government examination, both at departmental and political (cabinet) levels, contract tendering and actual implementation. This remains a constant challenge in environments where large projects are particularly subject to various checks and balances through expenditure control and management mechanisms, stoutly in place and actively operational. Just as in the case of income tax, VAT computerization has been a topic of constant attention. While the focus is usually on software to facilitate cross-checking by the tax administration, the newest proposals are being made in the direction of unilinear checking. Thus, only when a taxpayer A in the VAT chain credits the account of another taxpayer B (from whom A has collected VAT) and this is reported in the bank (where credit scrolls would be maintained for different taxpayers) would B’s account be credited. In this simplified method, the onus of cross-checking is effectively passed on from the government to the universe of taxpayers. As indicated above, therefore, computerization models of tax administration are in continuous development and represent one of the most dynamic areas in the taxation field.

58 Parthasarathi Shome A final point on the computerization theme that should be mentioned is the information relationship between the center and states (or provinces) in a federal fiscal system. Cooperation was already mentioned in the context of successful joint audits in China. In addition, it is imperative to develop information exchange systems to minimize tax evasion. In particular, information should be easily exchangeable if the states too are assigned important taxes that they themselves legislate, collect and appropriate, rather than being subject only, or mainly, to revenue sharing with the center. The majority of Indian states introduced a VAT on 1 April 2005, at the sub-national level. The center also has a VAT (called the CENVAT) up to the manufacturing stage. A computerized information exchange system (TINXSYS) is being developed on a cost-sharing basis between the center and states that will link various states in a common information exchange network. Eventually, such information exchange should also be carried out between the center, on the one hand, and the states, on the other. A successful TINXSYS should be able to identify and check tax evasion more efficiently in the future.

Design of tax policy to support tax administration Last, but not least, we touch upon some tax policy instruments that are being used as new innovations in tax administration control. While they do not yet feature in a large number of countries, there seems to be a spreading demonstration effect encouraging more countries to adopt such policies. Usually tax policy experts have considered them to be distortionary (Shome and Stotsky 1995); but it appears that tax administrators are using them successfully without their commonly believed distortive effects coming into view. The two most common ones are taxes on financial transactions and on fringe benefits of employees. These are examined next. Financial transactions taxes have been introduced by Brazil at 0.038 percent on an array of financial transactions, followed by Argentina (0.75 percent), Colombia (0.4 percent), Peru and a few other Latin American countries. All countries are reporting an increase in revenue collection. But its main benefit seems to be in the form of an enhanced flow of information for the tax administration. Brazil has experienced a 40 percent rise in tax assessments issued against taxpayers during the first nine months of 2004 as a result of the tax administration’s new strategy of cross-checking taxpayers’ records of the financial transactions tax (CPMF) with information provided by credit card companies and real estate brokers (Soares da Silva 2004). In the first three-quarters of 2004, assessments resulting from CPMF cross-checking totaled BRL28 billion, a record since the CPMF began as a tax audit instrument. An increase in assessments is projected for 2005, mainly reflecting a new 0.005 percent capital gains withholding tax on stock sales in stock exchanges. This tax

Control of tax evasion and tax administration 59 was created to cross-check stock sales with taxpayers’ tax returns (stock sales are exempt from CPMF). With exactly the same objective, in her annual 2005–6 central fiscal budget, India has just introduced a withholding tax of 0.1 percent on cash withdrawals from bank checking accounts above a specified threshold. To quote Chidambaram (2005), the Finance Minister: The National Common Minimum Program (NCMP) requires the government to introduce special schemes to unearth black money and assets. I am obliged to carry out the mandate, but without giving undeserved relief or amnesty. I am concerned about large cash transactions, especially withdrawals of cash, when there is no ostensible purpose to withdraw such large amounts of cash. These cash withdrawals leave no trail, and presumably become part of the black economy. Therefore, . . . I propose to levy a tax on withdrawal of cash . . . from banks at the rate of 0.1 percent. (Chidambaram (2005): 31) While initially it met with popular opposition, it has come to stay in a scaled-back form with respect to its base. Despite its clipped wings, if it could still be successfully used, the instrument could assist in tracking money laundering. A common form of money laundering appears to take place through false-account stock transactions and their subsequent channeling through multiple bank accounts, thereby legitimizing their existence without payment of tax. Taxation of fringe benefits (FBT) accruing to employees – that comprise non-salary benefits that may be cash or non-cash – can be carried out in one, or a mix, of three ways. Most appropriately, it can be taxed at the hands of the employee through his individual income tax return, thereby requiring voluntary declaration by the taxpayer. Developed countries use this method and tend to have stringent provisions to make it work. The second method is through disallowance from expenses under the corporate tax. If a company is declaring losses anyway, the disallowance would not be binding, however, and would not yield revenue. The third method is to have a positive list of fringe benefits and tax them at the hands of the employer at pre-specified presumptive rates. While the employer would be allowed to deduct all such expenses from his income tax, the more he deducts, the larger would be his tax base for the FBT. As in the case of the financial transactions tax, the FBT is also a relatively new tax and not yet à la mode. Nevertheless, some have extolled its virtues. To quote Brooks (2001): A second obvious example of a technical rule that is unenforceable is one that allows business people to deduct the cost of business meals and entertainment. Allowing business people to deduct the cost of

60

Parthasarathi Shome their business meals and entertainment makes absolutely no sense in terms of basic tax principles. Almost inevitably the personal value of a business meal, and in every case the personal value of a business entertainment expense, will equal its cost. The most basic tax principle is that taxpayers should not be allowed to deduct the cost of expenditures from which they derive personal value. But even if in some cases business people do not enjoy the business meal or entertainment that they inflicted on their clients, these expenses should still be nondeductible since they are impossible to verify. Who knows what business people talk about while enjoying a meal at a fancy restaurant. In Toronto, over 90 percent of Blue Jays’ (the cities National Baseball League team) season tickets are deducted as business expenses. It is impossible for the tax department to verify that these tickets were in fact used to take business clients to baseball games. Basically, for all so-called dual purpose expenses, from which people customarily derive some personal benefit, such as home-office expenses, education expenses, travel expenses, and automobile expenses, drafters should be ruthless in drafting enforceable brightline tests for distinguishing between personal and business expenses and disallowing all of the rest. (Brooks (2001): 23)

Australia and New Zealand are two developed countries that have introduced the FBT. India followed in the 2005–6 annual fiscal budget with initial sharp adverse reaction from the corporate and firm sectors that are subject to the tax. It has since been revised but is nevertheless in operation currently. The Indian version lists out fringe benefits that are primarily enjoyed jointly by employees. It also includes some benefits that are identifiable with individual employees, but it has not been easy to collect tax on them (through a perquisites tax that falls on the employee). The new FBT assigns a presumed percentage of such benefits – health club membership, entertainment expenses, travel, tour and lodging, telephone expenses, scholarships for children, optional pension schemes tailor-made for the upper end of the salary scale – as fringe benefits that do not represent business expenses. These vary between 100 percent for the pension scheme, 50 percent for entertainment and health club expenses, 10 percent for telephone and 20 percent for most others. The corporate tax rate of 30 percent is applied on that fraction and is to be collected as an annex to the corporate income tax return. Certain business activities such as airlines and tour companies, information technology, pharmaceuticals and construction in which travel forms a significant component of business expenses are given special treatment. The tax was essentially introduced as a device to reduce tax evasion and is expected to be revenue productive. The Indian FBT has been introduced pari passu with a major simplifi-

Control of tax evasion and tax administration 61 cation of and scaling up of individual income tax brackets that has reduced the income tax burden of the taxpayer across all income brackets and categories such as salary earners, non-salary earners, women and senior citizens. In combination, therefore, the FBT is expected to improve the equity of individual income tax: even though it will be collected from the corporate sector, its incidence is expected to be shared, if not passed on, to the upper-income salary earner enjoying fringe benefits without being adequately taxed on them. To conclude, a word of caution must be provided here. While theoretically fine tax structures may not be too helpful in successful tax administration, at the same time, tax administrators should be very cautious about introducing taxes that merely collect revenue but cause inefficiencies in resource allocation or are intrinsically inequitable. In the medium term, such taxes will arrest economic growth and, therefore, would have a dampening impact on tax revenue.

Note 1 The opinions expressed in this chapter are entirely those of the author and do not reflect those of any institution or government unless specified. The author appreciates helpful information support from Monica Bhatia and Chandrajit Singh.

References Aguirre, C. A. and Shome, P. (1988) “The Mexican value added tax (VAT): methodology for calculating the base,” National Tax Journal, 41: 543–54. Asher, M. G. (2001) “Design of tax systems and corruption,” Conference on Fighting Corruption: Common Challenges and Shared Experiences, Singapore: Konard Adenauer Stiftund and Institute of International Affairs, May 10–11. Baer, K. (2002) “Improving large taxpayer compliance: a review of country experience,” Occasional Paper 215, Washington, DC: IMF. Bagchi, A., Bird, R. and Das-gupta, A. (1995) “An economic approach to tax administration reform,” Discussion Paper 3, Toronto: International Center for Tax Studies, Faculty of Management, University of Toronto. Bardhan, P. (1997) “Corruption and development: a review of issues,” Journal of Economic Literature, 35: 1320–46. Bird, R. and Casanegra de Jantscher, M. (eds) (1992) Improving Tax Administration in Developing Countries, Washington, DC: IMF. Brooks, N. (2001) “Key issues in income tax: challenges of tax administration and compliance,” Manila: Asian Development Bank Tax Conference (unpublished). Chander, P. and Wilde, L. (1998) “A general characterization of optimal income tax enforcement,” Review of Economic Studies, 65(1): 165–83. Chidambaram, P. (2005) Speech, Budget 2005–6, New Delhi: Ministry of Finance, Government of India. Cullis, J. G. and Lewis, A. (1997) “Why people pay taxes: from a conventional economic model to a model of social convention,” Journal of Economic Psychology, 18: 305.

62

Parthasarathi Shome

Highfield, R. (2001) “Tax administration: understanding and using the cost of collection ratio,” Draft Note, Fiscal Affairs Department, Washington, DC: IMF. Kibria, M. (2004) “Bangladesh revenue authority pleased with progress of new tax evasion division,” Tax Analysts, August 6. Mookherjee, D. and Png, I. (1989) “Optimal auditing, insurance, and redistribution,” The Quarterly Journal of Economics, 104: 399–415. Mukhopadhyay, S. (2005) Economics of Value Added Tax: Theory and Practice, New Delhi: Centax Publications, PVT, Ltd. Pricewaterhouse Coopers (2003) Asia Pacific Tax Notes, Issue 16, New York: August. Roth, J. A. and Scholz, J. T. (eds) (1989) Taxpayer Compliance: Social Science Perspectives, Volume 2, Philadelphia: University of Pennsylvania Press. Sanchez, I. and Sobel, J. (1993) “Hierarchical design and enforcement of income tax policies,” Journal of Public Economics, 50: 345–69. Schneider, F. and Enste, D. H. (2000) “Shadow economies: size, causes and consequences,” Journal of Economic Literature, 38: 77–114. Shome, P. (1995) “Comprehensive tax reform: the Colombian experience,” Occasional Paper 123, Washington, DC: IMF. —— (2002) India’s Fiscal Matters, New Delhi: Oxford University Press. —— (2004) “Tax administration and the small taxpayer,” IMF Policy Discussion Paper, PDP/04/2: 32, Washington, DC: IMF. Shome, P. and Stotsky, J. G. (1995) “Financial Transactions Taxes,” IMF Working Paper 77, Washington, DC: IMF. Silvani, C. and Baer, K. (1997) “Designing a tax administration reform strategy: experiences and guidelines,” IMF Working Paper 30, Washington, DC: IMF. Slemrod, J. B. (ed.) (1992) Why People Pay Tax: Tax Compliance and Enforcement, Ann Arbor: University of Michigan Press. Soares da Silva, D. R. R. (2004) “Tax assessments soar as Brazil uses CPMF information,” Tax Analysts: December, 2. Tanzi, V. and Pellechio, A. (1995) “The reform of tax administration,” IMF Working Paper 22, Washington, DC: IMF. Tanzi, V. and Shome, P. (1993) “A primer on tax evasion,” Staff Paper 40/4, Washington, DC: IMF. Transparency International (2001) http://www.transparency.org and http://www. transparency.org/documents/cpi/2001/cpi2001.htm. Vasin, A. A. and Panova, E. I. (1998) “Tax collection and corruption in fiscal administration,” Final Report on EERC project. World Bank (1997) The World Development Report 1997: The State in a Changing World, New York: Oxford University Press. Zagaris, B. (2001) “Developments in international tax enforcement: a brave new world at the millennium,” Tax Analysts: August, 12.

3

Fiscal federalism in the big developing countries China and India Angela Fraschini

Introduction Our sample of South and East Asian countries is made up by the two biggest developing countries (China and India) at the rushing starting stage of their catching up, by two transition countries (Malaysia and Thailand) and, finally, by two industrialized countries (Japan and South Korea). According to different historical and cultural roots, a highly centralized government prevails in many of these countries, although recently some trends are moving toward a greater degree of fiscal decentralization. By adding some information presented inside the country chapters, a short synopsis is given in the next section of this chapter. A different situation characterizes China and India. In the following section we show that these enormous countries, although different as to their historical, institutional and cultural developments, cannot rely on a merely centralized government, owing to the extension and the striking existing differences of their regions, populated by some hundred million people. Therefore, both the countries share a greater or lesser degree of fiscal federalism or, better, of fiscal unionism. We compare the different institutional arrangements of fiscal decentralization in China and in India, focusing on the financing (own resources, share to central taxes, grants, equalization systems) of sub-national layers. The two countries have different typologies of decentralization. In China the local government system provides four levels: provincial level, which is the highest local level and includes provincial, autonomous regions and municipal governments; city level, which includes cities under the jurisdiction of the provinces, autonomous prefectures and districts under the jurisdiction of municipal governments; county level, which includes autonomous counties, counties and towns; township level, which is the lowest level and includes towns and villages. Intergovernmental fiscal relations were revamped by the 1994 reform that established a new tax-sharing system that fundamentally changed the apportionment of tax revenue between the central and provincial governments. The local financial revenue mainly derives from local taxes (such as business tax, personal income tax, tax on the use of

64

Angela Fraschini

urban land, tax on real estates, tax on agriculture and special agriculture products, etc.), shared taxes (value added tax, stamp tax and tax on resources other than the ocean petroleum resources) and non-tax revenue (fees, penalties, subsidies, other income, etc.). Notwithstanding the reforms, the fiscal system is still unitary. Nevertheless, local government is increasingly playing a part in local economic development and some local governments are beginning to exercise influence on central government. In India, the federal system is quite complex, as a consequence of regional disparities and of both vertical and horizontal fiscal imbalances. The center–states relations are envisaged in the Constitution also for financial aspects and the assignment of tax power is based on the principle of separation. Local governments have different institutional arrangements in rural and urban area, in accordance with the two 1992 constitutional amendments that made India one of the most politically decentralized countries among developing ones. However, in practice the implementation of the decentralization program is still lagging, especially in rural areas. Major taxes levied by urban local bodies, which have greater tax power than the rural ones, are tax on property including service levy for water supply, conservancy, drainage, lighting and garbage disposal; tax on entry of goods into a local area for consumption, use or sale therein, known as octroi; tax on professions; tax on vehicles (other than motor vehicles). Finally, in the last section we conclude by comparing fiscal federalism arrangements prevailing in these vast and still developing countries with the rules that should be followed to implement fiscal decentralization.

A short synopsis of administrative divisions and taxes by level of government Briefly, the governmental systems that characterize the different countries of our sample are quite different: China is a communist state that is increasingly opening to areas of free-market; India is a federal democratic republic; Japan is a constitutional monarchy with a parliamentary government; Malaysia is a federation of 13 states1 and it is a constitutional monarchy; South Korea is a democratic republic; and Thailand is a constitutional monarchy. Nevertheless, all these countries have different levels of government and administrative divisions, as shown in Table 3.1.

China and India: a comparison China and India are the world’s two largest nations2 and, from an historical point of view, there are a number of similarities between the two countries: ancient civilizations that were, at one time, the richest in the world, declining in the second half of the second millennium and starting their way to modernity in the middle of the last century. Moreover, many general similarities existed between the economies of China and India at

28 states, 7 union territories (Andaman and Nicobar Islands, Andhra Pradesh, Arunachal Pradesh, Assam, Bihar, Chandigarh, Chhattisgarh, Dadra and Nagar Haveli, Daman and Diu, Delhi, Goa, Gujarat, Haryana, Himachal Pradesh, Jammu and Kashmir, Jharkhand, Karnataka, Kerala, Lakshadweep, Madhya Pradesh, Maharashtra, Manipur, Meghalaya, Mizoram, Nagaland, Orissa, Pondicherry, Punjab, Rajasthan, Sikkim, Tamil Nadu, Tripura, Uttaranchal, Uttar Pradesh, West Bengal); and local governments (Nagar Panchayat, municipal councils, municipal corporations)

47 prefectures (Aichi, Akita, Aomori, Chiba, Ehime, Fukui, Fukuoka, Fukushima, Gifu, Gumma, Hiroshima, Hokkaido, Hyogo, Ibaraki, Ishikawa, Iwate, Kagawa, Kagoshima, Kanagawa, Kochi, Kumamoto, Kyoto, Mie, Miyagi, Miyazaki, Nagano, Nagasaki, Nara, Niigata, Oita, Okayama, Okinawa, Osaka, Saga, Saitama, Shiga, Shimane, Shizuoka, Tochigi, Tokushima, Tokyo, Tottori, Toyama, Wakayama, Yamagata, Yamaguchi, Yamanashi) subdivided into municipalities (cities, towns and villages)

13 states (Johor, Kedah, Kelantan, Melaka, Negeri Sembilan, Pahang, Perak, Perlis, Pulau Pinang, Sabah, Sarawak, Selangor, and Terengganu), one federal territory with three component (city of Kuala Lumpur, Labuan, and Putrajaya), and local governments (municipal councils and district councils)

9 provinces (Cheju-do, North Cholla, South Choll), North Ch’ungch’ong, South Ch’ungch’ong, Kangwon-do, Kyonggi-do, North Kyongsang, South Kyongsang), 7 metropolitan cities (Incho, Kwangj, Pusan, Seoul, Taegu, Taejon, Ulsan), and 232 lower-level governments (cities, counties and districts)

76 provinces (Amnat Charoen, Ang Thong, Buriram, Chachoengsao, Chai Nat, Chaiyaphum, Chanthaburi, Chiang Mai, Chiang Rai, Chon Buri, Chumphon, Kalasin, Kamphaeng Phet, Kanchanaburi, Khon Kaen, Krabi, Bangkok, Lampang, Lamphun, Loei, Lop Buri, Mae Hong Son, Maha Sarakham, Mukdahan, Nakhon Nayok, Nakhon Pathom, Nakhon Phanom, Nakhon Ratchasima, Nakhon Sawan, Nakhon Si Thammarat, Nan, Narathiwat, Nong Bua Lamphu, Nong Khai, Nonthaburi, Pathum Thani, Pattani, Phangnga, Phatthalung, Phayao, Phetchabun, Phetchaburi, Phichit, Phitsanulok, Phra Nakhon Si Ayutthaya, Phrae, Phuket, Prachin Buri, Prachuap Khiri Khan, Ranong, Ratchaburi, Rayong, Roi Et, Sa Kaeo, Sakon Nakhon, Samut Prakan, Samut Sakhon, Samut Songkhram, Sara Buri, Satun, Sing Buri, Sisaket, Songkhla, Sukhothai, Suphan Buri, Surat Thani, Surin, Tak, Trang, Trat, Ubon Ratchathani, Udon Thani, Uthai Thani, Uttaradit, Yala, Yasothon), districts, sub-districts and local governmentsb

India

Japan

Malaysia

South Korea

Thailand

Notes: a China considers Taiwan its 23rd province. Hong Kong and Macau are special administrative regions. b Urban-based forms of local government include: the Bangkok Metropolitan Administration (BMA); the Municipality, governing urban centers in the provinces; and the City of Pattaya. Rural-based forms of local government include: the Provincial Administrative Organization (PAO) constituting local government at a provincial level; the Tambon Administrative Organization (TAO), constituting local government at a sub-district level; and The Sukhapiban or Sanitary Committee, a local government in a rural center.

Source: Our adaptation from CIA, The World Factbook.

22 provincesa (Anhui, Fujian, Gansu, Guangdong, Guizhou, Hainan, Hebei, Heilongjiang, Henan, Hubei, Hunan, Jiangsu, Jiangxi, Jilin, Liaoning, Qinghai, Shaanxi, Shandong, Shanxi, Sichuan, Yunnan, Zhejiang); 5 autonomous regions (Guangxi, Nei Mongol, Ningxia, Xinjiang, Xizang [Tibet]); 4 municipalities (Beijing, Chongqing, Shanghai, Tianjin); and local governments (prefectures, districts, counties, cities, towns and villages)

China

Table 3.1 Government layers by countries

66 Angela Fraschini the time the Communists assumed power in China in 1949 and India achieved its independence in 1947. Until the 1980s the economic performance of China and India was not much different (also per capita GDP was similar) and both the countries experienced economic reforms that led to a growth acceleration. In particular, during the period 1952–80 the two countries grew at about the same GDP rate because the slow growth of the Chinese agriculture and service sectors smoothed its fastgrowing industrial sector. Under the leadership of Deng Xiaoping, in 1980 China initiated economic reforms, a decade earlier than India, and as a consequence its economy grew at double the rate of growth of India during the 1980s and the early 1990s. Thus for several years there has been a growing gap between the performance of the two giant countries and in 2003 the per capita GDP3 was estimated to be US$5,000 in China and US$2,900 in India (see CIA, The World Factbook). This result is not only imputable to a faster GDP growth, but also to a lower population increase, thanks to the onechild policy implemented in China. For a short comparison of selected items see Table 3.2. Moreover, the two countries have a different government type, though both have a multi-level government. China is a “socialist state under the people’s democratic dictatorship led by the working class and based on the alliance of workers and peasants” (art. 1 of the Chinese Constitution). According to article 30 of the Chinese Constitution, the administrative division of the People’s Republic of China is as follows: (i) the country is divided into provinces, autonomous regions and municipalities directly under the central government (these latter and other large cities are divided into districts and counties); (ii) provinces and autonomous regions are divided into autonomous prefectures,4 counties, autonomous counties and cities; and (iii) counties and autonomous counties are divided into townships, nationality townships and towns. The state may establish special administrative regions when necessary. As shown in Table 3.1, currently China has 22 provinces (considering Taiwan as the 22nd), five autonomous regions and four municipalities, while Macau and Hong Kong are special administrative regions.5 Therefore, the modern Chinese system6 includes a share of authority between the central and local governments, providing a partial basis for a special kind of federalism called “market-preserving federalism”7 (Weingast 1995). As regards India, we have already noted that it is a federal republic and its government consists of a central (union) government, 28 state governments and seven union territories. Many states have autonomous regions with regional councils and in different states there are three tiers of local bodies. There also are 602 districts administered by their respective state/UT government. In the following sub-sections we briefly outline the main features of intergovernmental fiscal relations in both the countries.

1990

1995

0.8 1.8

0.8 1.6

145,730.0 99,098.0 67.9 59.6

151,830.0 98,313.0 68.6 60.1

856 453

63.8 71.6

69.1 72.0

788 437

1,267.4 1,017.0

2000

1,259.1 998.0

1999

67.2 59.2

170,110.0 97,320.0

911 465

0.7 1.2

62.3 72.2b

1,276.3 1,027.0b

2001

66.4 –

168,538.0 –

966 471

0.7 2.4

– 71.0

1,284.9 1,052.0

2002

Notes a End of year. b Census figures as at 1 March 2001. Data for other years are from the United Nations Population Prospects, the 2002 Revision. c For 1985. d Data refer to labor force. e Gross domestic product per capita in US$ are derived by converting the national currency into US$ based on the average official exchange rates for the period. The official exchange rate might differ significantly from market exchange rates. f Data refer to total debt stocks, defined as the sum of public and publicly guaranteed long-term debt, private non-guaranteed long-term debt, the use of International Monetary Fund credit and short-term debt.

Source: Unescap Statistics Division. Database updated on 11 March 2004.

Total population (million) Chinaa 987.1 1,143.3 1,211.2 India 673.0 835.1 931.0 Rural population (% of total) China 80.6 73.6 71.0 India 76.9 72.8 73.0 Population growth rate (% per annum) China 1.2 1.4 1.1 India 2.1 2.1 2.0 GDP per capita (US$)e China 206 312 581 India 257 373 376 External debt (millions US$)f China 15,828.0c 52,545.0 106,590.0 India 20,581.0 83,628.0 94,464.0 Population employed in agriculture (% of population employed) China 75.2 73.7 72.1 Indiad 69.7 64.0 61.9

1980

Table 3.2 Comparison of selected items

68

Angela Fraschini

Intergovernmental fiscal relations in China8 China has a five-tiered administrative structure: apart from the central government, there are 31 provincial level units (22 provinces, five autonomous regions and four municipalities directly under the central government); 331 prefectures and municipalities at the prefectural level; 2,109 counties, autonomous counties and cities at the county level; 44,741 townships, towns and city districts. Intergovernmental fiscal relations have undergone substantial changes in the last 25 years (Bahl and Wallich 1992; Arora and Norregaard 1997; World Bank 2002). Until the beginning of the reform era and as a consequence of central planning (extending approximately from 1957 to 1979) public finances in China were rather centralized in that all taxes and profits accrued to the central government, which then transferred to local governments the funds they needed to meet the spending priorities set by the central government itself. Local governments were responsible for collecting all taxes – acting simply as agents of the center – and their spending autonomy was restricted to minor amounts covered with extra-budgetary funds. A revenue-sharing mechanism was introduced in 1980, partly with a view to providing local governments with incentives to improve tax collection. To this end revenues from each tax were classified as “central fixed revenue,” “local fixed revenue” or “shared revenue” (the shares were determined through negotiations between the central and provincial governments). It should be noted that this arrangement only involved the center and the provinces, which were left free to decide on revenue assignments at lower levels as they pleased, in line with the nature of the system as a “nested hierarchy.” To enable poorer provinces to cope with their expenditure needs, the system was revised in 1985 so that provinces where local fixed revenue exceeded local expenditures were required to remit part of that revenue to the center, while provinces where local expenditures could not be covered by local fixed revenue were granted a higher proportion of shared revenue or, in the event that even all of it was insufficient to break even, were given a transfer from the central government. A final change before the 1994 reform was made in 1988 with the establishment of the so-called “fiscal contracting system” (also known as the “fiscal responsibility system”), under which provincial governments agreed to transfer a fixed tax quota (i.e. a lump-sum remittance or subsidy, to increase annually at a single-digit rate) to the central government, while retaining all revenue in excess of it. The different terms of the revenue-sharing contracts negotiated by each province resulted in growing disparities among provinces and increased the elements of bargaining present in intergovernmental relations. In the words of Ahmad et al. (2002):

Fiscal federalism in China and India 69 the new system also created a strong incentive for local governments to conceal information about local revenue from the center, else they would face a “ratchet effect,” as this information would be valuable at the time the fiscal contracts were renegotiated. Furthermore, many of the new enterprises in the rapidly expanding township and village enterprise sector were joint ventures with local government ownership. With retained profits accruing to the benefit of “local shareholders,” there was a continued incentive to shift deficits to the center and hide profits from taxation. One of the main objectives of the 1994 reform was therefore to revamp intergovernmental fiscal relations. Consequently a new tax-sharing system (TSS) was established which fundamentally changed the apportionment of tax revenue between the central and provincial governments. Taxes were classified into three categories: central fixed taxes, local fixed taxes and shared taxes. Tax revenues assigned to the central government included mainly those from: customs duties, VAT and excise taxes on imports; consumption tax; income tax on all centrally owned state enterprises; taxes collected from the Ministry of Railroads and from the headquarters of banks and insurance companies; income tax, turnover taxes and resource tax from offshore oil activities; and enterprise income tax collected from banks and other financial institutions. Local governments were assigned revenues from: personal income tax; income tax on locally owned state enterprises, collectives and private firms; urban and rural land use tax; farmland occupation tax; land appreciation tax; house property tax; urban real estate tax; vehicle and vessel use tax; deed tax; agricultural and animal husbandry taxes; slaughter tax; and entertainment and banquet taxes. Shared taxes included: VAT (75 percent central, 25 percent provincial); the securities exchange tax (88 percent of the revenues from stock transactions to the central government; all the rest to provincial governments); and the natural resource tax (almost entirely local). To implement the new revenue assignment the tax administration was split: the bureaus of the state administration of taxation were charged with collecting all central and shared taxes, while local taxes were left to the responsibility of local tax bureaus. Despite the introduction of the TSS provincial governments were not given any significant degree of tax autonomy, since they can only set the rates of a few minor taxes, while every other revenue decision is to be taken in Beijing. The reform of intergovernmental relations was completed by redesigning the system of transfers, with a view to introducing a more rule-based mechanism. Now there are four types of grants in China: the “fixed subsidies under the old system” serve to provide local governments with the same (nominal) level of revenue as in 1993;9 the “revenue returned” is intended to allow provinces to share in the increase in the revenue from VAT and consumption tax; the “specific-purpose grants” (or earmarked

70 Angela Fraschini transfers) are administered by individual ministries and have a regulatory function, forcing local governments (which are also required to match the grants received with local funds) to comply with policy priorities set by the center; finally, the “transitional transfers,” introduced on a pilot basis in 1995, are designed to equalize fiscal resources across provinces. In view of the unique redistributive purpose of the latter, we look at them in more detail. The formula for computing the amount to be transferred to each province is made up of three components: the first one, objective in nature, is intended to measure the gap between “standard expenditures” and local fiscal capacity; the second one has policy motivations, tending to favor regions with large ethnic minority population; and the third one, added in 1996, should reward provincial tax effort (World Bank 2002). The respective weights of each of these types of transfers in the five-year period 1997–2001 are shown in Table 3.3. Fixed subsidies are a minor component of total transfers and their share halved from almost 4 percent in 1997 to 2 percent in 2001. Earmarked transfers amounted to more than one-third of total transfers in 2001 and they have nearly doubled in relative terms during the period involved; their increasing importance is the result of the proactive regional policy of the central government in recent years and of the necessity to respond to particular emergencies (the Asian financial crisis, the inadequacy of resources for local spending on social protection, the rise in pension benefits); however, in the absence of effective monitoring mechanisms to control how these funds are really used, they may be diverted by local governments to their own priorities. Transfers based on the “revenue returned” mechanism, though continuously declining, still represented almost 40 percent of total transfers in 2001 and, in view of the regressive nature of the formula for determining their amount (which favors the wealthier coastal provinces), their predominance is an enduring obstacle to the equalization of fiscal resources. Finally, general purpose grants (a composite item including the transitional period transfers) more than tripled their share in total transfers (from 7.5 percent in 1997 to 24.5 percent in 2001). However, during the transitional period, transfers are still a minor component, by accounting for around one-tenth of the total in 2000. An insight into the equalization properties of the actual transfer system in China is given by the simulations presented in Ahmad et al. (2004). First of all, the authors calculate expenditure needs for each province based on 2000 data: they group expenditures into seven categories, determine the share of each in total expenditure and apply specific weights for each category that should reflect factors likely to affect public services costs in each province (total population, degree of urbanization, presence of ethnic minorities, age structure of population). They then construct an indirect measure of each province’s “revenue capacity:” to this end they use each province’s GDP as a proxy of the tax base of the province and

0.8

0.8

Source: Ahmad et al. (2004).

100.0 62.7 3.4 7.5 26.4

100.0 70.5 3.9 7.5 18.1

Transfers from the central government to local governments Revenue returned Fixed subsidy under the old system General purpose grants Earmarked transfers Transfers from local governments to the central government Fixed subsidy under the old system

4.2 2.7 0.1 0.3 1.1

1998

3.8 2.7 0.2 0.3 0.7

1997

Transfers from the central government to local governments Revenue returned Fixed subsidy under the old system General purpose grants Earmarked transfers

Table 3.3 Intergovernmental transfers, 1997–2001

0.7

100.0 48.9 2.7 17.9 30.6

(% of total transfers) 100.0 53.0 2.9 10.2 33.8 0.7

5.2 2.6 0.1 0.9 1.6

2000

(% of GDP) 5.0 2.6 0.1 0.5 1.7

1999

0.6

100.0 38.9 2.0 24.5 34.6

5.8 2.3 0.1 1.4 2.0

2001

72 Angela Fraschini multiply it by a coefficient obtained by regressing provinces’ own revenue (i.e. revenue before transfers) against their GDP.10 The next step is subtracting the standard expenditure needs of each province from its revenue capacity to get the shortfall or excess of resources of each province. Finally, the authors consider three scenarios: in the first one, the amount of total transfers from the central government to provinces in 2000 is used for equalization purposes (provinces with a positive balance get nothing, while provinces with a “deficit” receive a grant proportional to their shortfall) and every other type of transfer is cancelled; in the other two scenarios, equalization is carried out only partially (absorbing respectively only 20 percent and 60 percent of total transfers), while the remaining amount is attributed to provinces according to the transfer pattern in 2000. By regressing per capita transfers against per capita GDP in each province, the authors find that the actual system has no equalizing effect, which is absent or insignificant also in the two hypotheses of partial equalization, while a positive and significant equalizing effect is present only in the first scenario. Intergovernmental fiscal relations in India As we have already noticed, India has a federal structure with peculiar features. The two essential features of Indian federalism are: (i) federalism is not the result of an agreement by the units; and (ii) the component units have no freedom to secede.11 Historical factors, mainly the colonial system, have played a strong role in making the Indian federal system quite centralized. Indian Constitution makers divided the government functions into three lists: federal, state and concurrent. Under the Seventh Schedule of the Indian Constitution, the central government has exclusive powers on foreign policy, defence, communications, currency, taxation on corporations and non-agricultural income, and railroads; while state governments have the exclusive power to legislate on such subjects as law and order, public health and sanitation, local government, betting and gambling, and taxation on agricultural income, entertainment and alcoholic beverages. On some issues both the central government and state governments may legislate, though a union law generally dominates a state one. Among these areas are criminal law, marriage and divorce, contracts, economic and social planning, population control and family planning, trade unions, social security and education. All residuary issues lie within the exclusive domain of the central government. An important power of the central government is that of creating new states, by combining states, changing their boundaries and terminating a state’s existence. The central government may also create and dissolve any of the union territories, which have more limited powers than those of the states. Although the states exercise either exclusive or joint control over a substantial range of issues, the Constitution establishes a more dominant role for the union government.

Fiscal federalism in China and India 73 The assignment of tax power is based on the principle of separation; most broad-based taxes are assigned to the center, whereas in practice the states have a narrower tax base12 and the consequence is a vertical fiscal imbalance. In 2002–3 the states, on average, raised about 38 percent of central revenues, but incurred about 58 percent of expenditures. The capacity of the states to finance their current expenditures from their own sources of revenues has declined from 69 percent in 1955–6 to 52 percent in 2002–3. Transfers from the center made up the balance (Singh 2004). The inadequacy of the states to meet expenditures from their own resources is recognized by the Constitution of India (articles 275 and 282) that provides principles for the sharing of resources between the center and the states, without specifying the revenue shares but providing for a finance commission, which is appointed by the President of India every five years, or earlier if needed.13 In other words, the finance commission is the body provided by the Constitution to regulate the flow of transfers from the central government to the states and their allocation among different states. Generally, the finance commission makes recommendations on the following matters: 1

2

3

The distribution between the union and the states of the net proceeds of taxes that are to be divided between them under chapter I part XII of the Constitution14 and the allocation among the states of the respective shares of such proceeds.15 The principles that should govern the grants-in-aid of the revenues of the states out of the consolidated fund of India16 and the sums to be paid to the states which are in need of assistance by way of grants-inaid of their revenues under article 275 of the Constitution.17 The measures needed to augment the consolidated fund of a state to supplement the resources of the Panchayats and municipalities in the state on the basis of the recommendations made by the finance commission of the state.18

Moreover, the commission reviews the financial situation of the union and the states and suggests a plan by which the governments, collectively and severally, may bring about a restructuring of the public finances to restore budgetary balance and to achieve macro-economic stability and debt reduction along with equitable growth. Over the last 50 years the finance commissions have elaborated a sophisticated methodology to deal with horizontal and vertical fiscal imbalances. To distribute horizontally the major taxes that are shared between the center and the states19 the finance commissions used a large number of criteria, among which: population, tax effort, collection assessment, income distance, income adjusted total population, indices of social and economic backwardness, territorial area, post-devolution deficits,

74 Angela Fraschini poverty and revenue equalization (Singh 2003). The Eleventh Finance Commission (2000–5) set a new benchmark in the center-state fiscal relations: it reduced weight of population from 20–30 percent in the recent past to 10 percent, maintained weight of income distance criterion at 62.5 percent and chose to allocate the remaining percent of the state share of pooled proceeds according to area (7.5), infrastructure (7.5), tax effort (5) and fiscal discipline (7.5). Besides the devolution of share in central taxes and duties, the central government gives the states grants-in-aid20 to cover their revenue deficit.21 In addition, specific grants to states are provided for their special problems and for upgrading administration standards in a number of sectors.22 Grants for local bodies are also provided.23 Moreover, the central government distributes substantial grants to the states through its development plans as elaborated by the planning commission.24 While the finance commission decides on tax shares and makes grants-in-aid, the planning commission allocates resources in accordance with the foreseen priorities, making grants and loans to implement development plans. It is worth noticing the problem of coordination between the two independent commissions that arises. The loan–grant composition of the assistance given to special category states is 10:90 while that to other states is 70:30 (the amount allocated to any recipient state includes grant and loan in the above proportion, and the state cannot accept the grant without accepting the loan). Before 1969, plan transfers were project-based; since then, the distribution has been done on the basis of a formula that takes into account population, per capita income, fiscal performance (tax effort, fiscal management, national objectives) and special problems (Singh 2004 and Ma 1997). Plan revenue grants make about 7–8 percent of the total revenue receipts of the union (Chaubey 2003). Summing up, the Indian intergovernmental transfer system consists of three elements: (i) a general purpose grants mechanism, based on a revenue-sharing scheme (at present general tax sharing), operated by the finance commission to assist the backward areas (equalization transfers, formula-based); (ii) formula-based unconditional transfers and specific purpose transfers operated by the planning commission to implement development plans; and (iii) specific purpose transfers with or without matching requirements. This system of intergovernmental fiscal relations, characterized by transfer dependence and soft budget constraint, has created adverse incentives for prudent fiscal behavior by the state sector. A recent study (Purfield 2004) confirms that transfer dependency, coupled with bailout expectations, contributes to the growth in states’ deficits. And, it is worth noting, high fiscal deficit of the states represents an important obstacle to the fiscal decentralization process begun in the 1990s. This process has been forced by economic and political events, such as liberalization and

Fiscal federalism in China and India 75 globalization on one hand, and the end of single party rule, with the emergence of coalition parties in power at the center and the increasing importance of regional parties in the political affairs of the country, on the other hand (Rao 2004). At present there are two types of local governments: urban local governments and rural local governments. In fact, in 1992 to bring to effect the process of decentralization – that is, the transfer of administrative, fiscal and political responsibilities to locally elected bodies25 – the government of India introduced two constitutional amendments: the Seventythird, for rural decentralization, and the Seventy-fourth, for urban decentralization. These amendments, which established the political decentralization, leaving the implementation of administrative and fiscal aspects to the states, provided for a uniform structure of Nagar Panchayats26 for areas in transition from a rural area to an urban area, municipal councils for smaller urban areas and municipal corporations for larger urban areas.27 Rural local governments operate through district (Zilla) Panchayats, intermediate28 (Taluka) Panchayats and village (Gram) Panchayats29 (see Table 3.4). Moreover, the amendments granted local self-governments a constitutional status and safeguarded their continued existence. Rules and institutions are different between the two types of local governments and generally the fiscal power is higher for urban than for rural governments, as shown in Table 3.5. As is recognized, an important indicator of fiscal autonomy is the share of the revenue expenditure covered with own resources and also the percentage of own resources on total resources. This is necessary to establish how autonomous or dependent local bodies are on external sources. Table 3.5 shows that the percent share of own resources in total revenue of the rural local bodies was only 6.72 in 1998–9, declining to 5.99 in 1999–2000 and increasing in the following years, reaching 6.85 in 2002–3. This implies that more than 93 percent of their total revenues were derived from external sources. On the other hand, the urban local bodies raised 59.69 percent of their total revenues from own resources in 1998–9, but this percentage declined to 58.44 in 2002–3. Also the percentage of revenue expenditure covered by own resources is very Table 3.4 Number of local bodies at different tiers Rural local bodies 1 2 3 4

Gram/village Panchayats Panchayats Samities Zilla Panchayats Autonomous district councils

Total Source: Government of India (2004).

Urban local bodies 236,350 6,795 531 9

1 Municipal corporations 2 Municipalities 3 Nagar Panchayats

109 1,432 2,182

243,685

Total

3,723

76 Angela Fraschini Table 3.5 Revenue and expenditure of local bodies (rural and urban) in % of total 1998–9

1999–2000

2000–1

2001–2

A) All India Revenue and expenditure of Panchayati Raj institutions (all tiers) Revenue Own tax 3.64 3.04 3.24 3.61 Own non-tax 3.08 2.95 2.86 2.77 Assignment ⫹ devolution 30.19 29.23 28.10 27.46 Grants-in-aid 56.34 58.92 57.76 58.85 Others 6.75 5.85 8.04 7.31 Total revenue 100.00 100.00 100.00 100.00 Expenditure Revenue expenditure 71.18 73.25 75.36 75.92 Capital expenditure 28.82 26.75 24.64 24.08 Total expenditure 100.00 100.00 100.00 100.00 Own revenue as % of 8.91 8.40 7.82 7.88 revenue expenditure B) All India Revenue and expenditure of urban local bodies (all levels) Revenue Own tax 41.30 39.10 38.53 38.85 Own non-tax 18.39 16.92 18.12 18.97 Assignment ⫹ devolution 19.18 20.09 20.45 18.12 Grants-in-aid 15.70 17.09 15.36 17.64 Others 5.43 6.80 7.54 6.42 Total revenue 100.00 100.00 100.00 100.00 Expenditure Revenue expenditure 75.28 73.97 74.10 76.69 Capital expenditure 24.72 26.03 25.90 23.31 Total expenditure 100.00 100.00 100.00 100.00 Own revenue as % of 75.87 69.03 70.81 71.78 revenue expenditure

2002–3

3.87 2.98 27.69 58.95 6.51 100.00 73.05 26.95 100.00 9.26

39.23 19.21 17.69 16.48 7.39 100.00 76.24 23.76 100.00 68.97

Source: Own calculations based on the report of the 12th finance commission (2005–10).

different for rural and urban local bodies (9.26 and 68.97 percent, respectively, in 2002–3). Again, the percentage of revenue derived from own taxes is much lower for rural local bodies than for urban local bodies (3.87 percent against 39.23 percent in 2002–3). At present the revenue of the rural local bodies is principally constituted by grants and the dependence upon the state government even for carrying out routine functions is quite heavy. Among the three tiers of Panchayats, the Gram Panchayats are comparatively in a better position because they have some taxing power of their own, while the other two tiers are dependent only on tolls, fees and nontax revenue for generating internal resources. Relative to the municipalities, even after the Seventy-fourth Amendment, the Constitution does not provide them with an autonomous domain of tax-raising power, which continues to be decided and regulated by the state governments that

Fiscal federalism in China and India 77 specify the taxes (taken from the state list in the Seventh Schedule)30 that the municipalities can levy and collect. Historically these taxes have comprised taxes on lands and buildings, on entry of goods into a local area for consumption, on animals and boats, on entertainment, on professions, trades, etc. There is an important variability among the states, but a uniform feature is represented by the significant control of the state governments in determining the tax, tax rates or even tax exemptions, since there is no distinct tax domain of the municipalities as such. Therefore, as recommended by the National Commission to Review the Working of the Constitution (Venkatachaliah Commission), a distinct and separate tax domain for municipalities should be recognized (Government of India, Ministry of Law 2002). Local government bodies are covered in the state list31 and are governed by the state statutes or, in the case of union territories, by the union parliament. Notwithstanding the above quoted amendments to the Constitution that made India one of the most politically decentralized countries among developing ones, in practice the implementation of the decentralization program is still lagging, especially in rural areas. In fact, while political decentralization has progressed satisfactorily – states modified their acts consistently with the requirements of the amendments, and most of them have carried out local elections – administrative and fiscal decentralization are taking place at a much more hesitant pace, also owing to the reluctance of some state governments to share their fiscal powers with the local self-government institutions. As in every decentralized system, in India local bodies represent the nearest government to the people, charged with the responsibility of providing most of the basic services. The core services that would be granted by local bodies are identified as primary education, primary health, rural or municipal roads, drinking water supply, sanitation and street-lighting. But few states have vested the local bodies with the necessary powers, funds and staff to enable them to perform the functions assigned to them under the statutes. It is obvious that the three “F”s (functions, functionaries and finances) have to go together for any process of devolution to be meaningful. Therefore it is clear that the failure to assign human resources affects the growth of Panchayats and municipalities as self-governing institutions. As made evident by the Venkatachaliah Commission: in the process of implementation of the Seventy-third and Seventyfourth Amendments, considerable gaps have been noticed. The union government and the state governments continue to exercise powers in planning and the Panchayats and municipalities do not enjoy autonomy – financial or administrative – as institutions of local self-government. While today Panchayats elect some 3 million members of whom one-third are women, the objectives envisaged in the amendments have not been fully achieved (. . .) Even in the states which have

78 Angela Fraschini shown political will to decentralize, devolution has not gone beyond entrusting to them responsibility for implementation of the schemes/ projects conceived by the state or union government. As a result, Panchayats have not blossomed into institutions of self-government. Instead they have been reduced to an implementing arm of the state government. (Government of India, Ministry of Law 2002) In several states all functions provided for devolution to local bodies, as envisaged in the constitutional amendments, have not been transferred. For example, out of 29 subjects, the government of Andhra Pradesh has transferred only 17 to rural local bodies and most of these transfers are partial, without the transfer of funds and functionaries. Functions like primary education, primary health care and drinking water supply have not been devolved to local bodies in rural areas, but are being looked after and operated by the line departments of the state government or by special boards/agencies. In many cases, even when a function is transferred to local bodies, the state governments do not vacate their operations from such areas, with the result that all local functions, in effect, have become concurrent. In other words, most of the states, despite transferring a number of functions, have not wanted to give up their involvement in such matters and continue to maintain a large staff at the state and district headquarters. Moreover, there is no specific accountability, in the sense that the role of the three tiers of local bodies has not been clearly defined in the state legislations. In fact, very few states have translated all the subjects into activities and specific functions for the three tiers of local bodies. Most states have listed all the functions as equally relevant for all the three tiers of local bodies. It is evident that in doing so there is a breach of the principle of subsidiarity, which establishes, as well known, that whatever can be done at one level should be done at that level and not at a higher one. In addition, in some states only Gram Panchayats have been entrusted with all the functions included in the Eleventh Schedule,32 while the intermediate level and the district level Panchayats have not been assigned any taxation powers and any functions except the supervisory ones (Mishra n.d). Another important function assigned to local bodies under the constitutional provisions is that of planning for economic and social development in their respective areas, but this function is not being performed in most of the states. This can be considered a violation of the spirit of the Constitution, though it is difficult to imagine an active role of local governments as to this function in a country where deprivation is endemic and regional disparities are growing (on regional disparities see, for example, Buddhadeb and Prabir 2005). In many states one of the main reasons for the limited devolution of functions and fiscal decision-making power is commonly imputed to the lack of reliable information on the spending and revenue of the local

Fiscal federalism in China and India 79 bodies33 that prevents the center and the states from implementing a serious empowerment of local governments. Moreover, the lack of accountability of local bodies, because of inadequate provisions in law relating to audit of accounts, does not support decentralization. In addition, the high fiscal deficit of the states represents an important obstacle to the fiscal decentralization process (World Bank 2004). To achieve the constitutional goal of making local bodies into units of self-governance, the Eleventh and the Twelfth Finance Commissions have had the mandate in their terms of reference to recommend measures needed to augment the consolidated fund of the states to supplement the resources of local bodies. In the view of the Eleventh Finance Commission, the states may take the following measures for augmenting their consolidated funds to supplement the resources of Panchayats and municipalities: 1

2

3

Land taxes: taxes on land/farm income in some form may be levied to strengthen the resource base of the local bodies. The amounts so collected may be passed on to the local bodies for improving and strengthening the civic services. Local bodies may also be involved in collection of these taxes. Surcharge/cess on state taxes: “cess” on land-based taxes and other state taxes/duties may be levied to mobilize resources for augmenting specific civic services and for improving their quality. For example, a cess or surcharge of 10 percent on sales tax, state excise, entertainment tax, stamp duties, agricultural income tax, motor vehicles tax, electricity duties etc. may give significant additional revenue, which could be devolved to the local bodies for improving the basic civic services and for taking up schemes of social and economic development. Profession tax: article 276 of the Constitution provides for levy of a tax on professions, trades, callings or employment for the benefit of the state or local bodies at a rate not exceeding RS2,500 per taxpayer per year. Many states either do not levy this tax or levy it at very low rates. States should levy this tax with a view to supplement the resources of local bodies or they should empower the local bodies to levy it.

Notwithstanding these recommendations, not much seems to have been implemented so far, because the state governments pay more attention to the devolution part of the recommendations than to the recommendations related to fiscal aspects. Also the local bodies are more interested in the devolution package rather than the reforms suggested in the fiscal system owing to their being closer to the people who are to be taxed. Therefore, the fiscal effort made by local bodies is very poor; particularly Gram Panchayats do not make any stated efforts to levy taxes that they are empowered to levy and collect. Also the performance of urban local bodies, though better than that of Gram Panchayats, is not satisfactory, as shown in Table 3.5.

80 Angela Fraschini

Fiscal decentralization China versus India During the 1990s a common trend seen in China and India is the impulse towards fiscal decentralization. In the case of China, the 1994 fiscal reform attempted to recentralize the tax system and to reform the tax-sharing system to place intergovernmental transfers on a more systematic footing. The main goals of the reform were to simplify the tax system introducing a revenue-sharing system, to raise the revenue to GDP ratio, to raise the central government’s ratio to total revenue to increase equalization transfers, and to make the fiscal federal system more stable by shifting from negotiated transfers to a rule-based tax assignment. Only the first objective seems to have been achieved, while transfers have been inadequate and are not based on expenditure needs, and expenditure assignments have not been basically changed since 1994 (Ahmad et al. 2002). Since 1992, India has tried to develop a three-tier federal system, strengthening the third level of government through Constitutional amendments to transform local bodies to units of self-government. However, the push to decentralize below the state level has been stronger on the center side than on the states side, so the process of local government reform is still under way and the local governments play a very limited role both in raising revenues and in spending (Rao 2002). It seems, therefore, that the two countries are not committed to decentralization. As it is well known, the standard economic rationale for decentralization rests on efficiency, equity and macro-stability grounds.34 In developing countries decentralization is also seen as a means to achieve several goals that government interventions have failed to attain, such as the stimulation of economic growth, the reduction of poverty and the reinforcement of democracy, but mainly the improvement of service delivery to large populations.35 The theory of fiscal federalism set forth in Oates (1972), derived from the classic Musgrave model of public sector responsibility for stabilization, distribution and allocation (Musgrave 1959), provides rationale and instructions to assign these functions to different levels of government. Though initially the conventional fiscal decentralization theory has been applied in industrialized countries, it does not seem difficult to justify its application in developing and transition countries: while the justifications for centralization of the stabilization and distribution functions are relatively straightforward, the issues concerning assignment of responsibility for both the expenditure and revenue dimensions of the allocation functions are more complicated, due to the potential undermining of a number of assumptions underlying public finance theory in general, and fiscal federalism in particular, so that the theory has to be adapted.36 Among the conditions of a successful decentralization, apart from an adequate institutional design,37 the following can be mentioned (Bahl 1999):

Fiscal federalism in China and India 81 •

•

•

•

•

•

•

•

•

•

•

•

Fiscal decentralization should be viewed as a comprehensive system, characterized by some key elements – such as elected local councils, locally appointed chief officers, significant local government discretion to raise revenue, significant local government expenditure responsibilities, budget autonomy, hard budget constraint, accountability, transparency, borrowing powers, freedom from expenditure mandates, unconditional transfers from higher levels of government. Finance follows functions, in the sense that first should come the assignment of expenditure responsibility to local governments, and then the assignment of revenue responsibility should be determined. There must be a strong central ability to monitor and evaluate decentralization, which implies the preparation of a fiscal analysis unit and an extensive data system that allows quantitative monitoring and evaluation. One intergovernmental system does not fit the urban and the rural sector, since sub-national governments have different capabilities to deliver and finance services, and to borrow. Fiscal decentralization requires significant local government taxing powers, which improves the accountability of elected officials but implies the correct identification of the sub-national tax bases. Central government must keep the fiscal decentralization rules that it makes, guaranteeing the transfer of power with constitutional changes if necessary. Intergovernmental fiscal arrangements must be simple, which requires the local governments to devote fewer resources to administration and the central government to face lower monitoring and evaluating costs. The design of the intergovernmental transfer system should match the objectives of the decentralization reform, considering that different kinds of intergovernmental transfers have different types of impacts on local government finances. Fiscal decentralization should consider all levels of government, to allow citizen participation at a level that insures that voter preferences matter, and to result in accountability of government officials. Impose a hard budget constraint, which implies that local governments with fiscal autonomy must balance their budget without recourse to year-end assistance from the central government. Recognize that intergovernmental systems are always in transition and plan for this, which implies that central governments must have flexibility in their fiscal decentralization plans to adjust to changes, for example, in disparities among regions, quality in basic infrastructures and technical capacity of local governments. There must be a champion for fiscal decentralization, that is, potentially strong supporters that must be identified in the people and their elected representatives, the president, the parliament or congress,

82 Angela Fraschini urban local governments and external advisors, such as international organizations that provide encouragement and some technical assistance for fiscal decentralization. Trying to evaluate several of the above described guidelines with respect to the Chinese and Indian systems, we can observe that both the countries have not followed the advice that fiscal decentralization should be viewed as a comprehensive system. The four more relevant dimensions of intergovernmental fiscal relations are, as known, the assignment of expenditure responsibilities, the assignment of revenue sources, the provision of intergovernmental fiscal transfers and the rules governing subnational borrowing and debt. Until now India seems to have considered decentralization mainly in terms of the local election system, without the transfer of all functions provided for devolution to local bodies and the assignment to them of the fiscal decision-making power. In fact, in order not to lose power, most state governments have been and are still reluctant to give up either significant control over the expenditure budget or part of their tax bases. As underlined by the planning commission (2002–7) in its report on urban development, state governments continue to take decisions on such matters as rates of user charges, property tax coverage, levy or withdrawal of “octroi,” role of parastatals in water supply and sanitation services, with little reference to the urban local bodies that are affected by these decisions. Hence, these decisions do not always have the effect of strengthening the constitutional role of the elected local bodies that often are a subordinate entity under the day-to-day control of the state governments, beholden to the state for not only development of the cities but often even for survival (Government of India 2001). The situation is even worse for rural local bodies. With regard to China, the 1994 fiscal reform greatly changed the national revenue-sharing system, giving local governments more control over the administration of local taxes but no significant degree of tax autonomy and no substantial expenditure assignments. It is true that the central government is expanding the financial capacity of local authorities but, notwithstanding this increased capacity, their budgets still need approval from higher levels of government. Also the second rule (finance follows function) seems to be neglected in both the countries. Relative to the central ability to monitor and evaluate decentralization, in India there is no mechanism for collection of data on the revenue and expenditure of the various levels of local bodies at a centralized place. The National Development Council38 meets infrequently and is ineffective as a monitoring institution (Rao 2004). In China local governments are the administrative organs of the state under the leadership of the state council and monitoring should be easier because local autonomy is of subordinate autonomy, that is lower levels of government must complete tasks derived from higher levels of government and

Fiscal federalism in China and India 83 what local autonomy they exercise can only be within the guidance of those higher levels (Unescap b, n.d.). In fact, the center has relatively limited information on local government finance. Only India set up a different system of local bodies in rural and urban areas with different expenditure responsibilities and financing powers, which means to recognize explicitly the differences among local bodies in the capabilities to deliver and finance services. On the contrary, China has a unitary fiscal system; the establishment of the taxation separation system in 1994 has not been equivalent to the reform of that unitary fiscal system, which requires to hand down part of fiscal power to local governments. Currently, both the countries have not given local government significant taxing powers. In China local governments are based on a hierarchical system with a characteristic of leadership at different levels, the lower level being subordinate to the higher level, and this feature is recognized in the Constitution that, however, has not been changed to allow local governments more independent power on fund raising under the existing centralized fiscal system. Still, a range of factors contributes to the durability of decentralization, limiting the discretion of the central government to attempt a reversal.39 On the contrary, the Indian central government amended the Constitution to establish local bodies in rural and urban areas, but it has left in the hands of the states the implementation of the decentralization process that until now has not been resolutely pursued. It is worth remembering that historically India has had a strong centralized system, which may justify the delay in achieving effective federal governance. About the design of the intergovernmental transfer system, it has been recognized (Rao 2004) that in India it is necessary to redesign the transfer system to improve accountability, incentives and equity.40 The reform of the transfer system must begin with the avoidance of overlapping roles of the finance commission and the planning commission – preferably leaving as the responsibility of the finance commission all transfers, while the planning commission should focus on physical infrastructure – and must offset both vertical and horizontal fiscal imbalances. Also the consolidation of the more than 220 centrally sponsored schemes is an urgent need, even though they represent only 5 percent of transfers. With regard to China, the fiscal revenue-sharing schemes limit intergovernmental budget transfers. Nevertheless, the system of earmarked transfers seems to be too complex and undersized to fill the gap between the excessive assignment of core expenditure responsibilities to lower level governments (counties and townships) and their limited fiscal capacity. Moreover, there often is substantial diversion of earmarked funds at sub-national levels to meet short-term cash outlay requirements. Also equalization grants are undersized and have a very poor impact on addressing horizontal fiscal disparities across sub-national governments (World Bank 2001). The need for revision is clear.41

84

Angela Fraschini

Finally, the rule of hard budget constraint in China is faced by all levels of government and sub-national governments are prohibited from borrowing.42 On the contrary, in India in fact sub-national governments face soft budget constraint, mainly because of the vertical fiscal imbalance whose size also depends on the possibility of having ex ante budget deficit that creates an incentive to increase expenditure and to undermine financial discipline. The consequence is a bail-out by the higher tier government.43 Sub-national governments have some freedom to borrow but the greater share of their borrowing comes from the central government or public financial institutions. Despite the differences between the two countries, there are important issues of common interest (Rao 2001). The challenges of fiscal decentralization in transitional economies concern, among other things, the development of an efficient fiscal system, the replacement of a command and control system with market-based instruments, and the evolution of responsive intergovernmental fiscal relations.

Notes 1 Administratively, Malaysia is organized along a three-tier type of government: federal, state and local government. Local authorities are classified into two groups: municipality, for large towns, and district council, for small urban centers. In 1997 there were 26 municipal councils and 113 district councils (Unescap a, n.d.). 2 According to the estimated data of The World Factbook, on July 2004 the population of China was around 1,299 million and that of India was around 1,065 million. 3 GDP on a purchasing power parity basis divided by population as of 1 July for the same year. 4 Autonomous prefectures are divided into counties, autonomous counties and cities. 5 The special administrative region is the product of the conception of “one country, two systems,” which means that the Mainland of China carries out a socialist system and Hong Kong, Macao and Taiwan a capitalist system. But in international affairs, the People’s Republic of China is the only country representing China. 6 Currently the government structure in China is the following: the People’s Congress is the supreme organ of state power and its permanent organization is the standing committee that exercises legislative power. The local People’s Congresses at different levels are the state power organs at local level. The State Council is the supreme administrative organ of the state and the executive organ of the supreme organ of state power. People’s Courts at different levels are the judicial organs. The People’s Courts at local levels, Special People’s Courts and Supreme People’s Courts exercise judicial authority. The Supreme, Local and Special People’s Protectorates at local levels are the organs of law supervision of the state. Local governments are the administrative organs of state under leadership of the State Council and are divided into autonomous governments of nationality regions and governments of special administrative regions. The organizational system of local government is divided into provincial, city, county and village level (Unescap b, n.d.).

Fiscal federalism in China and India 85 7 It has been noticed that, to the extent that federalism has played a helpful role in promoting China’s economic growth, the competitive benefits of “marketpreserving federalism” depend very much on political centralization (Blanchard and Shleifer 2000). 8 This sub-section has been written by Domenico D’Amico. 9 It should be added that “fixed subsidies under the old system” include also transfers from local governments to the center. 10 As for expenditures, data refer to year 2000. 11 For a brief description of the evolution of financial relations from 1858 up to the coming into force of the Constitution in 1950 see Vithal and Sastry (2001). 12 See Chapter 7. 13 The last finance commission appointed is the Twelfth and its report must cover a period of five years commencing on the 1 April 2005. 14 Before the Eightieth Amendment Act, 2000, the Constitution provided for sharing of two taxes, income tax and union excise duties, with the states. The relevant ratios determining the vertical allocation in tax devolution have remained for many years at 85 percent in the case of income tax and at 45 percent for union excise duties. The Tenth Finance Commission proposed a system of vertical resource sharing in which central taxes are pooled and a proportion of 29 percent of gross proceeds devolved to the states (26 percent to all states and three percent to those where sales tax on sugar, textiles and tobacco was not levied). That recommendation brought forth an amendment to the Constitution (Eightieth Amendment Act, 2000). The Eleventh Finance Commission recommended the devolution of 29.5 percent (28 percent to all states and 1.5 percent to those which did not levy sales tax on sugar, textile and tobacco) of net proceeds of all shareable taxes (Government of India 2000). About 20 percent of the revenue collected by the union is transferred under the tax-sharing mechanism (Chaubey 2003). 15 For example, the Eighth and Ninth Commissions determined the respective shares of states in the devolution of income tax and union excise duties on the basis of three allocating criteria: (i) population; (ii) distance (measured by the term (yn ⫺ yi) where yn is the highest per capita income among all the states); and (iii) inverse of income. 16 The consolidated fund of India is a part of the government accounts to which are credited all revenues received by government by way of taxation and other receipts flowing to government in connection with the conduct of government business, like receipts from railways, post, transport, etc. (non-tax revenues). Similarly, all loans raised by government by issue of public notifications, internal and external debt and all moneys received by government in repayment of loans and interest thereon is also credited into this fund. All expenditure incurred by the government for the conduct of its business including repayment of internal and external debt and release of loans to states/union territory governments for various purposes is debited against this fund. 17 For example, the Eleventh Finance Commission suggested giving grants-in-aid to the states equal to the amount of the deficits as estimated for each of the years during 1995–6 to 1999–2000. Under this head only 3–4 percent of the total revenue receipts of the union are transferred (Chaubey 2003). 18 For a brief description of the main recommendations with respect to local government see, for example, Rao and Singh (2004). 19 Under article 270 of the Constitution, as amended by the Constitution (Eightieth Amendment) Act 2000, a prescribed percentage of the net proceeds of all central taxes and duties (except union surcharge, cess levied for specific purposes and the duties and taxes referred to in article 268 and 269 – that is stamp duties, duties of excise on medicinal and toilet preparations, taxes on the sale

86

20

21 22

23

24 25 26

27

28 29

Angela Fraschini or purchase of goods other than newspapers and taxes on the consignment of goods, where such sale, purchase or consignment take place in the course of inter-state trade or commerce) is to be assigned to the states within which that tax or duty can be levied in that year and distributed among those states in terms of the recommendations of the finance commission. The Eleventh Finance Commission recommended that 29.5 percent of the net proceeds of central taxes/duties may be distributed amongst all such states where the central tax/duty can be levied. Grants-in-aid under article 275 of the Constitution are need-based, on the recommendations of the finance commission, while grants under article 282 are purpose-based, in the sense that the central government has the power to make discretionary grants to the states. For example, the Eleventh Finance Commission recommended grants-in-aid amounting to RS35,359.07 crore to 15 states equal to the amount of deficits assessed for each year during the period 2000–5. For example, the Eleventh Finance Commission recommended grants for the period 2000–5 amounting to RS3,843.63 crore to all the states for upgrading of standards of administration for the following sectors: district administration, police administration, prisons administration, fire services, judicial administration, fiscal administration, health services, elementary education, computer training for school children, public libraries, heritage protection, augmentation of traditional water sources. For example, the Eleventh Finance Commission recommended grants totaling RS10,000 crore for local bodies during 2000–5, to be utilized (except the amount earmarked for maintenance of accounts and audit – RS493.04 crore – and for development of database – RS200 crore) for maintenance of civic services in rural and urban areas. The annual grant recommended was RS1,600 crore for rural local bodies and RS400 crore for urban local bodies. The planning commission was not conceived in the Constitution but through a resolution of the cabinet, after 50 days of promulgation of the Constitution. In India, locally elected bodies are the Panchayati Raj Institutions (PRIs) (at the district, intermediate and village levels) and the urban local bodies at all levels. Panchayat means an institution of self-government constituted under article 243B of the Constitution. It is worth observing that what is referred to as local self-government is in actual fact a circumscribed space where there is no legislative and judicial authority and where the issues on which citizens can make decisions is limited (De Souza 2000). For the urban local bodies the Seventy-fourth Amendment provides for “. . . g) the devolution by the State Legislature of powers and responsibilities upon the Municipalities with respect to preparation of plans for economic development and social justice, and for the implementation of development schemes as may be required to enable them to function as institutions of self-government; h) levy of taxes and duties by Municipalities, assigning of such taxes and duties to Municipalities by State Governments and for making grants-in-aid by the State to the Municipalities as may be provided in the State law.” Intermediate level means a level between the village and district levels. For the rural local governments the Seventy-third Amendment states that “. . . subject to the provisions of the Constitution, the Legislature of a State may, by law, endow the Panchayats with such powers and authority as may be necessary to enable them to function as institutions of self-government and such law may contain provisions for the devolution of powers and responsibilities upon Panchayats at the appropriate level, subject to such conditions as may be specified therein, with respect to: a) the preparation of plans for economic development and social justice; b) the implementation of schemes for economic

Fiscal federalism in China and India 87

30

31 32

33

development and social justice as may be entrusted to them including those in relation to the matters listed in the Eleventh Schedule. The Legislature of a State may, by law, a) authorize a Panchayat to levy, collect and appropriate such taxes, duties, tolls and fees in accordance with such procedure and subject to such limits; b) assign to a Panchayat such taxes, duties, tolls and fees levied and collected by the State Government for such purposes and subject to such conditions and limits; c) provide for making such grants-in-aid to the Panchayats from the Consolidated Fund of the State; and d) provide for constitution of such funds for crediting all moneys received, respectively, by or on behalf of the Panchayats and also for the withdrawal of such moneys there from as may be specified in the law.” The items included in the Seventh Schedule under the state list are the following: (1) taxes on agricultural income; (2) duties in respect of succession to agricultural land; (3) estate duty in respect of agricultural land; (4) taxes on lands and buildings; (5) taxes on mineral rights subject to any limitations imposed by Parliament by law relating to mineral development; (6) duties of excise on the following goods manufactured or produced in the state and countervailing duties at the same or lower rates on similar goods manufactured or produced elsewhere in India: (a) alcoholic liquors for human consumption; (b) opium, Indian hemp and other narcotic drugs and narcotics, but not including medicinal and toilet preparations containing alcohol; (7) taxes on the entry of goods into a local area for consumption, use or sale therein; (8) taxes on the consumption or sale of electricity; (9) taxes on the sale or purchase of goods other than newspapers; (10) taxes on advertisements other than advertisements published in the newspapers [and advertisements broadcast by radio or television]; (11) taxes on goods and passengers carried by road or on inland waterways; (12) taxes on vehicles, whether mechanically propelled or not, suitable for use on roads, including tramcars; (13) taxes on animals and boats; (14) tolls; (15) taxes on professions, trades, callings and employments; (16) capitation taxes; (17) taxes on luxuries, including taxes on entertainments, amusements, betting and gambling. Recently, there has been a proposal to create a separate “local list” for the local bodies, so redesigning the present central and state jurisdictions (see Government of India, Ministry of Law 2002). The functions included in the Eleventh Schedule of the Constitution are the following: (1) agriculture, including agricultural extension; (2) land improvement, implementation of land reforms, land consolidation and soil conservation; (3) minor irrigation, water management and watershed development; (4) animal husbandry, dairying and poultry; (5) fisheries; (6) social forestry and farm forestry; (7) minor forest produce; (8) small scale industries, including food processing industries; (9) khadi, village and cottage industries; (10) rural housing; (11) drinking water; (12) fuel and fodder; (13) roads, culverts, bridges, ferries, waterways and other means of communication; (14) rural electrification, including distribution of electricity; (15) non-conventional energy sources; (16) poverty alleviation programs; (17) education, including primary and secondary schools; (18) technical training and vocational education; (19) adult and non-formal education; (20) libraries; (21) cultural activities; (22) markets and fairs; (23) health and sanitation, including hospitals, primary health centers and dispensaries; (24) family welfare; (25) women and child development; (26) social welfare, including welfare of the handicapped and mentally retarded; (27) welfare of the weaker sections, and in particular, of the Scheduled Castes and the Scheduled Tribes; (28) public distribution system; (29) maintenance of community assets. “In many States, the formats and procedures for maintenance of accounts by

88

34

35 36

37

38 39

40

41

Angela Fraschini these bodies prescribed decades ago, are continued without making any improvements to take into account the manifold increase in their powers, resources and responsibilities. Most village level Panchayats do not have any staff except for a full or a part-time secretary, because of financial constraints (. . .) There is no mechanism for collection of data on the revenue and expenditure of the various tiers/levels of the rural/urban local bodies at a centralized place where it could be compiled, processed and made available for use. In the absence of any reliable financial/budgetary data, no realistic assessment of the needs of the Panchayats and municipalities for basic civic and developmental functions can be made nor can any information be generated on the flow of funds to the local bodies for the implementation of various schemes for economic development and social justice” (Government of India 2000). A different approach to decentralization is followed by Breton (1996). In his seminal work he argues that the most important reason for decentralizing the public sector is that decentralization stimulates intergovernmental competition, and when competition breaks down, or produces undesirable outcomes, decentralization fails. On decentralization failures see also Breton (2002). The issues relevant to the question of whether fiscal decentralization generates the positive results that its supporters claim are addressed in Tanzi (2002). For example, a greater centralization of some functions may be justified on the grounds that widespread poverty may make preferences more homogeneous across local jurisdictions. For a discussion of these arguments see Smoke (2001) and bibliography therein. Recently the importance of the institutional design has been underlined: “decentralization is neither good or bad for efficiency, equity, or macroeconomic stability; but rather its effects depend on institution specific design” (Litvack et al. 1998). The National Development Council is presided over by the prime minister and comprises cabinet ministers, deputy chairman and members of the planning commission and the chief ministers of the states. For example, as a result of the reforms, new, rival power centers have emerged in China. Local governments, especially those in areas with the largest growth, now have substantial independent sources of revenue, authority and political support. Moreover, though local officials are still appointed and dismissed by the central government, their authority is implied and enhanced by their access to and control of local information and resources. Also the gradual decline of the personal authority of national leaders and the rise of local governments has weakened the reach of the Chinese Communist Party into the lower levels of government, and many lower government officials now bestow their loyalty to localities, not to central government. Finally, as the private market economy has expanded, the ability of the central government to monitor and control local economic behavior has weakened enormously (Montinola et al. 1995). According to Vaillancourt and Bird (2004), India’s complex system appears both to have been significantly equalizing and on the whole to have contributed to achieve a degree of cohesiveness in a large and diverse country, even if it may be criticized as providing some undesirable incentives with respect to fiscal management of the states. In recent years, the central authorities started to examine all the earmarked grants and some of them have been converted into equalization grants. For example, about RMB4 billion of subsidies for food in urban areas were transferred into the transitory equalization grants in 2001. The same was done with earmarked grants for the development of borders regions (Zhang and Martinez-Vazquez 2003).

Fiscal federalism in China and India 89 42 Before the 1994 reform, budget deficits were financed through a combination of credits from the People’s Bank of China and domestic and international borrowing as debt revenues. Since 1995 budgets at all levels of government have to be balanced and any violation of the balanced budget approved by the legal process results in administrative prosecution against parties directly responsible. In spite of that, the reform’s effectiveness on hardening the budget constraint has been undermined by several elements, and often local governments incur soft budgets. The central government now finances its budget deficits only through domestic and international borrowings, and in the state budget such borrowing is no longer counted as debt revenues (Jing and Zou 2003). 43 The expectation of bail-outs has encouraged Indian states and local governments to adopt high-risk deficit financing strategies. And whereas a major direct default has yet to occur, indirect defaults by urban development authorities, controlled by state governments, have occurred and bills owed by states sometimes go unpaid (McCarten 2003).

References Ahmad, E., Li, K. and Richardson, T. (2002) “Recentralization in China?,” in Ahmad, E. and Tanzi, V. (eds) Managing Fiscal Decentralization, London: Routledge, pp. 205–24. Ahmad, E., Singh, R. and Fortuna, M. (2004) “Towards more effective redistribution: reform options for intergovernmental transfers in China,” IMF Working Paper 98, Washington, DC: IMF. Arora, V. and Norregaard, J. (1997) “Intergovernmental fiscal relations: the Chinese system in perspective,” IMF Working Paper 129, Washington, DC: IMF. Bahl, R. (1999) “Implementation rules for fiscal decentralization,” Working Paper 99-1, Georgia State University: Andrew Young School of Policy Studies. Bahl, R. and Wallich, C. (1992) “Intergovernmental fiscal relations in China,” Policy Research Working Paper 863, Washington, DC: The World Bank. Blanchard, O. and Shleifer, A. (2000) “Federalism with and without political centralization: China versus Russia,” mimeo. Breton, A. (1996) Competitive Governments: An Economic Theory of Politics and Public Finance, New York: Cambridge University Press. —— (2002) “An introduction to decentralization failure,” in Ahmad, E. and Tanzi, V. (eds) Managing Fiscal Decentralization, London: Routledge, pp. 31–45. Buddhadeb, G. and Prabir, D. (2005) “Investigating the linkage between infrastructure and regional development in India: era of planning to globalization,” Journal of Asian Economics, 15–16: 1023–50. Chaubey, P. K. (2003) “Evolution of union–state fiscal relations in India: two steps forward and one step backward,” in Chaubey, P. K. (ed.) Fiscal Federalism in India, New Delhi: Deep & Deep Publications PVT, Ltd, pp. 21–44. CIAs The World Factbook, (www.cia.gov) De Souza, P. R. (2000) “Multi-state study of Panchayati Raj legislation and administrative reform,” in World Bank (2000), Overview of Rural Decentralization in India, Washington, DC: The World Bank, V, pp. 1–91. Government of India (2000) Report of the Eleventh Finance Commission, Delhi: Government Publication Division. Government of India, Planning Commission (2001) Report of the Steering Committee

90

Angela Fraschini

on Urban Development for the Tenth Five Years Plan (2002–2007), Delhi: Government Publication Division. Government of India, Ministry of Law (2002) Report of the National Commission to Review the Working of the Constitution, Delhi: Government Publication Division. Government of India (2004) Report of the Twelfth Finance Commission, Delhi: Government Publication Division. Jing, J. and Zou, H. (2003) “Soft-budget constraints and local government in China,” in Rodden, J. A., Eskeland, G. S. and Litvack, J. (eds) Fiscal Decentralization and the Challenge of Hard Budget Constraints, Cambridge, Massachusetts: The MIT Press, 289–323. Litvack, J., Ahmad, J. and Bird, R. (1998) Rethinking Decentralization in Developing Countries, Washington, DC: The World Bank. Ma, J. (1997) “Intergovernmental fiscal transfers: a comparison of nine countries,” Paper prepared for Macroeconomic and Policy Division, Economic Development Institute, Washington, DC: The World Bank. McCarten, W. J. (2003) “The challenge of fiscal discipline in the Indian states,” in Rodden, J. A., Eskeland, G. S. and Litvack, J. (eds) Fiscal Decentralization and the Challenge of Hard Budget Constraints, Cambridge, Massachusetts: The MIT Press, pp. 249–86. Mishra, C. S. (n.d.) “A study of the measures needed to augment the consolidated fund of the states for supplementing the resources of local bodies,” report for the 12th Finance Commission (2005–10). Montinola, G., Qian, Y. and Weingast, R. (1995) “Federalism, Chinese style: the political basis for economic success in China,” World Politics, 48(1): 50–81. Musgrave, R. A. (1959) The Theory of Public Finance: A Study in Public Economy, New York: McGraw Hill. Oates, W (1972) Fiscal Federalism, New York: Harcourt Brace Jovanovich, Inc. Purfield, C. (2004) “The decentralization dilemma in India,” IMF Working Paper 32, Washington, DC: IMF Rao, M. G. (2001) “Challenges of fiscal decentralization in transitional economies: an Asian perspective,” Conference Paper “Public Finance in Developing and Transition Countries: A Conference in Honor of Richard Bird,” Georgia State University: Andrew Young School of Policy Studies. —— (2002) “Fiscal decentralization in Indian federalism,” in Ahmad, E. and Tanzi, V. (eds) Managing Fiscal Decentralization, London: Routledge, pp. 286–305. —— (2004) “Changing contours in fiscal federalism in India,” New Delhi: National Institute of Public Finance and Policy. Rao, M. G. and Singh, N. (2004) “The political economy of India’s federal system and its reform,” Paper 04-07, UCSC, Santa Cruz: Centre for International Economics. Singh, N. (2004) “India’s system of intergovernmental fiscal relations,” Paper 578, UCSC: Department of Economics. Singh, S. K. (2003) “Federal transfer in India: an introduction,” in Chaubey, P. K. (ed.) (2003) Fiscal Federalism in India, New Delhi: Deep & Deep Publications PVT, Ltd, pp. 11–16. Smoke, P. (2001) “Fiscal decentralization in developing countries. A review of current concepts and practice,” Geneva: United Nations Research Institute for Social Development. Tanzi, V. (2002) “Pitfalls on the road to fiscal decentralization,” in Ahmad, E. and

Fiscal federalism in China and India 91 Tanzi, V. (eds) Managing Fiscal Decentralization, London: Routledge, pp. 17–30. Unescap a (n.d.) Local Government in Asia and the Pacific: A Comparative Study. Country paper: Malaysia. Unescap b (n.d.) Local Government in Asia and the Pacific: A Comparative Analysis of Fifteen Countries. Vaillancourt, F. and Bird, R. M. (2004) “Expenditure-based equalization transfers,” Working Paper 04-10, Georgia State University: Andrew Young School of Policy Studies. Vithal, B. P. R. and Sastry, M. L. (2001) Fiscal Federalism in India, Delhi: Oxford University Press India. Weingast, B. R. (1995) “The economic role of political institutions: market-preserving federalism and economic growth,” Journal of Law, Economics, and Organization, 11: 1–31. World Bank (2001) “The challenge of regional development in China: background note,” Washington, DC: The World Bank. —— (2002) “China: national development and sub-national finance,” Report 22951-CHA, Washington, DC: The World Bank. —— (2004) “Fiscal decentralization to rural governments,” Report 26654-IN, Washington, DC: The World Bank. Zhang, Z. and Martinez-Vazquez, J. (2003) “The system of equalization transfers in China,” Working Paper 03-12, Georgia State University: Andrew Young School of Policy Studies.

4

Democracy and welfare without welfare state Matteo Cacciatore, Paola Profeta and Simona Scabrosetti

Introduction South and East Asian countries have a “light” welfare state, characterized by low public spending on welfare (Jacobs 1998). Enterprises and families have traditionally played a major welfare role and have partially compensated for the low public spending. In some countries, such as Japan, enterprises have adopted a variety of flexibility measures to keep workers who are not necessarily profitable, while in all South and East Asian countries three-generation families substitute the public welfare system by pooling income between workers and economically inactive people. The absence of the welfare state is based on the common assumption that women are the main providers of personal care for children and the elderly at home. However, these forms of enterprise and family welfare are currently being challenged by the economic conditions (a recession in Japan, which will make enterprises unable to avoid massive layoffs any longer), the financial crisis (which has substantially raised unemployment in Korea), the falling fertility and aging process (in China and Thailand, but also in Korea and Japan), as well as by some common trends, such as urbanization, family nuclearization and the rise of female employment (which imply a reduced readiness of women to care for their parents or children). As a consequence, welfare public spending is expected to increase in South and East Asian countries. Some countries have already introduced important reforms in the last decade to strengthen their social welfare systems (Japan, Korea), while others have them in their agenda (China). The World Bank (1999) identifies “social protection” as a strategic sector for the structural long-term development of South and East Asian countries. This sector includes three areas, strictly interrelated: social safety nets (including social funds), labor market policies (including child labor) and pensions. This last area, pensions, is crucial, especially for countries in which the demographic transition is well advanced, such as China, Thailand, Korea and Japan. Current pension schemes are characterized by a general low level of

Democracy and welfare without welfare state 93 coverage (in 1996, 17.6 percent of the labor force in China and 25.1 percent in Thailand), the absence of social safety nets for the needy elderly, retirement schemes still in evolution and not mature and a large funded private component. In 1994 total pension expenditure represented 6.12 percent of GDP in Japan and in 2002 it only accounts for 3.5 percent of GDP in China. These factors introduce ample opportunities to design changes and reforms, which are best seized early while the schemes are still young and the aging process at the beginning. China emerges as a critical country, due to its additional problems, such as the differences between local and provincial levels, and a fragmented and non-unified system (Holzmann et al. 2000). In this chapter, we explore whether in addition to economic and demographic trends, political factors may also represent a crucial challenge for the development of a welfare state in South and East Asian countries. Many of these countries show a trend towards increased democratization and participation of civil society, which raise demands for government to assume more responsibility for the unemployed, sick, poor and the elderly. This political transition towards greater democracy and participation has already increased the government role in key social protection areas, for instance in Korea. Again, China represents an interesting case, since it does not seem to fit this pattern, because no democratization process is combined with the starting process of reforms aimed at raising the welfare state. This chapter contributes to the analysis of the relation between the democratization process and the emergence of a welfare state in South and East Asian countries. We argue that, although it seems not to be essential to have democracy in order to start reforms aimed at raising the welfare programs (e.g. in China), the interrelation between economic and political liberalizations may be important to characterize the outcome of the final stage of the reform process, with the best performances arising when political and economic development go hand in hand. The chapter is organized as follows: the next section provides evidence on democracy indicators in the six South and East Asian countries analyzed in this book. The third section focuses on the case of pensions in China. The final section introduces the political economy of democracies, by reviewing the existing literature and providing some conclusions for China.

Democracy in South and East Asian countries: some evidence Are economic and political outcomes related? In particular, may a specific economic outcome which broadly characterizes South and East Asian countries, mainly the absence of a comprehensive welfare state, be explained on political grounds?

94

Matteo Cacciatore et al.

In this section we collect data from the Polity dataset (2002) on the political characteristics of the countries analyzed in this book. The Polity dataset provides indicators of democracy, autocracy and other specific political characteristics for the period 1800–2003 for a large sample of countries. This dataset is widely used in political economy studies, in particular in a recent strand of the literature devoted to democracies and economic development (see Acemoglu et al. 2004 and 2005; Mulligan et al. 2004). We focus on the indicator called “democracy,” which represents an annual measure of institutionalized democracy. This measure is constructed based on three essential elements: (i) the presence of institutions and procedures through which citizens can express effectively their preferences about alternative policies and leaders, (ii) substantial institutionalized constraints on the exercise of power by the executive, and (iii) the guarantee of civil liberties to all citizens in their daily lives and in acts of political participation. The rule of laws, systems of checks and balances, freedom of the press and other aspects of democracies are included, because they are considered specific means of these three elements. The democracy indicator ranges from 0 (minimum democracy) to 10 (maximum democracy). Figures 4.1 and 4.2 show the values of the democracy indicator for the period 1800–20031 for China, Korea2 and India and for Japan, Thailand and Malaysia, respectively. Two results emerge: (i) China is characterized by the lowest absolute levels of the democracy indicator (scoring 1 from 1800 to 1950 and 0 from 1950 to 2003), Japan and India have a tradition of high democracy and Korea, Thailand and Malaysia have recently reached high levels of democracy; and (ii) China is characterized by the more stable trend with almost no variation (or a small negative variation) in the democracy indicator of the last two centuries, while the other countries, especially the ones which entered the mid-1990s with low levels of democracy, have experienced a certain variation of this indicator over time, mostly an ascendant path (see for instance Korea, Thailand and Malaysia). These simple graphs suggest two directions to explore. First, China is experiencing deep economic changes, while at the same time it is characterized by a stable non-democracy. At first sight, it seems that the economic development in China is unrelated to its democratic evolution, which is far from beginning. The next section is thus devoted to China, with a particular focus on its pension system, as a main category of a potential welfare system. Second, the development of a stronger welfare state in South and East Asian countries seems to follow the democratization process, with the exception of China. What are the determinants of democracy? Is there any relation between democracy and the level of welfare expenditures? These questions are essential to understand the future of the social and economic development of South and East Asian countries. A growing political economy

10 Democracy indicator (0–10)

9 8 7 6 5 4 3 2 1 0 1800

1850

1900

1950

2000

Year Japan

Thailand

Malaysia

Figure 4.1 Democracy in South and East Asian countries: Japan, Thailand and Malaysia (source: Polity dataset (2002)).

10

Democracy indicator (0–10)

9 8 7 6 5 4 3 2 1 0 1800

1850

1900

1950

2000

Year China

Korea

India

Figure 4.2 Democracy in South and East Asian countries: China, Korea and India (source: Polity dataset (2002)).

96 Matteo Cacciatore et al. literature is devoted to this topic. The final section, after reviewing the main results of the literature, explores some consequences of the relation between democracy and welfare for the future of China.

Welfare without welfare state: the case of pensions in China China represents the most relevant example of a country “without welfare state.” In particular, the public pension system, which typically represents the larger share of the welfare state in developed countries, is a very modest share of the economy, 3.5 percent of GDP in 2002. Official statistics report that only a small fraction of the Chinese workforce is currently part of the pension program: 55 percent of urban3 workers and only 11 percent of the rural4 workforce are covered by the public pension system and only 6 percent of workers in the private sector are covered by a pension scheme. However, these few pensioners receive quite a generous benefit: the replacement rate is about 80 percent of wages and it increases up to 90 percent for civil servants. The expansion of pension coverage, together with the unification of the program at a provincial level and the reduction of enterprise contribution share (Chen 2004), is at the center of the reform of the Chinese pension system. This reform is a critical component of China’s overall economic reform and one of the main challenges for the development of its economy. This challenge is exacerbated by additional risky circumstances, in particular a rapid aging of the population, the inability of state-owned enterprises (SOEs) to cover the pensions of current retirees, the underdevelopment of China’s capital market and deep social transformations. How China will meet with these problems will determine whether it becomes a prosperous developed country. One of the more relevant aspects of the Chinese pension reform concerns the implementation of a multipillar scheme, composed of three pillars: (i) a PAYG public pillar with flat rate benefits, financed by a 13 percent contribution from enterprises that goes into a municipal or provincial pooled fund; (ii) a mandatory defined contribution pillar with individual accounts for each worker, financed by a payroll tax of 11 percent; and (iii) a voluntary supplement pension pillar managed by each individual firm or private insurance company. The combined expected replacement rate of the first two pillars is 58.5 percent for a typical average-wage earner with 35 years of contributions. Currently, only the first pillar is employed and the implementation of the three-pillar scheme represents one of the most critical aspects of the reform process. This implementation implies a transition cost, which has not been fully addressed yet by the Chinese government (Chen 2004). The state council simply grouped benefit formulas into three main categories: (i) new workers, entering the labor force after 1997, who fully participate into the new three-tier system; (ii) middle workers, who started work before 1997,

Democracy and welfare without welfare state 97 receiving an additional transitional benefit; and (iii) old workers, retired prior to 1997 and entitled to benefits defined by the previous system. Although the implicit pension debt is small, due to the reduced size of the pension system, and thus the costs of the reform are expected to be low, the government has to provide a definite set of guidelines in order to implement effectively the pension reform.5 In particular, it has to address the transition problem, by recognizing that the cost of the transition cannot be supported only by new workers participating in the system. One of the main challenges for the future of the Chinese pension system and the overall economy is represented by the demographic transformation: “China is growing older” (Jackson and Howe 2004), as a combination of a falling fertility (due in part to the modernization policy and in part to the government control policy) and a rising longevity (due to the improvements of living conditions, especially in the urban areas). The United Nations (2003) assess that by 2050 the number of elderly aged 60 or more will exceed 30 percent of the population and that in 2050 there will be 18 percent fewer working age adults than today. In practice, each retiree in 2050 will be supported by 1.5 workers (today the ratio is one to three). Given that the current pension system covers only a small fraction of the population and that the modernization process is reducing family support in old age (which represents the only financial support to the elderly overall in the countryside), a reform of the pension system which introduces a substantial role of the public sector is inevitable. This reform should provide a decent standard of living for the elderly, without imposing a heavy burden on the younger generation. Otherwise, China will be the first country to “grow old before to grow rich” (Jackson and Howe 2004). In other words, China cannot continue to grow without developing a welfare state. The Chinese pension reform is strictly related to the SOE reform (Chen 2004). The ratio of contributing workers to retired beneficiaries has dropped from 5.4 in 1989 to three in 2003. On one hand, reforming the pension system will help the SOE reform process, since the creation of effective, and not just nominal, individual pension accounts can facilitate the choice of workers to move from the state to the private sector; on the other hand, reforming the SOEs will help the pension reform process, since when SOEs reduce their redundant workforce, the extension of pension coverage to non-state workers becomes urgent. Another reason for an urgent reform is the financial distress of the current Chinese social security system. According to China’s ministry of social security the deficit in social pooling funds was RMB18.7 billion in 1999, RMB35.7 billion in 2000 and it is expected to reach RMB50 billion in 2005. Several factors have contributed to this situation: (i) non-compliance and tax evasion, (ii) fragmentation of the pension system at local levels, and (iii) lack of capital market. As argued by Zhao and Xu (2002), non-compliance means that business traditionally excluded from pension

98

Matteo Cacciatore et al.

system, which typically have a relatively young workforce and a small number of retirees, refuse to agree to the program and the government cannot force them to do it (for several reasons, such as the fact that the social bureau is not able to collect levied administrative penalties, employment data are not available, etc.). This refusal is mainly due to the lack of incentives they receive from the government, due to the prohibitive cost of the system, the large redistribution operated at first pillar level and low rates of return on contributions to personal accounts. The second element concerns the fragmentation of pension pools at local levels, which is an obstacle in guaranteeing financial stability to the overall system. In fact, local governments have the systematic distorted practice of borrowing from personal account contributions to cover cash shortfalls in the social pools and use individual accounts to cover current expenses. Thus, pooling pensions at provincial levels is an inevitable solution. But pooling contributions is going to create a vicious circle, increasing moral hazard and principal agent problems and discouraging the participation in the system (Chen 2004), because if the center will cover any deficit or take away any surplus, the local units have incentives to underperform on collection and overperform on benefit expenditure (“soft budget constraint”). The third element contributing to the financial crisis of the pension system is the underdevelopment of the capital market. This implies that only a limited amount of pension assets in the individual accounts is available for capital market investment and that the opportunities and returns from contribution investment are reduced. Wang (2004) argues that a virtuous interaction between pension scheme and capital market is essential, especially if a funded component of the Chinese pension system is to be implemented eventually. To conclude, we expect the introduction of a welfare state, through a broad public pension program, to represent a major measure of the Chinese economic development process. Several factors make this process particularly difficult, such as aging, the reform of the SOEs, fragmentation and the underdevelopment of the capital market. Since at the same time we do not observe any ongoing democratization process, we ask whether this may represent an additional factor which affects the final economic outcome of the country. We explore this direction in the next section.

The political economy of democracies and non-democracies The existing political economy literature on democracies deals with two important questions: (i) what are the economic determinants of democracy (if any)? and (ii) does democracy affect public policy? In this section we provide a short review of the current theoretical and empirical findings on these two issues from which we can draw some lessons for China.

Democracy and welfare without welfare state 99 The economic determinants of democracy The first question, i.e. the identification of the economic determinants of democracy, is addressed by several contributions related to the “modernization theory,” which studies the link between democracy and economic development. Recent contributions in this tradition include Boix (2003), Barro (1996), Giavazzi and Tabellini (2005), Acemoglu et al. (2004, 2005) and Acemoglu and Robinson (2005). The modernization theory argues that markets can prosper only in a political framework characterized by constitutional liberties and democratic practices. In this sense, developed economies and political democracies should emerge and survive together, especially in the long run. According to Lipset (1959), only in a wealthy society can the mass of population intelligently participate in politics and avoid succumbing to irresponsible demagogues. Countries should become more democratic as they become richer. The process of economic modernization generally results in both the reduction of income inequality, which is a source of political conflict and brings authoritarian solutions, and the growth of a broad middle class, who act as a moderating political force. Moreover, the rise of education levels and the creation of a labor force required to make its own decision in the production process (that is an autonomous labor force) should increase the toleration of different values and options and the recourse to liberal democracy as the mechanism to settle disagreements. Education promotes democracy either because it enables a culture of democracy to develop or because it leads to greater prosperity. A recent complete theory of political transition and economic development is provided by Boix (2003). The author finds that democracy prevails when economic equality and capital mobility are high in a given country. First of all, economic equality promotes democracy. As the distribution of income becomes more equal among individuals, redistributive pressures from the poor on the rich diminish and the probability of a peaceful transition from an authoritarian regime to universal suffrage increases. The ultimate level of taxes becomes smaller than the cost of repression. Second, a reduction in the cost of moving capital away implies that government must curb taxes. As a consequence, the extent of political conflict among capital holders and non-holders declines and the probability of democracy rises. On the contrary, authoritarianism predominates in those countries in which both the level of inequality and the lack of capital mobility are high. The redistributive demands of the worse-off citizens on the rich are particularly intense and, as a result, the latter strongly oppose the introduction of democracy that would allow the majority of the population to establish heavy taxes on them. The presence of immobile capital makes the authoritarian solution worse. Given that they cannot escape the threat of high taxes by shifting assets abroad, capital owners want to block democracy. In this sense, the association

100 Matteo Cacciatore et al. between economic development and democracy comes from the transformation that capital experiences with economic modernization: from an economy based on fixed assets to an economy based on highly mobile capital in which the accumulation of human capital, harder to expropriate than the physical one, increases. Boix (2003) examines also how changes in the economy may affect the chances of democracy. The author underlines that economic growth is a necessary, but not sufficient, condition in order to secure a democratic outcome. The poor can commit themselves to moderate levels of redistribution according to the fact that low taxes stimulate faster economic growth. This reduces the opposition of the rich to universal suffrage and increases the probability of democracy. But if an organizational capacity of the poor to credibly commit to observe their promises is needed, then leftwing parties and unions may be instrumental to the success of democracy. Moreover, social mobility across classes fosters democracy by easing social conflict, that is by tending to equalize the income of individuals over time. Trade openness and democracy are also related. However, this relationship depends on the distribution of factors in a given economy. In countries where the poor are the abundant factor, trade openness equalizes conditions and favors the introduction of democracy. On the contrary, if the poor are the scarce factor, trade openness intensifies social conflict and raises the probability of authoritarianism. In order to assess the validity of his theoretical model and analyze the probabilities of transition to democracy and democracy stability, Boix (2003) refers to two different samples of countries, one from 1950 to 1990 with direct measures of inequality and the other from 1850 to 1980 with indirect measures.6 With reference to the first sample, the author takes different proxies for asset specificity such as per capita income, the average years of education, the share of agriculture as a percentage of GDP, the size of the oil sector and the level of economic concentration. Using the first measure of asset specificity, that is per capita income, Boix finds that a decline in income inequality is associated with both an increase in the likelihood of the democratization process and a lower probability of democratic stability. However, this latter conflicting result holds only with reference to very low levels of income. As pointed out by modernization theory, the author also finds that per capita income positively affects the probability of a democratic transition and in particular the stability of a democratic regime.7 Even when he comes to consider the second broader sample, Boix shows coefficients of the independent variables in line with his theoretical expectations, but it seems that per capita income is simply a proxy for other more important factors. In short, highly unequal countries remain authoritarian and whenever they go through a democratic phase, it is only a temporary phase. At the same time, countries with a limited share of mobile assets are unlikely to become democratic unless they show a fairly equal income distribution.

Democracy and welfare without welfare state 101 Another contribution on the relation between income and democracy is Barro (1996). In his econometric study, Barro (1996) finds that an improvement in the standard of living, measured by different indicators such as per capita GDP, infant mortality rate and education attainment, affects positively the probability of democratization over time. As in Lipset’s theory, prosperity stimulates the development of democratic institutions. Moreover, representative regimes are more likely to arise in a non-colony rather than in a colony, especially when it was not a British or Spanish possession. The religious traditions are also relevant in the sense that Protestant and Muslim countries are the most and the least democratic, respectively. However, when Barro simultaneously controls for both the standard of living and the colonial status (or religious affiliation), the latter stops being a key element for democracy. This means that the main effects on democratic regime apparently work indirectly through influences on the standard of living indicators. In this way, developed countries would export their economic systems rather than their political regimes. Democracy would catch on after reasonable standards of living have been attained, whereas would seem not to last without strong economic bases. This is also the idea of Giavazzi and Tabellini (2005). They find positive feedback effects between economic and political reforms even if it is difficult to establish the correct direction of the causality relationship. Furthermore, the interaction effects show that countries which undertake both reforms have better economic performance than countries which undertake only economic or political liberalization. In other words, the effects are not additive and moreover the sequence matters. Following the “easy path,” that is first becoming a democracy and then opening up the economy, leads to poorer economic payoffs in terms of growth, investment, trade volume and macro-policies. It is less likely that an authoritarian regime opens up the economy, but when it happens it is because interest groups opposing free trade and the market system have been crushed. Consequently, liberalization is more effective and devoid of compromises. On the other hand, it could be that better democracies arise in an open economic environment. Redistributive conflicts could weaken a young democracy characterized by a closed economy whereas openness to trade, competition and growth, which come from the economic liberalization, provide the resources for the redistribution that a democracy requires. In line with the modernization theory, the correlation between income per capita and democracy as well as the one between education and democracy are recurrent in many other cross-country analyses. But, according to Acemoglu et al. (2004, 2005), these works do not establish causation in the sense that, first of all, it could be that democracy causes enhanced income or education rather than the reverse and, in the second place, there is no relationship between changes in income or education and changes in democracy. Moreover, existing empirical results may suffer

102 Matteo Cacciatore et al. from omitted variables biases, i.e. factors determining both the political regime and the income per capita or the level of education could drive the findings. Because these omitted characteristics are either countryspecific or time-invariant, Acemoglu et al. (2004, 2005) include countryfixed effects in the regressions in order to improve inference on the causal relationship. The first and main measure of democracy is the Freedom House Political Rights Index, according to which a country gains a higher score if political rights are closer to the ideals coming from a checklist of different questions. By introducing fixed effects, the positive correlation between income and democracy as well as the one between education and democracy disappear: income and education are not determinants of democracy. These results hold with both indicators for democracy, with different econometric specifications and estimation techniques and in various subsamples. In order to explain the strong crosssectional relationship between income and democracy, Acemoglu et al. (2005) refer to historical factors which have influenced either the economic or the political development path of societies.8 Furthermore, contrary to the predictions of modernization theory, Acemoglu et al. (2005) show that economic crises, defined as a sudden and significant reduction in the growth rate, increase the probability of democratization. As a matter of fact it seems that economic crises do not affect transitions away from democracy but lead to dictatorship collapse. This empirical result fits in with the theory of democratization and democratic consolidation of Acemoglu and Robinson (2004). They suggest that some degree of development in civil society, some level of industrialization, a greater inter-group inequality, the middle class and the globalization process as well as economic or political crises are key factors for democratization. This latter can be thought as a credible promise by the elite endowed with de jure political power and faced with a serious revolutionary threat or significant social unrest from the citizens endowed with de facto political power of the majority.9 In this sense, when the citizens are not well-organized, the transition to democracy can be delayed or avoided. If social and political conflicts are damaging to physical and human capital owners, then democracy is more likely to arise when the elite are industrialists rather than landowners. Greater inter-group inequality leads to democratization because it makes revolution a chance to increase the citizens’ income. Democratic politics tend to be more conflict-ridden where there is no middle class acting as a buffer between the elite and the citizens. International financial integration, international trade and increased political integration are also important for democracy given that they discourage the elite from using repression. Finally, shocks and crises make revolutions easier and less costly and then force the elite to adopt representative regimes.

Democracy and welfare without welfare state 103 Democracy and public policy The second question, i.e. how democracy affects public policy, is addressed in Mulligan et al. (2004) and Boix (2003). Mulligan et al. (2004) underline that there are two very different perspectives on constructing positive theories of the public sector. The first one comes from the formal voting literature whereas the second one relates to the “Chicago Political Economic School.” In the formal voting literature three tenets of democratic decision-making would imply democratic–non-democratic policy gaps. In other words, it would be possible to predict public policy starting from a measure of democracy and holding constant economic and demographic variables. The first tenet says that in many formal models the voting process mitigates the expression of strong policy preferences which determine inefficient policy outcomes. The second tenet concerns the distribution of political power. This is more equal than the distribution of income or wealth and, as a consequence, democracy would massively redistribute from rich to poor. On the contrary, under authoritarian regimes the level of redistributive spending should be minimal. The third tenet of the formal voting theory emphasizes the importance of “the form of the voting game.” At the other extreme, there are positive theories of public policy such as that of Wittman (1989) that focuses on efficiency considerations as the main determinants of public policy. There is no room for political factors. These theories are also related to Stigler (1970) and Peltzman (1980), that is to the “Chicago Political Economic School.” When dealing with the link between democracy and the public sector, Boix (2003) initially specifies that as the choice of a political regime depends on its distributive implications, the economic and fiscal consequences coming with a democracy or an authoritarian system must be different. In other words, his analysis is consistent with the results of the formal voting literature rather than the ones of the Chicago School. Under a non-democratic regime the size of the public sector should be small, a substantial part of the electorate being excluded from the decision-making process. So, independent of the type of economy, the level of redistributive spending should be minimal. A transition to democracy, on the contrary, should raise taxes and public spending. Under the same level of inequality ex ante, the level of inequality ex post has to be lower in a democracy than in a non-democracy, i.e. the extent of redistribution increases according to the second tenet of the formal voting theory. However, even on the basis of its findings about the determinants of democracy, the author links the amount of redistribution to the underlying economic and social structure. Given that inequality is more limited in a democratic system, the public spending directed to redistribute actually remains small. Moreover, the electoral turnout plays a fundamental role since only when the number of low-income voters that vote is significant, is the level of taxes and transfers

104 Matteo Cacciatore et al. high. In this way, the structural basis of democracy can reduce its redistributive inclinations and, at the extreme, the size of the public sector in this political system can be the same as in non-democracy. In representative regimes redistribution also takes place depending on the extent of economic development, that is modernization. Democratic institutions can take root in farmer economies characterized by few differences among individuals. In this case, the public sector does not grow as the redistributive tensions are practically non-existent. But democracies can also develop in industrialized societies where income equality and capital mobility are high. By creating an urban working class and the basis for an older population who cannot receive informal family help any longer, the industrialization process increases pressure for intragenerational but also intergenerational transfers, that is for increasing public spending. Development increases the size of the public sector, because it stimulates demands for public programs that in turn foster the economic modernization process. Finally, the volatility of income also affects the magnitude of the welfare state. If income fluctuation increases, then voters who are averse to risk may want to stabilize their economic position by raising public spending. This will happen either in a democracy in which the risk is concentrated among the worse-off or in an authoritarian system in which the risk is concentrated among the well-off. Empirical evidence is not uncontroversial. In order to estimate the empirical importance of economic and demographic variables relative to the political institutional ones in determining public policies, Mulligan et al. (2004) focus on a sample of 142 countries and on the democracy index from the Polity IV dataset. The timeframe of the cross-country analysis goes from 1960 to 1990 and, among the other control variables, the most important from our point of view seems to be the dummy for whether a country has been communist for more than a few years.10 This dummy allows the authors to try to separate non-democracy from central planning. In their comparison of democratic and non-democratic public sectors, they consider the spending policy group consisting of government consumption, education spending and social spending (pension and non-pension programs) as a percentage of GDP. None of these three variables are statistically different in democracies and non-democracies. However if one refers to the communist dummy it seems that totalitarian countries spend more of their GDP on education, but also on pension and non-pension programs. Though there are no significant economic or social policy differences between democracies and non-democracies, the results change when Mulligan et al. consider as dependent variables different policies that might affect public office competition, erecting political entry barriers. Following Tullock (1987), five policy steps are important for blocking entry: torture, death penalty, press censorship, regulation of religion and maintaining an army. Democracies are less likely to use these anticompetitive policies than non-democracies.

Democracy and welfare without welfare state 105 A different result is reached by Boix (2003). The author shows that at low levels of per capita income, i.e. low levels of economic development, the public sector appears small. If per capita income increases, then public revenue also rises both in democratic and in authoritarian political systems. However, under an authoritarian regime the public sector expands at a slower rate than under a democracy, perhaps because the latter has to satisfy the increasing needs of the modernization process. In other words, under the same level of per capita income, government is larger in a democracy, but only when modernization starts. This is due to the interrelation of economic development and political regime, but also to the one of economic development and democratic participation.11 The economic development can also be associated with a reduced importance of the agricultural sector as a percentage of GDP and an aging population. In this sense, the industrialization process and an increasing proportion of elderly people cause a rise in public revenue at a higher rate in a democracy. On welfare state expenditure, such as pensions, health care and unemployment benefits, Boix finds a growth only after the introduction of a democratic system. Showing the strong effect of democratization on core programs of the welfare state, this result rejects the theory according to which poor countries cannot sustain a strong state whereas the rich ones can. To conclude, differently from Mulligan et al., Boix’ empirical study underlines that a significant share of the public sector depends on the political regime in place. Lessons for China This section aims at drawing some conclusions about the possible interrelations of economic development and political democratic transition in China, an interesting country to study in East Asia, owing to its stable absence of democracy and rising level of economic development. Are these phenomena related to the rise of a welfare state? On one side, an optimistic view suggests that if a country is characterized by a steady economic growth, based on a deep process of industrialization and urbanization, a peaceful political democratic transition is possible. For instance, Barro (1996) suggests that economic liberalization leads to political liberalization. In the long run, we should thus expect a democratic transition in China. However, there are fundamental problems in China which may affect this outcome and make a democratic transition more difficult. First, the existence of massive internal differences, in particular huge income disparities between coastal areas and interior regions, which may lead to territorial conflicts and fragmentation (Boix 2003). The risk is that coastal areas may oppose democracy because this would lead to interregional redistribution towards the rural interior regions. However, it may also happen that an authoritarian regime will expropriate coastal areas. A

106 Matteo Cacciatore et al. pessimistic view would thus predict a separation of areas. Boix (2003) identifies the historical roots of non-democracy in China with the existence of a unified empire, characterized by non-mobile capital, high taxes and inequality of income. Thus, provided that income inequality does not substantially change, together with fragmentation of areas, a democratic transition will be difficult to implement. Second, China is characterized by several barriers to political competition (torture, death penalty, press censorship, regulation of religion) which makes its political regime highly non-democratic and very stable (Mulligan et al. 2004). Thus, economic liberalization may not be sufficient if political competition is not pursued through the elimination of these constraints. An additional transition is necessary, namely the abolition of these constraints to the political competition, to make outcome in democracies and non-democracies comparable. Only after that, a democratic transition may be possible. In other words, China seems to follow a “hard way” of development by introducing economic liberalizations while still being an autocracy. According to Giavazzi and Tabellini (2005), if the process were successful, better final outcomes are expected, but the process itself is very challenging. To conclude, we argue that, as suggested by Mulligan et al. (2004), in the first phase of economic development, i.e. after modernization, a rising welfare state (public expenditures and revenues) in China, such as the development of the public pension system, does not need a democratization process. However, in the long run this democratization process may be important for the economic development of the country.

Notes 1 The missing years are interruption, interregnum or transition years. 2 From 1948 we consider South Korea. 3 Ten of this 55 percent is part of a separate and more generous pension system for civil servants. Jackson and Howe (2004) estimate that only 45 percent of the urban workforce participates in the basic public pension system. 4 The pension scheme for rural workers is extremely small and almost entirely beneficiary-financed. 5 For an analysis of the financing options of the transition see Li and Li (2003) and Zhao and Xu (2002). 6 Data on inequality are taken from Deininger and Squire (1996), see Boix (2003). In the second sample, inequality refers to two indicators that are well correlated to the Gini index: the distribution of agricultural property and the quality of human capital. 7 Boix (2003) also finds that higher levels of human capital contribute to the democratization process. Agricultural societies do not seem to affect the democratic transition but they increase the probability of democratic breakdowns. The presence of an oil economy reduces the possibilities of democratization, in this way accommodating the paradox of wealthy dictatorship, and finally the diversification of productive activities either raises the likelihood of a democratic transition or reduces the one of a democratic breakdown.

Democracy and welfare without welfare state 107 8 With reference to the sample of the former European colonies, Acemoglu et al. (2005) show that fixed effects explaining the mentioned cross-sectional correlation are related to these historical variables such as settler mortality rates, the density of the indigenous population before colonization, the constraint on the executive at independence and the date of independence. 9 The identity of the elite is not important. 10 This variable takes a value of 1 if the country is considered communist by Kornai (1992), and 0 otherwise. China is one of the 26 communist countries in the sample, see Mulligan et al. (2004). 11 The nature of the political regime does not affect on its own the size of the government because the public sector is not larger under a democracy at all income levels. Starting from the lowest level of per capita income the public revenue is higher in an authoritarian system because, for example, the latter has to finance its repressive apparatus.

References Acemoglu, D. and Robinson, J. A. (2006) Economic Origins of Dictatorship and Democracy, Cambridge University Press, forthcoming. Acemoglu, D., Johnson, S., Robinson, J. A. and Yared, P. (2004) “From education to democracy?,” MIT, Department of Economics, Working Paper 05. —— (2005) “Income and democracy?,” MIT, Department of Economics, Working Paper 05. Barro, R. J. (1996) “Democracy and growth,” Journal of Economic Growth, 1: 1–27. Boix, C. (2003) Democracy and Redistribution, Cambridge University Press. Chen, V. (2004) “A macro analysis of China pension pooling system: incentive issues and financial problems,” mimeo, Hitotsubashi University. Giavazzi, F. and Tabellini, G. (2005) “Economic and political liberalizations,” Journal of Monetary Economics, forthcoming. Holzmann, R., Mac Arthur, I. W. and Sin, Y. (2000) “Pension systems in East Asia and the Pacific: challenges and opportunities,” World Bank Social Protection Discussion Paper series 0014, Washington, DC: The World Bank. Jackson, R. and Howe, N. (2004) “The graying of the middle king: the demographics and economics of retirement policy in China,” The Center for Strategic International Studies (CSIS), Washington, DC. Jacobs, D. (1998) “Social welfare systems in East Asia: a comparative analysis including private welfare,” Centre for Analysis of Social Exclusion, Sticerd LSE, case paper 10. Li, D. and Li, L. (2003) “A simple solution to China’s pension crisis,” Cato Journal, 23(2): 281–9 Lipset, S. M. (1959) “Some social requisites of democracy: economic developments and political legitimacy,” American Political Science Review, 53: 69–105. Mulligan, C. B., Gil, R. and Sala-i-Martin, X. (2004) “Do democracies have different public policies than non democracies?,” Journal of Economic Perspectives, 18(1): 51–74. Peltzman, S. (1980) “The growth of government,” Journal of Law and Economics, 23: 209–87. Stigler, G. J. (1970) “Director’s law of public income redistribution,” Journal of Law and Economics, 13: 1–10. Tullock, G. (1987) Autocracy, Boston, MA: Kluwer Academic Publishers.

108 Matteo Cacciatore et al. United Nations (2003) World Population Prospects: The 2002 Revision, UN Population Division. Wang, X. (2004) “China’s pension reform and capital market development,” China & World Economy, 12(3): 3–16. Wittman, D. (1989) “Why democracies produce efficient results,” Journal of Political Economy, 97: 1395–424. World Bank (1999) “Towards an East Asian social protection strategy,” Working Paper, Washington, DC: The World Bank. Zhao, Y. and Xu, J (2002) “China’s urban pension system: reforms and problems,” Cato Journal, 21(3): 395–414.

Websites http://www.cidcm.umd.edu/inscr/polity/ – Polity dataset.

5

Features and effects of corporate taxation on FDI Francesco D’Amuri and Anna Marenzi

Introduction This chapter aims at giving an overall picture of the existing interactions between the economic structure, the corporate tax system and the attractiveness of FDI inflows in a selected sample of Asian countries. The economic structure of a country can explain many of the features of its tax system; following economic growth, social and institutional changes and fiscal systems change. New tax handles substitute the old ones. The systems become more complex and tax administration and compliance more demanding. Furthermore, both the level and the structure of taxation affect the level of private saving and thus the volume of resources available for capital formation. Corporate income taxation (CIT) plays a strategic role in this respect. In particular, differences in corporate taxation across countries can affect the allocation of foreign direct investment (FDI) by driving the wedge between the post- and pre-tax rates of return. In a world where an increasing number of countries compete hard to attract FDIs, tax authorities try to encourage capital inflows by offering generous investment tax incentives. Consequently, the fiscal incentives have become a relevant aspect in the design of national tax policy. Generally, developing and transition economies rely mainly on profit-based incentives, including tax holidays and partial profit exemptions, which are particularly prone to aggressive tax planning (Owens 2004), while industrialized countries more often use capital-based incentives which are intended to reduce the cost of qualified capital. The sample of countries analyzed in this chapter comprises the two biggest developing countries (China and India) at the rushing stage of their catching up; two countries in a middle stage of development (Malaysia and Thailand) and, finally, two industrialized countries (Japan and South Korea). The six countries are characterized by a different degree of economic development and by a different level of maturity in their corporate tax systems. Also the role played by FDI in the economic performance of these countries has been different: Malaysia, Thailand

110 Francesco D’Amuri and Anna Marenzi and Korea have experienced high levels of foreign capital inflows for a long time. For different reasons, India and Japan have had a poor performance in the attraction of FDIs. Finally China has opened up its markets in the last two decades, experiencing a huge inflow of foreign investment. The chapter is organized as follows. We start with an analysis of the degree of economic development of the countries considered in the sample, the openness of their markets, the level of industrialization and their capacity to attract FDI. Then we review the main ideas in the literature about the effects of the systems of corporate income taxation and tax incentives on economic development and capital inflows. We then move to analyze in detail the corporate tax structures in the selected countries, distinguishing between the main features of the basic tax regimes and the complex systems of tax incentives. In the conclusion we bring the various themes together, evidencing how the different stages of economic development of the various countries are reflected in their respective corporate tax systems. Moreover some critical aspects of the design of a tax incentive policy, namely the risks of tax competition and the need of a comprehensive cost–benefit analysis, are stressed.

External trade and FDI inflows in the countries of our selected sample The economies of the six South and East Asian countries whose fiscal systems are analyzed in this book differ dramatically (see Figure 5.1, and, more widely, Chapter 1). One of them, Japan, can be considered one of the most industrialized countries in the world, characterized by a mature economy and a level of GDP per capita not comparable with those of the other countries, with the exception of Korea, which has experienced a strong and stable economic growth in the last 20 years. In the period 1975–2003 Malaysia and Thailand show economic performances similar in levels and growth rates and can be considered two countries at a medium stage of economic development. Finally, China and India differ substantially from the others by their dimensions (respectively 1.3 and one billion inhabitants) and for their low but rapidly increasing level of economic development. In 1975 China had the lowest level of income per capita among the six countries, but, thanks to a strong growth, reached India in the early 1990s and it is, in 2003, the greatest of the six economies in absolute terms. The six economies are characterized by a relevant level of economic openness. Japan is an exporter of high-tech goods and cars, with a trade/GDP ratio around 20 percent. Gross FDI inflows are high in absolute terms (US53 billion in 2004), but low if compared with the size of the economy (0.97 percent on GDP in 2004, Figure 5.2). Many elements could have prevented multi-national enterprises (MNEs) from investing

Constant 2000 international $

25,000

20,000

15,000

10,000

5,000

0 1975

1980

1985

1990

1995

2000

Year China

India

Japan

Korea

Malaysia

Thailand

Figure 5.1 GDP per capita, PPP, 1975–2003 (source: World Development Indicators, World Bank).

10 9 8 7 6 5 4 3 2 1 0 1975 ⫺1 China

1980

1985

1990

1995

2000

Year India

Japan

Korea

Malaysia

Thailand

Figure 5.2 Gross FDI inflows as percentage of GDP 1975–2003 (source: World Development Indicators, World Bank).

112 Francesco D’Amuri and Anna Marenzi massively in Japan. Lawrence (1991, 1993) assumes that the prevalence of corporate groups or keiretsu could constitute an impediment to FDI, since they usually discriminate against foreign firms. Also the particular conditions of the Japanese labor market could partially explain the low level of FDIs: high-skilled Japanese workers prefer to work in domestic enterprises since they offer them long-term employment and better training opportunities. Korea can be considered an export-led growth economy. At the beginning of its long period of economic development (early 1960s), the country was characterized by heavy protection, with high levels of tariffs and quotas on imported goods. This did not prevent the country from engaging in international trade: instead of removing the anti-export bias due to the high level of internal market protection, the government started a policy of export promotion. In particular, firms fulfilling the export objectives were subsidized with cheap credit by the financial system. At this stage exports primarily constituted labor-intensive goods such as clothing and footwear, with FDIs playing almost no role in the economy. As the economy grew, the country had to shift its trade policy towards more openness, removing tariff and non-tariff trade barriers. Moreover, the production moved progressively from labor-intensive goods towards higher levels of quality and style. Thanks to the increasing liberalization, FDIs have been playing an increasing role since the second half of the 1980s, with a clear boom ten years later (gross FDI inflows were equal to 3.8 percent of GDP in 1998, see Figure 5.2), interrupted only by the financial crises. Also Malaysia has industrialized rapidly in the last 30 years, with an economic structure initially based on the production of mineral and agricultural export commodities, now relying primarily on electronics. Malaysia has a sincere orientation to trade: the trade/GDP ratio has been increasing steadily during the past 30 years, reaching a quota of 220 percent in 2004. During this period the Malaysian economy has been characterized by a high level of gross fixed investment (43.1 percent of nominal GDP in 1997), which decreased dramatically after the financial crises (22.1 percent in 2003) due to a severe contraction of domestic and foreign investment. Reaching the old levels of private investment is a primary target for government economic policy, but the increasing competition for FDIs makes this goal difficult to achieve. FDIs have had an important role in Malaysian economic development, attracted by good infrastructures and a highly educated workforce (expense in education is the largest item of the federal budget). FDIs are concentrated in manufacturing and in the service sector, with many MNEs reinvesting profits in the domestic market. Efforts have been made in the last few years to attract more foreign investment, as underlined for example by the abandonment of the policy of reserving quotas of corporate equity on an ethnic basis, which scared some foreign investment in the past. Notwithstanding this, the

Features and effects of corporate tax on FDI 113 control operated by the Malaysian Industrial Development Agency (MIDA) remains strong. This agency has the task of approving inward investments on the basis of equity stakes, products and processes concerned, and financing. In the last few years FDIs in low productivity industries have been discouraged, while stronger incentives and reduced restrictions on investment inflows have been offset by the increasing Chinese competition in this field. Thailand, as many of the other economies analyzed in this book, has progressively lifted restrictions on trade in the last 40 years. During the late 1950s, the government started its effort of raising the level of domestic and foreign investment, without renouncing a high level of protectionism and a strong commitment on import substitution. Since then Thai trade policy has progressively approached a more free market view, which has been partially criticized by the government in charge since 2001. Above all, Thailand remains a very open economy, with a trade/GDP ratio steadily increasing and equal to 123 percent in 2003. Also FDIs have increased steadily in the last 20 years, reaching a maximum in 1998 (6.6 percent of GDP) and decreasing sharply afterwards because of the financial crises. The competitive advantage of the Thai economy has been traditionally on labor-intensive goods, such as textiles and shoes, but since the mid-1990s these sectors are suffering from the competition of lower-cost producers such as China, India and Vietnam. Export performances remain high thanks to an increasing role of the high-tech and automotive industries. China is experiencing a different path of economic development and market liberalization, started in 1979 with the first stage of transition from a planning to a market economy. Before 1979 no FDIs were allowed to enter the country, while the trade regime was characterized by the presence of a small number of foreign trade corporations with monopolies in the import–export of selected goods. The planned level of imports was determined in order to cover the difference between domestic demand and supply for certain goods, while exports were set at the level strictly necessary to finance imports (Ianchovichina and Martin 2001). In 1979, together with the introduction of the first reform oriented to trade liberalization,1 the establishment of four Special Economic Zones (SEZs) was approved. Foreign companies investing in these coastal areas were allowed to exploit tax holidays, physical infrastructures and low labor costs, together with the possibility of importing intermediate and capital goods duty free. Foreign investments started to arrive in China, concentrated in small-sized export-oriented businesses. In 1986 restrictions on FDIs were relaxed2 and tax incentives augmented, while local governments were recognized as playing a role in the attraction of foreign investments. Foreign companies’ activities started to concentrate in manufactures, while their importance was rapidly increasing until the Tiananmen Square incident. But in the early 1990s new successful provisions were approved

114 Francesco D’Amuri and Anna Marenzi in order to attract foreign investments, which reached double-digit growth. In 2004 China was the third largest exporter in the world, with foreign-invested enterprises (FIEs) accounting for 58 percent of export, according to Chinese customs data. FDIs reached in the same year the maximum of US$70 billion, while exports were concentrated in textiles (15 percent of total exports in 2004) and machinery and electrical appliances (41.8 percent of total exports). Finally, India has a poor record in FDI attraction. The country was one of the first in the world to establish an Export Processing Zone in 1965, but many elements have prevented it from becoming an attractive destination for foreign investments. Acharya (2001) ascribes the not particularly high industrial performances of this country to the anti-export bias of foreign trade policies, strict labor laws, poor infrastructure and bureaucracy. Henley (2004) attributes the low attraction of FDI to causes such as the ineffectiveness of local government policies, the low level of public investment in infrastructure and lobbying activities against liberalization undertaken by domestic enterprises. The first comprehensive pro-trade reforms were approved in the late 1980s, slowly abandoning the import substitution policies of the 1960s and 1970s. The 1991 “New Industrial Policy” removed the public sector monopoly in many sectors and relaxed many restrictions imposed up to that time on foreign investment. In particular, the threshold of 40 percent on foreign equity share was removed, while the automatic approval of foreign investments was progressively introduced in many sectors. The introduction of liberal policies did not have a strong impact on FDIs, which are still very low if compared with the size of the economy (less than 1 percent of the gross FDI/GDP ratio in 2004). FDIs are concentrated in power generation, mobile telecommunications and software industries, while they are low in laborintensive manufacturing sectors, due to the well-established policy of reserving some sectors for small-scale domestic industries.

Interactions between corporate taxes and FDIs: a reminder of the literature Host country tax policy may have a relevant role in attracting foreign direct investments. As long as location decisions made by MNEs are driven by after tax profit maximization, the structure of the corporate income tax and tax incentives offered by the host country can influence the inflow of investments, determining the wedge between before and after tax profits. If investments in different locations are characterized by the same expected rate of return, a reduction of the tax burden borne by MNEs in a particular country could determine the final investment location decision. However, tax policy alone cannot be very effective in influencing FDI location choice in the absence of other elements that determine the expected rate of return of a company, such as good infrastructure, prox-

Features and effects of corporate tax on FDI 115 imity to final markets, stable social and political environment, good labor market conditions and many other economic fundamentals that could influence the desirability of an investment location. Economic theory suggests that tax incentives designed to encourage FDIs are justified by the presence of positive spillover effects of foreign investment on the host country economy. FDIs are in fact expected to raise the competitiveness of the host country economy by increasing the level of capital, transferring knowledge and enhancing productivity. Moreover, regional-specific incentives can be a tool of regional development, attracting investments into less developed areas. All these positive spillovers are not included in the objective function of MNEs when choosing the optimal level of investment. This therefore brings a sub-optimal level of FDI. To prevent this market failure, host country governments are then interested in offering tax incentives in order to subsidize the inflow of foreign investments. It is not often underlined in the literature that special tax incentives and corporate tax reductions targeted on foreign investments are justified, on this ground, only if foreign enterprises differ significantly from domestic ones (Blomstrom and Kokko 2003). In fact, foreign enterprises cause positive spillovers on the host economy as long as they can improve the economic specialization of the host country by producing new products or being characterized by elements such as higher productivity, better management, greater international orientation or more advanced technology. In the absence of these elements, preferential tax treatment for foreign enterprises would only distort competition and put local companies at a disadvantage. In this case, investmentsubsidizing policies are still justified by the aim of improving economic performance, but they should not differentiate between foreign and local investors. Host countries aiming at a reduction of the tax burden borne by income generated by FDIs have several options. The first possibility is a reduction in the statutory corporate tax rate level. It enables the government to increase the after tax rate of return, stimulating investment but letting the market define the most profitable investments without governmental influence. The reduction could be targeted on a particular source of income (foreign or domestic) or on a particular sector of activity. In this case administrative and compliance difficulties arise together with the possibility of tax avoidance. Another shortcoming of a reduction in corporate rates relies on the fact that the incentive favors indiscriminately all investments, even those made before its introduction. Such revenue loss has no effect in raising the investment level. A tax holiday is another instrument widely used to attract investments. Under a corporate tax holiday, newly established, qualifying firms are exempt from paying corporate income tax for a specified period; usually, once the basic tax holiday has expired, companies pay a reduced tax rate for additional years. Tax holidays have some advantages that make this

116 Francesco D’Amuri and Anna Marenzi form of incentive particularly attractive for developing and transitional countries; however, as it is well known, their costs might well outweigh their benefits as a means of attracting foreign investment (see, among others, Chalk 2001; Mintz 1990; Owens 2004; Zee et al. 2002). Tax holidays provide, with respect to other types of tax incentives, a simple regime both for foreign investors and for tax authorities. They allow qualified investors, in the first years of operation, to avoid interactions with the tax authorities, while, on the other hand, they relieve tax authorities from the burden of administering them. In addition, they are neutral in their impact on the use of productive factors (Wells and Allen 2001). However, CIT holidays provide large benefits if the company makes profits in the early years of operation. In this respect, they primarily favor short-term investments, which are often undertaken in so-called footloose industries characterized by companies that quickly disappear from one jurisdiction to reappear in another, and tend to penalize long-term projects with large start-up costs in the initial production years. Furthermore, tax holidays create competitive distortions between old and new firms, pushing them to redesign existing investments as new investments or to form a new company after the holiday expires. They are also likely to have a direct negative effect on fiscal revenues and create significant opportunities for aggressive tax planning (OECD 2003). The impact of tax holiday on the return of investment is quite difficult to evaluate since it depends on technical aspects of its design, that is, the starting period of the holiday, the tax treatment of losses incurred during the holiday period, and the duration of the incentive (for a rigorous demonstration of this result, see Mintz 1990). Other forms of incentives, such as investment allowances and tax credits, aim at lowering the corporate tax burden by a reduction in net costs of investment. These kinds of incentives should be preferred since they are targeted just on increases of the capital stock, and thus do not provide windfall gains to existing capital holders as in the case of a reduction in the corporate tax rate. Besides, they offer less possibility of tax avoidance. As with tax holidays, investment allowances and tax credits are more attractive for those companies that are incurring profits when the incentive is introduced. Start-up businesses or newly established long-run activities (that should be the main target of an investment incentive policy) are likely to incur losses in the first years of activity. For them, investment incentives would then be irrelevant. Finally, it has to be noted that the effectiveness of tax incentives in lowering effective tax rates borne by foreign enterprises depends on the interactions between the host and the home country tax systems. In the case of full exemption for corporate income earned abroad, tax incentives will be effective in lowering the total tax burden borne by MNEs. They could instead be ineffectual in the presence of tax credit schemes, in which a

Features and effects of corporate tax on FDI 117 decrease in corporate taxes paid by a MNE abroad would be offset by an equal increase in the home country tax liability. In order to prevent this, many governments provide “tax sparing” credits for investments in developing countries. Under “tax sparing” agreements, companies investing abroad are allowed to claim tax credits computed as if no tax incentives existed in the foreign country. In this way tax breaks offered by capitalimporting countries can enhance the after tax profitability of foreign investors (Hines 1998). Several empirical works have tried to assess quantitatively the impact of taxation on FDI inflows (for recent surveys of empirical findings: De Mooij and Ederveen 2003; Hines 1999). Most of the empirical work has been focused on the US, for reasons of data availability. In particular, investment decisions made by US companies investing abroad or by foreign enterprises investing in one of the 50 US states (corporate tax legislation differs among US states) are taken into consideration in the attempt to estimate quantitatively the impact of taxation on FDI inflows. As noted earlier, tax policy effectiveness in attracting FDIs depends on the interaction between home and host country tax legislation. Therefore, empirical findings obtained analyzing only US data should not be generalized. Empirical analyses use time-series and cross-sectional estimation of FDI response to changes of after tax rates of return. In both cases the main limitation relies on the fact that the variation in effective tax rates could be correlated with other important omitted variables, which are difficult to control for (Hines 1999). This makes difficult the interpretation of tax/FDI elasticities. Notwithstanding these limitations, evidence does support the hypothesis of a negative tax elasticity3 of investment, even if a consensus on its value is far from being reached. De Mooij and Ederveen (2003) construct a meta sample including the estimates of different empirical works on the effect of tax burden on FDIs. They homogenize different definitions of tax elasticity of investment across the 25 studies analyzed. It is equal to ⭸ ln(FDI)/⭸/t and measures the percentage change in FDI induced by a percentage change in the tax rate. The majority of the observations rank between ⫺5 and 0, giving support to the hypothesis of a positive effect of corporate tax cuts on FDIs. Moreover a certain consensus has been reached on the fact that FDI financed by retained earnings of foreign affiliates is more responsive to changes in the host country tax rate than FDI financed by transfers of parent company funds (Hines 1999). Another result common to many empirical studies is that the responsiveness of FDI flows to tax changes is increasing over time, consistent with the global reduction in non-tax barriers to FDI, including the suppression of investment and currency controls (OECD 2001). However, given the uncertainty on the quantitative impact of tax incentives on foreign investment, elasticity estimates should still be used with caution when trying to carry out cost-effectiveness analysis of a given tax

118 Francesco D’Amuri and Anna Marenzi policy in attracting FDI. Moreover, until now, all the empirical works have concentrated on the effect of overall corporate tax policy (i.e. average or marginal effective tax rates borne by companies) on FDI. Given that a tax cut may have a beneficial impact on foreign investment, there is no evidence on how host countries should reduce taxes in order to obtain the maximum beneficial impact in terms of real capital inflows. As noted earlier, in fact, a decrease in effective tax rates can be obtained by means of different options.

Corporate taxation and tax incentives in South and East Asia The systems of corporate taxation The current corporate tax systems of the selected countries are the result of several successive fiscal reforms mainly started in the beginning of the 1990s and, for certain countries, still underway. Generally, most countries have realized a consistent reduction in statutory tax rates in the last decade, while only partial efforts of broadening tax bases have been made. Table 5.1 summarizes the main features of the corporate tax systems of the countries analyzed in the sample. In considering the information reported in the Table 5.1, it is first of all important to point out that, generally, for tax purposes, a domestic company is liable to be taxed on its worldwide income (except Malaysia; see below), while the tax liability of a foreign company is normally limited to host-source income. Moreover, with regard to the treatment of foreign source income, in order to avoid double taxation, almost all countries adopt the “residence” or “worldwide” approach in computing tax liability, combined with tax credits. Under this system, foreign source income is subject to home country taxation, but a credit or deduction is allowed for taxes paid to the host government. The foreign tax credit is typically limited to the home country tax liability on foreign source income. In Japan and Korea any remaining excess of tax credit can be carried forward for crediting in succeeding years (three years in Japan and five years in Korea). In this respect, Malaysia represents a special case: income arising from foreign sources and received by a resident company4 is not taxed at all (“territorial” system). Concerning the structure of the corporate tax rates, Malaysia, Japan and Korea have a graduated structure while India, China and Thailand have adopted a flat rate. Most countries have a statutory rate of around 30 percent at the national level (see Table 5.1). In India, the rate depends on the nationality of the firm: for foreign companies it is set at 41 percent against 35.875 for domestic ones.5 In China, different tax codes are in force for domestic and foreign enterprises, even if, after WTO accession, there are increasing pressures to adopt a unified legislation. The national

Dividends are no longer taxed in the hand of recipient equity shareholders but subjected to DDT if distributed Short-term: CIT rate Long-term: 20.5%

Fully/partially excludede

20%

CIT rate

5 years carried forward

Inter-company dividends

Dividend withholding taxes

Capital gains

Treatment of losses

30%

28%b

Carried forward indefinitely

CIT rate

5 years carried forward

CIT rate

Included as a part Fully/partially of the taxable excludede income Included as a part 10% of taxable income for PIT

Thailand

Malaysia

Fully/partially excludede

27%d

South Korea

CIT rate surtax CIT rate of 5% on gains from land or similar properties 5 years carried 5 years carried forward; 1 year forward carried back

Partially included Included as a as a part of part of taxable taxable income income for PIT for PIT or 20%f

Fully/partially excludede

30%c

Japan

Notes a The tax rate of 35.875 is levied on retained earnings; dividends are taxed under the dividend distribution tax (DDT) at the rate of 12.81. Moreover, both resident and non-resident companies are liable to pay tax on their book profits where the tax liability of the year is less than 7.5% of the adjusted book profit; the tax rate is fixed at 7.688. b Companies with paid-up capital of RM2.5 million and below are taxed at a rate of 20% on chargeable income of up to RM500,000. The CIT rate on the remaining income is 28%. In the case of petroleum companies, the tax rate is 38%; different rates are applied to insurance companies. c If the corporation’s paid-up capital is more than YEN100 million the tax rate is 30%; otherwise, on taxable income up to YEN8 million a tax rate of 22% is levied; excess over YEN8 million is taxed at 30%. d The tax rate of 15% is levied on taxable income less than WON100 million; the excess is taxed at 27%. e Under specific conditions on the proportion of shares in the payer company and, in some cases, on types of subsidiaries. f If dividends are taxed as separate income, otherwise they benefit from a tax credit equal to 10% of dividend income.

Source: KPMG (2003) and others.

Domestic company: 35.875%a Foreign company: 41% Fully/partially excludede

33% (state tax of 30% and local tax of 3%)

Standard CIT rate

Business losses: 8 years carried forward Capital losses: carried forward indefinitely

India

China

Table 5.1 The main features of the corporate tax systems

120 Francesco D’Amuri and Anna Marenzi CIT rate (formally equal for foreign and domestic enterprises) is 30 percent plus a local surtax of 3 percent. However, the state rate is reduced to 24 percent for foreign investment enterprises (FIEs) operating in coastal regions; the rate even goes down to 15 percent for FIEs located in one of the special economic zones (as Shenzhen and Shanghai); moreover the local tax of 3 percent may be waived or reduced by the local government. As a result, generally, the Chinese domestic companies are penalized with respect to foreign-funded companies.6 Among industrialized countries, Japan has a low statutory rate; the picture substantially changes when corporate taxation levied by the central government is combined with local taxes to determine the overall statutory rate. Given the graduated rates in the calculation of both corporate and business taxes and the different local tax rates, the “all-in” statutory rate varies in Japan within the range of approximately 39 to 43 percent. Most of the countries in the sample adopt a broadly similar definition of taxable income; although certain relevant differences may be noted in relation to the types of deductions allowed, the amount of deductible expenses, and the types and the amounts of exemptions. In some cases, special rules in determining tax liabilities vary according to the size, location and industry of the companies (in particular in China, Malaysia and Korea). The depreciation system differs in many aspects from country to country. The Chinese depreciation system is calculated on the straight-line basis; accelerated depreciation may be conceded in a few specified circumstances. In India depreciation is calculated on the declining-balance method. The general rate of depreciation for plant and machinery is 25 percent. Additional depreciation of 15 percent on new machinery and plant is allowed. The higher rate of depreciation was initially adopted to offset the negative effect of the high corporate tax rate on internal accrual of resources for replacement and modernization. However, the cut in the tax rate realized in the last decade, and a new reduction currently under discussion, make less justified generous depreciations. In the absence of adequate profits, unabsorbed depreciation on both tangibles and intangibles can be carried forward indefinitely. Any accounting method of depreciation can be used under the Malaysian and Thai systems; in both countries, accelerated depreciation may be allowed for accessories used in research and technological development. The Japanese corporations may select either the straight-line method or the declining-balance method depending on the type of asset;7 special depreciation by means of either increased initial depreciation or accelerated depreciation is available for corporations in relation to specific IT-related assets and R&D-related machinery/equipment. Most of the countries permit losses to be carried forward for a maximum period of five years (China, Thailand, Korea, Japan), while Malaysia allows unabsorbed losses to be carried forward indefinitely.

Features and effects of corporate tax on FDI 121 Under the Indian income tax system rules change depending on the nature of losses: the operating business losses are allowed to be carried forward for eight years; capital losses arising from depreciation are carried forward indefinitely.8 In all countries of the sample, excluding Malaysia, inter-corporate dividends are partially or fully exempt from corporate income tax for the resident enterprise or company. The exemption is granted under specific conditions prescribed by the tax laws with respect to the proportion of the shares of subsidiaries owned by their company. For example, in Thailand and in Japan inter-corporate dividends are fully excluded from taxable income if the corporation owns more than 25 percent of the shares of the domestic corporation which pays dividends;9 otherwise, the amount excluded is reduced to 50 percent. In Korea, in addition to the proportion of shares of subsidiaries owned by the company, the amount of dividends excluded depends on the type of subsidiary (listed or non-listed on the Stock Exchange) and whether the company receiving dividends is a holding company or not. Generally, capital gains of corporations are included in ordinary taxable income and are subject to taxation in full as they are realized; in certain countries they are charged under corporate tax, but at a different rate. This, for example, applies to the Indian tax system, which uses different rules depending on whether they refer to short- or long-term capital gains: short-term capital gains are taxed at the normal domestic rate of 35.875 percent; long-term capital gains10 are taxed at the rate of 20.5 percent. In Japan, capital gains from short-term transactions of land carry an additional special tax burden (surtax of 5 percent). Finally, in most countries, the system of integrating personal and corporate taxation is fairly conventional, adopting essentially the “classical” model. Three countries (China, Thailand and Japan) apply final withholding tax on dividends paid to domestic individuals (the withholding tax rate varies from 10 to 20 percent). In order to mitigate the burden of double taxation, Japanese taxpayers may either benefit from a tax credit equal to 10 percent of their dividends if dividends are taxed as part of aggregate income, or, if dividends are taxed as separate income, pay a withholding tax at the same rate as capital gains and interest (20 percent). With respect to the classical model, India and Malaysia represent important exceptions. Under the Indian system, the company paying the dividends is subject to a dividend distribution tax (DDT) of 12.5 percent plus a surcharge of 2.5 percent on DDT; the tax rate on retained earnings is the standard CIT rate (35.875 percent). Dividends are exempt from income tax in the hands of shareholders, irrespective of their residential status. The taxation of companies in Malaysia is based on a full imputation system where the shareholders are taxed on the gross dividends at their own respective tax rates and are given full tax credit in respect of the tax deducted at source from the company.

122 Francesco D’Amuri and Anna Marenzi Of course, depending on the interactions between the different features of the corporate tax system, effective tax rates may differ substantially from the statutory rates. The differences depend on the determinants of the tax base and on the tax incentives granted by the tax systems.11 The erosion in the tax base is a serious problem for most countries, whether due to generous deductions and exemptions or to heavy use of a large number of multifaceted incentives. For example, in India, for manufacturing companies, the erosion in the tax base measured by the difference between the statutory and the effective corporate tax rate corresponded to about 22 percentage points in 1996–7 (43 against 21.36 percent) and about 13 points in 2002–3 (36.75 against 23.53 percent).12 Evidence on the distortions caused by the incentive regimes in force in Malaysia and Thailand in 1999 are reported by Chalk (2001). In the absence of any incentive, the effective tax rate was 30 percent in Malaysia and 46 percent in Thailand; the incentives (in the form of tax holidays and indirect tax concessions) reduced the effective tax burden by 8 percent points in Malaysia and by 39 percent in Thailand.13 Corporate tax incentives Following the standard classification (see, among others, Owens 2004 and Zee et al. 2002), corporate tax incentives can be broadly divided into two main categories: profit/income based and capital investment based. The first category includes: exempted profits (tax holidays), profits taxed at a lower nominal rate than the standard CIT rate, and losses carried forward or backwards. The second type of incentives are intended to reduce the cost of capital through investment and reinvestment allowances, investment tax credit and accelerated depreciation. Tables 5.2, 5.3 and 5.4 compare in some detail corporate tax incentives across the selected countries, distinguishing among the various incentive categories. Profit/income-based incentives TAX HOLIDAYS

As previously mentioned, all six countries, excluding Japan, adopt tax holidays, notwithstanding their numerous shortcomings. Furthermore, the reforms designed and implemented to mitigate the adverse impact of the Asian financial crisis of 1997 have reinforced the use of such incentives, rather than reduce it, as suggested by international bodies (see among others, OECD 2003 and Wells and Allen 2001). It is important to note that, in China, foreign investment enterprises enjoy tax holidays (and some other policy incentives) denied to local firms; on the contrary, in India, Malaysia and Thailand, tax holidays are

Refund of 40% (extended to 100% for exportoriented or technologically advanced enterprises) of tax already paid on the reinvested amount Double deduction of R&D if these expenditures exceed previous years by over 10%

– 2 years at 100% plus 3 years at 50% (Groups A, B, C) – Further reduction of 50% of CIT for 1 to 3 years (Group B); – Further 15/30% reduction for 10 years (Group C); – 5 years at 100% plus 5 years at 50% (Group D) Dividend incomes distributed during holiday period are exempt Special economic zones (SEZs): 15% for FIE and FE Other designed zones (ETDZs): 24/15% for 3 years Technologically advanced FIE: 10% Allowed in a few specified circumstances

Additional depreciation of 15% on new machinery and plant for new industrial undertaking and to existing one if the installed capacity increases by 25% or more in the same year

Dividend incomes distributed during holiday period are exempt

– 5 years at 100% plus 5 years at 30% (Group A); – 5–10 years at 100%, following years at 30/50/100%, further year at 50% (Group B); – 10 years at 100% rate of exemption (Group C)

New industrial undertakings located in specific backward states and districts (Group A); companies engaged in R&D, commercial production, developing and building and project, and various other activities (Group B); power generating and distributing companies, enterprises providing communication services, infrastructure facilities (Group C)

Newly established FIE with a term of not less than 10 years and engaged in productionoriented activities (Group A); technologically advanced and export-oriented enterprises (Group B); enterprises engaged in farming, forestry or located in remote and economically underdeveloped areas (Group C); port and wharf construction Sino-foreign joint ventures (Group D)

Sources: Chalk (2001), Fletcher (2002), KPMG (2003), UNCTAD (2000).

Other

Reinvestment allowance

Accelerated depreciation

Reduced CIT rate

Dividend

Tax holidays Duration and rate of tax exemption

Sector qualifying for incentives (not exhaustive)

India

China

Table 5.2 Summary of corporate tax incentives in China and India

ITA of 25% of the costs of installation infrastructure facilities for 10 years; ITC of 25% for ROH buildings costs for 10 years

AD for cash registering machines and machinery and/or accessories used in R&D technology

ITA of 60% of the qualifying capital expenditure to a maximum of 70% of the statutory income (5 years); more generous ITA (100% for 5 years) are allowed for specific activities (industrial adjustment allowance, infrastructure allowance)

AD of computer, technology and environmental protection equipment, and capital exp. on approved agriculture projects

15 years of 60% of expenditure on a factory, plant, machinery or other apparatus used for qualifying project

Double deduction for promotion of exports, publicity and advertisements, approved R&D expenditure and other specific expenses

Investment tax allowance and credit

Accelerated depreciation

Reinvestment allowance

Other

Sources: Chalk (2001), Fletcher (2002), KPMG (2003), UNCTAD (2000).

SMEs and companies listed on the Stock Exchange and on the Market for Alternative Investment: 20–25% Regional operating headquarters (ROH): 10%

Offshore companies in Labuan: 3% Operational headquarters companies: 10%

Reduced CIT rate

Double deduction of cost of transportation, electricity and water supply expenses

Excluded from taxable income

Excluded from taxable income

Dividends distributed during holiday period

5 years carried forward after the end of holiday period

3–8 years at 100%, plus 5 years at 50% for enterprises located in less developed regions

Not permitted to be carried forward after the end of holiday period

5–10 years at 70–100% of CIT reduction of tax exemption, depending on location, specific activities or industries

Priority activities (use of domestic resources, create employment opportunities, develop infrastructure; conserve natural resources, strengthen industrial and technological capability; develop basic and support industries) and companies in certain geographical area

Companies involved in promoted activities, located in certain promoted areas, that provide intermediate goods, or are of national and strategic importance (heavy capital investment and high technological projects); contract and R&D companies, export-oriented manufacturing

Treatment of losses during holiday period

Tax holidays Duration and rate of tax exemption

Sector qualifying for incentives (not exhaustive)

Thailand

Malaysia

Table 5.3 Summary of corporate tax incentives in Malaysia and Thailand

Full exemption for first 7 years; 50% for next 3 years

Advanced technology and high-technology service businesses that support domestic industry; companies located in Korea’s Free Trade Zone and Customs Free Zone or designated Foreign Investment Zone (FIZs)

Total R&D expenditure; increase in R&D expenditure; improving technological infrastructure; IT investment

Japan

Note a Incentives reported on the table refer only to tax incentives for FDI. In addition to these specific industrial incentives, the Korean tax system also provides various tax abatements that are available to all companies including a lower corporate tax rate for SMEs, tax credits for investment in machinery or energy saving technologies, tax holidays on withholding on dividend distributions, tax credits to relocate companies to less congested areas, tax deferrals for investment reserve provisions, and others.

Sources: Chalk (2001), Fletcher (2002), KPMG (2003), UNCTAD (2000).

Tax holidays Duration and rate of tax exemption Investment tax credit

FDI qualifying for incentives

South Koreaa

Table 5.4 Summary of corporate tax incentives in South Korea and Japan

126 Francesco D’Amuri and Anna Marenzi granted to both domestic and foreign companies, but, generally, additional benefits are limited to foreign enterprises. Differential regimes apply in Korea: tax holidays are explicitly targeted on FDI and are limited to foreign investment enterprises; less generous tax holidays are allowed for domestic small and medium sized enterprises. As shown in Tables 5.2 to 5.4, in almost all countries tax holidays are used to support priority, hi-tech and pioneer industries.14 In addition, incentives may be adopted in different countries to support more specific sectors and/or geographical areas or to encourage investments by exportoriented firms. For example, in China, tax exemption is granted mainly to export-oriented manufacturing; in Thailand the incentive regime is explicitly targeted by geographical area and is progressively more generous the more underdeveloped is the region; companies that benefit from tax holidays in Korea are high-technology service businesses and different types of advanced technology FDI. There are important differences across countries in terms of holiday periods and percentage of statutory CIT rates exempted. In China, a newly established foreign investment enterprise is eligible for a tax holiday as long as it is engaged in production or business operations for the specific minimum period of ten years. The standard holiday entails a tax exemption for the first two profit-making years15 and allowed a 50 percent reduction in the third to fifth year, upon approval by the appropriate authority; further reduction may be allowed for technologically advanced and export-oriented firms and for enterprises engaged in farming, forestry and other low-profit industries or that are located in remote and economically under-developed areas (15–30 percent reduction in state CIT rate for ten years). Finally, a ten-year tax holiday (five years at 100 percent plus five years at 50 percent) is granted to Sino-foreign joint ventures engaged in port and wharf construction. An initial period of tax holiday, followed by rebates at a decreasing percentage in later years, is allowed to Indian companies. The full tax exemption period varies from sector to sector (see Table 5.2). In India, the negative impact of tax incentives, in general, and of tax holidays in particular, has been well documented in the numerous reports of committees, task forces and study groups. The resulting common view is the need for a new redesign of Indian corporate tax where, specifically, the generous tax holiday should be progressively reduced; however, the fiscal policies implemented by the government in the last few years have only partially gone in this direction. In fact, rather than minimizing them, the recent coalition governments have gone about proliferating tax incentives that complicate the tax system, creating a wide wedge between the nominal and effective corporate tax rates. A company granted pioneer status16 in Malaysia can choose to receive an income holiday of 70 percent of statutory income for a period of five years. The benefit is partially reduced as unabsorbed losses are not permit-

Features and effects of corporate tax on FDI 127 ted to be carried forward to the post-pioneer period. Furthermore, a Malaysian company granted the pioneer status incentive is not eligible for investment tax allowance. More generous tax holidays (100 percent reduction in standard CIT for a period of ten years) are allowed for high technology companies, strategic projects, manufacturers of specific machinery and equipment, and R&D companies. Companies with Multimedia Super Corridor status17 enjoy similar incentives or an investment tax allowance of 100 percent for five years. Thai and foreign companies satisfying a minimum requirement of investment capital, a minimum Thai shareholding in the project company and minimum levels of products destined to export (not less than 80 percent of its total sales) enjoy progressively more generous tax holidays the more underdeveloped the region where the company is located. The holiday CIT periods range from three up to eight years (extended for an additional five years with 50 percent of CIT rate reduction). Losses may be carried forward and deducted as expenses for up to five years after the end of the income tax holiday period. Furthermore, companies engaged in priority activities (agricultural products, technological and human resource development, public utilities and infrastructure, environmental protection and conservation, and other targeted industries) are eligible for a CIT exemption of eight years. Tax holidays are not granted in Japan. However, there are certain exemptions from fixed assets tax and business tax for local purposes. Also, a special taxable income deduction has been granted for qualifying financial institutions operating in a designated financial operation zone in Okinawa. While tax holidays have been strongly used in the past in Korea, currently such incentives are only granted to advanced technology FDI, new high-technology service businesses that are expected to support domestic industries and newly established small- and medium-sized enterprises. The period of time during which tax exemptions and reductions stay in effect for FDI has been recently extended from eight to ten years (full exemption for the first seven years and 50 percent reduction for the next three years). A reduction of 50 percent in the CIT rate for the first six years is granted for newly established SMEs.18 REDUCED CIT RATE

As shown in Tables 5.2 and 5.3, preferential corporate income tax is allowed in China, Malaysia and Thailand. Foreign investment enterprises (FIEs) and foreign enterprises (FEs) in Chinese Special Economic Zones (SEZs) are taxed at the reduced rate of 15 percent; other reduced tax rates are allowed for technologically advanced FIE and for companies located in designated zones. In Malaysia, a tax rate well below the standard CIT rate is accorded to offshore companies in Labuan and to companies that set up operational

128 Francesco D’Amuri and Anna Marenzi headquarters in the country (10 percent for five years with the possibility of renewal for a further five years). Similarly in Thailand, new tax and non-tax incentive packages were introduced in 2002 to attract foreign Regional Operating Headquarters that provide services, including management, technical support, research and development and training, to subsidiary companies or branches in the host country. Tax incentives include a preferential tax rate of 10 percent on net profits. It should be noted that, with increased globalization, many countries have been adopting new tax policies for multinationals to establish headquarters and financial and trading operations in their jurisdictions; these policies seem to be have become common in the last few years in many developing countries (see Mintz 2004). Capital/investment-based incentives Particularly generous investment tax allowances (ITA) are authorized in Malaysia. Companies that are not granted pioneer status are eligible to apply for the investment tax allowance instead of a tax holiday. A company granted ITA is permitted to offset an amount equal to a percentage of the capital expenditure incurred on a factory and the provision of plant and machinery against its taxable profits. Any ITA that cannot be utilized against taxable income may be carried forward indefinitely or offset against future taxable income derived from the same project. In Korea research and development expenditures are favored by the presence of a tax credit allowance: companies may choose the larger amount between a flat tax credit (15 percent of expenses for technology and human resources development) or an incremental tax credit (50 percent of the expenses exceeding the average amount disbursed during the preceding four business years). Less generous tax credits (ranging from 3 to 10 percent of the investment amount) are allowed for specific investments.19 Corporate tax incentives in Japan are mainly offered through investment tax credit for R&D expenses. In this respect, and for strengthening the global competitiveness of Japanese business, a new proportional R&D tax credit was introduced by the 2003 tax reform as an alternative to the previous one. Corporations are granted a tax credit of 8 percent plus 2 percent (applicable only for FY 2003 to FY 2005) of the amount of R&D expenditures.20 Moreover, a tax credit (10 percent) or special allowance for accelerated depreciation (50 percent) is applied for qualifying IT investment. Two different forms of reinvestment incentive, reinvestment tax refund and investment tax allowances, are employed in China and Malaysia. China offers to foreign investment firms a tax refund of 40 percent (100 percent if reinvested in an exported-oriented or technologically advanced enterprise) on profits that are reinvested for at least five years to increase the capital of the firm or launch another firm. A more traditional reinvestment allowance is granted in Malaysia. Indeed, companies engaged in

Features and effects of corporate tax on FDI 129 manufacturing or agricultural activities are eligible for reinvestment allowance for qualifying expenditures in plant, machinery and industrial buildings. Finally, double deduction of specific expenses are available for corporations in Malaysia and Thailand (see Table 5.3).

Tax policy considerations As mentioned above, FDIs of the countries analyzed vary greatly in their level. Moreover they are also different in their composition. In particular, there are different types of FDI, according to the economic rationale driving them: resource and asset and capabilities-seeking investments, aiming at securing specific resources (raw materials, highly skilled workforce, intangible assets) present in the host country; market-seeking investments, which have the objective of producing locally for the host country’s market instead of serving it with exports; finally, efficiency-seeking investments, aiming at cost minimization. These kind of investments typically exploit the presence of very low labor costs and less regulation in certain developing countries. When assessing the sensitiveness of FDIs to tax considerations, FDI typologies play an important role: resource and asset and capabilities-seeking investments are obviously the least sensitive to fiscal considerations; even if tax incentives are considered not very effective on market-seeking investments, the growing number of regional free trade areas could generate fiscal policy competition between member countries in order to attract headquarters and production plants. Finally, fiscal incentives could have a major impact in attracting efficiency-seeking investments: given the aim of cost minimization of these investments, a reduction of the tax burden could enhance the attractiveness of a particular country. Japan and Korea are characterized by the presence of market and capabilities-seeking FDIs (as typical in developed countries). The minor role of tax incentives in these countries is then justified by the fact that these kinds of investments are less sensitive to tax considerations. Incentives are offered only to enterprises that could trigger positive spillovers on the domestic economies (typically IT investment). More complex is the situation in Malaysia and Thailand, where FDIs have contributed to economic growth permitting the achievement and maintenance of very high levels of investment. While FDIs in these countries were typically driven by efficiency considerations, tax incentives played a major role in their attraction. But economic development has boosted costs and the initial comparative advantage in labor-intensive goods has progressively been lost. This has forced Malaysia and Thailand to change their economic specialization, moving to sectors characterized by higher technological content. Tax incentives are in fact now targeted (especially in Malaysia) on IT and R&D enterprises. In this case incentives cannot be the only policy used for the attraction of investments, but they should be only an

130 Francesco D’Amuri and Anna Marenzi element of a comprehensive industrial policy including long-term investment in infrastructure and human capital. The last two countries analyzed have similar economic characteristics, and are attracting (or trying to attract) primarily efficiency-seeking investments. In China, thanks to the strong comparative advantage in laborintensive goods, tax incentives play a role in attracting efficiency-seeking investments, cutting further costs. On the other hand, the high potential of the Chinese consumer market could cause, in the medium term (if strong economic growth continues as expected), a surge in marketoriented investments that need not be subsidized with tax incentives. Finally, India has an economic structure similar to the Chinese one, but no success in attracting foreign investment inflows. The home bias of the domestic tax system (underlined by a substantially higher statutory corporate tax rate for foreign enterprises) and its complexity are just two elements of an economic policy that does not show a clear will to make the country attractive to foreigners. Fiscal reforms should be part of a wider process of normative simplification and market liberalization. Particularly advisable in this perspective is the abolition of the discrimination in tax rates between domestic and foreign enterprises and a comprehensive simplification of the overall fiscal system. Coming back to more general considerations, two more points have to be noted. By now most of the empirical and theoretical work on tax incentives and FDIs has analyzed only the beneficial side of the problem. In this direction many studies (see the section on interactions between corporate taxes and FDIs) have tried to quantify the tax elasticity of investment or the extent of the positive spillovers generated by foreign investment on the host economy. However, there is a lack of data on the opportunity costs of these kind of incentives. In designing an expensive policy of investment incentives, this kind of consideration should be kept clear in mind, since incentives often imply high costs for scarce public funds. Incentives offered to a company for an investment that would have taken place even without it constitute a clear revenue loss for the host country. Moreover, tax incentives usually create loopholes in the tax system, giving a chance for aggressive tax planning. Incentives not justified by the presence of positive externalities in foreign investment would also introduce costly distortions in the host economy, disadvantaging existing domestic firms (Zee et al. 2002). In a cost–benefit analysis, positive results of a tax incentive policy should be weighted against opportunity costs, that is, gains that would have occurred with alternative uses of public funds (for example, policies enhancing infrastructures and human capital). A comprehensive cost–benefit analysis of different foreign investment-promoting policies is not easy to put in place, due to the difficulty of estimating the complex and multifaceted elements on the ground (OECD 2003). Notwithstanding this, in the design of a costly incentive policy, all the elements mentioned above should be kept in mind by analysts and policy-makers.

Features and effects of corporate tax on FDI 131 Finally, in the assessment of tax incentives to foreign investment, the possible rise of tax competition between countries should be kept in account. Evidence emerging by interviews with corporate managers (Oman 2000) suggests that, while investors attach increasingly more importance to economic fundamentals than to incentives in the location decision for a long-term investment, they also tend to create a shortlist of preferred sites that have all the required economic characteristics and facilities. Then they usually negotiate incentives and other conditions with each government. This practice is clearly conducive to a prisoner’s dilemma in the host countries perspective: for each country it is optimal to offer higher incentives in order to attract more investment, but all of them end up with a similar amount of aggregate investment and with heavy revenue losses. Tax competition can be particularly strong in countries included in free trade areas, such as ASEAN. The fact that Malaysia and Thailand have introduced a special corporate tax rate reduction for MNEs that establish their regional operating headquarters is a clear signal that the South and East Asian countries are not immune to tax competition and they should try to subscribe to agreements in order to contain it.

Notes 1 In 1979 the number of enterprises allowed to trade increased, indirect trade policy instruments (such as quotas and tariffs) were introduced, and restrictions on prices and exchange rates were removed. 2 According to Sun et al. (2002), in 1986 China “relaxed the restrictions regarding expatriation of profits and dividends, and allowed foreign nationals to be the chairman of the board of directors in Foreign Invested Enterprises.” 3 Studies reporting a positive tax elasticity of investment are not absent. See Scholes and Wolfson (1990) and Swenson (1994). 4 The exemption is not allowed for companies whose business is in banking, insurance, sea and air transport. 5 If the tax liability for the year is less than 7.5 percent of the adjusted book profits, both Indian resident and non-resident companies are liable to pay tax on their book profits; the current tax rate of the minimum alternative tax (MAT) is 7.688 percent, including 2.5 percent surcharge. 6 Huang (2003) gives evidence that the business environment is more “friendly” to FIEs than domestic firms. 7 More precisely, both methods are allowed for tangible fixed assets; intangible assets must be amortized using the straight-line method. 8 According to the Kelkar Task Force (Government of India 2004) the differential treatment between unabsorbed depreciation and unabsorbed business should be removed and business loss, like unabsorbed depreciation, should be allowed to carry forward indefinitely. 9 In Thailand, full exemption is allowed for companies listed in the Stock Exchange of Thailand irrespective of the proportion of shares owned by the company. 10 Long-term capital gains on sales of shares in a government-approved enterprise wholly engaged in specified infrastructure sectors are exempt from corporate income taxation. 11 See later for a detailed description of tax incentives.

132 Francesco D’Amuri and Anna Marenzi 12 The data are taken from the Report of the Task Force on Implementation of the Fiscal Responsibility and Budget Management Act, 2003 (Government of India 2004). 13 This comparison is with respect to the most generous incentive regime in Thailand that applied to exporters located in underdeveloped regions. 14 Industries in sectors that are not sufficiently developed in the host countries. 15 The definition of the “first profit-making year” is the first year during which all prior losses are used, leaving a taxable profit. 16 Includes companies involved in promoted activities, located in certain promoted areas, that provide certain types of intermediate goods, or that are of national and strategic importance (heavy capital investment and high-technology projects). 17 Applying to ICT enterprises operating in a special area in the south of Kuala Lumpur. 18 In addition, SMEs are eligible for a 30 percent reduction if located in nonmetropolitan areas; for small-sized enterprises located in metropolitan areas the cut in CIT rate is 20 percent. 19 A tax credit of 3 percent of the investment amount is granted for investment in anti-pollution and non-pollution facilities, mine safety facilities, welfare-increasing facilities for employees, facilities for manufacturing business using advanced technology and skills, facilities for improvement of the distribution industry and other kinds of investments; for SMEs and for specific investment among the above mentioned, the percentage grows up to 5 percent; where a corporation invests in energy-saving facilities, the amount that can be credited increases to 10 percent. 20 The percentage is augmented for corporations with a higher proportion of R&D expenses, for R&D activities conducted jointly by academic, business and government circles, and for R&D expenses of SMEs. The amount of the credit shall not exceed 20 percent of the amount of corporation tax; under certain conditions, there is the possibility of carrying forward the unused credit for one year.

References Acharya, S. (2001) India’s Macroeconomic Management in the Nineties, Indian Council for Research on International Economic Relations: New Delhi. Blomstrom, M. and Kokko, A. (2003) “The economics of foreign direct investment incentives,” NBER Working Paper 9489, National Bureau of Economic Research, Cambridge, Massachusetts. Chalk, N. A. (2001) “Tax incentives in the Philippines: a regional perspective,” IMF Working Paper 181, Washington, DC: IMF. De Mooij, R. A. and Ederveen, S. (2003) “Taxation and foreign direct investment: a synthesis of empirical research,” International Tax and Public Finance, 10: 673–93. Fletcher, K. (2002) “Tax incentives in Cambodia, Lao PDR, and Vietnam,” Paper presented at the IMF Conference on Foreign Direct Investment: Opportunities and Challenges for Cambodia, Lao PDR and Vietnam, Hanoi, Vietnam, August 16–17. Government of India, Ministry of Finance (2004) Report of the Task Force on Implementation of the Fiscal Responsibility and Budget Management Act, 2003. Henley, J. S. (2004) “Chasing the dragon: accounting for the under-performance

Features and effects of corporate tax on FDI 133 of India by comparison with China in attracting foreign direct investment,” Journal of International Development, 16: 1039–52. Hines, J. R. (1998) “ ‘Tax sparing’ and direct investment in developing countries,” NBER Working Paper 6728, Cambridge, Massachusetts. —— (1999) “Lessons from behavioral responses to international taxation,” National Tax Journal, 52: 305–22. Huang, Y. (2003) “One country, two system: foreign-invested enterprises and domestic firms in China,” China Economic Review, 14: 404–16. Ianchovichina, E. and Martin, W. (2001) Trade Liberalization in China’s Accession to the World Trade Organization, Washington, DC: The World Bank. KPMG International (2003) Asia Pacific Taxation 2003, (www.kpmg.com). Lawrence, R. (1991) “Efficient or exclusionist? The import behavior of Japanese corporate groups,” Brooking Papers of Economic Activity, 1: 311–31. —— (1993) “Japan’s low levels of inward investment: the role of inhibitions on acquisitions,” in Froot, K. A. (ed.) Foreign Direct Investment, Chicago: University of Chicago Press. Mintz, J. M. (1990) “Corporate tax holidays and investment,” The World Bank Economic Review, 4(1): 81–102. —— (2004) “The changing structure of tax policies for foreign direct investment in developing countries,” Paper presented at the Annual Conference on Public Finance Issues in an International Perspective, Georgia State University. OECD (2001) “Corporate tax incentives for foreign direct investment,” OECD Tax Policy Studies, Paris: OECD. —— (2003) Checklist for Foreign Direct Investment Incentive Policies, Paris: OECD. Oman, C. (2000) Policy Competition for Foreign Direct Investment, Development Centre Studies, Paris: OECD. Owens, J. (2004) “Competition for FDI and the role of taxation: the experience of South Eastern European Countries,” Working Paper SIEP 316/204, Pavia: Società Italiana di Economia Pubblica. Scholes, M. S. and Wolfson, M. A. (1990) “The effects of changes in tax law on corporate reorganization activity,” Journal of Business, 62: 141–64. Sun, Q., Tong, W. and Yu, Q. (2002) “Determinants of foreign direct investment across China,” Journal of International Money and Finance, 21: 79–113. Swenson, D. L. (1994) “The impact of US tax reform on foreign direct investment in the United States,” Journal of Public Economics, 54: 243–66. UNCTAD (2000) Tax Incentives and Foreign Direct Investment: A Global Survey, Geneva. Wells, L. T. and Allen, N. J. (2001) “Tax holidays to attract foreign direct investment,” in Wells, L. T., Allen, N. J., Morisset, J. and Pirnia, N. “Using tax incentives to compete for foreign direct investment: are they worth their costs?,” FIAS Occasional Paper 15, Washington, DC: The World Bank. Zee, H. H., Stotsky, J. G. and Ley, E. (2002) “Tax incentives for business investment: a primer for policy makers in developing countries,” World Development, 30(9): 1497–516.

Part II

Country studies of tax systems and tax reforms in South and East Asia

6

China Domenico D’Amico

Introduction, contents and main conclusions1 The People’s Republic of China (PRC) is one of the largest countries in the world, extending over 9,597,000 km2, and the most populous with almost 1.3 billion inhabitants, more than one-third (37.7 percent in 2001) living in urban areas. Despite its size and the presence of ethnic minorities, China is a unitary state, comprising 22 provinces, five autonomous regions and four municipalities directly under the central government (Beijing, Shanghai, Tianjin and Chongqing). Political power is held by the Communist Party of China: the National People’s Congress (NPC) does not usually deviate from the directives of the party’s politburo. After more than two decades of central planning, in the late 1970s China inaugurated the so-called “reform era” and at the same time started establishing a formal legal system; since then the areas of criminal, civil, administrative and commercial law have been codified. Domestic economic reforms have been supplemented with the gradual opening of the Chinese economy to the rest of the world, which culminated in the accession of China to the WTO in December 2001. The spectacular economic growth of the country has been such that China has not been harmed by the recent East Asian financial crisis and indeed is promoting economic recovery in the region. Economic growth, however, has not always resulted in sustained socioeconomic progress. In the second half of the 1990s poverty reduction, especially in rural areas, seems to have slowed and, furthermore, the slowdown has coincided with a rise in inequalities (World Bank 2003).2 Interprovincial disparities are a major factor in explaining inequality (Wang 2001). During the 1990s, over 90 percent of all foreign direct investment (FDI) accrued to coastal provinces, while central and western regions still suffer from poor infrastructure and the absence or paucity of markets. Therefore, in March 2004 the NPC, in delineating the government’s priorities, set out a more balanced development agenda aimed at reducing the rural–urban divide, promoting growth in western regions and in the old industrial bases in the northeast, and laying more emphasis on

138 Domenico D’Amico health, education, social security reform and the protection of urban migrants (ADB 2004). Such a change in fiscal policy will require that the government rationalize the country’s tax system, especially to further enhance tax collection and to rectify the present mismatch between expenditure responsibilities and revenue sources at the sub-national levels of government, by means of a more equalizing and targeted transfer system. In this chapter we intend to provide a brief introduction to the Chinese tax system. In the second section, after a quick look at China’s recent macroeconomic developments and a concise description of the main items in public expenditure and their assignment to different layers of government, we will consider the level and composition of tax revenue in 2002, with a view to emphasizing the still inadequate ratio of tax revenue to GDP as well as the inappropriate reliance of the Chinese tax system on indirect taxation (mainly on VAT). We will then examine fiscal trends over the 12-year period 1991–2002; the effects of the 1994 reform will be apparent with regard to both the level of tax revenue and its apportionment between the central government and the provinces. The institutional features (tax base, rates, exemptions and so on) of the chief direct and indirect taxes levied in China will be described in the third section. Though China has decidedly moved towards a uniform tax system and it will have to do so even more owing to its recent accession to the WTO (which requires compliance with the principle of “national treatment”), there are still notable exceptions to equality of treatment of foreign and Chinese taxpayers. Disparities are present both in income taxation (where foreigners are allowed, for instance, more generous monthly deductions) and especially in corporate taxation (where the drive for modernization and the desire to attract FDI and favor technology transfer have led to preferential treatment for foreign-related enterprises). There are also weaknesses in the design of indirect taxation, due to the simultaneous presence of a tax on gross turnover (the business tax) and of an almost standard VAT, engendering inefficient cascading effects. In the fourth section an attempt is made to evaluate the Chinese tax system with regard to the efficiency of its design and the equity of its effects. It will be shown that the level of tax revenue is inadequate to the growing expenditure needs of China, even in comparison with other developing countries. Furthermore, income taxation and consumption taxation are not well balanced, with too heavy a reliance on the latter. The large weight of indirect taxes (especially VAT) also has inequitable effects, further widening the gap between the wealthier coastal regions and the northern and western provinces. The final section records the notable improvements that China has notwithstanding managed to achieve in the design of its tax system. We will describe the reform carried out in 1994, a comprehensive package of measures that have affected every major aspect of taxation, leading to the introduction of a modern VAT (based on an appropriate invoice

China 139 mechanism) and to the almost complete unification of the income tax regimes for foreigners and Chinese citizens (with the caveats mentioned above). There is, however, room for further improvements and the final section examines some proposals for reform, especially with regard to the shift from the current production-based VAT to a consumption-based VAT. To conclude, the chief weaknesses of China’s fiscal system may be summarized in the following way: (i) the tax revenue to GDP ratio is still low, even in comparison with other developing countries, and is insufficient in view of the direct and contingent (non-performing loans connected with banking system reform, unfunded pension obligations, restructuring of state-owned enterprises) liabilities3 the government will face in the near future; (ii) the weight of indirect taxes is disproportionate and has adverse effects on both enterprises and households; (iii) the current production-type VAT should be replaced with a consumption-type VAT and extended to services now subject to the business tax; (iv) changes should be made to personal income tax so as to exploit fully its potentially redistributive function; and (v) a better balance should be found between expenditure responsibilities and revenue assignments at sub-national level.

A broad view of the tax system and its development from the early 1990s A short reminder of the Chinese economy and public sector outline Though still a developing country with a relatively low per capita income (per capita GDP was CNY8,999 in 2003 or around US$ 1,087), China has made remarkable economic progress since the late 1970s. The average growth rate of real GDP was 9.9 percent in the ten-year period 1986–95 (compared with the average 7.7 percent for developing Asia countries) and, according to IMF estimates and projections for 2004 and 2005, it is predicted to be 8.2 percent in the ten-year period 1996–2005 (6.6 percent for developing Asia countries).4 China’s weight and role in the global economy has correspondingly widened: the country is now the sixthlargest economy in the world (at market exchange rates) and the fourthlargest trader. China’s remarkable economic performance has its roots in the sweeping structural reforms that begun in 1978 under the auspices of Deng Xiaoping. The changes that occurred since then led to the gradual replacement of central planning by market mechanisms, to the opening of the economy to the rest of the world, to the transformation of the country’s productive structure, with the secondary and tertiary sectors gaining ground at the expense of the primary one. It should be noted, however, that agriculture, being characterized by the scarcity of land in

140 Domenico D’Amico relation to labor and little mechanization, still employs around 44 percent of the country’s workforce. As regards China’s public finances, during recent years there has been a marked reversal of previous trends. After declining until the mid-1990s, revenue and expenditures started expanding in 1996; the latter, however, has grown faster than the former, resulting in larger budget deficits (3.4 percent of GDP in 2002). Revenue increases (1.1 percentage points of GDP in 2002) were chiefly the result of a large growth in tax collections (especially in VAT and income taxes), which, in turn, may be attributed to two main factors: the reform introduced in 1994 and the enhancements in tax administration (IMF 2004b). As to the latter, the implementation of the 1994 reform required two separate channels for tax collection to be set up, under the auspices of the state administration of taxation (SAT). A national tax service was established, charged with collecting central and shared taxes, while local tax bureaus were made only responsible for collecting local taxes. This measure substantially reduced the opportunities for local governments to appropriate central revenue by manipulating tax assessments. Another important contribution to improved tax administration came from the computerization of VAT collection. Expenditure, after falling to a minimum of around 13 percent of GDP in 1996, grew up to 21.7 percent of GDP in 2002 (see Table 6.1). The largest increase (in percentage terms) concerned expenditure on pensions and social welfare programs, as state-owned enterprises were relieved of spending mandates in such social areas as pensions and the provision of education and health services, which have gradually become responsibilities of local governments. Expenditure on administration and defense more than doubled over the period, while the government expansionary fiscal policy (initiated in the wake of the late 1990s Asian financial crisis) and the development strategy for central and western regions pushed up capital expenditure, which rose by 1.7 percent of GDP. Spending

Table 6.1 General government expenditure in China, 1997–2002 (% of GDP)

Total expenditure and net lending Current expenditure, of which: Administration and defense Culture, education, public health and science Pension and social welfare relief Capital expenditure Unrecorded expenditure Source: IMF (2004b), table 5.3.

1997

1998

1999

2000

2001

2002

14.0 11.5 2.7

16.1 12.6 3.8

18.3 13.7 4.1

18.9 14.7 4.4

20.1 16.1 4.9

21.7 17.3 5.5

2.7 0.2 2.2 0.3

2.7 0.2 2.6 0.9

3.1 1.0 3.5 1.2

3.2 1.7 3.3 0.8

3.6 2.0 3.6 0.4

4.0 2.5 3.9 0.4

China 141 Table 6.2 Shares of central and local governments in selected budget expenditures in China, 2001 Share of central government (% of item) Total Culture, education, science and health Capital construction National defense Administration Agriculture Policy subsidies Social security and welfare Debt service

Share of local governments (% of item)

30.5

69.5

10.8 34.3 99.2 2.7 10.9 40.3 0.7 100.0

89.2 65.7 0.8 97.3 89.1 59.7 99.3 0.0

Source: IMF (2004b), table 6.3 and World Bank (2003), statistical appendix, tables 26–7.

responsibilities are highly decentralized: in 2002 the shares in total expenditure of central (31.3 percent) and local governments (68.7 percent) were in the proportion 1:2 (World Bank 2003).5 As shown in Table 6.2 (which refers to 2001), there is a certain differentiation in expenditure assignments: the central government is entirely responsible (or almost so) for defense and debt service, while local governments are more or less exclusively in charge of spending on agriculture, culture, education and health, social relief and welfare. The tax system The current structure The structure of the Chinese tax system in 2002 (last available data) is displayed in Table 6.3. Total revenue (CNY1,917.4 billion) amounts to 18.3 percent of GDP, while tax revenue (CNY1,763.2 billion) accounts for 16.8 percent of GDP and 92 percent of the total receipts. There are no social security contributions, since pensions and other social benefits are restricted to selected categories of workers, and 80 percent of the latter are not covered at all by any kind of social protection. Taxes on income and profits make up a not irrelevant share of tax revenue (as compared with previous years), adding up to 25 percent of the total yield and 4.2 percent of GDP. Contrary to most Western countries, corporate taxes are the main item in this category, accounting for 17.1 percent of the total tax revenue. They are mainly levied on domestic enterprises, since foreign investment enterprises benefit from generous preferential treatment. On the other hand, the individual income tax (assigned to local governments by the tax sharing system introduced in

14.7 12.4 3.0

2.7 – 0.3

7.7 2.6 2.6 2.5

0.8 0.8 2.4

3.2 9.2

17.0 13.8 3.8

3.4 – 0.4

7.4 2.9 1.9 2.6

0.9 1.8 3.2

3.6 10.2

1992

2.6 9.7

0.7 0.9 1.4

8.3 2.4 3.1 2.8

2.0 – 0.3

13.7 12.3 2.3

1993

6.1 4.9

0.6 0.8 0.9

7.4 1.1 4.9 1.4

1.5 0.2 0.5

11.9 11.0 2.2

1994

5.5 4.8

0.5 0.8 0.8

6.8 0.9 4.4 1.5

1.5 0.2 0.5

11.1 10.3 2.2

1995

5.1 5.1

0.4 0.7 1.1

6.9 0.9 4.4 1.6

1.4 0.3 0.5

11.3 10.2 2.2

1996

Note a The following figures refer to tax revenue.

Source: author’s calculations based on World Bank (2003), statistical appendix, tables 1, 18, 19, 20.

Total revenue Tax revenue Taxes on income and profits, of which: Enterprises income tax Personal income tax (other) Agricultural income tax Taxes on goods and services, of which: Product/consumption tax Value added tax Business tax VAT and excises on imports Customs duties Other taxes Non-tax revenue Administrative levela Central government Local governments

1991

5.7 5.4

0.4 1.5 1.0

7.1 0.9 4.4 1.8

1.3 0.3 0.5

12.1 11.1 2.1

1997

6.1 5.7

0.4 1.7 1.2

7.7 1.1 4.6 2.0

1.2 0.4 0.4

13.0 11.8 2.0

1998

7.0 6.0

0.7 2.2 1.3

7.8 1.0 4.7 2.1

1.5 0.5 0.3

14.3 13.0 2.3

1999

7.7 6.4

9.8 0.9 5.1 2.1 1.7 0.9 0.5 1.2

1.9 0.7 0.3

15.3 14.1 2.9

2000

Table 6.3 Structure and developments of consolidated general government revenue in China, 1991–2002 (% of GDP)

8.6 7.1

10.3 1.0 5.5 2.1 1.7 0.9 0.5 1.4

2.7 1.0 0.3

17.1 15.7 4.0

2001

10.1 6.7

10.6 0.9 5.9 2.2 1.6 0.7 1.3 1.5

2.9 1.0 0.3

18.3 16.8 4.2

2002

China 143 1994) is only 7.9 percent of the aggregate tax revenue and 1 percent of GDP. As one could suspect, the Chinese tax system is based mainly on taxes on goods and services, which add up to 67 percent of total tax revenue (11.3 percent of GDP). Apart from the value added tax (VAT), which alone accounts for around 7.3 percent of GDP (and more than 43 percent of total tax revenue), we should also mention the business tax – levied on services not covered by VAT – and the consumption tax – an excise duty applying to particular categories of goods (most of them classified as “luxury” goods). Even if the average tariff rate has substantially declined, as a consequence of the choice of the country to open its economy to the rest of the world (and it will be even more so after China’s accession to the WTO), custom duties still stay at 4 percent of total tax revenue and at 0.7 percent of GDP. Finally, a few words on the other taxes, which account for 7.9 percent of the total tax revenue and 1.3 percent of GDP. The main items are the city maintenance and construction tax, the stamp tax, the house property tax and the urban real estate tax. One of the top priorities of the 1994 reform was to enable the central government to increase its share in fiscal resources, which previously had substantially declined to the advantage of local governments. A new tax sharing system was established which resulted in the central government collecting 60 percent of the total tax revenue in 2002 (see Figure 6.1).6 As a consequence, local governments, which are in charge of almost 70 percent of total expenditure (and in essential sectors such as health and education) (see Table 6.2), are forced to rely heavily on extra-budgetary funds. A brief overview of the 1990s changes in the tax system and the 1994 reform In reviewing and assessing the development of the Chinese tax system during recent years, one can begin with the sweeping tax reform launched in 1994, which to some extent represented an expected corollary to the social and economic changes China had been experiencing since the beginning of the reform era in 1978. Several objectives were pursued, a prominent one being that of countering the steady decline revenue had recorded since the peak attained in 1985. Actually, after reaching a minimum of 10.2 in 1996, the ratio of tax revenue to GDP has been rising since then up to 16.8 percent, as a consequence of an annual yield growth rate of 14.5 percent. There has been some increase in the ratio to GDP of the taxes on income and profits, which in 1994 provided 20 percent of aggregate tax revenue, while they reached one-quarter of the total in 2002. The 1994 reform introduced a revised personal income tax, which replaced the

144 Domenico D’Amico former one as well as some other previous levies. At the same time corporate taxation was made more uniform by abolishing the previous system, where enterprises were charged differently according to their ownership. Notwithstanding, foreign investment enterprises are still liable to a different tax than domestic businesses, benefiting from so-called tax holidays, i.e. tax exemptions and reduced rates. However, the most notable changes probably involved turnover taxes, which during the years after the reform, increased at a 5 percent annual growth rate. More importantly, there has been a radical reallocation of the relative weights of the three levies which make up the bulk of indirect taxation. The scope of VAT has been substantially broadened so as to cover the production and distribution of goods, imports and the provision of some services. As the final outcome it currently amounts to nearly 70 percent of turnover taxes and is the single largest source of revenue for the government. Complementary to the VAT, the business tax is levied on most services; while losing ground to the advantage of its counterpart during the first years after the reform was launched, it has later shown an increasing trend at a growth rate of about 60 percent, when compared with the trough in 1994. On the contrary, a marginal role is played in China by excise duties, represented by the consumption tax (which replaced the older product tax). It is levied on cigarettes, alcoholic beverages, motor vehicles and some other “luxury” goods, but its relevance dramatically reduced, since its share in GDP decreased by more than two-thirds from 2.9 percent in 1991 to 0.9 percent in 2002, always hovering more or less around this value. Our general description of the Chinese tax system would be incomplete if we did not consider, though briefly, the development of intergovernmental fiscal relations. According to several indicators, China is a highly decentralized country. Local governments are responsible for essential 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Central government share

Local government share

Figure 6.1 The distribution of tax revenue in China (source: see text).

China 145 public services such as education and health, but the assignment of resources established by the tax sharing system introduced in 1994 has put heavy pressure on their budgets which are often unbalanced. Clear evidence of this is provided in Figure 6.1, where we plotted the respective shares in tax revenue of central and local governments. In 1993 only around 20 percent of tax revenue accrued to the central government, while in 2002 this proportion reached 60 percent. The result of the 1994 reform was immediate and the share of the central government jumped suddenly to more than one half of the total.

Some quantitative and institutional features of the main taxes Direct taxes Personal income tax (PIT) Personal income tax (PIT) was introduced by the 1994 reform and it applies in a uniform way to both Chinese nationals and foreigners. The tax is essentially schedular, since only certain listed types of income are taxed and the tax liability relative to each type is computed separately, without aggregation of an individual’s different sources of earnings. Furthermore, there is no system of personal deductions nor aggregation of the income, or joint taxation, of spouses. Finally, income is taxed differently depending on its nature, either at progressive rates or at a flat rate. An individual is liable to tax in China if s/he is domiciled in China or resides in China for a full year or else derives income from Chinese sources. An individual domiciled in China is taxable on his/her worldwide income; s/he is considered to be domiciled if s/he “habitually” resides in China as far as evidence thereof is given by household registration, family or economic interests. In establishing whether or not a person is domiciled in China, citizenship or nationality plays an important role, since Chinese citizens are generally required to have a household registration in China. However, since domicile entails habitual abode or permanent place of residence, citizens who live permanently abroad are not considered to be domiciled in China. On the other hand, even foreigners could be regarded as domiciled in China if they chose to make the country their permanent home rather than a place of work or temporary stay; however China has seldom granted foreigners permanent residence status. Therefore foreigners are usually taxed on the basis of residence, that is “physical stay.” Their tax liability depends on the length of the period of residence, which is computed by counting the numbers of days of actual presence in China in a financial year.7 Chinese source income is defined to include income earned by an individual from a position, employment or fulfillment of contracts within

146 Domenico D’Amico China; income derived from letting or leasing of property for use in China; income from transfer of buildings, land-use rights or alienation of other property within China; income from royalties on technology used within China; interest, dividends and bonuses received from companies, enterprises, other economic entities and individuals located in China. Tax on employment income is calculated on a monthly basis. Losses are not deductible, however a standard monthly deduction of CNY800 is allowed. An additional deduction of CNY3,200 is granted to the following specified categories of taxpayers: foreign employees who work in foreignfunded enterprises or foreign firms within China; invited foreign experts working in enterprises, institutions, social organizations and state organs in China; resident individuals who obtain wages and/or salaries from their post or employment outside China; others as determined by the Ministry of Finance. The tax payable by employees is withheld at source by their employers. The tax rates applicable to wages and salaries are displayed in Table 6.4. “Remuneration for personal services,” that is consideration for providing such services as medical treatment, law practice, accounting, consulting, lecturing, intermediary services, agency, brokerage and other services, is taxed on a per-payment basis. Allowable deductions are as follows: • •

if a single payment is CNY4,000 or less, the deduction is CNY800; if a single payment is more than CNY4,000, the deduction is 20 percent of the payment.

Income from such professional services is taxed at a flat rate of 20 percent, however a surtax is charged for “abnormally high” payments (exceeding CNY20,000). The same tax regime applies to royalties (including author’s royalties). The income derived by individual industrial and commercial households from production or business operations is taxed at progressive rates as shown in Table 6.5. Table 6.4 PIT rates on employment income in China, 2002 Monthly taxable income (CNY)

Tax rate (%)

up to 500 501–2,000 2,001–5,000 5,001–20,000 20,001–40,000 40,001–60,000 60,001–80,000 80,001–100,000 over 100,000

5 10 15 20 25 30 35 40 45

Source: IBFD data (2004).

China 147 Table 6.5 PIT rates on household business income in China, 2002 Annual taxable income (CNY)

Tax rate (%)

up to 5,000 5,001–10,000 10,001–30,000 30,001–50,000 over 50,000

5 10 20 30 35

Source: IBFD data (2004).

Income from interest and dividends, income from the lease of property, income from the transfer of property, incidental income and other income are taxed at a flat rate of 20 percent. It should be noted that there is no imputation system to alleviate double taxation of dividends. Some types of personal income are exempt from income tax: • •

• • • •

monetary awards from provincial-level people’s governments, the state council’s ministries and certain units of the People’s Liberation army; monetary awards given by foreign international organizations in the fields of science, education, technology, culture, health, sports and environmental protection; subsidies and allowances paid in accordance with uniform state stipulations; welfare benefits, pensions for the disabled and for survivors and relief payments; military severance pay and decommission or demobilization pay for soldiers; family allowances, job discharge fees, retirement wages and supplementary retirement fees paid to cadres and workers in accordance with uniform state stipulations.

Agriculture tax Our overview of personal income taxation in China would be incomplete if we did not mention the agriculture tax, which is levied on farmers. Farmers represent a separate category of taxpayers, being generally exempt from income tax and VAT or business tax. The agriculture tax is a mixed levy, having elements of both property and income taxation. The tax base is the normal yield for agricultural crops and gross production for other native agricultural products. The normal yield is the estimated average annual yield for grains determined on the basis of the natural conditions of the land, ordinary level of management and production level of normal years. Incentives for enhancing productivity are provided, since the base of assessment is left unchanged for three years if

148 Domenico D’Amico production increases on account of improvements in farming conditions (such as the construction of irrigation works) and for five years if the reason of the increase is better management or cultivating techniques. The average national rate is 15.5 percent; local governments however are free to set the actual rate, within the limit of 25 percent. Corporate taxation FOREIGN INVESTMENT ENTERPRISES

Entities liable to the foreign enterprise income tax (FEIT) comprise: foreign investment enterprises (FIEs), foreign enterprises (FEs) with establishments in China, and foreign companies without establishments in China deriving income from Chinese sources. FIEs consist of Chinese–foreign equity joint ventures, Chinese–foreign contractual joint ventures and wholly foreign-owned enterprises formed in China. These are all legal persons under Chinese laws and residents for income tax purposes, having established their head office in China;8 being residents, FIEs are taxed on their net worldwide income. FEs with an establishment in China are taxed on their Chinese sourced income. “Establishment” is defined broadly to include management offices, business sites, offices, factories, places of natural resources extraction, sites for contracted projects such as construction, installation, assembly or exploration or for the furnishing of services, and business agents. In addition to the income from the production and business operations of the establishments or places in China of FIEs or FEs, income from Chinese sources comprises any dividends, interest, rental, royalty and other income derived both inside and outside China that is actually attributable to those establishments or places. Finally, foreign enterprises without establishments in China are charged withholding taxes on the gross receipts of their passive income from Chinese sources. Taxable income is generally income less allowable deductions and expenses. The relevant laws provide formulas for computing the net taxable income for businesses in different types of industries, such as commerce, manufacturing and service industry. Some expenses, however, are not deductible, including: expenditures on fixed assets, expenditures on intangible assets, interest on capital, income tax payments and royalties paid to head offices. Fixed assets (valued at original cost) may be depreciated. Depreciation is calculated annually on a straight-line basis, assuming a residual salvage value of 10 percent of the original cost. Minimum depreciation periods (5 to 20 years) are prescribed for various classes of fixed assets. Accelerated depreciation may be conceded by the state administration of taxation (SAT) in a few specified circumstances, for example when machinery and equipment is in operation 24 hours a day throughout the year or when the fixed assets are contributed to a joint venture with an operation term

China 149 shorter than the prescribed depreciation period of the assets and the Chinese party to the venture takes title to the assets upon termination of the joint-venture contract. Amortization of intangible assets (such as patents, proprietary technology, trademarks, copyrights and site use rights) is also allowed on a straight-line basis, generally over a ten-year period. The FEIT is levied at a (national) flat rate of 30 percent plus a local surtax of 3 percent on taxable income. However, in recent years China has lured foreign investors by granting them generous tax incentives. In order to qualify for these incentives some criteria must be satisfied. First, the investment must be carried out through a FIE or specially through an enterprise invested by Taiwanese or Hong Kong residents. Second, the enterprise must be engaged in a designated industry.9 Third, the location in which the enterprise is established (or going to be established), the operating period and the level of investment are all relevant factors. Notable tax incentives are the so-called “tax holidays,” granted for periods varying from three to ten years (a prolongation may be authorized for huge infrastructure projects); a ten-year tax holiday, for example, entails a tax exemption for the first five profit-making years and a 50 percent tax rate reduction for the following five years. An enterprise is eligible for a tax holiday as long as it is engaged in production or business operations for the specified minimum period of ten or 15 years. In addition to tax holidays, the foreign income tax law provides for reduced tax rates of 24 percent and 15 percent for enterprises running a business in special areas of China. Furthermore, local governments normally waive their 3 percent surtax. Incentives are also available for profits reinvested in China in the form of a refund of the income tax paid on them. A resident enterprise or company is generally exempt from income tax on dividends received from a resident company, while non-residents, whether or not they have an establishment in China, are liable to withholding tax at the rate of 10 percent (before 1 January 2000 it was 20 percent) on all dividends paid by resident companies. With regard to dividends received from foreign companies, resident companies are not entitled to exemption, but receive a credit for withholding tax paid and for tax levied on profits out of which the dividends are paid, up to the amount of the Chinese tax payable. Resident enterprises are subject to tax on interest from all sources at the usual rate (20 percent). Non-resident enterprises are subject to income tax only on interest that has a Chinese source and is earned in carrying on business in China at or through an establishment. A non-resident foreign enterprise which derives interest from China without having a permanent establishment or place of business in China or, though having one, earns interest income that is unconnected to it is subject to withholding tax at 10 percent. The same provision also applies to royalties.

150 Domenico D’Amico DOMESTIC ENTERPRISES

The enterprise income tax (EIT) applies to state-owned enterprises, collective enterprises, private enterprises, joint-stock companies and other organizations which derive income from production or business operations. The taxable income is the gross income less allowable deductions and expenses. The taxpayer’s total income includes: income from production or business operations; income from interest, dividends, royalties and leases; income from the transfer of property; and other income. Permissible deductions include: the interest on loans taken out for the benefit of production and business; the payments to workers’ trade union funds, welfare funds and educational funds, for a maximum amount of 2, 14 and 1.5 percent of total taxable wages, respectively; and donations to charities and relief funds, up to a maximum of 3 percent of annual taxable income. Losses may be carried over for five years. The tax rate is 33 percent. CAPITAL GAINS

Capital gains are taxed as “other income” at the regular rate (20 percent) when they are derived by a FIE, while non-residents are subject to withholding tax. In addition capital gains from the transfer of real estate are subject to the land appreciation tax. Since under the Chinese Constitution land is publicly owned and cannot be transferred, the tax is levied only on buildings and land-use rights. In calculating the taxable gains one may deduct: the amount paid for land-use rights; costs and expenses incurred on land development; costs and expenses on the construction of new buildings and related structures, or the assessed value of old buildings; and taxes related to the transfer of real estate. Taxation is progressive, with rates varying from 30 to 60 percent. Taxes on property The revenue of property taxes accrues to local governments, but their amount is rather small (0.35 percent of GDP in 2001).10 They are the urban real estate tax, the farmland use tax, the urban land use tax and the house property tax. The urban real estate tax only applies to FIEs, FEs, foreign individuals and overseas Chinese; for buildings the tax base is either the residual value after deducting 10 to 30 percent of the original value or the standard rental, while land is taxed on its assessed value variable with the location and condition of the land. Exemptions are allowed for newly constructed buildings and for renovating old buildings. The house property tax is levied on houses and other buildings, mainly on Chinese institutions and individuals. The urban land use tax applies to all units and individuals using land located in cities and towns; the tax rates depend on the size and location of cities and towns and may be reduced,

China 151 upon approval, in less developed inland areas and autonomous regions. Finally, the farmland use tax (introduced in 1987) applies to all individuals, businesses and institutions using farmland for construction or other non-farming purposes with rates depending on the population density of the farmland location. Indirect taxes Value added tax (VAT) The 1994 reform introduced in China a standard value added tax (VAT), whose implementation required the removal of price controls on most goods and services by the end of 1993. Taxable persons are all individuals and enterprises that either supply goods, provide specified services or are engaged in importing. “Goods” comprise all tangible movable property, including power, heating and gas. Intangible property, on the other hand, such as patents, copyrights or proprietary technology, is exempt from VAT. In addition to industrial businesses, the new VAT applies also to primary producers in the sectors of agriculture, forestry, fishing and mining. However, agricultural products manufactured and sold directly by farmers (as opposed to agricultural enterprises) are exempt from VAT. As concerns services, taxpayers are liable to VAT only for the processing of goods and repair or maintenance services. Imported goods are charged, while services purchased from overseas are not, since they are not provided within China. For the purposes of taxation the special administrative regions of Hong Kong and Macau are both considered not to be part of China. The tax base is the amount paid for the sale of goods or supply of services, including any additional fees received by the seller or supplier. The taxable value for imports is the customs value plus customs duty, consumption tax and insurance and freight costs incurred in delivering the goods to China. A mechanism of input tax credit allows taxpayers to be refunded the VAT paid or payable on purchases of goods and services in China and on goods imported into China as long as the goods are used in producing taxable goods or services. For agricultural products that are exempt from VAT, the purchaser is notwithstanding entitled to an input tax credit equal to 10 percent of the purchase price. However, no input tax credits are available with regard to the following items: • •

•

fixed assets; goods and services that are used in making non-taxable supplies such as non-taxable services, the transfer of intangible assets and the sale of immovable property; goods and services used in making exempt supplies;

152 Domenico D’Amico • •

goods and services available to employees for personal consumption or welfare; goods and services for personal consumption.11

Some supplies are exempt from VAT, which means that the recipient does not have to pay tax on their purchase; at the same time, the supplier cannot claim any tax credit on the purchases of inputs used in the making of the exempt supply. Exempt supplies include: • • • •

• •

agricultural products manufactured and sold directly by farmers; contraceptive medicines; antique books; imported instruments and equipment for scientific research and educational goods or equipment given to the taxpayer by foreign governments or international organizations; used goods; other exemptions permitted by the state council.

A simplified regime is provided for small taxpayers, that is wholesale or retail businesses with an annual turnover less than CNY1.8 million; their tax liability is 4 percent of the total turnover, but they are not entitled to input VAT credits. In addition, individuals with a monthly turnover of less than CNY2,000 from the sale of goods or less than CNY800 from the provision of taxable services are fully exempt. As concerns VAT rates, the standard rate of 17 percent applies to most taxable goods and all taxable services. A reduced rate of 13 percent is charged on the sale or importation of some goods that are considered to be essential.12 Exports are taxed at zero rate, which implies that, even if no output VAT is chargeable, input VAT is still creditable; here lies the difference from a business providing only exempt supplies, which cannot reclaim or recover input tax it has paid. Business tax The business tax is a local turnover tax payable by all domestic enterprises, FIEs, FEs, institutions and individuals engaged in taxable business activities. It is levied on gross revenue and input tax credits are not allowed. Business tax applies to the sale or transfer of immovable or intangible property and to the following business activities: transportation (by land, air, water and pipe), communications, construction, banking and finance, insurance, post and telecommunications, cultural and sports activities, and entertainment and services (such as agency, hotels, catering, tourism, advertising, leasing and warehousing). Major exemptions concern: services rendered by nurseries, kindergartens, welfare institutions for dis-

China 153 abled people, old people’s homes, matchmaking agencies and funeral homes; services rendered by disabled people; medical services provided by hospitals, clinics and other medical institutions; educational services; services related to agriculture including insurance; royalties paid to domestic scientific research institutions; and income derived by FIEs, FEs and foreign financial institutions located in SEZs within five years of establishment. Furthermore, taxable business revenue below CNY200 to 800 is generally exempt from business tax, the exact amount being determined by local governments. Tax rates vary depending on types of business as shown in Table 6.6. Consumption tax The consumption tax is an excise duty payable by domestic enterprises, FIEs, FEs, institutions and individuals producing or importing taxable goods that are held to be luxury goods. The tax is levied on 11 categories of products including petrol (CNY0.2 per liter) and diesel (CNY0.1 per liter), cars (with three rates of 3, 5 or 8 percent depending on type and cylinder capacity) and motorcycles (10 percent rate), alcohol (25 percent rate) and tobacco (with two rates of 30 or 45 percent plus a fixed amount per 50,000 cigarettes), and jewelry (5 percent on gold or silver jewelry and 10 percent on other jewelry). Customs duties Customs duties comprise export and import duties. Export duties are levied on a number of goods which are normally the raw materials restricted for export. In 2001 only 36 types of goods were subject to export duties, with five rates varying from 20 to 50 percent. Import duties are levied on most imported goods at either general or preferential tariff rates. The latter apply only to goods imported and originating from countries with which China has a trade agreement containing reciprocal

Table 6.6 Business tax rates in China, 2003 Business activity Transportation, construction, post and telecommunications, cultural and athletic activities Transfer of intangible and immovable property, service industry, banking and insurance services Entertainment Source: IBFD data (2004). Note a The applicable rate is determined by local governments.

Rates (%) 3 5 5–20a

154 Domenico D’Amico favorable tariff clauses, while the general tariff rates apply in other cases. From 1 January 2002 the average import tariff rate is 12 percent.

The fiscal burden Efficiency and equity of the Chinese tax system In this section we offer a qualitative evaluation of the Chinese tax system, focusing on the efficiency of its design and the equity of its effects. In a paper devoted to tax policy in developing countries, Tanzi and Zee (2000) suggested that in evaluating a country’s tax system from a macroeconomic perspective one should consider in the first place (i) whether the overall tax level (generally defined as the ratio of tax revenue to GDP) is appropriate and (ii) to what extent the composition of tax revenue (usually in terms of the respective weights of income and consumption taxation) is well balanced. As explained by the authors, the reason why these features are (or should be) of interest to both researchers and policy-makers is the widespread belief that taxation entails the welfare costs of resource misallocation and affects income distribution. Lacking unequivocal and practical criteria from optimal tax theory for determining the optimal tax level and mix for a country, one may get some help from international comparisons. Looking at Chinese data, we observe that in 2002 – at the peak of an increasing trend following the 1994 reform – the ratio of tax revenue to GDP was 16.8 percent, still lower than the average figure (18.2) for a sample of developing countries13 as measured in the three-year period 1995–7 and less than half the corresponding figure (37.9) for OECD countries (excluding the Czech Republic, Hungary, Poland, Korea and Mexico).14 In addition, it should be noted that, while no significant changes had occurred in the ratio for the two groupings with respect to the three-year period 1985–7, China exhibits quite a different trend. Largely as a result of the “fiscal contracting system” regulating intergovernmental fiscal relations, the ratio of tax revenue to GDP declined continuously after the peak of 1985 (22.8 percent) until 1996 (when it reached the lowest value during the period: 10.2 percent), before rebounding when the 1994 reform started producing its effects. Another important feature of a country’s tax system is, as already mentioned, the composition of tax revenue. Both efficiency and equity considerations are relevant. As to efficiency, life-cycle models suggest that taxing consumption should be preferred to taxing income. Income taxes are usually levied on both labor and capital income and a labor tax, while equivalent to a consumption tax over the whole life-cycle, creates an additional distortion in savings decisions. Equity, on the other hand, comes into play as a corollary to the presumed more regressive nature of consumption taxes compared with income taxes, even if recent literature has shown that taxing consumption is far less regressive from a life-cycle

China 155 perspective than is the case in a static framework. Chinese data show that the ratio of income taxes to consumption taxes was 0.4 in 2002 and 0.3 in the three-year period 1995–7, significantly lower than the prevailing figures for developing countries and especially for the subgroup of Asian developing countries (0.5 and 0.6 respectively) and utterly distant from the value of 1.2 for OECD countries (these figures all refer to the threeyear period 1995–7).15 Furthermore, in line with similar trends in developing countries, though to a much greater extent, in China the revenue from corporate income taxes is larger than that from personal income tax (their ratio was 2.9 in 2002 and averaged 5.5 in the three-year period 1995–7). As observed by Tanzi and Zee (2000), this may depend on several factors including differences in wage income, the inadequacy of tax administration and the strong political influence from the richest deciles of taxpayers. A negative effect of the composition of tax revenue is the relatively heavy burden on enterprises, which are the bulk of taxpayers in China. Data for Qinghai province shows that the tax burden on industrial enterprises increased by almost one-third in the four-year period 1996–9 (Bao 2003).16 In addition, the tax burden is not evenly distributed, since it is heavier on SOEs and domestic enterprises than on non-state-owned enterprises and FIEs; it also increases with the size of the firm. Furthermore, by considering that the Chinese tax system relies disproportionately on indirect taxes17 largely not well linked to tax base values, one would expect tax revenue to increase at a slower pace than GDP. However, an important result of the 1994 reform has been the enhancement of tax administration: the GDP elasticity of tax revenue has been almost always higher than one in the years after the reform, with an average value of almost two.18 The buoyancy of tax revenue may be also attributed to the higher growth rate of the services sector in comparison with that of GDP (Toshiki 2004). Tax revenue could be further increased if China succeeded in reducing the size of its underground economy. By using the so-called cash ratio method, which is based on the ratio of cash in circulation to bank deposits, it has been estimated that the underground economy averaged 15.3 percent of Chinese GDP in the eight-year period 1991–8 (Bao 2001). Since during the same time the tax revenue to GDP ratio was on average 11.6 percent, annual tax revenue losses may be estimated at around 1.8 percent of GDP, that is more than 15 percent of actual tax receipts. Unfortunately, adequate data are not available to assess the impact of taxes on income (re)distribution in China. A recent paper (Chu et al. 2000), which deals with changes in income distribution in developing countries from the 1970s through the 1990s and examines a large number of studies on the incidence of taxes and government spending policies in some of these countries, has only a single piece of information on China, relative to the before-tax Gini coefficient in the 1990s.

156 Domenico D’Amico However, income gaps have widened in China during recent years in both urban and rural areas (Wang 2001). As to urban households, the average income of the highest 20 percent was 4.2 times higher than that of the lowest 20 percent in 1990, while their ratio more than doubled in a few years, reaching the value of 9.6 in 1998. Indeed, the richest decile accounted for 38.4 percent of total income in 1998 in comparison with 23.6 percent in 1990. A similar trend, though not so marked, emerged also among rural households: in 1990 the ratio of the top quintile’s average income to the bottom quintile’s was 6.3, while it rose to 9.5 in 1998. In addition, the distribution of fiscal resources across provinces has worsened (World Bank 2002). Even though market reforms and the development strategy for attracting FDI that favored coastal regions are the main reason for growing horizontal imbalances, the fiscal system – far from countering these unequalizing factors – has contributed to the growth of regional disparities. For example, comparing the ratio of per capita expenditure to the national average for selected provinces in 1990 and 1998, we note that in Shanghai it rose from 2.4 to 4.5 and from 2.5 to 3.1 in Beijing, while it fell from 0.85 to 0.61 in Gansu and even halved in Hebei (from 0.48 to 0.24).19 Over the same period, while the percentage of revenue collection for the five richest provinces declined from 26 to 23 percent, it fell from 12.3 to 9.8 percent for the five poorest ones. These widening disparities are a serious cause for concern, since local governments, as shown in Table 6.2, account for almost 70 percent of budgetary expenditure. Social spending in poor provinces is so low that most basic services such as health care and education are provided at extremely insufficient levels.

Tax reforms A quick glance at macroeconomic and budget outlook 20 Economic growth in China accelerated from 8 percent in 2002 to 9.1 percent in 2003. Growth in 2003 was driven mainly by investment, with public sector investment (mostly by local governments) accounting for 72.1 percent of total investment. The upsurge in investment in conjunction with rising prices of raw materials and shortages in electricity and oil have raised concerns about an overheating of the Chinese economy, threatened also by the rapid increase in bank lending, particularly in real estate. The buoyancy of domestic demand and a rise in oil prices resulted in imports growing more than exports, causing a decline in the trade surplus and a current account balance equaling 2.2 percent of GDP. There was a marked slowdown in the growth of foreign direct investment (FDI), depressed by the SARS outbreak and the global economic downturn.

China 157 The consumer price index rose by 1.2 percent in 2003, reversing the 2002 deflationary trend and the rate of inflation at the time of writing is forecast to hover around 3 percent in both 2004 and 2005. Official statistics recorded an increase in the registered urban unemployment rate from 4 percent in 2002 to 4.3 percent in 2003; however, including laid-off SOE workers who had not been reemployed would have given an adjusted figure of around 8 percent. Contrary to targets and estimates of the government, which had planned a rise by 8 percent in both revenues and expenditures in 2003, fiscal revenues increased by 14.7 percent (slightly less than in 2002, when the increase was by 15.4 percent), partly as a result of the buoyancy of the economy; public expenditures, on the other hand, grew by 11.6 percent. The deficit therefore amounted to 2.7 percent of GDP – a slightly better result than targeted by the government (2.9 percent of GDP); there is, however, a general belief that official figures largely underestimate the real deficit. The proactive fiscal policy the government has adopted in recent years with a view to bolstering domestic demand and promoting employment is expected to be phased out in the near future; signs of the government’s intentions were already visible in 2003: spending on capital construction declined, while there was a sharp rise in expenditures on administration, social security and health care (partly due to measures to contain the SARS epidemic). At the time of writing, in 2004 and 2005 GDP is expected to grow respectively by around 8.3 and 8.2 percent; however, the dependence of economic growth on public sector investment may be an obstacle to its continuance in the long run. Greater stimulus from private consumption, on the other hand, would facilitate the phasing out of the expansionary fiscal policy that government has initiated in the wake of the Asian financial crisis. Fiscal deficit in 2004 is targeted at CNY319.8 billion, the same as in 2003, which would lower the deficit-to-GDP ratio to 2.5 percent (as against 2.7 percent in 2003). Furthermore, government spending will be directed mainly to economic restructuring and social development rather than economic growth. In any event, in 2004 the issue of special long-term treasury bonds should be worth only CNY110 billion, less than in 2002 and 2003 (CNY150 billion and CNY140 billion, respectively). Tax reforms of recent years, under way and planned The most important tax reform carried out in China in recent years occurred in 1994.21 Despite the radical changes to the tax system introduced in the 1980s, several obstacles to a structure more in keeping with the workings of an expanding market economy needed to be removed. Domestic enterprises were subject to different regimes with regard to income taxation depending on their ownership; furthermore, their tax treatment differed considerably from that applying to foreign invested

158 Domenico D’Amico enterprises. Both these circumstances resulted in iniquity and created distortions in resource allocation and competition between firms. Indirect taxation, on the other hand, was complicated by the existence of four categories of taxes (product tax, value added tax, business tax and salt tax) and the coverage of VAT was quite limited. The government launched the 1994 reform with the declared intention of “raising the two ratios,” i.e. the revenue to GDP ratio and the central share in total revenue. Other important factors were the need for a simplification of the tax system and harmonization of tax regimes as well as the necessity of establishing new, more rule-based criteria for apportioning revenues between the central and local governments. Major changes concerned VAT. The coverage of the tax was broadened so as to comprise every industrial product, commercial sales, import of goods and some specified services, to which the Chinese legislation refers collectively as “processing, repair or maintenance services.” Tax rates were reduced to three: apart from the standard 17 percent rate and the zero percent rate on exports (the latter in line with general international practice), a 13 percent rate was provided to apply to a restricted group of goods regarded as essential. In addition, since every enterprise dealing with taxable goods or services would be subject to the reformed VAT, the consolidated tax on industries and commercial entities, which had been levied until then on foreign-related enterprises, was abolished. Other important changes affected corporate taxation for domestic enterprises and personal income taxation. A unified “enterprise income tax” was introduced with a standard rate of 33 percent applying to all domestic enterprises regardless of ownership; however, preferential treatments for foreign invested enterprises were retained. The “individual income adjustment tax,” which applied to individuals, the “tax on private businesses in urban and rural areas” (also known as the “urban and township individual industry and commercial income tax”), which was levied on household businesses, and the former “individual income tax” were all replaced by a simplified “personal income tax,” applying in a more or less uniform way to Chinese citizens and foreigners. As concerns changes introduced more recently, the main ones were designed to alleviate the so-called “peasant burden.” The “tax-for-fees” reform was launched in 2000: it should lead to the elimination of several unofficial fees levied by local governments and to the replacement of the remaining ones with taxes subject to ceilings mandated by the central government. In the past four years up to 20 provinces have adopted this policy, which is estimated to have reduced the financial burden of 620 million farmers by at least 30 percent.22 Since the majority of the country’s poor live in rural areas, these measures are expected to result in a parallel decline in poverty; furthermore, as remarked by the IMF (IMF 2004b), they should improve budget management through the reduction of extrabudgetary funds. Starting in 2004, the agricultural tax rate will be reduced

China 159 on average by more than one percentage point a year and agricultural taxes will be removed in five years. Planned reforms involve mainly VAT. The government is reported to be on the verge of launching an experimental reform of VAT in the three north-eastern provinces of Liaoning, Jilin and Heilongjiang, moving from the current production-type VAT to a consumption-type VAT in eight main industries, including chemicals, metals, oil and the automotive industry.23 It should be added, however, that such a reform in the VAT structure has been repeatedly announced (and postponed) in the recent past and that in the present circumstances the central government has behaved in a somewhat cautious and passive way, since the chief taxpayers in northeast China are large SOEs and revenues collected from them accrue largely to the center. Evidence of another prospective reform may be viewed in the decision taken in October 2003 by the third plenum of the 16th central committee of the Chinese communist party to introduce in the near future a unified property tax, which would replace several fees on real estate transactions. Suggestions for further improvements One of the top priorities in reforming the Chinese tax system should be the replacement of the current production-type VAT (P-VAT) with a consumption-type VAT (C-VAT). The distinguishing feature of a P-VAT is that the tax paid on purchases of capital goods is not creditable; when VAT was introduced in 1994, China was experiencing a double-digit inflation (with a peak of 27 percent in late 1994) and thus the government opted for a PVAT with the intention of restraining capital investment and reducing the risk of overheating the economy (OECD 2002). Since then, however, there has been a rapid decline in inflation and even two episodes of deflation (during the years 1998–9 and 2001–2) (IMF 2004b), which have removed the chief reason for the adoption of a P-VAT. On the other hand, maintaining a P-VAT would perpetuate a situation that penalizes capital-intensive industries and blocks investment in research and development, which is of paramount importance for economic growth, especially in view of the increasing international competition to which domestic firms will be exposed after China’s WTO accession (Toshiki 2004). It may therefore be useful to illustrate the likely effects of the move to a C-VAT, on the basis of the simulations presented in Ahmad et al. (2004). These authors use data from the China Statistical Yearbook to estimate the losses in VAT revenue for each Chinese province consequent on the introduction of a C-VAT. The authors’ calculations show that the revenue losses expressed as a ratio to VAT revenue in 2001 would vary from a minimum of 13.7 percent for Beijing to a maximum of 53.9 percent for Hubei province, with an average of 30.9 percent and a coefficient of variation of about 0.3 (these

160 Domenico D’Amico values refer to the hypothesis of 100 percent efficiency in revenue collection). Losses to the central government would amount to CNY125.6 billion (under the same hypothesis). Then the authors weigh the pros and cons of the possible compensatory mechanisms: increasing the share of the VAT revenue accruing to provinces so as to leave their aggregate revenue unchanged or raising the standard VAT rate to the same effect. Another reform which is widely regarded as desirable on grounds of economic efficiency is the replacement of the business tax, which is currently applied to most services, with VAT (OECD 2002; World Bank 2002). The business tax is levied generally on gross turnover and is not creditable against VAT, thus producing a cascading effect and generating distortions in the choice of inputs; this, in turn, leads to an arbitrary pattern of effective tax rates on different consumer goods (Ahmad et al. 2004). In their paper Ahmad et al. examine the consequences of the extension of VAT to services (excluding finance and insurance, due to the difficulties in levying a value added tax on transactions in this sector) and the simultaneous (partial) cancellation of the business tax (which would continue to apply to the finance and insurance sector). Such a reform would result in severe losses for the provinces,24 since the business tax is a local tax and provinces only get 25 percent of VAT revenue, while the central government would gain considerably25 and could use this additional revenue to compensate the provinces. Personal income tax could also be profitably reformed (OECD 2002). In the first place, in calculating the amount to be paid by a taxpayer it would be advisable to aggregate all of his/her different sources of income and tax them together. Second, it is widely believed that the threshold for taxation should be lowered: actually the monthly deduction of CNY 800 for employment income is around 1.3 times greater than the 12th part of per capita income in 2002; an exceedingly high personal exemption could prevent the statutory progressive rate structure from generating a really progressive pattern of taxation (Tanzi and Zee 2000). Finally, the World Bank (2002) has judged the current assignment of PIT to provinces to be inappropriate in view of the redistributive function of such a tax and its role in macroeconomic stabilization. It has therefore suggested that PIT be split into two new taxes: a progressive one assigned to the central government and a proportional one allotted to local governments and piggybacking the national tax. Such a change could also provide the opportunity for giving a certain degree of tax autonomy to local governments, which could be allowed to set the flat rate of their PIT between minimum and maximum rates legislated at the national level.

Notes 1 Financing from the Fondazione Cassa di Risparmio delle Provincie Lombarde is gratefully acknowledged.

China 161 2 The Gini index, which had already risen from around 20 in the first half of the 1980s to 42 in 1993 (the fastest increase among all countries), even worsened in 1999–2000, reaching 43.7. 3 On these definitions see World Bank (2003), p. 59, note 77. 4 IMF (2004a), Statistical appendix. Unless otherwise stated, data in this section are taken from IMF publications. 5 IMF (2004a), statistical appendix, Tables 26–7. 6 However, it should be added that a large share of central government revenue is transferred to provinces in consequence of the “revenue returned” system (on which see below). That share averaged 40 percent of central government own revenue in the years 1996–2001 (Ahmad et al. 2004). 7 Foreign individuals who live in China for one year or more are generally taxed on their worldwide income (so-called “one-year rule”); however if their stay does not span a full five-year period, they may, upon concession, pay Chinese tax on foreign source income only if it is derived from a company, enterprise or individual in China. In applying the one-year rule, temporary absences are not subtracted if they amount to less than 30 days at one time or less than 90 days in total. Foreigners who reside in China for less than one year are liable to tax only on Chinese-source income. If their actual presence in China does not exceed 90 days, they are exempt from tax on any compensation paid by an employer outside China for services performed within China as long as the compensation is not borne by the employer’s Chinese establishment. Double taxation treaties may modify the 90-day rule resulting in the exemption of a taxpayer’s employment income if s/he is not in China for more than 183 days in a calendar year and the employer is a non-resident enterprise. 8 For corporate entities residence is determined according to the location of their head office defined as the chief establishment responsible for the operation, management and control of an enterprise. 9 The list includes production-orientated, advanced technology-orientated and export-orientated industries plus the financial industry; other sectors, such as infrastructure development as well as agriculture, forestry and animal husbandry, may also be designated. 10 See the final statement of government revenue by region (2001), in China Statistical Yearbook (2001). 11 Fixed assets include machinery, transportation vehicles, equipment, instruments and appliances with a useful life of more than one year and goods worth CNY2,000 or more with a useful life of at least two years. The Chinese VAT is therefore a product-type one; indeed when it was introduced China was experiencing a two-digit inflation and the government aimed at curbing capital investment. 12 The list includes: grain and edible oil; running water, hot water, gas, residential coal products, air-conditioning and central heating; books, newspapers and magazines; feed, chemical fertilizers, pesticides and farm machinery as well as other goods stipulated by the state council. 13 The sample consists of eight African countries, nine Asian countries, seven Eastern countries, and 14 Western hemisphere countries. 14 Tanzi and Zee (2000), p. 8. For China refer to Table 6.3. 15 Tanzi and Zee (2000), p. 13. 16 However, we note that in the same period the overall tax level rose by 27.5 percent. 17 Including customs duties, they accounted for more than two-thirds of tax revenue in 2002 and their share averaged almost 70 percent in the 12-year period 1991–2002 to which data in Table 6.3 refer. 18 These figures are our own calculations based on the 2000 and 2001 editions of

162 Domenico D’Amico

19 20 21 22 23 24 25

the China Statistical Yearbook as well as on the statistical appendix in World Bank (2003). See Table 2.5 in World Bank (2002). There was also an increase in the coefficient of variation of provincial per capita expenditure. Data in this section are taken from ADB (2004). For a description of the tax system in the 1980s and until the mid-1990s, see Heady and Mitra (1992). See ADB (2004). Unless otherwise stated, information on tax reforms included in this section is taken from this ADB publication. This news is from an article that appeared in the China Daily issue of 22 July 2004. It is worth noting that reforms in China are usually tested through extensive pilot projects (Ahmad et al. 2000). These losses would vary from a minimum of CNY44.3 billion (based on 2001 data) to a maximum of CNY88.4 billion depending on the level of efficiency in revenue collection. From a minimum of CNY132 billion to a maximum of CNY264 billion.

References Asian Development Bank (ADB) (2004) Asian Development Outlook 2004, on-line edition. Ahmad, E., Li, K. and Richardson, T. (2000) “Recentralization in China?,” Paper presented at the IMF Conference on Fiscal Decentralization, Washington. Ahmad, E., Singh, R. and Lockwood, B. (2004) “Taxation reforms and changes in revenue assignments in China,” IMF Working Paper 125, Washington, DC: IMF. Bao, L. (2001) “China’s underground economy: loss of tax revenue and tax countermeasures,” Asia-Pacific Tax Bulletin, 12: 311–16. —— (2003) “Current situations in tax burden and future tax reform in China,” International Tax Review (Intertax), 31(4): 168–83. China Statistical Yearbook (2000), Beijing: China Statistics Press, 2001. —— (2001), Beijing: China Statistics Press, 2002. Chu, K., Davoodi, H. and Gupta, S. (2000) “Income distribution and tax and government social spending policies in developing countries,” IMF Working Paper 62, Washington, DC: IMF. Heady, C. J. and Mitra, P. K. (1992) “Taxation in decentralizing socialist economies: the case of China,” Policy Research Working Paper 820, Washington, DC: The World Bank. International Monetary Fund (2004a) World Economic Outlook: April 2004, Washington, DC: IMF. —— (2004b) “China’s growth and integration into the world economy: prospects and challenges,” IMF Occasional Paper 232, Washington, DC: IMF. OECD (2002) China in the World Economy: the Domestic Policy Challenges, Paris: OECD. Tanzi, V. and Zee, H. H. (2000) “Tax policy for emerging markets: developing countries,” IMF Working Paper 35, Washington, DC: IMF. Toshiki, K. (2004) “Fiscal reform in the People’s Republic of China: current issues and future agenda,” ADB Institute Research Paper 55, Tokyo: ADB Institute. Wang, S. (2001) “Openness and inequality: can China compensate the losers of its WTO deal?,” Paper presented at the Conference on Financial sector reform in China, Cambridge (MA), 11–13 September.

China 163 World Bank (2002) “China: national development and sub-national finance,” Report 22951-CHA, Washington, DC: The World Bank. —— (2003) “China: promoting growth with equity,” Report 24169-CHA, Washington, DC: The World Bank.

Websites http://www.adb.org – Asia Development Bank. http://www.ibfd.org – International Bureau of Fiscal Documentation. http://www.imf.org – International Monetary Fund. http://www.mof.gov.cn/ – Ministry of Finance of China, Chinese only. http://www.stats.gov.cn/english – National Bureau of Statistics of China, English version. http://www.worldbank.org – World Bank.

7

India Luigi Bernardi and Angela Fraschini

Introduction, contents and main conclusions India is a federal republic with an area of 3,287,590 km2; the population in 2002 stood at about 1.052 billion persons with a population density (people per km2) of 320. At the turn of the century, GDP totaled about US$440 billion. This aggregate figure is the fourth largest in the world, but its per capita counterpart is less than US$450. Poverty remains an enormous problem: according to the World Bank, India has some 433 million people living on less than US$1 a day, 36 percent of the total number of poor worldwide. Non-income poverty also represents a huge problem and many social indicators are worsening. This notwithstanding, India’s economic performance has been impressive over the last two decades. Total GDP grew yearly at about 6 percent, per capita GDP at about 4 percent. During the 1990s the average gross fiscal deficit level was 7.2 percent of GDP, while in recent years the gross fiscal deficit went up to 9–10 percent of GDP. The general government debt fell from 72.5 percent of GDP in 1990–1 to 65.1 percent in the middle of the 1990s, but then rose to 79.8 percent in 2001–2. In the late 1990s Indian general government public expenditure totaled nearly 25 percent of GDP. This figure is not low with respect to the Indian per capita income. However the joint share of education and health is poor and that of social security is still poorer. Around the turn of the century, about 80 percent of public expenditure is financed by means of taxes, whose total pressure stays around 15 percent of GDP. Central government collects more than half of tax revenue, but about a quarter is transferred to the states. Taxes on goods and services are definitively the dominant item (60 percent of total taxes). Direct tax revenue is far lower and barely reaches 20 percent of total taxes. On the contrary, import duties are still not a negligible share (about 12 percent) of total taxation’s yield. Finally, social contributions are wholly lacking. Beginning in the early 1990s, the Indian tax system entered a series of reforms. At present the accomplishments are of some relevance, but

India 165 neither radical nor rapid. They are largely still under way or planned for the coming years (see below). Obviously one must be aware that the economic structure as well as administrative difficulties severely constrain features of the tax system in developing countries. Therefore, in India the change over to the typical “modern” system (PIT ⫹ CIT ⫹ VAT ⫹ few large excise duties) has not been accomplished yet, but is expected to take place in the near future. India has a three-tier federal tax structure (the union government, the state governments and the urban/rural local bodies). The main taxes/duties that the union government is empowered to levy are: income tax, customs duties, excise duties, sales tax and service tax. It is worthwhile to note some peculiar features of this tax bundle. Income tax is a single tax that is levied on a comprehensive basis both on persons and companies. The structure of custom duties is particularly intricate and the burden hits almost all the import side of international trade. Also internal indirect taxes are a complicated matter: they are separately levied on goods, services and intra-state sales. Over time, excise duties and service tax have benefitted from a credit system compensating tax paid on inputs and capital goods that has grown to avoid cascading effects and to approximate a VAT-type structure. The main taxes levied by the state governments are sales tax, stamp duty, state excise, land revenue, duty on entertainment and tax on professions and callings. The local bodies are empowered to levy tax on properties, octroi, tax on markets and tax/user charges for utilities. The Indian tax system is still predominantly made up of a large, complex and obsolete bundle of excises and sales taxes. The scope of direct taxation, both on individuals and companies, is much smaller. Consequently a number of issues arise. The share of personal income tax is more limited than in countries where the value of the per capita income is similar to the Indian one. This may be aimed at preserving poverty income from taxation, but unavoidably the poor are then hit by regressive consumption taxes. Furthermore, the argument that the collection costs are higher for the direct than for the indirect taxes does not seem well demonstrated. Finally, some traditional arguments of economic theory – i.e. that income tax does not preserve savings and might induce supply disincentives – is not consistent with the situation of the Indian economy. In regard to company taxation, its basis should be broadened through the reduction of a large number of multifaceted incentives. To conclude on this point, a great deal of rationalization is still mandatory to avoid cascading effects between national excises and states’ sales tax, random “all in” rates charged on final goods and high costs of collection. For a few years the long-term strategy has moved towards a double VAT system (union and states), leaving just a few excises on specific goods. As we have already noted, India is a federal republic (union and states) with a complex structure of local authorities. The assignment of tax

166 Luigi Bernardi and Angela Fraschini powers is based on the principle of tax separation and the consequence is a vertical fiscal imbalance. The inadequacy of the states to meet expenditure from their own resources is recognized by the Constitution, which provides for grant-in-aid both purposed-based and need-based. Over the last two decades impressive growth has been occurring in the Indian economy. In 2003–4 real GDP at factor cost has been estimated to have grown by 8.2 percent and was accompanied by a relative stability of prices. The fiscal deficit of the central government, in GDP terms, after declining from 6.6 percent in 1990–1 to 4.1 percent in 1996–7, started rising to 5.3 percent in 2002–3. The combined fiscal deficit of the center and the states increased from a level of 9.4 percent of GDP in 1990–1 to a level of 10.1 percent of GDP in the revised estimates for 2002–3. In 1991, in reaction to a severe macroeconomic crisis involving high fiscal deficits, India carried out a series of economic reforms, including tax reform. The main proposals comprised: (i) reduction in the rates of the most important taxes; (ii) enlargement of all taxes’ bases; (iii) transformation of the taxes on domestic production into something similar to a value added tax; and (iv) simplification of laws and administrative procedures. Most of the recommendations have been implemented over the years, at least at the central level. In the case of the states, the reforms of their tax systems did not proceed. In September 2002 the government set up a new task force on tax reforms headed by V. Kelkar. The Kelkar committee suggested sweeping reforms including: (i) raising the limits of income tax exemption and twotier brackets; (ii) a cut in corporate tax rate; (iii) three-rate basic customs duty structure; (iv) service tax levied in a comprehensive manner; (v) repeal of wealth tax; (vi) removal of tax exemptions, rationalization of incentives for savings and simplification of procedures; and (vii) gradual move to the destination-based, consumption-type value added taxes at the state level. The introduction of VAT was repeatedly postponed, mainly because of the lack of administrative preparation and of disagreements between union government and some states. In July 2004 a task force on implementation of the Fiscal Responsibility and Budget Act, also headed by V. Kelkar, has come up with a proposal for an integrated VAT on goods and services to be levied by the central government and the states in parallel, removing all cascading taxes. Of course, dismantling and deeply reforming about 60 percent of taxation will be an operation neither easy nor without the risk of losing yield.

India 167

A broad view of the tax system and its development since the early 1980s A short reminder of the Indian economy and public sector outline India was a British colony and it gained its independence on the 15 August 1947. It is a republic with a federal constitution, consisting of 28 states and seven union territories. India has an area of around 3.3 million km2, and it is the seventh largest country in the word. The population rose by 184.4 percent between 1951 and 2001 and on 1 March 2001 stood at about one billion and 27 million persons. Notwithstanding an infant mortality rate of 6.4 percent in 2002, the population is still rapidly increasing (2.4 percent in 2002 and 1.44 percent estimated in 2004). The country is characterized by striking contrasts, with huge linguistic, religious and cultural diversity. Poverty remains an enormous problem: according to World Bank figures, India has some 433 million people living on less than US$1 a day, 36 percent of the total number of poor in the world. Also non-income poverty represents a big problem: about 25 percent of the world’s deaths due to childbirth, about 23 percent of the world’s under-five child deaths every year and about 20 percent of the world’s children (aged 6–14) out of school are estimated to be in India. The general UN index of human development stands at the 111th place in the world, with a value equal to 0.590 in 2001; endemic diseases are widely dispersed. This notwithstanding, India’s economic performance has been impressive over the last two decades, while in the previous period its rate of economic growth appears ordinary (see Table 7.1). In the mid-1980s and in the 1990s India has been one of the fastest growing economies in the world, with a doubling time for average GDP per capita of only 16 years. A large share of GDP originated in the services sector that accompanied the relative decline in the share of agriculture (Table 7.2). While conventional wisdom imputes the growth acceleration to neoliberal economic reforms of the 1990s, according to DeLong (2001) the sources of that acceleration have to be found in some relatively minor reforms undertaken by Rajiv Gandhi’s government. Other scholars attribute the acceleration in growth in the 1980s to liberalization of trade and industry (see, for example, Pursell 1992 and Desai 1999). Table 7.1 Indian yearly percent rates of economic growth

Total real GDP Per capita GDP

1950–80

1980–90

1990–2000

3.7 1.5

5.9 3.8

6.2 4.4

Source: IMF data quoted in DeLong (2001).

168 Luigi Bernardi and Angela Fraschini Table 7.2 Percent shares of Indian GDP

Agriculture Industry Services

1980

1990

2000

38.6 24.2 37.2

31.3 27.6 41.1

24.9 26.9 48.2

Source: World Bank data quoted in Panagariya (2004).

During the 1980s growth was also driven by fiscal expansion financed by borrowing abroad and at home. But this was unsustainable and led to a crisis in 1991. General government’s (center plus states consolidated) gross fiscal deficit averaged 9 percent of GDP in the second half of the 1980s. The period 1992–3/1996–7 (Eighth Plan) has been one of high growth (the average real GDP growth rate was 7.1 percent yearly) and of fiscal restraint (the average gross fiscal deficit level was 7.2 percent of GDP against 9.3 percent in 1990–1). In the same period, the average revenue deficit (current spending minus revenues) was 3.6 percent of GDP (4 percent in 1990–1) and the average primary deficit (fiscal deficit minus interest payments) was 2.1 percent of GDP (4.8 percent in 1990–1). During the Ninth Plan period (1997–8/2001–2) gross fiscal deficit returned to the 9–10 percent of GDP range, the level of the mid-1980s; the revenue deficit and the primary deficit rose, respectively, to 6.1 and 3.5 percent of GDP. The general government debt fell from 72.5 percent of GDP in 1990–1 to 65.1 percent at the end of the Eighth Plan period, but then rose to 79.8 percent at the end of the Ninth Plan period. In 1998 (IMF’s Government Finance Statistics Yearbook 2001: last data available), Indian general government public expenditure totaled nearly 25 percent of GDP, net from intra-national transfers. The shares of central (11.9 percent) and state government (the remaining 13.1 percent) were rather even. However the two layers’ specific items were rather different. The central government was engaged mainly in defense, economic affairs and services (agriculture, transports, communications and so forth) and paying a huge amount of interest (4.1 percent of GDP) on public debt. State governments devoted their resources mainly to education (3.5 percent of GDP), considerably less to health (0.5 percent) and, on the contrary, somewhat more to economic services, especially agriculture (1.4 percent). Surprisingly enough, social security and welfare programs were virtually absent at the central level, and small also at the state tier (0.6 percent of GDP).1 On the contrary, general administration, law and order, and defense together reach more than 7 percent of GDP, i.e. nearly a third of total spending. A comparison with all developing countries (Burgess and Stern 1993) shows that total spending is not low in India, especially with respect to per

India 169 capita income, and general administration and defense spending is about average. However the share of spending on joint education and health is poor and it is still poorer for social security, taking into account the caveat reported on the reliability of the data. This happens in a country where many social indicators are worsening, as we have already seen. The broad structure of the tax system and its development Tables 7.3, 7.4 and 7.5 show the broad quantitative structure of the Indian tax system and its developments since the late 1980s until the first years of the current century. Data are shown for consolidated general, central and state governments.2 At the turn of the century, tax revenue amounts to about 80 percent of total general government revenue (Table 7.3). The remaining 20 percent is a not-homogeneous item, made up of tariffs, foreign grants, interest and so forth. Its level is not negligible, higher than usually experienced in more advanced countries. By considering the average values among those of the years shown by Table 7.3, total tax pressure is something less than 15 percent of GDP. Taxes on goods and services – almost entirely by excise duties – are definitively the dominant item (near 10 percent of GDP and 60 percent of total taxes). Direct tax revenue is far lower and barely reaches 3 percent of GDP and not more than 20 percent of total taxes. One may calculate that this figure can be split almost evenly between individuals and corporations, as they are both subject to the same single personal tax.3 Property taxes are very low, especially considering that they also include capital transaction tax yield. On the contrary, import duties accounted on average for a significant share (about 20 percent) of total taxation. Finally, social contributions are entirely lacking, like it is common in developing countries and as one could suspect from the low level of social services and welfare programs we have already spoken about. A comparison between the Indian tax structure and its counterparts prevailing in other developing countries (e.g. Burgess and Stern 1993; Tanzi 1994) shows a somewhat outdated picture. The total Indian fiscal pressure is just slightly lower than the figure prevailing in those countries where per capita income is near to the Indian value. However, in comparison, Indian direct taxes appear very low, while indirect ones are quite high and the share of import duties is relatively low. The development of general government tax structure does not show striking changes from the late 1980s to the turn of the century, but some ups and downs and some increasing/decreasing trends as to certain tax items. The total fiscal pressure went down from the beginning of the 1990s. A recovery emerged at the end of the decade but it has not been confirmed in recent years. This movement is explained almost completely by the analogous one of excise duties and was due to the rate reductions that were adopted at that time (see below). Personal taxes gained one

17.1 2.9 14.2 2.0

1.0 1.0

0.5 8.0 7.8 3.8 3.8

19.9 3.9 16 2.1

1.1 1.0

0.6 9.4 9.2 3.8 3.8

1990

0.5 8.1 7.9 3.5 3.5

1.1 1.2

17.9 3.5 14.5 2.4

1991

0.6 9.0 8.8 3.3 3.2

1.1 1.2

19.3 4.0 15.3 2.4

1992

0.6 8.6 8.4 2.7 2.6

1.1 1.2

18.2 4.0 14.2 2.4

1993

0.6 8.5 8.4 2.3 2.2

1.0 1.2

17.7 4.0 13.7 2.2

1994

0.6 8.1 8.0 3.0 3.0

1.3 1.4

18.4 3.9 14.6 2.8

1995

0.6 8.0 7.8 3.1 3.1

1.3 1.4

18.0 3.5 14.5 2.8

1996

0.6 8.0 7.9 2.6 2.6

1.1 1.3

17.3 3.5 13.8 2.5

1997

0.5 7.9 7.8 2.3 2.3

1.2 1.4

17.4 3.9 13.4 2.6

1998

0.6 9.1 9.0 2.5 2.5

1.4 1.5

18.4 3.2 15.2 3.0

1999

0.7 10.3 10.2 2.3 2.2

1.5 1.7

19.6 3.1 16.5 3.3

2000

0.6 8.4 8.3 1.8 1.7

1.4 1.6

17.4 3.6 13.8 3.0

2001

0.7 8.9 8.8 1.8 1.7

1.4 1.8

18.3 3.8 14.6 3.2

2002

Notes Preliminary data for 2002. Fiscal year starting on 1 April.

Sources: Own calculations (see text) according to: IMF, Government Finance Statistical Yearbook, 2000 and 2001; Baird and Ferro (World Bank 2003); IMF (2003: India Selected Issues); Ministry of Finance of India (Union Budget: various years).

Total revenue and grants Non-tax revenue and grants Tax revenue Income and profits, of which: Individual (income tax and other) Corporate Property and capital transaction Goods and services, of which: Excises International trade, of which: Import duties

1989

Table 7.3 Structure and developments of consolidated general government revenue in India, 1989–2002 (% of GDP)

10.7 2.8 10.5 2.0

1.0 1.0 4.7 4.5 3.8 3.8 2.6 7.9

9.0 0.8 11.0 2.1

1.1 1.0 5.0 4.8 3.8 3.8 2.8 8.2

1990

1.1 1.2 4.7 4.5 3.5 3.5 2.7 8.0

11.6 3.6 10.7 2.4

1991

1.1 1.2 4.5 4.3 3.3 3.2 2.8 7.7

11.1 3.4 10.6 2.4

1992

1.1 1.2 3.9 3.7 2.7 2.6 2.7 6.3

9.6 3.3 9.0 2.4

1993

1.0 1.2 3.2 3.1 2.3 2.2 2.1 6.7

10.3 3.1 7.8 2.2

1994

1.3 1.4 3.5 3.4 3.0 3.0 2.4 6.9

10.2 3.3 9.4 2.8

1995

1.3 1.4 3.4 3.3 3.1 3.1 2.6 6.9

10.1 3.2 9.4 2.8

1996

0.0 1.3 3.3 3.2 2.6 2.6 2.9 6.3

9.3 3.1 9.1 2.5

1997

1.2 1.4 3.2 3.1 2.3 2.3 2.2 6.0

9.1 3.1 8.3 2.6

1998

1.4 1.5 3.3 3.2 2.5 2.5 2.2 6.6

9.9 3.3 8.9 3.0

1999

1.5 1.7 3.4 3.3 2.3 2.2 2.5 6.5

9.8 3.2 9.0 3.3

2000

1.4 1.6 3.3 3.2 1.8 1.7 2.3 5.9

9.4 3.5 8.1 3.0

2001

1.4 1.8 3.4 3.2 1.8 1.7 2.2 6.3

9.6 3.4 8.4 3.2

2002

Notes Preliminary data for 2002. Fiscal year starting on 1 April.

Sources: Own calculations (see text) according to: IMF, Government Finance Statistical Yearbook, 2000 and 2001; Baird and Ferro (World Bank 2003); IMF (2003: India Selected Issues); Ministry of Finance of India (Union Budget: various years).

Total current revenue and grants Non-tax revenue Gross tax revenue Income and profits, of which: Individual (income tax and other) Corporate Goods and services, of which: Excises International trade, of which: Import duties Less state share Net tax revenue

1989

Table 7.4 Structure and development of consolidated central government revenue in India, 1989–2002 (% of GDP)

11.2 1.2 1.8 8.1 2.8 5.3 4.7 0.6

11.6 1.3 2.3 8.0 2.6 5.4 4.8 0.5

1990 11.9 1.4 2.4 8.1 2.7 5.4 4.8 0.6

1991 12.1 1.3 2.5 8.2 2.8 5.4 4.8 0.6

1992 11.8 1.3 2.5 8.1 2.6 5.4 4.8 0.6

1993 11.5 1.3 2.4 7.9 2.5 5.4 4.8 0.6

1994 10.6 1.2 2.0 7.4 2.4 5.1 4.5 0.6

1995 10.5 1.1 1.7 7.6 2.5 5.1 4.5 0.6

1996 10.5 1.0 1.6 7.9 2.6 5.3 4.7 0.6

1997 9.8 0.9 1.5 7.4 2.3 5.1 4.5 0.6

1998

9.9 1.0 1.5 7.4 2.2 5.2 4.6 0.6

1999

10.6 1.1 1.7 7.8 2.4 5.4 4.8 0.6

2000

10.6 1.1 1.6 7.8 2.4 5.5 4.9 0.6

2001

11.3 1.1 1.8 8.3 2.5 5.8 5.1 0.7

2002

Sources: Own calculations (see text) according to: IMF, Government Finance Statistical Yearbook, 2000 and 2001; Baird and Ferro (World Bank, 2003); IMF (2003: India Selected Issues); Ministry of Finance of India (Union Budget: various years).

Total current revenue and grants Non-tax revenue Grants from national government Tax revenue, of which: State shares in central taxes Own taxes of which: Goods and services Property and capital transaction

1989

Table 7.5 Structure and developments of state government revenue in India, 1989–2002 (% of GDP)

India 173 point of GDP – i.e. 50 percent of the starting value – while the share of import duties more than halved. Notwithstanding this somewhat static picture, a lot of reforms were introduced into the Indian tax system during the 1990s, on the wave of the proposals of the Tax Reforms Committee, chaired by R. J. Chelliah in 1991 (Shome 1997; Rao 2000; Sarma and Gupta 2002). Broadening the bases, reducing tax rates, simplifying the system, making it more supply friendly and attractive for FDI were the main aims of the suggested reforms. But the accomplishments were somewhat more limited than the initial challenging design, as we will see at the end of this chapter. During the 1990s the central government collected nearly two-third of consolidated tax revenue, but about a quarter was transferred to the states, so that the net tax revenue was almost evenly distributed between the two main government layers. The tax mix of central government was far more balanced than it was at general government level. The union budget appropriates all direct taxes and import duties, and about half of total excise duties. The relative weight of these three items is about the same. During the 1990s, excise duties went down, as did import duties. Notwithstanding the increase in direct taxes, central government total tax pressure showed an overall decreasing trend, albeit with ups and downs. A more flat level of total tax revenue is on the contrary shown by state accounts, with the exception of some decrease during the second half of the 1990s. The two main headings of tax financing – both quite stable during the decade – are the shares in central taxes (about one-third) and states’ own taxes. The latter are almost all excise duties so that this is the dominant item also in tax financing of the states. Finally, it is worthwhile to note that the share of taxes is just a half of state total revenue, once the former is removed from the contribution of central government’s taxes. To sum up, the Indian tax system seems to be slowly moving away from the traditional features of Musgrave’s “early stage” (Musgrave 1969). Obviously the economic structure and administrative capabilities severely constrain features of tax systems in developing countries. The large share of agriculture and the prevailing small scale of early manufacturing prevent the determination of business income. Therefore income tax could be effectively applied just to wage income of the civil servants and to the employees of large firms. Retail or multiple-stage sales taxation is difficult to implement effectively. However in India the changeover to the typical “modern” system (PIT ⫹ CIT ⫹ VAT ⫹ few large excise duties) has not been accomplished yet or is just at the starting stage.

Some quantitative and institutional features of the main taxes India has a tax structure with a three-tier federal structure (the union government, the state governments and the urban/rural local bodies).

174 Luigi Bernardi and Angela Fraschini The power to levy taxes and duties is distributed between the union government and the state governments in accordance with the provisions of the Indian Constitution.4 The state government may delegate any of its fiscal powers to local authorities, which do not have any constitutionally reserved powers of taxation. The main taxes/duties that the union government is empowered to levy are: income tax (except tax on agricultural income, which the state governments can levy), customs duties, excise duties (except on alcoholic liquors or narcotics), sales tax and service tax. The principal taxes levied by the state governments are sales tax (tax on intra-state sale of goods), stamp duty (duty on transfer of property), state excise (duty on manufacture of alcohol), land revenue (levy on land used for agricultural/non-agricultural purposes), duty on entertainment and tax on professions and callings. The local bodies are empowered to levy tax on properties (buildings, etc.), octroi (tax on entry of goods for use/consumption within areas of the local bodies), tax on markets and tax/user charges for utilities like water supply, drainage, etc. Direct taxes Income tax Income tax is charged under the Indian Income Tax Act, 1961. It is an annual tax on income of both individuals and companies5 levied by the union government. Every person (individuals, Hindu undivided families, companies, firms, association of persons or bodies of individuals and all other artificial juridical persons), whose total income exceeds the maximum exemption limit, is subject to income tax at the rates prescribed in the Finance Act passed each year by parliament. The income tax is paid on the total income of an individual, determined on the basis of her/his residential status in India.6 The tax is charged in respect of the income of the previous year – that is the financial year, beginning on 1 April and ending on 31 March – and the same is chargeable in the assessment year – that is the next financial year. Taxable income includes the following categories: salaries, income from house property (determined by reference to the annual value of property), profits and gains of business or profession, capital gains and income from other sources (including interest).7 The basis of taxation is the gross receipts after deducting the related expenses incurred in connection with earning such receipts. Such deductions, determined according to rules that vary from different categories of income, are allowed from the aggregate of income and are in the nature of incentive provisions of different kinds. The main deductions are the following:

India 175 • •

• •

•

• •

•

Standard deduction – available to certain taxpayers receiving salary or pension.8 Entertainment allowance – provided to government employees who may claim a deduction up to 20 percent of their salaries or RS (rupees) 5,000, whichever is lower, for certain entertainment allowances granted by the employer. Tax – for any sum paid by an employee on account of state or municipal tax on employment. Annuity and insurance payment – up to RS10,000 per annum for payments made in respect of an annuity contract in order to receive a pension. Repayment of loan – deduction of up to RS40,000 per annum for the repayment, including interest, of loans used to finance higher education (available for eight assessment years); deduction of up to RS150,000 in respect of interest on capital borrowed to purchase or construct owner-occupied residential property. Donation – to charities approved for tax purposes up to the limit prescribed by the tax authorities. Investment income – up to RS9,000 per annum in respect of interest received from certain specified investments, including dividends from co-operative societies and interest on bank deposits; up to RS3,000 per annum for interest received from government securities. Permanent physical disability – up to RS40,000 for a permanent physical disability (which occurred in the previous year) certified by a competent person.

Other deductions are granted as tax incentives, for example: new export-oriented undertakings are entitled to an exemption from income tax; new industrial undertakings that fulfill certain conditions are entitled to a deduction (25 or 30 percent in the case of a company) of the profits for a period of ten consecutive assessment years (or 12 for a co-operative society); a deduction of 50 percent of profits is available to hotels in hilly/rural areas and pilgrimage centers (except Calcutta, Madras, Delhi and Mumbai), and so forth. After reducing the gross total income by the amount of deductions, what is left is the total income that is the basis for taxation.9 If the total income is below the basic exemption limit, no tax is chargeable. All receipts having the character of income are taxable unless they are specifically exempt from taxation. For tax purposes spouses are treated separately and generally their income is not clubbed together. On the contrary, income of all minors, except handicapped ones, is assessed with the income of their parents, unless the income is derived from manual work or an activity involving skill, specialized knowledge and experience. The tax rates for the assessment year 2004–5 are listed in Table 7.6.

176 Luigi Bernardi and Angela Fraschini Table 7.6 Tax rates, assessment year 2004–5, India (Finance Bill) a) Individuala Net income range (RS) up to 50,000 50,001–60,000 60,001–150,000 150,001 and above b) Co-operative society Net income range (RS) up to 10,000 10,001–20,000 above 20,000

Rateb nil 10% of the amount exceeding RS50,000 RS1,000 plus 20% of the amount exceeding RS60,000 RS19,000 plus 30% of the amount exceeding RS150,000 Ratec 10% of total income RS1,000 plus 20% of the amount exceeding RS10,000 RS3,000 plus 30% of the amount exceeding RS20,000

c) Firm and domestic company

35% of the total income plus a surcharge of 2.5% of the income tax

d) Local authority

30% of the total income plus a surcharge of 2.5% of the income tax

Notes a The tax rates applicable to individuals are also applicable to Hindu Undivided Family (HUF), Association of Persons (AOP) and Body of Individuals (BOI). b A surcharge of 10% of the income tax is levied (except by non-residents) where taxable income exceeds RS85,000. c A surcharge of 2.5% of the income tax is levied.

Different types of assessments are provided: 1 2

3

Self-assessment (the taxpayer is required to make a self-assessment and pay the tax on the basis of the returns furnished). Regular assessment (on the basis of the return of income chargeable to tax furnished by the taxpayer an intimation is sent to her/him informing about the tax or interest payable or refundable). Best judgment assessment (the assessing officer bases the assessment on her/his best judgment).

Wealth tax Wealth tax is charged under the Indian Wealth Tax Act, 1957 and the union government levies it. The tax is charged on individuals, Hindu undivided families (HUF) and companies in respect of the net wealth held by them during the assessment year. Indian citizens, resident companies and HUF are charged in respect of their worldwide assets, whereas non-residents are charged in respect of assets located in India. Net wealth is the aggregate of the assets owned by the taxpayer,10 less

India 177 the debts owed by her/him relative to the taxable assets. From the computation of net wealth some assets are excluded (for example, the value of one house or plot of land for an individual or a HUF). Among the assets subject to wealth tax there are, for example: buildings, or land belonging with them, used for residential or commercial purposes or as a guest house or farm house, within 25 km of the local limits or cantonment board; motor cars (other than those used in a business carhire or which are stock-in-trade); jewels or precious metals (unless they are stock-in-trade); yachts, boats and aircraft (unless used for commercial purposes); urban land (with some exclusion); and cash in hand in excess of RS50,000. Among the entities that are exempt from the wealth tax there are: any social club; any political party; or any cooperative society; any company whose object is the promotion of art, science, religion, charity, commerce, etc. The rate is 1 percent on net wealth exceeding RS1.5 million. There is a self-assessment scheme. Other taxes on capital and property Other taxes on capital and property are levied by the states and the local authorities. The states impose: a land tax on the value of land (the methods of valuation and the rates vary from state to state); and a tax on motor vehicles, whose yield is used for the development and maintenance of state roads. The local authorities impose: land cesses in the form of a surcharge on land revenue; a tax on land and buildings, generally based on the annual rental value; betterment taxes, based on increases of land value caused by town planning and town improvement; and taxes on the transfer of immovable property, based on the value of the property and in addition to state stamp duty. Expenditure tax The expenditure tax is charged under the Indian Expenditure-Tax Act, 198711 and it is imposed by the union government. The tax is charged at the rate of 10 percent on any chargeable expenditure incurred in a hotel wherein the room charges12 for any unit of residential accommodation are RS3,000 or more per day. The expenditure tax is collected by the person who carries on the business of such a hotel. The tax collected during any calendar month is paid to the credit of the union government by the 10th of the month immediately following the said calendar month. Any person responsible for collecting the expenditure tax who fails to collect it shall pay, in addition to paying the tax, a sum equal to the amount of tax that s/he failed to collect.

178 Luigi Bernardi and Angela Fraschini From the 1 October 1991 to the 31 May 1992 a tax at the rate of 15 percent of the chargeable expenditure incurred in a restaurant was levied. Indirect taxes Customs duties The Constitution has given to the union the right to legislate and collect duties on goods imported into or exported from India. The Customs Act, 1962 is the basic statute, effective from 1 February 1963. The categories of items and the rates of duties which are leviable have been specified in two schedules to the Customs Tariff Act, 1975. The first schedule specifies the various categories of import items, in accordance with an international scheme of classification of internationally traded goods (Harmonized System of Nomenclature (HSN), established by the World Customs Organization). All goods are classified into categories, called “headings” and “sub-headings;” for each sub-heading, a specific rate of duty is prescribed. The duties are levied both on a specific and ad valorem basis, while there are a few cases where at times specific-cum-ad valorem duties are also collected on imported items. Where ad valorem duties are collected, the value of the goods has to be determined for customs duty purposes according to WTO valuation agreement. Under the Custom Tariff Act, 1975 and other laws, the following types of duties are leviable: •

• •

•

Basic customs duty – that is duty specified against each heading or subheading in the first schedule. There are different rates of duty for different commodities and there are preferential rates for goods imported from certain countries in accordance with bilateral agreements with such countries. The duty may be ad valorem or specific. Surcharge – that is levied at the rate of 10 percent of the basic customs duty on imported goods, unless exempted by a notification. Additional duty of customs – equivalent to the excise duty leviable on goods produced or manufactured in India. Generally it is on an ad valorem basis, though specific rates are prescribed for some items. Imported goods that are to be used as inputs for manufacture of other goods are generally eligible for a credit (called CENVAT credit) equal to the additional duty of customs paid on the imported goods. This credit can be used for paying central excise duties. Special additional duty – whose amount is computed by applying the specified rate13 on the total of the assessable value, the basic customs duty and the additional duty of customs described above.

There are also additional levies on particular items and other levies which are specific to the country of origin. Among the latter there are

India 179 anti-dumping duty, on specified goods imported from specified countries to protect indigenous industry, and safeguard duty, applicable to certain goods for specified periods in order to check their excessive imports which may damage the Indian industry. The rates vary and are based on official notification. The custom duty on exports is levied on items listed in the second schedule to Customs Tariff Act, 1975. Currently, the rates vary from 10 to 60 percent and they are either ad valorem, specific or a combination of both. Very few items are subject to customs duties on their export. In order to make exports more competitive, a duty exemption scheme is provided for registered exporters so that they may import the inputs required for export production at international prices and free from duty. Imported items that are exempt from customs duty are raw materials, components and consumables. Central excise duties Central excise duties are charged under the Central Excise Act, 1944 at the rates specified in the schedules to the Central Excise Tariff Act, 1985. They are an indirect tax levied on goods produced or manufactured in India, excluding those produced or manufactured in special economic zones. There are several types of duties which become payable at the time of clearance of such goods. These duties are: •

•

•

•

Basic excise duty (specified against each sub-heading in the First Schedule to the Central Excise Tariff Act, 1985), actually called the Central Value Added Tax (CENVAT).14 Special excise duty (leviable only on a few items, in addition to CENVAT, at the rate specified under the second schedule to the Central Excise Tariff Act, 1985). Additional duties of excise (leviable on various commodities, such as specified textiles and textile articles, or on sugar or tobacco products in lieu of sales tax). Cess (on different items – for example, spices, agriculture and processed food products, coffee, marine products which are exported – through special enactment).

The duty is payable by the manufacturer at the time of removal of goods from the factory premises or warehouse; the taxable base is the wholesale price of the goods manufactured. However, to achieve particular objectives – for example, to promote exports, to avoid multiple taxation, to promote educational and research activities, or to encourage the use of specified raw materials – it is in the power of the central government to exempt certain excisable goods from the whole or part of the duty leviable on such goods.

180 Luigi Bernardi and Angela Fraschini The general rate of the basic excise duty in 2002–3 is 16 percent, but there are a number of items that are subject to either a “nil” or ad valorem rate. In the same fiscal year the special excise duty is levied at a rate of 16 percent, although there are a number of items subject to a rate of 8 percent; additional duties of excise are imposed at rates ranging from 5 to 18 percent, or at ad valorem rates. Cess is applied at varying rates (for example, 0.5 percent on spices and agriculture and processed food products, RS25.00 per quintal on coffee). Unless the assessee is a cigarette manufacturer (in which case the assessment is carried out by the authorities), a self-assessment procedure is provided. Service tax Service tax was introduced in India for the first time in 1994. It extends to the whole of India except the states of Jammu and Kashmir. It is levied, collected and appropriated by the union government. Service tax is levied on specified taxable services and the responsibility of payment of the tax falls on the service provider. The Finance Act 2001 introduced self-assessment for service tax returns, which are expected to be filled in half yearly and by the 25th of the month following the half-year.15 This replaces the monthly/quarterly returns prescribed earlier. Initially the service tax was imposed on the following services: telephone, stockbroker and general insurance. Over the years it has been extended to other services, such as advertising agencies and courier agencies. At present the total number of services on which service tax is levied has gone up to 58, despite withdrawal of certain services from the tax net or grant of exemptions. At the time of writing, in the budget 2003–4 more services16 have been added to the tax net and the levy of service tax on these services is effective from 1 July 2003. Service tax is levied on the gross or aggregate amount charged by the service provider to the receiver; only in particular cases is the tax permitted to be paid on the value received. Since the 14 May 2003 it is collected at a rate of 8 percent, while the previous rate was 5 percent. To reduce the cascading impact of tax on tax and to help restore competitiveness of the service sector, a credit of the service tax paid on the input service has been allowed since 2002. At present the assessee can avail himself of input credit in respect of any of the categories of the services and utilize the said service tax credit for payment of service tax on any of the output services. Sales tax Sales tax is charged under the Central Sales Tax Act, 1956. It is levied on the sale or purchase of goods. There are two kinds of sales tax: (i) central

India 181 sales tax (CST), imposed by the union government; and (ii) sales tax, imposed by each state. Central sales tax is generally payable on the sale of all goods by a dealer in the course of inter-state trade and commerce and it is levied in the state where the movement of goods commences. Although the tax is imposed by the central government, the revenue is administered by the state in which it is levied. The taxable base is determined by applying the appropriate rate, depending on the type of transaction, to the dealer’s turnover.17 Sales tax on intra-state sale or purchase of goods (other than newspapers) may only be imposed by the state in which the sale or purchase takes place. Nearly all the states impose sales taxes at rates that range from 4 to 15 percent.

Some critical issues of the Indian tax system The predominance of a complex system of indirect taxes We have already noted that the Indian tax system is still predominately made up of a large, complex and entangled bundle of excises and sales taxes. The amount of direct taxation, both on individuals and companies, is very small. Formal rates are not particularly low, especially for corporations (see previous section) but they have a small base. Such a system is not just a consequence of free tax policy choices, but is defined by the severe constraints – economic, social and administrative in nature – which limit the room for maneuver to build the tax systems of developing countries (Musgrave 1969; Burgess and Stern 1993; Tanzi 1994). The predominance of a complex system of indirect taxes, however, raises a number of critical issues that require some discussion. The limited share of direct taxes on individuals The per capita income is not the only explanatory factor of the total fiscal pressure,18 as well as of the level of a particular tax. However, it is a key factor and a starting point to compare different countries’ tax levels, not to be disregarded. If we put the issue in that way, the current level of personal income tax in India stands about one-third below the figure computed by the prevailing literature (e.g. Burgess and Stern 1993)19 with reference to the bracket that encompasses the Indian per capita income. Why not a larger amount of PIT? Why do just 32 million taxpayers file income tax returns, in a population of more than one billion?20 An initial argument to limit the scope of income tax could be the need to preserve poverty incomes from taxation. Otherwise avoiding any taxation of consumption by the poor is virtually impossible. It is also commonly recognized that consumption taxes tend to be regressive, especially

182 Luigi Bernardi and Angela Fraschini in developing countries, if they are not coupled with commanding and administratively costly measures of price subsidies, in-kind rations and transfers. Furthermore the rates should be scheduled according to a steeply – revenue reducing – increasing tax rate profile (e.g. Burgess and Stern 1994). It might well be that a properly structured, more comprehensive income taxation could perform a better redistributive job. Of course agricultural incomes should also be taxed at a non-negligible level, but without burdening poor peasant households. Land tax may be a good solution in terms of both efficiency and equity (Burgess and Stern 1993). It also is often assumed that the costs of administration and compliance are higher for direct than for indirect taxes, especially inside the informal setting of developing countries’ economies. Here the true difficulty is more specific but common to the two kinds of taxation: the need to improve the capabilities of both the tax administration and of the taxpayers. In this connection, it has been demonstrated that simple reforms of personnel policy inside the Indian income tax administration can produce significant enforcement and compliance gains (Das-Gupta et al. 2004) so reducing the cost of collection per unit of yield. Furthermore, the complex system of Indian indirect taxes seems not to cope with an inexpensive system of tax collection. Finally, according to standard economic theory, consumption taxes would be more saving preserving and income taxes could induce supply disincentives, especially as their rate schedule becomes steeper. Note, however, that these arguments mainly apply to countries where the per capita income is already higher than it is today in India. On the contrary, we may infer that taxation has little to do with savings in a still so impoverished country and with the labor supply in a country where the employed workers are not more than 24 percent of the total population.21 Broadening companies’ tax base The native literature unambiguously recognizes (e.g. Kwatra 1997) that the corporate backward effective rates are far lower than the legal ones. This is mainly due to tax holidays for new small undertakings and venture capital, to incentives to export, to many allowances for FDI and to deductions for particular sectors (e.g. power plants, infrastructures, industrial research). This happens when it is well recognized by a general authoritative opinion (rooted especially inside the international organizations – OECD/WB/IMF) that playing the field and reducing the standard rate is more incentive inducing than the sector allowances, in transition and developing economies too (e.g. Owens 2004). Rationalizing indirect taxes We have already noted that the present extensive system of indirect taxes is quite complicated and confused. A state sales tax is added to the

India 183 national excise duties. The national excises hit goods at the production stage, while the services are subject to a different separate tax. Furthermore, the national excises are organized in a multi-class structure (basic, special, additional excises and cess), according to various types of goods. In its turn, the sales tax (anyway payable to the state) is set by the central government for inter-state trade and by the states themselves for intrastate trade. This system unavoidably raises cascading effects (although partially mitigated by widespread deductions of the taxes paid on inputs and capital goods in the case of the national excises and service taxes) and may result in random “all in” rates charged on final goods and services. Many recent government reports (e.g. Government of India 2002) underline the need for in-depth rationalization of the structure of indirect taxation and improvement of tax administration and taxpayer compliance, along with reduction in the costs of collection. For some years the longterm strategy has been in the direction of a double VAT system (at the central level on manufactured goods and at the state level on retail sales). Only some excise duties on particular goods (tobacco, alcoholic beverages, energy) should survive. Unambiguously the literature favors this move (e.g. Shome 1997; Rao 2000), whose current steps are reported below. However, the setting of rates will not be easy: the unavoidable trade-off between equity and efficiency (and yield) clearly emerged in a seminal paper devoted just to India (Ahmad and Stern 1984). Finally, an extensive reform of what accounts for about 60 percent of total taxation will not be easy or without risk of, at least temporary, revenue losses. All this would make unsustainable the Indian budget position, already suffering a high-level deficit (see next section).

Tax reforms A quick glance at the macroeconomic and budget outlook 22 Over the last two decades the Indian economy has made significant improvements with an annual average growth rate rising from 2.9 percent in the 1970s to 5.8 percent in the 1990s. Notwithstanding this improvement, the per capita income remains very low in comparison with other East Asian countries, particularly China, which had the same level of per capita income as India in the 1970s.23 The progress in growth was accompanied by structural changes quite different from those experienced in other developing countries, where a decline in the share of agriculture in GDP was coupled by a remarkable expansion of industry. In India between the 1970s and 2003–4, the share of agriculture and allied sectors in GDP declined from an average of 42.8 to 22.1 percent, while that of services rose from an average of 34.5 to 51 percent. The share of industry showed a small increase from an average of 22.8 to 26.9 percent. In 2002–3 the GDP growth decelerated from 5.8 percent of the previous

184 Luigi Bernardi and Angela Fraschini year to 4 percent, mainly because of a heavy decline of 5.2 percent in the agriculture and allied sectors, due to a severe drought (the industrial sector growth was 6.4 percent, whereas the services sector growth was 7.1 percent). In 2003–4 real GDP at factor cost has been estimated to have grown by 8.2 percent, sustained by a growth of 9.1 percent in agriculture and allied sectors (aided by an abundant monsoon), 6.7 percent in the industry sector and 8.7 percent in the services sector. The growth recovery in 2003–4 was accompanied by a relative stability of prices; inflation, as measured by the wholesale price index (WPI), was 5.5 percent on average, while retail price inflation, as measured by the consumer price index for industrial workers (CPI-IW), declined from 5.1 percent in April 2003 to 2.2 percent in April 2004. Higher growth rates are needed for the rapid elimination of poverty, in spite of the fact that there already was a significant decline in the poverty ratio from 36 percent in 1993–4 to 25 percent in 2001–2. In recent years domestic demand was the main driver of growth; during the period 1998–9 to 2002–3, on average, the contributions of private final consumption expenditure and investment to growth of GDP at current market prices were 64.2 percent and 21 percent, respectively. Since 2001–2 the current account of India’s balance of payments recorded a surplus, indicating that the rest of the world has contributed to support aggregate demand; moreover, a strong balance of payments position in recent years resulted in a steady accumulation of foreign exchange reserves. Public finances, which have been under pressure since 1997–8 on account of the pay revision of government employees and the economic slowdown, showed a further deterioration. The fiscal deficit of the central government, in GDP terms, after declining from 6.6 percent in 1990–1 to 4.1 percent in 1996–7, started rising to 5.3 percent in 2002–3. The deterioration in revenue deficit was sharper: in 1990–1 it reached 3.3 percent of GDP, then declined to 2.4 percent in 1996–7 and started rising to 4.4 percent in 2002–3. The main factors that have contributed to this deterioration have been rising expenditure on salaries, interest payments (higher fiscal deficits have resulted in higher government borrowings), unfounded pensions, improperly targeted subsidies and stagnation in the tax–GDP ratio that continues to remain at a lower level even compared with the pre-reform year of 1990–1. The tax–GDP ratio for the central government declined from 10.1 percent in 1990–1 to 8.8 percent in 2002–3 and the decline was entirely due to the sharp decrease in the ratio of indirect taxes to GDP (7.9 percent in 1990–1; 5.3 percent in 2002–3). That decrease was mainly imputable to the reduction of customs duty provided by the tax reforms to improve both resource allocation and efficiency and to make Indian manufacturing competitive. The total expenditure–GDP ratio of the central government, after declining from 17.3 percent in 1990–1 to 13.9 percent

India 185 in 1996–7, started rising from 14.2 percent in 1997–8 to 16.2 percent in 2002–3. The fiscal situation improved in 2003–4: in the revised estimate, the fiscal deficit and the revenue deficit came down, respectively, to 4.8 and 3.6 percent of GDP. The deterioration in the fiscal situation of the states was even sharper. The fiscal deficit increased from 3.3 percent of GDP in 1990–1 to 4.7 percent in the revised estimates for 2002–3. In the same period the revenue deficit deteriorated from 0.9 to 2.5 percent of GDP on account of the growing burden of interest payments, pension liabilities and administrative expenditure. Losses of state-owned public enterprises, inappropriate user charges and deceleration in central transfers finished the job of deteriorating revenue deficit. The combined fiscal deficit of the center and the states increased from a level of 9.4 percent of GDP in 1990–1 to a level of 10.1 percent of GDP in the revised estimates for 2002–3, whereas the combined revenue deficit as a proportion of GDP was 4.2 and 6.7 percent in the same years. Tax reforms of recent years, under way and planned In 1991, in reaction to a severe macroeconomic crisis involving high fiscal deficits, India carried out a series of economic reforms,24 among them a tax reform. The tax reform strategy was largely based on the Raja Chelliah Committee Report. The main proposals put forth by the committee comprised: (i) the reduction in the rates of the most important taxes – namely individual and corporate income taxes, excises, customs – still maintaining the progressivity of the system but not such as to induce evasion; (ii) the enlargement of the tax base of all taxes by reducing exemptions and concessions; (iii) the transformation of the taxes on domestic production into a value added tax; and (iv) the simplification of laws and procedures to make the administration and enforcement of the tax system more effective. Most of the committee’s recommendations have been implemented over the years, at least at the central level. Relative to personal income tax, to try to decrease tax evasion there has been a strong reduction in tax rates, both in the number (to three) and in the value (10, 20 and 30 percent25), the tax threshold was raised from RS40,000 to 50,000,26 and the number of brackets was reduced from seven to three. Also the rates of corporate income tax were reduced (from 40 to 35 percent for domestic companies and from 50 to 48 percent for foreign companies), but there was no wide broadening of the corporation tax base, mainly due to tax holidays and rapid depreciation given to investments in many activities. Notwithstanding the reduction in the marginal tax rates, the revenues from personal and corporate income taxes increased after the reforms and therefore the share of revenue from direct taxes as a proportion of

186 Luigi Bernardi and Angela Fraschini GDP showed a significant increase (from less than 14 percent in 1990–1 to 24 percent in 1997–8). A not unimportant share of that increase was due to the voluntary disclosure scheme (VDIS)27 that was introduced in 1997–8 to provide an opportunity for individuals, companies and non-resident Indians to declare their concealed income and assets by paying 30 percent tax. There were also reductions in the wealth tax rate (only one rate equal to 1 percent of the amount by which net wealth exceeded RS1.5 million ⫽ about C30,000) and in the basic exemption of the gift tax (from RS20,000 to 30,000). As regards the tariffs, both the average and peak tariff rates were drastically reduced, whereas in terms of rate differentiation the number of tax rates remained high and in more recent years has even increased. In the case of union excise duties there was an important simplification and rationalization: the number of rates was reduced and, in respect of the majority of commodities, the tax was progressively transformed from a specific into ad valorem levy. Exemptions and the lowest rate (8 percent) were removed thus broadening the tax base. A tax on specific services (telephones, non-life insurance and stock brokerage) was introduced in 1994–5 and that tax was extended successively to a large number of services. The reforms’ effect on revenue was to reduce it: the improvement in direct tax revenue only partly compensated for the decline in indirect tax revenue, mainly due to the reduction in import duty rates and in excise duty rates for items of mass consumption. While the reform of the central government’s tax system has been implemented during the 1990s, although not completely, in the case of the states the reforms of their tax systems did not proceed, notwithstanding the recommendations of the study group appointed by the government of India to rationalize and harmonize the state tax systems themselves. In September 2002 the government set up a new task force on tax reforms and successively a task force on implementation of the Fiscal Responsibility and Budget Management Act, 2003 (FRBM Act), both headed by Vijay Kelkar.28 The Kelkar committees had suggested sweeping reforms including: (i) raising income tax exemption limit to RS100,000 and a two-tier rate structure (20 percent for income of RS100,001 to RS400,000 and 30 percent for income above RS400,000); (ii) a cut in corporate tax rate from 35.875 to 30 percent for domestic companies – to remove the gap between the peak rate for personal income tax and the corporate tax rate – and a cut in depreciation rate for plant and machinery to 15 percent from 25; (iii) a three-rate basic customs duty structure (raw materials 5 percent, intermediate goods 8 percent and finished goods 10 percent); (iv) service tax levied in a comprehensive manner,29 leaving out only a few services (public utilities and social services) to be included in a

India 187 negative list; (v) abolition of wealth tax; (vi) merging of tax on expenditure in hotels with service tax; (vii) abolition of the concessional treatment of long-term capital gains through a reduced scheduler tax rate; (viii) removal of tax exemptions, rationalization of incentives for savings30 and simplification of procedures; and (ix) gradual moving over to the destination-based, consumption-type value added taxes at the state level. The decision to introduce VAT was discussed first at a conference of state chief ministers and finance ministers in 1999 and the deadline of April 2002 was decided to bring in the tax. However the introduction of VAT was postponed to April 2003 and successively to April 2005, mainly because of the lack of administrative preparation of some states. Moreover, there was no agreement between the central government and the states on the system of compensating the states that incur revenue loss on account of VAT’s implementation. Only on 2 November 2004 was an agreement reached after all states, except three, declared they were ready with the necessary legislation. Therefore, the sales taxes of states are going to be replaced with a harmonized VAT from April 2005, based on a blueprint finalized by the empowered committee of state finance ministers.31 Meanwhile, in July 2004 the above-quoted task force on implementation of the FRBM Act has come up with a proposal for an integrated VAT on goods and services to be levied by the central government and the states in parallel, removing all cascading taxes, such as, for example, octroi, central sales tax, state level sales taxes etc. The task force proposed a “grand bargain” whereby the states would have the power to tax all services currently with the center,32 and therefore both central and state government would exercise concurrent but independent jurisdiction over common tax bases extending over all goods and services. The new goods and services tax (GST) would have three ad valorem rates, in addition to the zero rate. The proposed rate structure considers a floor rate, equal to 6 percent for the center and 4 percent for the states, a standard rate, equal to 12 percent for the center (to replace the CENVAT of 16 percent)33 and 8 percent for the states, and a higher rate, equal to 20 percent for the center and 14 percent for the states. Under this proposal, the total tax burden on most goods and services would work out to 20 percent, comparable with the standard VAT rates in OECD countries. Moreover, the treatment of imports and exports should be fully integrated with the dual-GST system. In particular, for imports a two-part levy should replace the countervailing duty (CVD) with the first part reflecting the central GST and the second reflecting state-level GST. All imports should be charged to the central and state GST at the same rate applicable to domestic goods. According to this proposal, the states would obtain revenues from taxation of services and from access to GST on imports, but, in our opinion, their fiscal autonomy would be undermined owing to the uniform rates across the states.

188 Luigi Bernardi and Angela Fraschini According to the task force, the reforms proposed would have great positive implications for India’s outlook and would make the most of the tax system, as part of efforts to cancel revenue deficit and lower fiscal deficit to less than 3 percent of GDP by 2009. Moreover, the implementation of the proposed fiscal reforms should reduce both tax evasion and costs of compliance, and should eliminate most of the distorted behavior coming from tax avoidance.

Notes 1 The figures in the text, however, may be under-reported. Some welfare spending could in fact be hidden inside certain agricultural or other economic activities’ support. Furthermore, welfare services are largely supplied by local authorities whose accounts are largely defective. 2 The main source is the IMF Government Finance Statistics Yearbook (2000, 2001). It has been integrated especially with the aid of Baird and Ferro (2003), IMF (2003) and India Union Budget (1995–2002). The general government account is our estimate, this account not being available in official data sources. We gleaned central government accounts from state share taxes and highlighted state grants revenue from the union. These data are mainly drawn from Baird and Ferro (2003), from whom the GDP series is drawn too. Pay due care to the fact that these sources are not perfectly homogeneous and that there is some evidence that last years’ data are better than previous ones. 3 A Kaldor-type expenditure tax was levied in 1956–7, but it was repealed three years later because of its revenue shortage (Rao 2000). 4 The Constitution points out three lists of legislative fields: (1) the union list (in which the central government has exclusive jurisdiction); (2) the state list (in which the state governments have exclusive jurisdiction); and (3) the concurrent list (in which the union government and the state governments have concurrent jurisdiction, subject to the power of the union government). 5 In the statistics, income tax payable by corporate bodies goes under the head “corporation tax.” 6 An individual is considered a “resident” if s/he stays for the prescribed period during a fiscal year either for 182 days or more, or 60 days or more (182 days or more for non-residents) and has been in India for an aggregate of 365 days or more in the previous four years. Any person who does not satisfy these norms is considered as a “non-resident.” 7 However, interest derived in the course of a business is taxed as business profits under the head “profits and gains of business or profession.” 8 From the assessment year 2001–2 the standard deduction is equal to: 33.33 percent of salary up to RS150,000 or RS30,000, whichever is lower; RS25,000 for salary ranging from RS150,001 to RS300,000; RS20,000 for salary ranging from RS300,001 to RS500,000; no deduction for salary over RS500,000. 9 In certain cases (for example, interest, winnings from lotteries, horse races, card games and other games of any sort), income tax is deducted at source at the rates in force. For the assessment year 2004–5 the tax rate for interest is 10 percent, whereas for the other items above it is 30 percent, in the case of a person (other than a company) that is resident in India. 10 The net wealth computation of an individual must include assets transferred to a spouse or minor child. 11 The act extends to the whole of India except the states of Jammu and Kashmir.

India 189 12 In the case where a composite charge is payable in respect of residential accommodation and food, the room charges included therein shall be determined by deducting from the composite charge, the charges for food in the following manner: (i) where the composite charge includes the charge for breakfast: 10 percent of the composite charge; (ii) where the composite charge includes the charge for breakfast and one meal: 25 percent of the composite charge; (iii) where the composite charge includes the charge for breakfast and two meals: 40 percent of the composite charge. As of 1 June 2002 the definition of “chargeable expenditure” excludes payments made to the hotel in respect of food, drinks or any other services. 13 The maximum rate permissible is 8 percent. Currently there are various rates, with the maximum rate being 4 percent. 14 The duty paid on specified inputs and capital goods used in relation to the manufacture of specified final products can be claimed, under specific conditions, as a credit (CENVAT credit). 15 The individual assessees are required to pay the levy only once in a quarter. 16 The services are the following: (i) commercial vocational institute, coaching centers and private tutorials; (ii) technical testing and analysis (excluding health and diagnostic testing), technical inspection and certification service; (iii) maintenance and repair services; (iv) commission and installation services; (v) business auxiliary services, namely business promotion and support services (excluding information technology services); (vi) Internet cafe; (vii) franchise services; (viii) foreign exchange broking services; (ix) maxi cab repair services; and (x) minor ports (other than major ports). 17 The turnover is defined as the aggregate of the sale prices received by a dealer net of sale tax less the sale price of goods returned within six months of the date of delivery. 18 Tanzi’s (1994) equation adds the weight of agriculture sector (⫺), the ratio of imports (⫹) to GDP, and the level of public debt (⫹). Musgrave (1969) and Burgess and Stern (1993) furthermore add qualitative factors of the same kind. Recall from the previous sections that the agricultural share has been declining for some decades relative to other economy sectors, while the debt/GDP ratio stays high in India. 19 A figure of 2.53 GDP points as to the bracket US$ (360–750) from a sample of 82 developing countries. The average figure for all Asian and all African countries is about the same. 20 The ratio taxpayers/population is then around 3 percent. 21 UN data for the late 1990s: 33 percent active population with 9 percent total unemployment. 22 The contents of this section are mainly based on government of India, Union Budget, various years, and Economic Survey, various years. 23 According to the estimates of CIA, The World Factbook, in 2003 the GDP per capita in India and in China was US$2,900 and US$5,000, respectively. 24 Those reforms included liberalized foreign investment and exchange regimes, significant reductions in tariffs and other trade barriers, reform and modernization of the financial sector, and significant adjustments in government monetary and fiscal policies. 25 During the year 1999–2000 a surcharge of 10 percent was levied on income above RS60,000 and in 2000–1 that surcharge was further increased to 15 percent on income above RS150,000; consequently the tax rates increased. 26 About C1,000 or US$1,300. 27 The scheme was very successful and the amount collected (RS100,500 million) exceeded the expected one (RS70,000 million). 28 The Kelkar task force on FRBM has fine-tuned the previous reports of the

190 Luigi Bernardi and Angela Fraschini

29 30 31

32

33

Kelkar task force on tax reforms, particularly on indirect taxes (see Government of India 2004). Service tax is currently levied on 51 services only. The savings incentives should be rationalized into a single “EET” (exempt during collection, exempt during accumulation and taxed during withdrawal) system, where savings up to RS100,000 a year would be eligible for this deduction. On 17 January 2005, a white paper on value added tax was released by the Empowered Committee of State Finance Ministers. The white paper on VAT, which would replace the sales tax regime in states, was drawn up after all states except Uttar Pradesh were prepared to implement VAT from 1 April 2005. The white paper lays down a roadmap for levy of a uniform state-level tax on over 500 items, exempts 46 local and social items (comprising natural and unprocessed products in the unorganized sector) and gives states an option to exempt food grains for a year. Under the VAT system there will be only two basic VAT rates of 4 and 12.5 percent, plus a special VAT rate of 1 percent for gold and silver ornaments (see Ministry of Finance 17 January 2005). Within the limits of the existing taxation of goods and services, the tax base is fragmented between the center (that levies tax on goods at the manufacturing level and tax on services) and the states (that levy tax on goods at the point of sale). The reduction would be possible as a consequence of the broadening of the tax base.

References Ahmad, S. E. and Stern, N. H. (1984) “The theory of reform and Indian indirect taxes,” Journal of Public Economics, 15(3): 259–98. Baird, M. and Ferro, M. (2003) “India: sustaining reform, reducing poverty,” Report 25797, Washington, DC: The World Bank. Burgess, R. and Stern, S. (1993) “Taxation and development,” Journal of Economic Literature, 31(2): 762–830. CIA, The World Factbook, (www.cia.gov). Das-Gupta, A., Gosh, S. and Mookherrjee, D. (2004) “Tax administration reform and tax payer compliance in India,” International Tax and Public Finance, 11(5): 575–600. DeLong, J. B. (2001) “India since independence: an analytic growth narrative,” http://www.j-bradford-delong.net. Desai, A. (1999) “The economics and politics of transition to an open market economy: India,” OECD Economic Working Papers VII, 100, Paris: OECD. Government of India, Ministry of Finance (2002) Final Report of the Task Force on Direct and Indirect Taxes. —— (2004) Report of the Task Force on Implementation of the Fiscal Responsibility and Budget Management Act, 2003. —— (various years) Economic Survey. —— (various years) Union Budget. International Monetary Fund (2000 and 2001) Government Finance Statistics Yearbook, Washington, DC: IMF. —— (2003) “India: selected issues and statistic appendix,” IMF Country Report 261, Washington, DC: IMF. Kwatra, G. K. (1997) “An overview of the Indian tax system,” International Bulletin of Fiscal Documentation, 458–68.

India 191 Musgrave, R. A. (1969) Fiscal Systems, New Haven: Yale University Press. Owens, J. (2004) “Competition for FDI and the role of taxation: the experience of South Eastern European countries,” Working Paper 316, Pavia: Società italiana di economia pubblica. Panagariya, A. (2004) “India in the 1980s and 1990s: a triumph of reforms,” IMF Working Paper 43, Washington, DC: IMF. Pursell, G. (1992) “Trade policy in India,” in D. Salvatore (ed.) National Trade Policies, New York: Greenwood Press, pp. 423–58. Rao, M. G. (2000) “Tax reform in India: achievements and challenges,” Asia-Pacific Development Journal, 7(2): 59–74. Sarma, A. and Gupta, M. (2002) “A decade of fiscal reforms in India,” Working Paper 04, Georgia State University: A. Young School of Political Studies. Shome, P. (1997) “The 1990s revolution in tax policy,” International Bulletin of Fiscal Documentation: 431–7. Tanzi, V. (1994) “Taxation in developing countries,” in L. Bernardi and J. Owens (eds) Tax Systems in North Africa and European Countries, Deventer: Kluwer, pp. 1–22.

Websites http://indiabudget.nic.it – Indian Ministry of Finance. http://www.mospi.nic.in/ – Indian Ministry of Statistics. http://www.imf.org – International Monetary Fund. http://www.wb.org – World Bank.

8

Japan Luigi Pascali

Introduction, contents and main conclusions This chapter briefly describes and evaluates the Japanese tax system. The next section is devoted to a short reminder of the Japanese economy and to an outline of the development of the Japanese tax system during the last decades. The Japanese fiscal burden has traditionally been modest compared with other G7 countries. The sizeable reductions in statutory marginal tax rates in both personal and corporate tax during the 1990s have strengthened this feature. Significant loopholes and non-neutralities are in place in key parts of the tax system, leading to potentially substantial efficiency losses. The third section examines the main Japanese national taxes. Particular attention is paid to personal income tax, corporate income tax and value added tax. The last was only introduced in Japan in 1989, after longstanding hesitations and controversies and has some particular features. First of all, the tax rate’s schedule has a single rate which stays among the lowest in the developed countries while the tax base is the broadest. Second, Japan employs the subtraction method instead of the credit invoice method that is used by all the other OECD countries: what is taxed is the difference between a taxpayer’s sales and his taxed inputs. The fourth section evaluates the tax system in terms of its redistributive impact, as to the horizontal and vertical equity and evaluates marginal effective tax rates on labor and capital. Although the distribution in both pre- and post-tax terms has become more concentrated since the 1950s, the Japanese society is still among the most equitable inside the OECD area, apart from Nordic countries and The Netherlands. Since the pre-tax income distribution is already fairly even, the total amount of tax redistribution is relatively low, roughly the same as that of the United States. On the other side, the tax burden seems not to be evenly distributed. The subsequent tax reductions of the 1990s have compromised the vertical and horizontal equity of the system. Moreover the favorable treatment of public pensions, which has no comparison with the systems of the other major countries, determines an uneven distribution of tax burden among

Japan 193 generations. The end of the section examines the effects of the tax system on labor and capital in terms of marginal effective tax rates and elasticity. The evidence is that the tax system does not pose major obstacles to a high utilization of the labor force, and also the influence of tax burden on investment seems to be low. The final section presents the current situation of public finance and the prospects of reform of the tax system. The final objectives of the new medium–long term tax strategy are common to the reforms undertaken in many other countries: to secure stable tax revenue for public services, to increase the fairness of the system, to reduce distortions and inefficiencies and make the system simpler.

A broad view of the current structure of the tax system A short reminder of the Japanese economy and public sector outline Japan is the second greatest economic power in the world after the United States with the GDP being approximately 45 percent of that of the United States. Its production constitutes about 75 percent of the production of the entire Asian market. For three decades overall real economic growth had been spectacular: a 10 percent average in the 1960s, a 5 percent average in the 1970s and a 4 percent average in the 1980s. Growth slowed markedly in the 1990s (Figure 8.1) largely because of the after effects of overinvestment during the late 1980s and contractionary domestic policies intended to wring speculative excesses from the stock and real estate markets. Subsequent government efforts to revive economic growth have met little success and were further hampered in late 2000 by the slowing of the US and Asian economies. Some changes in the economic situation occurred in 2003. Recently announced real GDP in the first quarter of 2004 grew an annualized 5.6 percent over the preceding quarter. Likewise, nominal GDP was up an annualized 3.2 percent (Figure 8.1). In the corporate sector, export volume was increasing, reflecting recovery of overseas economies. Industrial production has also expanded, though with some fluctuations. Figure 8.2 summarizes the rates of growth of the main components of GDP. Since 2001 private consumption has started to grow at significant levels and since 2002 the foreign balance displays an important enhancement. As a result, after an uninterrupted growth of 12 years, the unemployment rate will decrease from 5.4 to 4.6 percent, between 2002 and 2005 (Figure 8.3). However, even if the worst period seems to be finished, significant sources of concern are still valid for Japan. The financial position of the country has deteriorated significantly from the early 1990s in terms of fiscal deficit and debt accumulation, reflecting the sluggish economy and successive expansionary policies. Over the long term, a further element of pressure on public finances will come from the aging

194 Luigi Pascali 8 6 4 2 0 ⫺2 1990

1992

1994

1996

Real GDP

1998

2000

2002

2004

Nominal GDP

Figure 8.1 Rates of economic growth in Japan, 1990–2005 (%) (source: OECD (2004)).

9

6

3

0

⫺3

⫺6 1990

1992

Private consumption

1994

1996

1998

Public consumption

2000

2002

Investment

2004 Foreign balance

Figure 8.2 Rates of growth of the main components of GDP in Japan, 1990–2005 (%) (source: OECD (2004)).

of the population, as the ratio of elderly to younger people continues to increase more rapidly than in other OECD countries. Public expenditure on pensions has doubled from 6 to 12 percent of national income in the past decade and will rise to 17 percent by 2060 if no further action is taken.

Japan 195 6 5 4 3 2 1 0 1990

1992

1994

1996

1998

2000

2002

2004

Figure 8.3 Rate of unemployment in Japan, 1990–2005 (%) (source: OECD (2004)).

A second source of concern for Japan is the persistent deflation. Three sets of factors can be recognized as the cause of that. There are supplyside factors such as the increase in low-priced imports and the progress in technological innovation, demand-side factors stemming from weakness on the business front and, finally, monetary policy factors. These are linked with the fact that Japan also attempted to boost its economy by gradually lowering the official interest rate from its high of 6 percent in August 1990 right down to the adoption of a zero interest rate policy in February 1999. The structure of the tax system and its development during recent decades The current taxes levied by national and local governments in Japan can be classified into three groups: taxes on income, taxes on property and taxes on consumption. In 2002, of total tax revenue collected in Japan, 49.7 percent came from taxes on income, 17.6 percent came from taxes on property and 32.7 percent from taxes on consumption (OECD 2002). The taxes on income at the national level are the individual income tax and the corporation tax, while the taxes on income at the local level are the prefectural inhabitants tax, the municipal inhabitants tax and the enterprise tax. The taxes on property at national level are the inheritance tax, the gift tax, the land value tax, the registration and licence tax and the stamp tax. On the other side, the main taxes on property at local level are the automobile tax, the property tax and the special landholding tax. The taxes on consumption are given by the general consumption tax and several excises (e.g. on liquor, tobacco, gasoline, local roads, petroleum and coal, and so forth). Figure 8.4 compares the total fiscal revenue and its composition as a

196 Luigi Pascali

50%

40%

30%

20%

10%

0%

Japan

Korea

UK

Canada France Germany

Italy

US

Income – individual

Income – corporate

Social security contributions

Goods and service – general

Goods and service – specific

Property

Other revenues

Figure 8.4 A comparison between the fiscal revenues in G7 countries and Korea, 2000 (source: OECD (2002)).

percentage of GDP between Japan, the other G7 countries and Korea. The Japanese tax system shows some typical features. •

•

•

•

The fiscal burden is one of the lowest among the G7 countries. The ratio between tax revenue and GDP is 27.1 percent. Among the comparison group, only Korea features a lower ratio (26.1 percent). The relative share of taxes on income is the lowest among the G7 countries. The ratio between the fiscal revenue from these taxes and the GDP is 9.2 percent. Again, only the Korean tax system features a lower ratio, 7.5 percent. The lower tax burden on income in Japan is mainly due to the low tax burden on individuals, while the tax burden on corporations is substantially in line with the other developed countries. Social security contributions are relatively high. The ratio between social security contributions and total fiscal revenue is the highest in our sample, perhaps a reflection of the fact that Japan faces the aging problem earlier than most countries do. The relative share of taxes on goods and services is the lowest among the sample. Among major advanced countries, Japan is the only one that

Japan 197

•

had not imposed a general consumption tax until 1989. Even in 2000, the relative share of this tax was at the lowest level between OECD countries; revenues from indirect taxes rely heavily on specific excise taxes. Minor sources of revenue are obtained from death and gift taxes, while property taxes, a main source of local government revenue, occupy a relatively higher share on the total.

The process of developing a modern-type tax system in post-war Japan was initiated by the US. In 1947, several important reforms were undertaken under the influence of US occupation authorities. The schedular tax on individual income was replaced by a unified tax on an aggregate basis with progressive tax rates. In 1949, a tax mission headed by Carl S. Shoup came to Japan with the task of reorganizing the tax system as a whole. Essentially, the Shoup recommendations placed more importance on direct taxes, mainly income taxes on individuals and corporations. The most important issue in the income tax field was the introduction of a concept that the corporation tax was an advanced payment of individual income tax by shareholders. In order to avoid double taxation, dividends received by corporations were exempt and a 25 percent credit with respect to income tax was granted for dividends received by individuals. On the other side, a new law was enacted in 1950 to introduce a value added tax under the local tax system. The tax base of the value added tax was the gross receipts of an enterprise less its purchases from other enterprises. Consequently, wages payable to its employees as well as interest were included in the base. The ideal tax system achieved by the initiative of US influences was just temporary: many of these taxes were abolished or modified soon after their enactment. Two tendencies emerged from these modifications: several tax-cutting policies to maintain a lower tax burden and an incentive tax policy to achieve specific policy goals (Ishi 2001). The growing economy of post-war Japan generated a large increase in annual tax revenue. From 1950 to 1970 the GNP rose by an average of 15 percent by year. As a consequence, the fiscal revenue should have risen by an average of more than 20 percent by year (the elasticity of tax revenue on income exceeds 1). However, until 1965, the Japanese government adopted a taxcutting policy in order to keep constant the ratio of public expenditure on national income. A large reduction of the marginal tax rates was combined with a reduction of the tax base. Especially before the oil shock in 1973, there was a wide agreement in Japanese government and business circles that the tax system should be actively employed to promote economic growth. Based on tax incentive policies a number of special measures were formulated to promote exports, private savings and investment, housing, environmental quality and technological development. These usually included tax exemption, accelerated depreciations and lower rates on selected incomes.

198 Luigi Pascali A radical change of this policy occurred in 1976. In accordance with the new direction of fiscal consolidation, annual tax policy has adopted the opposite stance towards tax increases. The trend of tax increase, however, was ended in fiscal year 1987 and thereafter tax reduction policy was initiated once again by using increased revenue due to the bubble boom until 1991. A drastic tax reform took place in Japan between 1988 and 1989. First of all, there was a strong mitigation of the progressive tax rate structure for personal income tax: from 12 brackets with marginal tax rates between 10.5 and 60 percent to five brackets with marginal tax rates between 10 and 50 percent. Also the personal tax base was strongly reduced as a consequence of a great increase in standard personal exemption. Corporate income tax featured a great reduction in the tax rate for the ordinary corporations from 42 to 37.5 percent as well as for the small and mediumsized corporations from 30 to 28 percent. Finally, a general consumption tax was introduced. Since the introduction, its tax revenue has increased steadily, now accounting for about 15 percent of total tax revenue. In the 1990s, after the collapse of the bubble, fiscal stimuli packages led to successive rounds of large reductions in direct taxes to buoy up the depressed economy. In 1994, from the standpoint of realizing a welfare state embodying fairness and dynamism, the government decreased the tax burden by changing the tax structure of individual income tax and by relaxing progressiveness in order to alleviate the perception of a progressive tax burden, especially among the middle-income class. At the same time, the government increased the weight of the VAT by increasing the tax rate from 3 to 4 percent. In 1999, a new reform has strongly reduced the tax rates of the direct taxes. The top rate bracket of individual income tax has been cut from 50 to 37 percent and the corporate income tax rate has decreased to 30 percent. The result of these subsequent tax reductions can be inferred from Table 8.1. From 1990 to 1999, the total tax revenue decreased from 21.3 percent of GDP to 16.3 percent and the total fiscal revenue decreased from 30.1 to 26.1 percent. The revenues from personal income tax and from corporation income tax were almost halved. On the other side, there has been a sharp increase in the revenue from indirect taxes, due to the rise of the consumption tax rate in 1994 and an increase in the social contributions on account of the rapidly aging population.

9.5 4.3 5.2 1.5 4.5 0.0 4.2 15.5 4.5 1.7 2.3 20.0

– – –

8.1 4.0 4.1 1.5 4.8 0.0 4.6 14.4 4.0 1.3 1.7 18.3

– – –

1970

Note a Fiscal revenue by sub-sectors of government.

Source: OECD (2004).

Direct taxes, of which: personal income corporation income Taxes on property Taxes on goods and services: general consumption specific consumption Total tax revenue Social contributions: employees employers Total fiscal revenue Administrative level:a Central government Local government Social security funds

1965

45.4 25.6 29.0

9.5 5.1 4.4 1.9 3.7 0.0 3.2 15.1 6.2 2.3 3.2 21.2

1975

– – –

11.6 6.1 5.5 2.1 4.1 0.0 3.5 17.8 7.3 2.6 3.7 25.1

1980

43.7 26.0 30.3

12.4 6.7 5.7 2.6 3.8 0.0 3.3 18.8 8.2 2.9 4.2 27.2

1985

– – –

14.6 8.1 6.5 2.7 4.0 1.3 2.2 21.3 8.7 3.3 4.5 30.1

1990

– – –

10.1 5.9 4.2 3.2 4.2 1.4 2.2 17.5 10.0 3.9 5.1 27.7

1995

Table 8.1 Structure and development of fiscal revenue in Japan, 1965–2002 (% of GDP)

– – –

8.9 5.2 3.7 2.9 5.2 2.5 2.1 17.0 9.9 3.8 5.1 26.9

1998

– – –

8.2 4.8 3.4 2.9 5.2 2.5 2.1 16.3 9.7 3.8 5.0 26.1

1999

37.2 25.1 37.7

9.2 5.6 3.6 2.8 5.1 2.4 2.1 17.1 9.9 3.8 5.0 27.1

2000

– – –

9.0 5.5 3.5 2.8 5.2 2.4 2.1 17.0 10.3 4.1 5.1 27.4

2001

35.7 26.0 38.3

7.9 4.7 3.2 2.8 5.2 2.4 2.1 15.9 9.9 4.2 4.5 25.8

2002

200 Luigi Pascali

Some quantitative and institutional features of the main taxes Direct taxes Personal income tax (PIT) In common with all the other developed countries, Japan depends upon individual income tax for a significant portion of its tax revenues. In principle, the Japanese tax system still maintains a global system of individual income taxation, although in actual practice, separate taxation methods have been introduced as exceptional cases. Each income earner first adds up his income from all taxable sources, then subtracts allowable exemptions and deductions to attain the amount of taxable income. After applying a single progressive rate schedule to such taxable income, the taxpayer uses tax credits to arrive at the final amount due. Under the Income Tax Law, taxable income is classified for the purpose of calculation into the following ten categories in accordance with the nature of income: 1 2 3 4 5 6 7 8 9 10

Employment income Business income Retirement income Occasional income Real estate income Timber income Miscellaneous income Interest income Dividend income Capital gains

Fundamental personal deductions for households consist of the basic exemption, the exemption for spouse and the exemption for dependants. The basic exemption is equal to YEN380,000 and it is guaranteed to all taxpayers. The exemption for spouse is provided to the taxpayers who live with a spouse whose annual income does not exceed YEN380,000. It is equal to YEN380,000 (YEN480,000 if the spouse is older than 70 years old). Finally, the taxpayer receives an exemption for dependants. This exemption is equal to YEN380,000 for each relative supported except the spouse (YEN480,000 if the relative is older than 70 years old). In addition to the personal exemptions, which are applied to the taxable income as a whole, numerous deductions are applied to the specific sources of income. Employment income is given a special deduction. The main aim is to put wage and salary workers, the only categories to whom this deduction is

Japan 201 available, on a more equal level with the self-employed, who can treat their personal expenditure as business expenses. A variety of special treatments is also involved in the determination of business income. In particular, a special deduction for wages paid to family employees is allowed. Even the proprietor’s own remuneration is deductible from his business income and can benefit from the special deduction on employment income. Retirement income is taxed separately from other income. Only 50 percent of the retirement income is taxed. Moreover, there is a deduction (from YEN400,000 to YEN8 million) and a minimum guaranteed amount of YEN800,000. Finally, only 50 percent of occasional income is taxed after subtracting YEN500,000 as a special deduction. Interest income and some specific capital gains are taxed separately at the source at special rates; also, dividend income may be taxed in this way. Interest paid to residents is normally subject to withholding at the source at the rate of 15 percent (a 5 percent local inhabitants tax is levied in addition). Some kinds of interest are still excluded from taxation, for example, interest from current bank deposits and, when received by certain categories of individuals (such as the elderly until the end of 2005) interest from small deposits, from central and local government bonds and from postal savings. Capital gains derived from the sale of securities are also taxed separately. The tax rate applied is 20 percent (7 percent from 2003 to 2007 for listed stocks). The losses that are not deductible during the relevant year may be carried forward to be deducted within the next three years. The gross personal income tax due is computed by applying tax rates to the taxable income. Income tax rates are progressive ranging from 10 to 37 percent (Table 8.2). Finally, to compute the net tax liability, several tax credits should also be included, such as credit for dividend, credit for foreign taxes, credit for incremental research and experimental expenditure, and credit for acquisition of a house. The Income Tax Law allows a taxpayer to deduct 10 percent of dividend income from income tax. However, if ordinary taxable income, Table 8.2 Progressive tax rates on personal income in Japan Tax base (in YEN)

Tax rate (%)

up to 3,300,000 3,300,000–9,000,000 9,000,000–18,000,000 over 18,000,000

9 18 27 36

Source: MOF (2003).

202 Luigi Pascali including dividends, exceeds YEN10 million, a tax credit of only 5 percent is applicable to the fraction of dividend income equal to the ordinary taxable income. Foreign income taxes of residents may be credited against their Japanese income liability. This credit cannot exceed the result of product between the Japanese income tax and the share of the total income outside Japan on the entire income subject to Japanese income tax. Corporate income tax (CIT) Corporate income tax is levied on the net income earned by both domestic and foreign corporations in each accounting period or in liquidation. The computation of the net income accords with the actual practices of the modern business accounting system; however, some adjustments on capital incomes and investments should be made to obtain the taxable income. The capital gains accrued by corporations are subject to taxation in full as they are realized. These capital gains are taxed at the same rates as operating profits, although capital gains from short-term transactions of land are additionally levied a special tax burden. On the other side, 50 percent (in the case of specific shares, 100 percent) of dividends less a portion of the interest paid by the corporation on the borrowed funds, is excluded from revenue for tax purposes. Depreciation is allowed on the basis of the acquisition cost, salvage value (10 percent in the case of tangible assets and nil otherwise) and the statutory useful lives or the numbers of years during which such assets are serviceable. If the acquisition cost is less than YEN100,000, the corporation may deduct it in the accounting period of the acquisition. Generous provisions have been allowed for accelerated depreciation, increased initial depreciation and special tax free reserves to stimulate particular types of investments. Table 8.3 Tax rates on corporate income in Japan Type of income Ordinary income Corporations (capital ⬎ YEN100,000,000) Corporations (capital ⬍ YEN100,000,000) Cooperative associations and public service corporations Liquidation income Corporations Cooperative associations and public service corporations Retirement pension funds Special medical corporations Source: MOF (2003).

Tax rate (%) 30.0 30.0 22.0 27.1 23.1 1.0 22.0

Japan 203 The corporation tax for each accounting period is computed by multiplying the taxable income of the corporation by the tax rates listed in Table 8.3. Family corporations are subject to an additional special tax on retained earnings exceeding a prescribed level.1 This additional tax is designed primarily to deter shareholders of closely held corporations to avoid the progressive individual income tax that would be imposed on dividend distributions. A corporation can carry forward losses for five years following the loss year or carry back losses one year preceding the loss year. Indirect taxes Value added tax (VAT) The value added tax was finally introduced in the Japanese tax system in April 1989, after long-standing hesitations and controversies. The Japanese government started to stress the importance of a value added tax in the late 1970s. However, the first attempts to adopt it were frustrated by the strong opposition of influential parts of Japanese society. Ishi (2001) found three main reasons for this vigorous resistance to VAT. First of all, there was a general opposition to the regressive effects of this tax in a country with a strongly progressive tax structure. Second, small traders feared that the new tax would force them to reveal all of their transactions to the tax offices and that they would therefore be unable to avoid part of their income tax. Last, there was a general perception of inefficient use of public funds and therefore it was a political taboo proposing a new indirect tax. Whenever attempts to adopt VAT were frustrated, different names were applied to it so as to defuse political tensions, i.e. the general consumption tax in 1979, the sales tax in 1987 and the consumption tax in 1989. In 1989, in order to minimize the political opposition, the tax base of VAT was conceived to be as broad as possible. In this way, the tax would have been simpler and less onerous to business and with a lower tax rate. Given its history, it is not surprising that the coverage of the Japanese consumption tax seems to be among the broadest in the world and that the tax rate is a single one and the lowest in the world. Another peculiarity of Japanese VAT is that no invoices are provided for. Invoices admit the use of the tax-credit method, universally preferred in all VAT countries. Each invoice for a purchase from another firm indicates the total amount of the tax already paid. Firms collect all invoices and aggregate the input tax shown on them. This is the amount credited against the firm’s own gross tax in order to calculate the VAT payable. VAT without invoices must rely on the accounts method. Total purchases

204 Luigi Pascali are subtracted from total sales by using the bookkeeping records and the tax rate is applied on this difference. The Consumption Tax Law applies to domestic transactions, made as business activity, effected for compensation and categorized as sales of assets or provisions of services. The main non-taxable transactions are the following: 1 2 3 4 5 6

Sales and leasing of land Sales of securities, money lending and other financial transactions Medical services Social welfare services Educational services Housing rent

In principle, consumption tax seeks to impose tax on goods and services in the country where they are consumed (principle of taxation at consumption place). Therefore, transfer of taxable assets that taxable business operators engage in trades, similar to export trading and international transportation, is exempt from consumption taxation. The “base period” is the period that determines the existence or absence of tax liability.2 Taxpayers are the enterprises and the individuals who receive foreign goods from bounded areas. In other words, taxpayers in domestic transactions are limited to enterprises. In contrast, consumers can become taxpayers when they receive taxable cargoes from bounded areas. Small enterprises with taxable sales less than YEN10 million in a base period are exempt from tax liability on sales of taxable assets during that tax period. The current tax rate is 4 percent (a 1 percent local consumption tax has also been introduced, for a total current rate of 5 percent). A simplified tax system has been designed to help small enterprises (taxable sales are less than YEN50 million) easily calculate deductible tax on purchases. These enterprises can choose to compute the tax deduction for purchases by multiplying their taxes on purchases by a deemed rate. The deemed rate can vary from 90 percent (businesses corresponding to wholesalers) to 50 percent (businesses corresponding to real estate, transport and service industry). Once the simplified system is chosen, it cannot be changed for two years. Excise duties Japanese excise taxes can be classified into two groups: excises on alcohol and tobacco and earmarked excises. The first group includes the liquor tax, the tobacco tax and the special tobacco tax. These excises are normally justified with two reasons. The first one is that they correct a negative externality, limiting the disec-

Japan 205 onomies generated by the consumption of alcohol and tobacco. The second is that they are taxes on non-essential or luxury items considered proxies for taxpaying taxation. The liquor tax has traditionally been one of the most important in the national tax system. In the 1950s, it produced 18.5 percent of total national tax revenue. However, its importance has gradually decreased accounting now for not more than 3 percent of total tax revenue. It is levied on domestic alcoholic beverages shipped from manufacturing premises and on imported ones, which are drawn from bounded areas. Alcoholic beverages are classified in five categories and a specific tax rate is applied to each category. The second group of excises includes gasoline tax, liquefied petroleum gas, aviation fuel tax, petroleum and coal tax, motor vehicle tonnage tax and promotion of power resource development tax. These taxes have been conceived as service charges for the use of road, airports and power plants. Since the consumption of these services is exclusive to some extent, it seems fair that beneficiaries should pay for them in proportion to their use, even if the government does provide services.

An evaluation of the tax system in terms of equity and efficiency Redistributive impact of the Japanese tax system It is widely accepted that taxation can reduce the inequality of income distribution. A determinant role is here played by individual income tax, as it is typically equipped with progressive tax rates. For studying inequality in distribution, we may look at the Gini coefficient. Before directly investigating the effects of income taxes on distribution, consider the long-term variation in terms of pre-tax income distribution for self-employed and wage earnings in Japan (Figure 8.5). An interesting feature of these statistics is that while selfemployed incomes move towards a greater inequality, wage earnings move in the opposite way. Particular attention should be paid to the sharp rise in the inequality of self-employed incomes during the 1980s due to the rise of capital gains. However, after the collapsing bubble phenomenon in 1991, this type of income has moved rapidly toward more even distribution. The effects of taxation on income distribution are studied by computing the percentage difference between the pre-tax and post-tax Gini coefficient. The results are depicted in Figure 8.6. The most evident conclusion is that the redistributional effects of income taxes have strongly decreased over the past four decades. With an important exception between 1969 and 1973, the extent of redistributive effects had been lower for wage and salary earnings (withholding tax) than for self-employed and other incomes (self-assessed tax) until the late 1980s. The reason is that the former include all property

60%

50%

40%

30%

20% 1950

1960

1970

1980

1990

2000

Fiscal year Self-employed and other incomes Wage and salary earnings

Figure 8.5 The movement of income distribution in terms of Gini coefficient in Japan, 1951–2000 (source: NTA, in Ishi (2001)).

12% 10% 8% 6% 4% 2% 0% 1950

1960

1970

1980

1990

2000

Fiscal year Self-assessed tax on self-employed and other incomes Withholding tax on wage and salary earnings

Figure 8.6 Redistributive effects of income tax (percentage difference of the Gini coefficient before and after taxes) in Japan, 1951–2000 (source: NTA, in Ishi (2001)).

Japan 207 income and capital gains. These items of income tend to be concentrated at higher income brackets: thus, the income tax levied on them has greater power to equalize income distribution through its progressive structure. The equalizing power of income tax on self-employed and other incomes has fallen in the 1990s because of the combined effects of two factors: a great reduction in the capital gains due to the explosion of the financial bubble and a reduction in the progressivity of the tax system. As we have seen, the post-war history of income taxes had been characterized by annual reductions until 1979. Those reductions consisted of increasing exemptions and deductions, lowering progressive tax rates and enlarging special tax measures. Undoubtedly, annual tax reductions must have substantially affected the redistributive effect of income taxes. Raising the levels of exemptions and deductions implies that a certain amount of income is removed from the tax rolls, thus narrowing the whole scope of taxable income. More importantly, such tax cuts tends to weaken the equalizing power of a progressive tax rate. On the other side, it is difficult to assess the effects of tax rate adjustments on redistribution. Those measures have been less frequent than increasing the exemptions and with no clear directions. Only in the 1990s, a sharp decrease in the top marginal rate of individual income tax and a decrease in the rate of corporate tax can be considered as the main cause of the reduction of the redistributive impact of the tax system. A more significant impact on the distribution of income has been made by the introduction of special tax measures to achieve specific policy goals (especially capital accumulation). Two special measures merit particular consideration: partial exclusion from the tax base (as for dividend income) and separate taxation under reduced tax rates (as for interest and capital gains). Some authors believe that special tax measures have been a major contributory cause of the decreasing redistributive effects of income tax, especially with respect to self-employed and other incomes. The great reduction of the redistributive effects of self-assessed tax between 1968 and 1969 can be explained by some special tax measures (the typical example, in these years, is the tax concessions for capital gains on land granted in 1969). The OECD statistics revealed that, although the distribution in both pre- and post-tax terms has widened since the 1950s, Japanese society is still among the most equitable in the OECD, apart from the Nordic countries and the Netherlands (Table 8.4). Since the pre-tax income distribution is already fairly even, the total amount of redistribution is relatively low, roughly the same as that of the United States. Horizontal and vertical equity The Japanese tax system has developed in such a way as to deviate substantially from the original coherent Shoup design of the 1950s. After the continuous tax cuts implemented in the 1990s, to face deflation and

208 Luigi Pascali Table 8.4 A comparison between the Gini coefficient before tax and after tax among some OECD countries Country

Gini before taxes

Gini after taxes

% change

Australia, 1994 Belgium, 1995 Denmark, 1994 Finland, 1995 Germany, 1994 Italy, 1993 Japan, 1994 The Netherlands, 1994 Sweden, 1995 US, 1995

46.3 54.5 42.0 39.2 43.6 51.0 34.0 42.1 48.7 45.5

30.6 29.9 21.7 23.1 28.2 34.5 26.5 25.3 23.0 34.4

⫺33.9 ⫺48.4 ⫺48.3 ⫺41.0 ⫺35.3 ⫺32.4 ⫺22.0 ⫺39.8 ⫺52.9 ⫺24.5

Sources: Oxley et al. (1997) and Burniaux et al. (1998).

stimulate economic recovery, the tax system lost its internal coherence and even though it raises a very low amount of revenue, compared with other countries, it does so by concentrating the burden on an exceptionally narrow group of taxpayers (Giannini and Guerra 2004). Almost 80 percent of personal taxpayers pay their income tax at the lowest rate of 10 percent and 16 percent fall in the next 20 percent bracket; moreover, 70 percent of corporations do not pay any corporation tax at all. As it is well known, tax experts view equity and fairness in two dimensions, vertical and horizontal. Vertical equity is concerned with the distribution of tax burden among different income classes. Progressive taxation of all income on an all-inclusive basis is an essential condition in allocating different tax burdens fairly between the rich and the poor: we have already seen how much tax reductions in the last years have compromised the vertical equity of the Japanese tax system. Horizontal equity is concerned with equalizing the tax burden among people in similar economic conditions. This clearly requires that the income tax base is as broad as possible. However, several tax deductions are granted for personal income tax (basic exemption, exemption for spouse, exemption for dependents, employment income deductions and pension and retirement income deduction). Tajika and Furutani (2002) have studied the horizontal equity. Consider Table 8.5. Deductions are not only very large in comparison with the income earned but their variance is significant as well. The average deduction rate of income class 6 is 62.5 percent; however, it varies from 0–20 percent to over 80 percent, with the two largest groups, 39.4 percent and 33 percent, respectively, in the ranges of 60–80 percent and 40–60 percent. Moreover, 16.8 percent of those in the average income group have deducted more than 80 percent of their income to reach their taxable income.

Japan 209 Table 8.5 Dispersion of deduction rate by income class in Japan (%) Income class

Average income

0 to 20%

20 to 40%

40 to 60%

60 to 80%

over 80%

1 2 3 4 5 6 7 8 9 10

1,430 2,229 2,852 3,488 4,157 4,893 5,774 6,857 8,257 12,479

2.2 2.2 1.9 3.2 3.8 2.3 2.2 1.5 1.6 5.8

6.0 9.4 7.5 5.0 6.8 8.5 8.7 7.6 12.1 31.0

16.8 20.6 26.1 40.1 36.4 33.0 34.8 42.2 54.5 43.3

39.9 51.0 47.4 35.7 34.9 39.4 41.7 39.7 27.0 17.4

35.0 16.8 17.2 16.0 18.1 16.8 12.6 9.0 4.8 2.5

Source: Tajika and Furutani (2002).

In sum, Table 8.5 seems to suggest that granting deductions on personal income tax may not be as fair as it intends to be. The tax burden is also unevenly distributed between generations, particularly because of the very favorable treatment of public pensions, which has no comparison with the systems of the other major countries. As in most developed countries, in Japan contributions to public and private pension schemes are exempt. However, in contrast to other countries, lump sum payments and annuities paid by qualifying pensions are also exempt or at least lightly taxed. Lump sum payments are classified as retirement income. After a special deduction that increases from YEN4 to 20 million, only half of these items of income is taxed. Annuities are classified as miscellaneous income and enjoy a preferential treatment as well. In particular, they benefit from a generous deduction that is similar to the employment deduction for pensioners younger than 65 years old and much higher for older pensioners. An OECD survey (2002) compared the fiscal treatment of pension and labor income, computing the percent difference between the effective tax rate on workers and the effective tax rate on pensioners for fiscal year 2000. The relative advantage attributed to Japanese pensioners is very high with respect to other countries. Only Germany and the US are relatively more generous than Japan to a pensioner with income equal to or higher than the average worker income. Combined with the rapid aging of the population, the spontaneous result of this highly favorable treatment of retirement and pension income would be an increasing intergenerational inequity and the shrinking of tax revenue (Giannini and Guerra 2004). According to Morinobu (2002), the future income tax base to national income in Japan estimated to be 29.4 percent in 2000 will drop to 25.8 percent in 2005, 22.3 percent in 2010 and 14.2 percent in 2025 solely as a consequence of the

210 Luigi Pascali interaction between the favorable treatment of pensions and the rapidly aging society. Incentive to work, save and invest The total labor market distortion caused by taxation is given by the combination of marginal tax rates and labor supply elasticities. Table 8.6 reports the marginal tax rate on labor in some OECD countries: Japan features the lowest value among G7 countries. Only Korea does better in our sample. Moreover, the progressivity of the tax schedule unfolds relatively smoothly compared with many other countries, implying a virtual absence of problems related to unemployment and poverty traps, which are often caused by abrupt changes in tax rates along the income schedule (Dalsgaard and Kawagoe 2000). The labor supply elasticities are determined by a host of factors, including the wage bargaining framework, labor market policies and the degree of competition in the product markets. Most cross-country studies find that labor supply elasticities are low and often insignificant for primary earners, whereas they tend to gain in importance for secondary households earners, typically women. This pattern also appears to exist in Japan, although the evidence is scarce (Tachibanaki 1997). The phasing out of spousal allowances as well as the introduction of pension contributions when spouses earn more than YEN1.3 million annually create high marginal tax rates for the households when the spouse enters the labor market. The elasticity of Japanese married women with respect to own wages is 1.1 and with respect to family income is ⫺0.2. These elasticities are similar to estimates found for France and Germany but higher than those found in the United States (Tachibanaki 1997). Finally, the average personal tax rate in Japan has been traditionally low compared with other developed countries. Taxpayers with an annual income below YEN12 million (around 95 percent of all personal Table 8.6 Total tax wedge on labor in 2003 in G7 countries and Korea (marginal rate in %)

Canada France Germany Italy Japan Korea UK USA

Single

Married (principal earner)

Married (spouse)

33.9 52.5 65.3 57.9 31.8 17.1 40.6 34.1

37.7 48.7 59.7 57.9 29.2 17.1 40.6 34.1

43.9 38.3 59.7 31.8 31.4 12.0 40.6 34.1

Source: OECD (2003).

Japan 211 taxpayers) face average rates, including social contributions, below 20 percent. Dalsgaard and Kawagoe (2000) conclude that the tax system, unlike those in many other OECD countries, does not pose major obstacles to a high utilization of labor force. This conclusion is underpinned by the fact that Japan has one of the highest market participation ratios in the OECD, although this is due to a combination of very high male participation and a female participation ratio closer to the OECD average. Participation rates of the 55–64-year-old group are also relatively high and hence so is the average effective retirement age. The incentive to save and invest in physical capital also benefits from the overall level of taxation. To what extent the tax burdens for corporations and individuals influence aggregate investment levels is a question of how much taxation changes the cost of capital and to what degree the cost of capital determines investment. Table 8.7 shows the marginal effective tax wedge in manufacturing, computed using the KingFullerton methodology. Most OECD tax systems favor debt finance since corporate interest payments are deductible from the corporate tax base and because effective tax rates on personal interest income are often low. This is also the case in Japan, where the marginal tax wedge on debt is the lowest among G7 countries while the marginal tax wedge on retained earnings and new equity is higher and substantially in line with the other G7 countries. The influence of tax burden on investments seems to be low in Japan. Tachibanaki (1997) concludes that “the effect of various tax policies on investment was very minor.” The Economic Planning Agency (1998) likewise concludes that even a significant cut in corporate tax rates would imply only a small change in the cost of capital.

Table 8.7 Marginal effective tax wedge in manufacturing in G7 countries in 1998 (%) Source of financing

USA Japan Germany France Italy UK Canada

■

Physical assets

Overall weighted average

Retained earnings

New Equity

Debt

Machinery

Building

Inventories

1.9 3.5 1.5 3.7 2.3 2.2 4.3

5.0 5.6 1.0 6.9 2.6 2.8 5.0

1.7 0.3 1.4 0.8 0.6 1.8 1.3

1.5 1.4 1.0 2.2 1.0 1.7 2.2

2.8 4.1 1.7 3.8 1.8 2.1 4.1

2.7 3.4 2.1 3.8 3.3 3.1 4.8

Source: OECD.

2.1 2.6 1.4 3.0 1.7 2.2 3.3

212 Luigi Pascali

Tax reforms A quick glance at the macroeconomic and budget outlook A significant source of concern for Japan is its fiscal deficit. The budget position has deteriorated significantly from the early 1990s in terms of fiscal deficit and debt accumulation. Probably the Japanese economy has passed the most difficult economic period of its post-war years. After enjoying great economic performance in the 1950s, 1960s and 1970s, it entered into an era that was known as the “bubble economy” around the late 1980s. After hitting its peak in 1991, the bubble burst triggering Japan’s plunge into a period of stagnation of unprecedented length. The annual rate of growth rate during the 1990s was a mere 1 percent on average, in contrast to the 6 percent level of the 1980s. In order to tackle the sluggish economy, successive expansionary policies took place in Japan. These policies are summarized in Table 8.8: YEN136 trillion was injected as a continuous fiscal stimulus in the form of public investment and various tax incentives. However, these Keynesian fiscal policies have not brought about the desired effect of boosting the economy. Since 1991, the difference between total expenditure and tax revenue began to grow: from 1999, the total expenditure has been almost double the tax revenue (Figure 8.7). As a result, since 1991, Japanese debt as a percentage of GDP started to increase exponentially and in 1999 it became the highest among G7 countries (Figure 8.8). Nowadays it accounts for more than 150 percent of GDP. The worsening of the Japanese financial position is not only a result of the expansionary policies of the last decade, but is also related to the Table 8.8 Expansionary fiscal packages since 1992 in Japan (trillion YEN) Total August April September February April September April November November October November February Total Source: OECD (2003).

1992 1993 1993 1994 1995 1995 1998 1998 1999 2000 2001 2002

Investment

Tax cut

Others

10,700 13,200 6,200 15,300 4,600 12,800 16,700 23,900 18,000 11,000 1,000 2,600

6,250 7,620 1,950 4,500 1,077 6,540 7,700 8,100 6,800 5,200 400 2,600

0 150 0 5,850 0 0 4,600 6,000 0 0 0 0

4,450 5,430 4,200 4,900 3,543 6,270 4,350 9,800 11,200 5,800 600 0

136,000

58,737

16,600

59,543

Japan 213 120

90

60

30

0 1990

1992

1994

1996

1998

2000

2002

Year Total expenditure

Tax revenue

Government bond issue

Figure 8.7 Revenue, expenditure and bond issuance in Japan, 1990–2003 (source: OECD (2004)).

180 150 120 90 60 30 0 1990

1992

1994

1996

1998

2000

2002

2004

Year Japan Italy Canada

USA Germany France

UK Korea

Figure 8.8 General government gross financial liabilities, 1990–2005. A comparison between some OECD countries (source: OECD (2004)).

impact of Japan’s aging population. Japan’s population is becoming the most aged among G7 countries. The ratio of the working population (aged 20–64) to the elderly population has been falling rapidly, from 7.7 in 1975, to 3.6 in 2000, to a projected 1.9 in 2025. This implies that, in 2025, for every two persons in the working population, there will be one

214 Luigi Pascali elderly person who needs support. As a matter of fact, this determines lower fiscal entrance, due to the very favorable treatment of public and private pensions and higher social cost. Pension and health insurance benefit payments currently amount to roughly YEN70 trillion per year: this amount will double by 2025. The Japanese government is still sticking to its policy that will not achieve a primary balance surplus until late 2010. According to Dekke (2002), under unchanged spending policies, for the government to be solvent, taxes would need to increase from the current 28 percent of GDP to over 40 percent by 2020. Even considering the officially stated commitment to rationalize and cut public expenditure, it is increasingly recognized that a higher national tax burden in the future will be unavoidable (Giannini and Guerra 2004). Reforming direct taxation in Japan The drastic change in the population structure will have an enormous impact on Japan’s economy and society. The declining labor force due to the declining birthrate and aging of the population and the lower saving rate may cause a drop in the economic growth rate and an increase in the social security burden on working generations. As a consequence the tax system needs to be reformed. The major objectives of the reform, as stated in the Tax Commission Report, are the following: 1 2 3

Restoring a tax system that inspires confidence for the future. Improving tax fairness. Encouraging the productivity of individuals and corporations.

As the population ages, increases in public expenditure are inevitable. As both national and local finances are experiencing difficulty at the moment, the people have a sense of uncertainty regarding increases in tax or social security contributions in future. Therefore, even a low tax burden has no effects in boosting the economy, if it is perceived as temporary. As we have seen, the tax burden is unevenly distributed between generations, particularly because of the very favorable treatment of public pensions which are almost tax exempt. As the ratio of the elderly population increases rapidly, excess burden on the working generations will take away their motivation to exercise their social responsibility as long as the current structure is maintained. The working generations will not be able to expect large increase in salaries, but they will still be responsible for bearing the burden of the public service costs, while such costs are projected to increase rapidly in the future (Tax Commission 2003). An important task is to formulate a tax system that encourages the par-

Japan 215 ticipation of elderly persons and women to the economy of the society, so as to establish a lifelong working society. Furthermore, improvement in productivity is a key element for generating economic wealth in a society with a smaller population. Corporations, which play a central role in productivity, need to develop environments where they can cope flexibly with foreseeable structural changes such as globalization. The tax system should be reviewed in a simple, neutral and clear way so that it does not hinder individual employment or corporate choice (Tax Commission 2003). Concerning individual income tax reform, the major objectives stated in the Tax Commission’s Reports are restoring the major role of this tax as a revenue-raising instrument and restoring its redistributive function. The leitmotiv in the reform debate to pursue such objectives is the need to broaden the tax base by simplifying, consolidating and in some cases withdrawing the existing exemptions and deductions (Giannini and Guerra 2004). According to the estimates of Tajika and Furutani (2002) a cut in deductions of 30 percent generates an increase in revenue of 45.1 percent. A total deletion of the deductions triples the fiscal revenues, considering that tax relief could even approach 10 percent of GDP. On the other side, removing or reducing existing allowances hurts vested interests and for this reason the debate on the subject is particularly alive. The only measure taken in the 2003 tax reform has been the reduction, from the year after, of the special allowance for spouses. Concerning capital income taxation, the tax treatment of interest, capital gains from stocks and dividends on listed stocks has been unified at the rate of 20 percent. However, there are still several loopholes in the tax base and significant discrimination among different assets and types of subscribers. First, some kinds of interest are excluded from taxation (such as interest from current bank deposits), as well as those received by certain categories of individuals (such as the elderly). Second, dividends may be partially double taxed. The taxpayer can decide to tax them at source at the rate of 20 percent, without tax credit, or to include them in the income tax base and benefit from a tax credit equal to 10 percent. Third, capital gains on bonds, other than convertible bonds, are not taxed. This approach reflects the idea that capital gains should be taxed only when deriving from a speculative activity. The efforts of capital income tax reforms in the fiscal year 2002 and 2004 regard mainly capital gains and aim to ensure tax neutrality and equity. Nowadays, the innovation in the financial markets makes it possible to speculate on the bond market as well as on the stock markets. Moreover, the exemption of capital gains on bonds can easily promote tax avoidance behaviors, consisting of transforming interests (which are taxed) into capital gains on bonds (which are exempt). For this reason the orientation of the Tax Commission is toward a generalized taxation of all kinds of capital gains at the rate of 20 percent. Another problem

216 Luigi Pascali regarding capital gains concerns tax enforcement, since the acquisition price necessary to determine the tax base is not easily known. In order to solve this problem, Japan has experienced a fixed tax rate on the selling price. The most important shortcoming of this solution is that it deeply distorts the functioning of financial markets since it penalizes trading (Giannini and Guerra 2004). An alternative way to face this problem is by getting intermediaries directly involved: all agents trading on financial markets communicate the identity of taxpayers involved in transactions to the tax authorities. This solution is adopted in the majority of OECD countries, but not in Japan. The difficulties of adopting it in Japan concern the fact that this country has not yet introduced a taxpayer identification code. The tax reform in 2002 has introduced into Japan a different and innovative solution to this problem: it makes it possible to compute capital gains on listed stocks held in designated account, opened with a brokerage house, independently from those held in other accounts, and to opt for capital gains taxation withheld at source.

Notes 1 A “family corporation” is a company of which 50 percent or more of all capital stocks are owned by no more than three shareholders and persons or corporations connected to them. 2 For individual proprietors, it is the calendar year two years before the current tax year. For corporations, with an accounting period of one year, it is the period two years before the current accounting period.

References Burniaux, J. M., Dang, T. T., Fore, D., Forster, M., Mira d’Ercole, M. and Oxley, H. (1998) “Income distribution and poverty in selected OECD countries,” OECD Economics Department Working Papers 189, Paris: OECD. Dalsgaard, T. and Kawagoe, M. (2000) “The tax system in Japan: a need for a comprehensive reform,” OECD Working Paper, Paris: OECD. Dekke, R. (2002) “The deteriorating fiscal situation and an aging population,” NBER Working Paper 9367. Economic Planning Agency (1998) Economic survey of Japan 1997–1998, Tokyo. Giannini, S. and Guerra, M. C. (2004) Reforming Direct Taxation in Japan, mimeo. Ishi, H. (2001) The Japanese Tax System, Oxford University Press, Oxford. MOF (2003) Taxes on Income, an Outline of the Japanese tax system – 2003, Tokyo: Ministry of Finance of Japan. Morinobu, S. (2002) “Comparison of Japanese and U.S. tax bases and changes of tax base in Japan,” Joint Seminar on Income Taxation From an International Perspective, IMF, Business Law Center of the University of Tokyo, and Policy Research Institute of the Ministry of Finance of Japan. OECD (2002) Taxing wages 2000–2001, Paris: OECD. —— (2003) OECD Economic Surveys, Japan, Paris: OECD. —— (2004) Revenue Statistics 1965–2003, Paris: OECD.

Japan 217 Oxley, H., Burniaux, J. M., Dang, T. T. and Mira d’Ercole, M. (1997) “Income distribution and poverty in 13 OECD countries,” OECD Economic Studies 29, 1997/II, Paris: OECD. Tachibanaki, T. (1997) “Public policies and the Japanese economy: savings, investments, unemployment, inequality,” London: Macmillan. Tajika, E. and Furutani, I. (2002) “Distribution of personal income tax in Japan: evidence from a microeconomics survey,” Paper presented at the ESRI Workshop on Tax reform in Japan, December. Tax Commission (2003) A sustainable tax system for Japan’s aging society, www.mof.go.jp/tax.

Websites http://www.mof.go.jp/english – Ministry of finance, Japan. http://www.oecd.org – OECD. http://www.go.jp/english/index.htm – Statistical Bureau, Japan. http://www.kkc.or.jp/english/ – Keizai Koho Center, Japan institute for social and economic affairs.

9

Malaysia Gianpaolo Fanara

Introduction, contents and main conclusions This chapter aims to discuss the case of Malaysia, showing the most significant features of its tax system and its more recent changes and reforms. Interest in the Malaysian case is due to the fact that it is actually one of the fastest developing countries of the Asian region; so it is relevant to present some data and features concerning its physical, political and economical context before going on to discuss taxation issues. Malaysia is strategically located in the heart of Southeast Asia, one of the world’s fastest growing regions; Malaysia is well serviced by all major air and shipping lines, and this easy access to the rest of the world has today made it an attractive center for trade, investment and tourism. Malaysia is composed of a peninsular region, which stretches from Thailand in the north to Singapore in the south, and an insular part (the states of Sabah and Sarawak), which straddles the northern and western coasts of Borneo; about four-fifths of Malaysia’s land area (around 330,000km2) is covered by tropical rainforests while major land-uses include cultivation of rice, rubber and oil palm. Concerning political features, Malaysia gained independence from the United Kingdom in 1957 through peaceful negotiations. Malaysia is a constitutional monarchy made up of states and federal territories; the states have sovereign monarchs or sultans: every five years, a system of rotation allows the sultans to elect the King (“the Yang Pertuan Agong”) among them. A federal form of government exists with a bicameral parliament consisting of a House of Representative, a body of 180 members fully elected on the basis of universal adult suffrage, and a Senate, whose members are appointed by the King and the State Legislatures. The Barisan National, the coalition representing the multiracial composition of the country, commands a two-thirds majority in parliament: this has ensured Malaysia a strong and stable government committed to the development of the country. This stability helps the design and development of short- and long-term policy plans such as the Third Outline Perspective Plan 2001–5, which projected the year 2020 as the target for the nation to achieve the “developed-country” status.

Malaysia 219 This chapter is organized as follows: the next section outlines a brief reminder of the Malaysian public sector, the current structure of the tax system and its development during the last decade, and concludes with a short comparison with other countries (developing, such as Mexico and Vietnam, and developed, such as the United States, Japan and Germany). The third section is devoted to the institutional features of the main Malaysian taxes (both direct and indirect, such as PIT, CIT, taxes on good and services, excises as so on). Then in the fourth section Malaysian fiscal burden by economic functions and by implicit rates is analyzed. The final section concludes by discussing the most relevant tax reforms of recent years and those expected for the future. The main proposals are the reviewing and rationalizing of the incentive structure, the modernization of fiscal administration (through self-assessment and transparency of the system) and the reduction of the fiscal burden on firms. These mid-term objectives are consistent with the aim of the Malaysian government to enhance the growth, competitiveness and resilience of the economy.

A broad view of the tax system and its development from the early 1990s A short reminder of the Malaysian economy and public sector outline During the last three decades, the Malaysian economy has been characterized by fast development. From 1971 to 1990 the GDP grew at an average yearly rate of 6.7 percent, but growth was especially impressive during the last years of the 1980s, the 1990s (except in 1997–8, because of the Asian financial crisis) and the 2000s, when growth rates reached nearly 12 percent, both in 1996 and 2000. As a consequence of this growth, during the last 40 years Malaysia became an industrialized economy. In fact, in 1960 about 44 percent of GDP was produced by the agricultural sector and only 12 percent by the industry, while in 2002 these percentages were, respectively, of 8 and 45 percent (Table 9.1). From the expenditure side, (year 2000) consumption was about 57 percent (of which 11 percent public consumption and 46 percent private consumption), investment 31 percent and net exports close to 12 percent (IMF 2001). Table 9.1 Malaysian GDP by sector of origin, 1960–95 (% of GDP)

Primary Industry Services

1960

1970

1980

1990

1995

2002

44 12 44

39 18 43

31 25 44

28 30 42

21 36 43

8 44 56

Source: IMF for 1960–95; Department of Statistics of Malaysia for 2002.

220 Gianpaolo Fanara In line with the growing trend of the GDP and the development of the economy, since the late 1980s, the main public finance indicators of the Malaysian federal government show a steep increasing trend both for total revenue and total expenditures (current and development components). Except during the early and middle 1990s, the government budget was imbalanced for most of the years here considered, and the deficit reached a huge amount from the late 1990s. In 2002 public sector expenditure was 40.3 percent of GDP; in particular, federal government expenditure represents 28.8 percent of GDP, while state and local government expenditure represents about 12 percent of GDP. Federal government expenditure is composed of: expenditure for general services (18.8 percent of GDP), including general public services, defense, education, health and social security; expenditure for economic development (5.3 percent of GDP), including agriculture, industry, transport and so on; and expenditure for interest payment and transfers to state and local government (about 5 points percentage). In 2002 per capita GDP was about US$3,450 (not PPP corrected). In the UN’s 2004 Human Development Report, Malaysia occupies the 59th place in the High Human Development index, with a life expectancy at birth of 73 years, an infant mortality rate of 8 per 1,000 live births (46 in the 1970s), an adult literacy rate of 88.7 percent and a total fertility rate of 2.9 births per woman in the period 2000–5 (it was 5.2 in the period 1970–5). In 2002 the population is 24 million, of which 14.8 percent are under the age of 15 and 16.3 percent are above 65 years of age; labor force participation is 64.5 percent with an unemployment rate of 3.5 percent; for the period 2002–15 the estimated annual population growth rate is 1.6 percent (it was 2.5 percent in the period 1970–2002). The health situation is not very good: public–private health expenditure is about 4 percent and about 20 percent of the population has no access to essential drugs; the HIV incidence is not very high (0.4 percent for 15–69year-old people), but there is a quite high incidence of malaria cases (57 per 100,000 people) and tuberculosis (120 per 100,000 people). The tax system The current structure and its development In 2002 Malaysian GDP reached the value of 217,453 million Ringgits (RM).1 Total revenue was RM73,167 million (of which approximately RM39,000 was from direct taxes, RM18,000 from indirect taxes and RM15,000 from non-tax revenues). Total expenditure was close to RM94,000 million (of which RM65,342 was current expenditure and RM28,382 development expenditure). The budged deficit reached RM18,857 million. The Malaysian structure of taxation and expenditure in percentage of

Malaysia 221 Table 9.2 Fiscal revenue and public expenditures in Malaysia, 2002 (% of GDP) % of GDP Total revenues, of which: Direct taxes Indirect taxes No-tax revenues Expenditures, of which: Current expenditures Development expenditures Deficit

33.6 17.9 8.4 6.9 43.2 30.1 13.1 8.6

Source: Own elaboration on data from IMF and the Department of Statistics.

GDP is depicted in Table 9.2. The main features are the high level of budget deficit, the relevant amount of development expenditure (about 33 percent of total spending) and the low value of total fiscal pressure (about 26 percent both from direct taxes – including income tax and petroleum tax – and indirect taxes – including sales taxes, excises on goods and duties). The light fiscal burden is possible because Malaysia social contributions are irrelevant (about RM800 million). Pension benefits are limited to public sector employees and in addition are very low: the maximum pension treatment is half of the last drawn salary. The current structure of the Malaysian tax system may be better understood by looking at its development from the early 1990s. A first general feature is the decreasing trend of revenue (either tax or no-tax components) during most of the 1990s, when compared with GDP. In Malaysia the ratio between total revenue and GDP decreased from 25.7 percent in 1992 to 18.3 percent in 2000 but increased again to 33.6 percent in 2002,2 in particular, the tax revenue share of total revenues decreased from 20.3 percent in 1992 to 14.3 percent in 2000. Thereafter a sudden jump increased tax revenue to 26 percent in 2002 (see Table 9.3). A possible understanding of the trends above described is on one hand the Asian crisis of 1997–8 and, on the other hand, the shift (from 1998) from a restrictive fiscal policy to an expansive one in order to reverse the slowdown of economic activity. A second general feature of Malaysian tax system development is the change in the share of direct vs. indirect taxes. In particular, in 1992 direct and indirect taxes were still more or less similar in their percentage of total revenues (both at about 38 percent of total revenues), but in 2000 direct taxes made up a much larger proportion (about 53 percent of total revenues) than indirect taxes (only 25 percent). A possible explanation for the increasing trend of direct taxes on total revenues is the general reduction of the incidence of poverty and its counterpart of the increase

9.8 ⫺ ⫺ 9.8 ⫺ ⫺ 20.3 5.4 25.7

9.6 ⫺ ⫺ 9.6 ⫺ ⫺ 20.1 5.1 25.2

1993 9.2 ⫺ ⫺ 9.9 ⫺ ⫺ 20.5 4.3 24.8

1994

Source: IMF since 1992 to 2000; Department of Statistics of Malaysia for 2002.

Direct taxes, of which: Tax on net income and profits Petroleum tax revenues Indirect taxes, of which: Sales taxes and excises Tax on international trade Total tax revenue Total non-tax revenues Total revenue

1992 9.4 ⫺ ⫺ 9.4 ⫺ ⫺ 19.8 3.6 23.4

1995 9.1 ⫺ ⫺ 9.1 ⫺ ⫺ 19.5 3.9 23.3

1996

Table 9.3 Structure and development of government revenue in Malaysia (% of GDP)

9.9 ⫺ ⫺ 9.2 ⫺ ⫺ 19.6 3.7 23.5

1997 10.0 8.2 1.7 6.0 3.3 0.9 16.7 3.3 20.0

1998

8.5 7.1 1.3 6.7 3.7 1.1 16.0 3.7 19.7

1999

8.1 5.8 2.0 6.0 3.9 1.1 14.3 4.0 18.3

2000

11.8 9.5 2.3 6.0 3.9 1.0 17.8 5.7 23.5

2002

Malaysia 223 of personal incomes due to the growth of the economy and the Malaysian government’s policy of eradicating poverty and improving social and economic status of the population. An international comparison In this section the Malaysian fiscal pressure and its composition of GDP is compared with that of some developing countries (e.g. Vietnam and Mexico) and the most industrialized countries (e.g. the United States, Japan and Germany); this choice is made in order to make a comparison between countries with different levels of development and, consequently, to underline their peculiar features (see Table 9.4). First, we may note the gap between Malaysian fiscal pressure (23.6 percent on GDP) and that for some of the most developed countries (e.g. 36.8 percent on GDP in Germany) due in part to differences in taxation but also to the considerable gap in social contributions, which are irrelevant in Malaysia as in other developing countries where social security systems are not well established yet. Second, it is interesting to observe the difference between the PIT and CIT relevance in the developing countries (Malaysia and Vietnam) and the most industrialized countries: in fact, in the developing countries PIT incidence of GDP is very low (2.8 percent in Malaysia and 0.4 percent in Vietnam), while CIT incidence is higher (6.8 percent in Malaysia and 5.6 percent in Vietnam); on the contrary, in the most developed countries CIT incidence is lower than PIT incidence (e.g. in United States, PIT is 12.2 percent while CIT is only 1.9 percent of GDP). These data show that in Malaysia, personal income taxation is still a weak and marginal source of revenue. To conclude, by considering the taxation of goods and services, it is possible to see that the incidence in Malaysia (about 4 percent on GDP) is Table 9.4 Comparison with the fiscal revenue structure of some selected countries, 2001 (% of GDP) Malaysia Vietnam Japan Hungary Germany Mexico USA Fiscal revenues Tax revenues Taxes on net income and profits PIT CIT Taxes on good and services Social contributions

23.6 18.4

n.a. 16.6

27.3 17.0

39.0 27.5

36.8 22.2

18.9 15.7

28.9 21.8

12.6 2.8 6.8

n.a 0.4 5.6

8.9 5.5 3.5

10.0 7.6 2.4

10.6 10.0 0.6

5.3 n.a n.a.

14.1 12.2 1.9

4.0 n.a.

9.7 n.a.

5.2 10.3

15.1 11.6

10.6 14.6

9.7 3.2

4.6 7.1

Sources: own elaboration on Malaysian Department Statistics’ data, OECD 2001, IMF 2003.

224 Gianpaolo Fanara very similar to the one of some developed countries (e.g. 5.2 percent in Japan and 4.6 percent in the United States) and very dissimilar from the ones of developing countries (e.g. about 10 percent in Vietnam and Mexico); this feature is probably due to the fact that the Malaysian government prefers a low taxation of goods and services in order to improve economic activity.

Some quantitative and institutional features of the main taxes Direct taxes Personal income tax (PIT) Personal income tax is an annual tax on the taxable income from all sources accrued or derived from Malaysia and income received in Malaysia from outside Malaysia by a resident. These sources include gains or profits from any profession, vocation or employment, pension or annuity and rent. Dividend income is taxed at gross and the tax deducted at source by the company under an imputation system will be given as a tax credit to the shareholders. The residence status of an individual determines her/his claim for personal reliefs and tax at graduated rates; in particular, an individual is regarded as a resident if: (i) s/he is in Malaysia for at least 182 days in a calendar year; (ii) s/he is in Malaysia for a period of less than 182 days but that period is linked to another period of presence of at least 182 consecutive days in an adjoining year; and (iii) s/he is a resident for the immediately following year and also for each of the three immediately preceding years. In ascertaining the taxable income of a resident individual, various forms of tax relief are given in order to reduce her/his tax liability. The different forms of relief are shown in Table 9.5. A tax rebate of RM350 is given to an individual whose chargeable income does not exceed RM35,000; a further rebate of RM350 is given if a husband/wife has no income. Also a tax rebate of RM400 will be granted to taxpayers who purchase a personal computer; this tax rebate can only be enjoyed every five years and is limited to one computer per family. Table 9.6 shows the income tax rates for resident individuals. Interest received by a resident from licensed banks and financial institutions in Malaysia is subject to taxation at the rate of 5 percent. Non-resident individuals are subject to taxation on income at the rate of 28 percent; for certain sources of income there are particular tax rates: in particular, the rate for royalties and technical fees is 10 percent, and for interest is 15 percent. Also, non-resident public entertainers are taxed at 15 percent of gross income.

Malaysia 225 Table 9.5 Different forms of relief for the determination of taxable income in Malaysia RM Personal Wife/husband Disabled taxpayer (additional) Disabled spouse (additional) Each childa Disabled child Parents’ medical expenses Equipment for disabled individuals Life insurance premiums Insurance premiums for education or medical benefits Medical expenses for serious diseases Contributions in setting up public facilities for disabled people Donations to seriously ill individuals needing financial assistance

8,000 3,000 5,000 2,500 800 5,000 (max) 5,000 (max) 5,000 (max) 5,000 (max) 3,000 (max) 5,000

Source: see text. Note a If the child is over 18 years of age and is receiving higher education at local university or college s/he is eligible for four times the normal relief of RM800; if the higher education is received overseas, only normal relief is granted.

Table 9.6 Income tax rates for resident individuals in Malaysia Taxable income (RM)

Rates (%)

up to 2,500 more than 2,500 to 5,000 more than 5,000 to 20,000 more than 20,000 to 35,000 more than 35,000 to 50,000 more than 50,000 to 70,000 more than 70,000 to 100,000 more than 100,000 to 250,000 more than 250,000

0 1 3 7 13 19 24 27 28

Source: see text.

Corporate income tax (CIT) Before the assessment year 1995, the scope of income tax in Malaysia was on a derived and remittance basis, except for banking, insurance, air and sea transport for which the scope of taxation is based on worldwide income. However, with effect from the assessment year 1995, the scope of company income tax in Malaysia is on a derived basis. For non-resident companies, tax is only imposed on income accrued in

226 Gianpaolo Fanara or derived from Malaysia, but not on income received in Malaysia from outside sources. A company is considered resident in Malaysia if its management and control is exercised in Malaysia; management and control are normally considered to be exercised at the place where the directors’ meetings are held. The taxation of companies is based on a full imputation system, i.e. the tax of a company is in fact an advance tax of the shareholders who receive dividends from the company; in other words, the shareholders are taxed on the gross dividends at their own respective tax rates and are given full tax credits in respect of the tax deducted at source by a company; effective from the assessment year 1998, the company income tax rate is 28 percent. Deductions from the company income are given for: 1

2

3

Allowable Expenses, including the expenditure incurred wholly and exclusively in the production of the gross income, cash contributions to local, state or federal authorities or governments, cash contributions to build or equip public libraries, contributions of local artworks to the state or National Art Gallery, costs of the equipment supplied by the employer for the disabled employee to perform her/his duties, donations to a seriously ill person needing financial assistance, scholarships granted to students in local institutions and entertainment and promotional expenses. Capital Allowances, including new plant and machinery, office equipment, information technology equipment, motor vehicles and environmental protection equipment. Industrial Building Allowances (IBA) are granted to companies incurring capital expenditure on construction or purchase of a new building which is used for production purposes.

Real property gains tax (RPGT) Real property gains tax is charged on gains accruing on the disposal of any real property situated in Malaysia or an interest, option or right over a real property company (RPC).3 Every person or company resident or non-resident in Malaysia is chargeable to the RPGT in respect of any gain accruing on the disposal of any real property or RPC shares in Malaysia. A chargeable gain arises if the disposal price exceeds the acquisition price and an allowable loss is incurred if the disposal price is less than the acquisition price. Table 9.7 shows the rate of the RPGT, which depends on the period of ownership of the property or RPC shares. These rates are applicable for disposals made on or after 27 October 1995 by citizens and permanent residents. With effect from 27 October 1997, disposal within five years from the date of acquisition of chargeable

Malaysia 227 Table 9.7 The rate of the RPGT in Malaysia Category of disposal

Individual (%)

Company (%)

Disposal within 2 years Disposal in the 3rd year Disposal in the 4th year Disposal in the 5th year Disposal in the 6th year and subsequent years

30 20 15 5 0

30 20 15 5 5

Source: see text.

assets by non-citizens and a non-permanent resident are subject to 30 percent tax rate (but only 5 percent after the fifth year). Indirect taxes Taxes on international trade of goods and services Import duties are usually ad valorem even if some specific duties are imposed on a number of items. Ad valorem rates of import duties are from 0 percent on basic foods to 300 percent on some motor cars; margin of preference (MPO) from 25 to 59 percent is given for selected goods of ASEAN origin. Export duties ad valorem are imposed on crude palm oil, processed palm oil and other oils. Excise duties, sales tax and service tax The excise duties are levied on locally manufacture cigarettes, intoxicating liquors, motor vehicles and playing cards. The sales taxes are ad valorem single-stage taxes imposed at the import (all the exports are not taxed in this case) and manufacturing levels. Current rates are shown in Table 9.8.

Table 9.8 Current rates of sales taxes in Malaysia General rate on all goods Fruits, certain foodstuffs and building materials Cigarettes and tobacco products Alcoholic beverages Petrol Diesel Source: see text.

10% 5% 25% 29% RM0.5862 per liter RM0.1964 per liter

228 Gianpaolo Fanara The service tax is a form of indirect tax on taxable services such as: • • • • • •

hotels (having more than 25 rooms) restaurants, bars, snack-bars and coffee houses night clubs, dance halls, health and massage centers and beer houses private clubs private hospitals professional services

Other taxes Other less relevant Malaysian taxes are the stamp duties, the tax/license/duties on gaming activities and the road tax. The stamp duties are imposed ad valorem on certain written documents and vary accordingly to the nature of the documents and values referred to; however, some specific documents attract specific rates of stamp duties. The gaming tax is imposed on punters placing their bets or investments in respect of any gaming authorized under any law and is collected by the promoter of the gaming so authorized and paid to the revenue of the federation. The road taxes are imposed on passengers cars, motorcycles, taxis, buses and tractors and are based on the engine capacity, the ownership (individual or company) and the fuel used (gasoline or diesel).

The fiscal burden The analysis of the distribution of taxation charge is focused on the impact of direct and indirect taxes on main components of GDP (economic function method); in particular, on the percentage impact on GDP of the consumption taxation (general and specific indirect taxes), of the labor taxation (direct taxes paid by employees and social contributions) and of the capital taxation. During the last decade, consumption taxation was relatively higher before the 1997 Asian crisis (above 9 percent) and lower after this year (between 6 and 7 percent, with the exception of 2002): this in part is due to the economic measures adopted by the government to mitigate the impact of the crisis. As to capital taxation, the average figure looks somewhat high. The trend is decreasing until 2000 (with the exception of 1998) but from 1999 it is increasing (above 9 percent); nevertheless a lighter capital taxation is forecast for the future, in order to attract more foreign direct investments to improve economic development. Labor taxation stays at an average value. Its trend is quite regular, in fact the range of the rates is 2–3 percent during the last decade; this is in part due to the very low amount of social contributions.

Malaysia 229 It is also possible to analyze the distribution of fiscal burden by means of the implicit rates method; in this case, the three main indicators are the implicit tax rates on consumption, on labor and on capital and business. They are obtained using, respectively, as numerators the same values used for the economic function method, and, as denominators, three economic aggregate values drawn from the national accounts: private consumption, compensation of employees (wage and salaries) and gross operating surplus.4 As Table 9.9 shows, the fiscal burden on consumption was high before the 1997 Asian crisis (between 19 and 20 percent), but lower in recent years (between 11 and 13 percent), while the fiscal burden on labor is increasing: this trend can be seen as an indicator of the growing scope for personal income tax inside the Malaysian tax system, which has been taking place for a number of years.

Tax reforms A quick glance at the macroeconomic and budget outlook Malaysia sustained a robust growth in 2003, driven by strong domestic demand and sturdy export performance. Real GDP grew by 5.2 percent in 2003, up from 4.1 percent in 2002. Private consumption is expected to rise by 8.1 percent this year, compared with 5.1 percent in 2003. Private investment is projected to grow by 11.5 percent in 2004, substantially higher than the 1.1 percent in 2003. The overall inflation rate, as measured by the annual change in the Consumer Price Index (CPI), remained low at 1.2 percent in 2003. From the supply side, all sectors of the economy expanded in 2003; the manufacturing sector saw 8.2 percent growth in constant prices compared with 4 percent in 2002, the finance and business sector expanded by 4.9 percent and the agriculture sector expanded by 5.5 percent in real terms. From the demand side, final consumption expenditure achieved a 5.7 percent growth in constant prices in 2003, up from 4.6 percent in 2002; this increase was attributed primarily to a 7.9 percent growth in real government expenditure and second to a 5.1 percent growth in real private final consumption. Despite increased expenditures in 2003, public debt remains within a manageable level (48.2 percent of GDP); given the high savings rate and the excess of liquidity in the financial system, it can be financed from non-inflationary domestic sources (debt servicing is at 14 percent of operating expenditures and the total external debt has fallen to 9.5 percent of GDP, lowering exposure to international risks). In 2003, fixed capital formation rose 2.7 percent in real terms, compared with a marginal increase of 0.3 percent in 2002. The strong export performance was attributed to the manufacturing sector, particularly electrical and electronic products and the primary

9.6 2.6 6.9

19.8 7.7

19.4 7.0

1993

9.8 2.3 7.3

Source: own elaboration on IMF data.

Economic functions Consumption Labor Capital and business Implicit tax rates Consumption Labor

1992

20.5 7.3

9.9 2.4 6.7

1994

19.5 8.3

9.4 2.8 6.3

1995

19.7 7.6

9.1 2.5 6.6

1996

20.1 7.2

9.2 2.4 6.1

1997

11.3 7.4

6.0 2.5 8.1

1998

12.7 6.7

6.7 2.2 6.9

1999

Table 9.9 Structure and developments of taxation by function and by implicit tax rates in Malaysia, 1992–2002

11.1 6.3

6.0 2.1 6.4

2000

11.5 8.6

7.0 2.9 9.7

2001

13.0 8.5

8.4 2.8 9.5

2002

Malaysia 231 commodities of palm oil, crude petroleum and liquefied natural gas. In 2003, Malaysia’s external trade balance recorded a merchandise trade surplus of US$19.7 billion, up by 45.6 percent compared with US$13.6 billion in 2002; in 2003, the overall balance of payments recorded a surplus of US$10.3 billion, up by 175.4 percent compared with US$3.7 billion in 2002. Macroeconomic policy in 2003 was focused on attenuating the impact of SARS and geopolitical concerns. A series of monetary and fiscal measures were implemented during the second quarter of 2003 in response to the SARS outbreak and the weaker than expected global economic conditions, namely slow growth in the US. As a result of these developments, Bank Negara Malaysia reduced its policy rate and the government announced a comprehensive economic stimulus and relief package of US$1.9 billion (2 percent of GDP), encompassing fiscal and financing measures that were designed to provide immediate relief to affected sectors, sustain a high level of consumer spending and contribute to stronger wage growth. Tax reforms of recent years, under way and planned The Malaysian economy has recovered from the Asian financial crisis of 1997 with a strong growth of 8.3 percent in 2000; however, in subsequent years, growth was again affected by the global economic slowdown, in part due to the September 11 Twin Towers incident in the US. To mitigate the adverse impact of the external slowdown, the government has designed and implemented some policies and strategies through the Pre-emptive Stimulus Package in 2002 and the Package of New Strategies in March 2004, containing some relevant fiscal measures to improve investment. Some of these fiscal measures are the following. The existing incentives for small companies will be improved. Small companies before the package reform were given the following incentives: (i) pioneer status with 70 percent income tax exemption for five years or (ii) investment tax allowance of 60 percent on capital expenditure incurred within five years to be offset against 70 percent of statutory income. The 2002 package has improved existing incentives through: (i) increasing income tax exemption under pioneer status from 70 to 100 percent and (ii) increasing statutory income that can be offset for investment tax allowance from 70 to 100 percent. Under the pre-package reform, pioneer status with 100 percent tax exemption for ten years or investment tax allowance of 100 percent for five years was given on a case-by-case basis. The 2002 package has improved as follows: (i) extending the maximum period for pioneer status from ten to 15 years and (ii) extending the period for investment tax allowance from five to ten years. Before the 2002 reform the Malaysian international trading companies

232 Gianpaolo Fanara (MITCs) were provided with income tax exemption on statutory income equivalent to 10 percent of their increased export value. To promote export of locally produced goods, the rate of income tax exemption for MITCs is increased from 10 to 20 percent of their increased export value. The 2002 package also contains some measures in order to improve the skilled human resource and the productivity of the workforce; in particular, a series of fiscal measures are implemented to facilitate Malaysian education programs, training of domestic and overseas human capital, and the construction of new schools, universities and libraries. Finally, the 2002 reforms provide relevant incentives to some sectors considered strategic to the Malaysian economy, such as tourism, construction, manufacturing (e.g. electronic) and agriculture: in particular a series of expenditure allowances, deductions and reliefs are implemented. Suggestions for further improvements Prospects for the next few years are very promising for Malaysia: the economy is projected to grow by 5.5 percent in 2004 and 2005. Consequently, the Malaysian government’s structural agenda for the next few years includes a further enhancement of competitiveness and resilience of the economy in order to provide a business-friendly environment in which domestic and international firms can flourish and prosper; the main actions include: (i) improving fiscal administration and simplifying fiscal procedures through self-assessment, (ii) reviewing and rationalizing the tax incentive structure, (iii) reducing the regulatory and fiscal burden to promote private consumption and investment, (iv) improving the labor market and ensuring the supply of a skilled workforce and (v) promoting greater usage of ICT by firms and increasing their ability to innovate through particular fiscal allowances and deductions.

Notes 1 RM/US$ ⫽ 0.28. 2 Data from 1992 to 2000 are from the IMF Statistical Appendix, while for 2002 they are from the Department of Statistics of Malaysia. 3 A RPC is a controlled company that owns real properties or shares with a defined value of not less than 75 percent of its total tangible assets. 4 The figures here calculated have to be considered as approximated measures of taxation distribution by the implicit tax rate method, because of the lack of detailed data; in particular the implicit tax rate for capital and business could not be calculated due to the lack of data on the gross operating surplus.

Bibliography International Monetary Fund (1998) Malaysia: selected issues, Washington, DC: IMF. —— (1998) Malaysia: recent economic development, Washington, DC: IMF.

Malaysia 233 —— (1999) Malaysia: selected issues, Washington, DC: IMF. —— (2000) Malaysia: statistical appendix, Washington, DC: IMF. —— (2001) Malaysia: statistical appendix, Washington, DC: IMF. —— (2003) Vietnam: statistical appendix, Washington, DC: IMF. —— (2004) Malaysia: statistical appendix, Washington, DC: IMF. OECD (2001) Revenue statistics, Paris: OECD. —— (2003) Revenue statistics, Paris: OECD. United Nations (2004) Human Development Report.

Websites www.statistics.gov.my – Department of Statistics of Malaysia. www.gov.my – Government of Malaysia. www.imf.org – International Monetary Fund. www.finance.gov.my – Ministry of Finance of Malaysia. www.oecd.org – Organization for Economic Co-operation and Development (OECD). www.un.org – United Nations (UN). www.worldbank.org – World Bank.

10 South Korea Lidia Ceriani

Introduction, contents and main conclusions This chapter is devoted to the analysis of the Korean tax system. The study has been elaborated using the most recent statistics accessible from different sources, from international institutions to Korean sources. The rapid growth beginning in the 1960s has transformed Korea from one of the poorest countries in the world to an industrialized nation with a per capita income of more than half the OECD average. This study is particularly interesting because it permits an investigation of the character of the fiscal system of a country that has experienced such a rapid change and that is still growing fast. The next section, after a brief analysis of Korea’s economic and political environment, is dedicated to the description of the current structure and the development of the tax system from the beginning of the 1980s. Korea is characterized by a relatively heavy dependence on indirect taxation. Among the indirect taxes, the most important is the value added tax (VAT), while earmarked taxes together reach a sum that is 4 percent of the gross domestic product. The third section describes the framework of the main Korean taxes, focusing on personal income tax, corporate tax and VAT. After that, there is a section that analyzes, according to the available data, the distribution of fiscal burden, giving particular stress to tax wedges on labor, capital and consumption, and to the distributional effects from taxes and transfer. The most striking positive features of the Korean tax system are the low overall tax burden and the limited labor market distortions. The tax system is also relatively neutral with respect to income distribution, leaving almost unaffected the fairly even pre-tax income distribution. The last section begins with an outlook of the future trend of the main economic indicators. After a sharp downturn experienced by Korea in the first half of 2003, the economy, according to the OECD estimates, should grow to around 4.75 percent in 2004 and 5.5 percent in 2005. The main problem that Korea must face is the aging population, which will increase public expenditure significantly. Other major weaknesses include the

South Korea 235 narrow bases, the excessive complexity of the system and the perceived unfairness of tax enforcement. The section then analyses some of the main challenges in tax reforms, in particular: improving the budgeting and expenditure management system, enhancing the efficiency and accountability of public service delivery and securing tax base. Reform is strongly needed in a number of areas of the tax system in order to improve its revenue-raising capacity, equity, efficiency and simplicity. Tax reform should preferably cover a broad range of measures to make all groups contribute to the future increase in the tax burden and to reap the full economic benefits of the various potential improvements and their mutual interactions. Moreover, reforms should be implemented in a transparent and timely fashion, allowing the private sector time to adjust. By broadening the tax bases, simplifying the system and making the structure and enforcement more equitable, the tax system in Korea could become substantially more neutral across different types of income and less distortionary to economic activity.

A broad view of the tax system and its development from the early 1980s A short reminder of the Korean economy and public sector outline 1 As one of the four tigers of East Asia, since the early 1960s, South Korea has achieved an incredible record of growth and integration into the hightech modern world economy. Four decades ago GDP per capita was comparable with the levels of the poorer countries of Africa and Asia. Today its GDP per capita is 18 times North Korea’s and equal to the lesser economies of the European Union. Growth plunged to a negative 6.6 percent in 1998, then strongly recovered to 10.8 percent in 1999 and 9.2 percent in 2000. Growth fell back to 3.3 percent in 2001 because of the slowing global economy, falling exports and the perception that much-needed corporate and financial reforms had stalled. Led by consumer spending and exports, growth in 2002 was an impressive 6.2 percent, despite anemic global growth, followed by a moderate recession (2.8 percent growth) in 2003. In 2003 the six-day working week was reduced to five days. The composition of GDP by sector shows in first place services (58 percent) followed by industry, mostly electronics, telecommunications, automobile production, chemicals, shipbuilding and steel (37.6 percent) and then agriculture (4.4 percent). The unemployment rate, 3.4 percent in the first quarter of 2004, is about the same as the average of the past three years, while core inflation (which excludes petroleum-based fuels and non-grain farm products) has remained around the midpoint of the medium-term target range of 2.5 to 3.5 percent, despite the economic downturn.

236 Lidia Ceriani The general government spending as share of GDP amounts to 23 percent of GDP and its main categories are consumption (10.1 percent) and net capital outlays (8.3), while income transfers are only 3.6 percent, compared with the 10 percent of Japan and the 12.4 percent of the OECD area. The population of the Republic of Korea as of 2003 was 47.9 million. Fast population growth was once a serious social problem in the Republic, as in most other developing countries. Due to successful family planning campaigns and changing attitudes, however, population growth has been curbed remarkably in recent years. The annual growth rate was 0.6 percent in 2003. A notable trend in the population structure is that it is becoming increasingly older. The 2003 population estimate revealed that 8.2 percent of the total population was 65 years old or over. The number of people in the age range 15–64 years accounted for 71.4 percent. The life expectancy at birth is 75.6 for the total population. The distribution of family income (calculated as Gini index)2 is a low 31.6 and 4 percent of the population is below the poverty line. The tax system The current structure The tax pressure in Korea is among the lowest in OECD. Total tax revenue amounts to only 27 percent of GDP. The low tax burden mainly reflects a relatively small government sector, as well as the lack of a social safety net, comparable with those existing in many other OECD countries and despite a recent significant expansion. However, compared with the other low-income OECD countries, the tax burden in Korea is about average. The tax-to-GDP ratio has risen slightly over the past 20 years (Table 10.1). Most of the increase of the tax burden took place during the 1990s, reflecting high growth rates of personal incomes, increasing social contributions and higher property taxes. Over the past two decades, Korea’s tax system has developed along lines similar to the other OECD countries. In particular, tax-bases have been broadened and rates lowered, thereby reducing the distortionary cost of taxation. But a number of features still continue to hamper equity, efficiency and redistribution, in addition to making the system excessively complex. Korea is characterized by a relatively heavy dependence on indirect taxation. During the 1970s and 1980s the share of indirect taxes to total taxes was recorded as about 60 percent. Although this share gradually decreased in the 1990s, it is still high, when compared with that of other OECD member countries: in 2001 indirect taxes amounted to 39.6 percent of the total tax revenue, while direct taxes only amounted to 26.4. Among the indirect taxes, the most important is the value added tax,

4.5 2.3 1.9 1.5 10.1 3.6 6.3 0.6 16.7 0.3 0.0 0.2 17.0 ⫺ ⫺ ⫺

⫺ ⫺ ⫺

89.0 10.1 0.9

1985

4.5 2.0 2.0 1.4 11.1 3.9 7.0 0.5 17.5 0.2 0.0 0.2 17.7

1980

3.7 1.3 1.4 1.5 9.3 1.9 7.2 0.7 15.2 0.1 0.0 0.1 15.3

Note a Fiscal revenue by sub-sectors of government.

Source: OECD (2002).

Direct taxes of which: personal income corporation income Taxes on property Taxes on goods and services: general consumption specific consumption Others Total tax revenue Social contributions: employees employers Total fiscal revenue Administrative level:a Central government Local government Social security funds

1975

⫺ ⫺ ⫺

6.2 3.5 2.7 2.5 9.2 3.9 5.1 0.2 18.1 1.0 0.3 0.7 19.1

1990

⫺ ⫺ ⫺

6.5 3.9 2.5 3.0 8.8 3.9 4.5 0.8 19.1 1.4 0.4 1.0 20.5

1995

⫺ ⫺ ⫺

7.0 4.3 2.6 2.5 8.7 3.5 4.6 0.9 19.1 3.8 1.7 2.1 22.9

1998

⫺ ⫺ ⫺

5.8 3.7 2.1 3.2 9.5 4.2 4.8 1.0 19.5 4.1 2.0 2.1 23.6

1999

68.2 15.1 16.7

7.5 3.8 3.7 3.2 10.0 4.4 5.1 1.1 21.8 4.4 2.4 1.9 26.2

2000

Table 10.1 Structure and developments of consolidated general government revenue in South Korea, 1975–2001 (% of GDP)

⫺ ⫺ ⫺

7.3 ⫺ ⫺ 3.1 10.9 ⫺ ⫺ 1.1 22.4 5.0 ⫺ ⫺ 27.4

2001

238 Lidia Ceriani which represents the biggest proceeds for the government, standing for 5.4 percent of GDP. The second most important tax is income tax: its revenue is about 15 percent of the total. The yield collected from transportation tax and custom duties is also significantly high: these make up roughly the same revenue gathered from corporation tax. A unique factor of the Korean tax system is the importance of earmarked taxes, among which the most important are education tax, transportation tax and rural development tax. Combined with some minor local earmarked taxes, total revenue from earmarked taxes amounts to nearly a fifth of total general government tax revenue, or almost 4 percent of GDP. Developments of the system from 1975 to 2002 The trend of fiscal revenue in Korea was affected by a continuous succession of tax reforms between 1974 and 2000: during this period tax reforms occurred almost once a year. In December 1976, the government carried out a large-scale tax reform and introduced the value added tax (VAT) and the special excise tax. In addition, 18 new tax laws were enacted or amended under the reform. This tax reform was mainly aimed at the modernization of the tax system and the creation of a better social environment. By this reform, the traditional indirect tax system, which included a cascade-type business tax, was replaced by a consumption-type VAT and supplementary special excise taxes. This was primarily to simplify tax administration and promote exports and capital investments. General consumption taxes on good and services, in fact, were enhanced by 2 percentage points between 1975 and 1980. There was a dramatic economic policy change in the 1980s. Following the second oil crisis in 1979 and the political turmoil of 1979 and 1980, the Korean economy was plunged into a deep recession and total tax revenue was reduced by 0.8 percent. Under these new circumstances, the Korean government fundamentally switched its economic policy directions: from protection to international competition and openness, and from regulation to liberalization and bold privatization. In this regard, the government began to reduce its intervention in the private sector. As a result, not many new functions were demanded from taxation. Therefore there were no relevant tax changes during the 1980s. The major content of tax reforms introduced from 1989 to 1992 was principally the reduction of the burden on wage and salary earners and an enhancement in tax equity. In order to achieve these results, the government lowered the fiscal rates on wage and salary income, increased the limits on tax credits for wage and salary earners and strengthened taxation on property. In fact, social security contributions over the three years here

South Korea 239 considered grew 0.5 percent points faster for employers than for employees, and there was also a rise of rates on financial assets and on inheritance and gift taxes. The 1993 tax reform was aimed at reaching the following objectives. First, rates of corporation tax and individual tax were lowered: this strategy had the consequence of reduction of yield from income and corporate taxation. Then the taxation method on capital gains was modified. A tax credit system on VAT was introduced, and there was an adjustment of the rates of excise duties and the liquor tax: this resulted in the lowering of the revenue from indirect taxation. The 1994 tax reform was designed to establish a tax system characterized by lower tax rates and broader tax bases. The Korean government planned to set up a fair tax system in terms of horizontal equity. The Korean government finally introduced the inclusion of the financial income in excess of WON40 million into the global income tax base. The Korean government also hoped to improve the efficiency of the economy by mitigating distortions caused by government intervention and by encouraging market competition. In order to strengthen the international competitiveness of domestic industries, corporation tax rates were reduced again. To overcome inequities in the fiscal charge on different goods, indirect taxes were redefined, by simplifying tax rates and structure. The decrease of corporate tax experienced from the second half of the 1990s is a consequence of the 1995 tax reform. In order to strengthen the competitiveness of Korean firms, the government alleviated the tax burden, adjusting the corporation tax rate downwards by 2 percentage points. During the same period a decrease of income tax was recorded: this was the effect of the broadening of individual income tax brackets. The economic crisis of Korea in late 1997 has forced the government to start a series of comprehensive economic reforms to overhaul the economy. At such a troubled time, it was essential that no tax liability should either discourage or prevent companies and financial institutions from undergoing necessary restructuring. Above all, the Korean government introduced tax measures to restructure the banking and business sector. Taxes on asset transaction for the purpose of corporate and financial reorganization were removed or reduced. Tax incentives to encourage and accelerate restructuring were mostly granted to transaction-related taxes such as capital gains tax: they were intended to encourage sales of business assets and corporate mergers and acquisitions. Tax revenue has declined significantly since the beginning of 1998: among all, the slowdown of income-elastic tax bases was particularly pronounced. Therefore, the government chose to increase taxes on incomeinelastic goods. To broaden tax bases, the government curtailed tax exemptions and reductions: a notable example was the abolition of the VAT exemption on some services. Another example of the government

240 Lidia Ceriani effort to broaden tax bases was the significant increase of the transportation tax, which reached its top value of 1.9 percent of GDP in 2001.

Some quantitative and institutional features of the main taxes Direct taxes Personal income tax (PIT) Korean citizen and individuals considered as residents for tax purposes are subject to Korean income tax on worldwide income. Korea’s income tax is a global tax, under which most incomes are aggregated and taxed at progressive rates. The income tax is collected either by withholding at the source or by filing a tax return based on voluntary self-assessment for the income during a calendar year. Personal incomes are classified into the following ten categories for the purposes of tax computation: interest, dividends, real-estate income, business income, wage and salary income, temporary property income, retirement income, timber income, capital gains, and other income (such as prize winnings, royalties, rewards, etc.). Tax is withheld at the source of income for the following items: interest, original issue discount premiums of non-interest-bearing bonds, dividends, employment and retirement payments, remuneration or fees for professional services, and fees for public performances. Tax on wages and salaries is withheld at the source, and adjustments between withheld amounts and final tax liability are made at the first payment of such wages and salaries in the next calendar year. For income from the lease of real estate and income from business activities, gross receipts minus all necessary expenses are taxable. Capital gains are added to the aggregate income to be taxed at progressive rates. There is a separate tax for capital gains on land or buildings for which the rate varies, depending on whether ownership has been long or short term. All income, except retirement income and timber income which are taxed separately, is added together. Subtracted from this total income are WON1 million for each exemption: the basic exemption, the exemption for spouse and the exemption for each dependant. When a taxpayer, spouse or dependant is handicapped, elderly (65 years or older), a widow/widower or a working student, an additional exemption is available. The basic exemption is WON1 million times the number of persons in the taxpayer’s family, plus another million if the family is formed by one person alone, or plus WON500,000 if the family is formed by two persons. A person who is 65 years or older, a handicapped person, a female worker or a single male worker who has a lineal descendant under six

South Korea 241 years of age are eligible for an additional exemption and may deduct WON1 million (plus WON500,000 if the person belongs to the third group) per year. Casualty losses, medical expenses, life insurance premiums, fire and casualty insurance premiums, social insurance contributions and donations may be deducted from taxable income, subject to certain limitations. Particularly, wage and salary income earners may deduct an amount equal to the sum of the following from their wage and salary income. 1 2 3

4 5 6

Insurance premiums paid, up to WON700,000, and total medical care insurance contribution. Medical expenses in excess of 3 percent of the taxpayer’s annual salary, up to WON1 million. Domestically incurred educational expenses up to WON700,000 annually per student for kindergarten and nursery school expenses and up to WON2.3 million for college education expenses. Forty percent of the loan interest (for a total up to WON720,000 per year) allotted to the purchase or lease of a house of appropriate size. Interest of a mortgage loan with the duration of more than ten years. Amounts donated to qualified institutions: up to 5 percent of the taxpayer’s salary and wage income of the year. This limit may be raised by an additional 5 percent of the salary and wage income in cases of donations made to privately operated schools.

A taxpayer may elect to choose an annual standard deduction if he fails to claim the deductions in question or accrues only global income without wages or salaries earned. Finally, a taxpayer may deduct the total amount of their pension contribution up to WON2.4 million. Recently, the amount of expenditure used by credit card was added as a deduction item, which was intended to promote the use of credit cards. The recent surge in the use of credit cards by this type of incentive is believed to have contributed in making business income more exposed to the tax authority. A taxpayer may deduct 20 percent of the amount of payment by credit card, exceeding 10 percent of income, up to WON5 million. Effective from 1 January 2002, individual income tax rates on global income range from 9 to 36 percent (Table 10.2). Inheritance and gift taxes The inheritance tax is imposed on the recipient of assets from an inheritance, bequest or devise. If any heir or legatee has received property by gift from the legator within ten years of death, the value of such property is added to the total

242 Lidia Ceriani Table 10.2 Personal income tax rates in South Korea Tax base (thousands WON)

Tax rate (%)

up to 10,000 10,000–40,000 40,000–80,000 over 80,000

9 18 27 36

Source: Jae Jin Kim (2002) p. 37.

taxable assets. Life and personal accident insurance payments and retirement allowances received by heirs up to a certain amount are also added to the total taxable assets. From the value of properties, WON200 million as the basic exemption and WON30 million for each statutory heir are subtracted to obtain the taxable value. The value thus obtained is divided among the statutory heirs in accordance with the shares provided by the Civil Code. The tax rates are progressive, beginning at 10 percent for a taxable estate of WON100 million or less; the maximum rate is 50 percent for taxable assets over WON3 billion. The tax rates are applied to the total share. Some allowances are granted for inheritance and gifts, varying according to the category of beneficiaries or donees. Besides a general basic deduction of WON200 million, in fact, there are also additional deductions for inherited family businesses (up to WON100 million), inherited farms, fisheries and forestry (up to WON200 million), deductions for dependants amounting to WON30 million per person, deductions for minors (an annual deduction of WON5 million is granted to the minor until he or she becomes 20 years old) and deductions for the elderly (a deduction of WON30 million is granted to the inheritor or legatee, or a member of the inheritor or legatee’s family when he is over 60 years old). Effective from 1 January 1997, gift tax rates are the same as those for inheritance tax (Table 10.3).

Table 10.3 Inheritance and gift tax rates in South Korea Tax base (millions WON)

Tax rate (%)

up to 100 100–500 500–1,000 1,000–3,000 over 3,000

10 20 30 40 50

Source: Beom-Gyo Hong (2002), p. 23.

South Korea 243 Corporation tax Companies that are subject to corporation tax in Korea can be classified into two types: domestic or foreign, and for-profit or non-profit. For taxation purposes, a company with its head or main office in Korea is deemed a domestic company and is liable to tax on its worldwide income. Otherwise, it is considered to be a foreign company, and the tax liabilities of foreign companies in Korea are limited to Korean-source income. The income of a domestic corporation during each business year is the amount remaining after deducting the gross amount of losses from the gross amount of gains during each year. To derive the tax base, the deficits carried over from the previous five years should be deducted from this income. Other deductibles specified in the law may apply. The Korean corporation tax rates have gradually been reduced in the last ten years: from 34 to 32 percent in 1994, to 30 percent in 1995, and further to 28 percent in 1996. For corporations with income of less than WON100 million, the tax rate on profits has been 16 percent since 1996 and that rate is also applied for non-profit organizations. Both rates were reduced by 1 percent in 2001 (Table 10.4). Indirect taxes Value added tax (VAT) In its 1976 tax reform, Korea adopted the value added tax (VAT) on the delivery of goods, the rendering of services and importation. The tax replaced the following taxes as of 1 July 1977: business tax, commodity tax, textile products tax, petroleum products tax, electricity and gas tax, travel tax, admission tax and the entertainment and food tax. VAT is imposed at the uniform rate of 10 percent on a wide range of goods and services. The tax is levied at each stage of sale on domestically produced or imported goods and services by business entities. The consumption tax paid to suppliers is credited against the tax on sale by the business entities in order to avoid tax cascading

Table 10.4 Corporation tax rates in South Korea Tax base (millions WON)

Tax rate (%)

up to 100 over 100a

15 27

Source: Beom-Gyo Hong (2002), p. 21. Note a For corporations with income of more than WON100 million, the tax amount is computed as follows: WON15 million plus 27 percent of tax base.

244 Lidia Ceriani Exports are exempted from tax (zero rate), and daily necessities such as unprocessed foodstuffs and tap water, financial services and so on are exempted. In addition to the zero rating and other exemptions, there was, until July 2000, a special treatment called simplified taxation. A trader whose turnover (or proceeds including VAT) of the supply of goods or services during the immediate preceding year is less than WON48 million falls under that category. If a trader is eligible for simplified taxation, the tax amount payable is calculated as follows: aggregate amount of supply during the concerned taxable period ⫻ average rate of value added as prescribed by the presidential decree for each category of business (ranging from 20 to 40 percent) ⫻ 10 percent. This simplified taxation tended to be abused because small business such as self-employed are prone to underreport their turnover. The elimination of this scheme implies that 0.5 million taxpayers will be included in the ordinary VAT regime, increasing the ordinary taxpayers by 40 percent. However, the taxable turnover in the ordinary regime will only increase by about 1 percent, making the immediate effect on total VAT revenue quite small. Excise taxes As in most other countries, a number of excise taxes are imposed on consumption. In Korea these are on transportation, liquor, telephone service and the special consumption tax, which imposes excise taxes on 27 luxury goods. Tax rates differ according to the type of commodity. Percentage rates vary from 10 to 20 percent except for the rate applying to horse race admissions. Taxable objects and tax rates are classified into four classes, each of which is subdivided into groups. The first class of goods includes slot machines, golf requisites, movie projectors, and so on. Class two includes jewelry worth over WON2 million, luxury cameras, watches, carpets and furniture, etc. The tax rate applied on these first two classes is 20 percent. Cars are classified in class three. Tax rates of class three depend on the size of engine and currently are in the range of 7 to 14 percent. Class four includes fuels such as gasoline, diesel, LPG, LNG, etc. A fixed amount of tax is applied to those items. Together, these excise taxes raise revenue equivalent to 2.5 percent of GDP. Compared with many other OECD countries, where excise taxes tend to be concentrated on a few items in three product groups (mineral oils, tobacco and alcoholic beverages), the Korean excise tax system is much more widespread and complex.

South Korea 245

The fiscal burden The general distribution of taxation charge The tax mix in Korea relies more heavily on property and consumption taxes than in most other OECD countries. The share of consumption taxes (40 percent on total revenue) is significantly above the level of countries such as Japan, the United States and those in the EU, though this share has strongly fallen during recent years and is now closer to the OECD average. VAT revenue constitutes under half of total consumption tax revenue, a relatively low share compared with other OECD countries. This is not desirable since VAT tends to be less distortive than other kinds of consumption taxes. Social security contributions are still fairly low, but have risen substantially, primarily because of the introduction of the national pension scheme in 1988. The share of individual income tax has also increased, in line with the government’s long-term objective of raising the ratio of direct to indirect taxes. However, the share of individual income tax is still somewhat lower than in many other OECD countries (Dalsgaard 2000). Measured by average ex-post effective (⫽implicit) tax rates, taxation of labor is much lower in Korea than elsewhere in the OECD, while average taxation of consumption and capital appears to be close to the OECD average (Table 10.5). Taxation of capital has risen sharply over the past 20 years, mainly as a consequence of soaring property tax revenue because of land and property price increases, but also a declining trend in the economy-wide operating surplus. When measured against the net operating surplus, the increase is even larger since depreciation of fixed capital, as measured in the national accounts, has increased more than proportionately with income (Dalsgaard 2000). Although these average tax rates should be interpreted with caution, they indicate that the rise in the average effective tax burden on labor relative to that on capital which took place over the past couple of decades in most other OECD countries did not occur in Korea. The rising trend in the OECD area is thought also to reflect the fact that capital taxation tends to be more prone to base erosion than labor and consumption taxes. However, since a significant share of Korea’s capital taxes are based on property, erosion of the tax base is less of a problem. Distributional effects from taxes and transfer The tax and transfer system has only a marginal influence on income distribution in Korea. Income taxes slightly reduce income inequalities whereas social security contributions and the value added tax act in the other direction. The special consumption tax, which was introduced along with the VAT to counterbalance expected adverse distributional effects of

Source: OECD (2000).

United States Japan Korea OECD average

28.3 39.1 13.4 25.1

1980–5

29.2 42.4 16.2 26.7

1986–90 31.1 32.6 26.8 26.6

1991–7

Capital based on gross operating surplus

21.6 20.1 3.5 30.0

Labor ■ 1980–5 22.1 23.1 5.0 32.2

1986–90

Table 10.5 Average effective tax rates on capital, labor and consumption in South Korea

22.6 24.0 7.7 33.4

1991–7

6.3 6.4 17.2 16.1

5.9 6.2 17.1 17.2

Consumption ■ 1980–5 1986–90

6.1 6.7 16.0 17.1

1991–7

100.00 0.3092

2.69 4.81 5.99 7.04 8.08 9.24 10.59 12.30 14.83 24.43

Gross income after income tax

100.00 0.3169

2.63 4.70 5.88 6.93 7.96 9.14 10.53 12.32 14.97 24.94

Gross income after SSC

100.00 0.3182

2.57 4.68 5.86 6.93 7.96 9.13 10.58 12.33 15.02 24.95

Gross income after VAT

100.00 0.3160

2.63 4.71 5.88 6.94 7.99 9.15 10.57 12.33 14.99 24.82

Gross income after SCT

100.00 0.3131

2.59 4.76 5.96 7.01 8.03 9.22 10.60 12.29 14.89 24.66

Net income after all taxes and SSC

Notes SSC ⫽ social security contributions; SCT ⫽ special consumption tax.

Source: National Statistical Office, Monthly Income and Expenditure per Household by Income Decile for Salary and Wage Earners Households of All Cities (1998); OECD calculations.

100.00 0.3157

2.64 4.71 5.89 6.95 7.99 9.15 10.56 12.32 14.97 24.83

1 2 3 4 5 6 7 8 9 10

Total Gini index

Gross income

Decile

Table 10.6 Distributional effects of taxes and social security contributions in South Korea, 1998 (%)

248 Lidia Ceriani the VAT itself, has only a negligible effect on the income distribution. The low overall redistributional effects from the tax system mainly reflect the low tax burden in Korea (Table 10.6). Progressivity of the personal income tax system is roughly similar to that in other OECD countries. The distance between the marginal tax rate for an average production worker (APW) and for a top income earner is higher in Korea than elsewhere. However, since the top marginal tax rate, unlike in many other OECD countries, is only applied at very high income levels, according to Dalsgaard (2000), this is not representative of the potential increase in tax burdens facing the average income earner. When measured at the margin, the income tax system in Korea appears to be no more progressive than other OECD countries, including Japan and the United States.

Tax reforms A quick glance at the macroeconomic and budget outlook According to the latest OECD Economic Survey of Korea (OECD 2004a), Korea has been one of the fastest growing economies in the OECD area over the past five years, with an annual growth rate of about 6 percent. This excellent performance has sustained the convergence process, lifting per capita income to two-thirds of the OECD average. Korea has made a stunning recovery from the Asian crisis. Within one and half years, output had already regained its pre-crisis level; within five years, the economy was one-quarter larger than before. Korea’s performance has been better than any other Asian country affected by the crisis. The return of high growth rates corrected some of the weaknesses that made it vulnerable to the Asian crisis. Rapid growth also reflects the country’s underlying economic dynamism, particularly in the information and communications technology sector. Korea has also benefitted from strong demand from China, which has emerged as its biggest trading partner. Korea’s outstanding performance is underpinned by a large input of labor and capital, reflecting a still-rapid population growth, rising labor force participation rates and high level of investments. Indeed, remaining weaknesses in the economic framework contributed to the economic recession in the first half of 2003, which slowed growth for the year to around 3 percent. Private consumption was negatively affected by the liquidity and solvency problems of the credit card companies. The instability in financial markets, resulting from the problems of the credit card companies, had a negative impact on business investments. This negative impact was magnified by the deterioration in already difficult industrial relations and labor strikes at major firms. In addition to these structural weaknesses, Korea was hit by a series of external shocks, notably SARS and the North Korea nuclear issue, which weakened confidence.

South Korea 249 Public expenditure as a share of GDP is relatively low in Korea: using national accounts data, total public spending of the general government amounts to 22.5 percent of GDP (OECD 2003), the lowest level among the OECD countries. The major factor that underlies the comparatively low spending level in Korea is that income transfers are limited by the lack of a well-developed social safety net. This low level is partially offset by a relatively high level of privately funded social expenditure, both mandatory and voluntary. This is largely due to the retirement allowance that firms are legally required to pay and voluntary social benefits, such as family allowance – paid by many employers. Even so, total social expenditure, public and private, is still only 8 percent of GDP, compared with 15 percent in Japan, and 20 percent or more in all other OECD countries. Government consumption is well below the OECD average, reflecting the small size of the public sector in Korea. Upward pressure on spending is likely to gain further momentum in Korea. One factor is the WON157 trillion (27 percent of GDP) program on financial-sector restructuring. Pressure to increase spending will also arise from the expansion of the social safety net. Social welfare spending increased as a result of the financial crisis in 1997 in order to cope with the sharp rise in unemployment and to expand the safety net. Over the longer term, spending pressure will likely accelerate as a result of population aging and the consequent demands that this implies on social spending, notably on pensions, healthcare and associated personal services. After Mexico and Turkey, Korea has the youngest population among OECD members, but its population will age more rapidly. According to the social expenditure database of OECD (2000), around 2030 the portion of pension and security expenditure will take the largest part of the total social welfare expenditure (about 48.4 percent). That is, pension and security expenditure will take about 10 percent of GDP in 2030, which is larger than the OECD average in 1995. Expenditures related to health, including health insurance, will take the second largest part of the social welfare expenditure in 2030 (6.4 percent of GDP). In sum, expenditures related to pensions and health will take 60 to 65 percent of welfare expenditure in 2030 and produce a considerable burden on the government budget. Tax reforms of recent years, under way and planned The direction of tax reforms was lending support to mid- and low income classes in 1999, bringing the energy tax regime in line with international standards in 2000, reducing tax incentives and lowering tax rates in 2001. The direction of tax reforms in 2002 was supporting mid- and low-income classes, giving stimulus to local economy, enhancing corporate competitiveness and transforming Korea into a regional hub of business. The main pillars of international tax reforms during this period have

250 Lidia Ceriani been regulated by the Foreign Exchange Transaction Act (1999) and its subsequent revision (2000), following three main directions. First, by allowing an exchange of financial transactions with other countries; second, by subjecting all international transactions to transfer pricing rules; and finally by revising rules on thin capitalization and controlled foreign corporations (CFC rules). With respect to exchanging information on financial transactions with other countries, the Ministry of Finance and Economy plans to provide the information on a reciprocal basis and upon request from foreign revenue authorities on a condition that the released information is strictly limited for the purpose of imposing tax. Non-residents living in Korea as well as foreign companies and branches operating in Korea will come under the scope of the new measure. As for subjecting international transactions to transfer pricing rules, some international transactions between related enterprises (or associated enterprises) falling under the corporate income tax law and not covered by the international tax law used to be subject to “the rules on denying deductibility of improper transactions” in the past. This rule has been changed and all international transactions have become subject to transfer pricing rules, effective from 2003. Korea expanded the scope of its major foreign shareholders by subjecting sister companies of a domestic company to thin capitalization rules. As of April 2003, foreigners owning more than 50 percent of shares of domestic corporations and those foreign shareholders effectively determining the course of the domestic company are classified as major foreign shareholders subject to thin cap rules. Korea has also expanded the scope of major foreign shareholders of domestic permanent establishments subject to thin capitalization rules. Foreign shareholders owning more than 50 percent of a sister company of the firm concerned are now subject to thin capitalization rules. Another revision the country is going to introduce into the rules on thin capitalization is about deemed foreign loan. In the past, only loans either issued by foreign shareholders or by the third party and guaranteed by the shareholder used to be subject to thin capitalization rules. Now other legal documents such as a comfort letter, effectively guaranteeing payment of the issued loan, shall be deemed as falling under the scope of foreign loans subject to thin cap rules. Thin cap rules prevail over other tax laws or transfer pricing rules in cases of divergence of interpretations. The CFC rules have also undergone some changes. The purpose of CFC rules is to impose tax on unreasonably retained profits of subsidiaries located in “tax havens” by treating them as notional dividends paid to the Korean parent. The concept of “tax havens” under the Korean tax law refers to jurisdictions with no tax or those exempting 50 percent or more of income from tax or with less than 15 percent tax rate. The companies falling under the scope of CFC rules are subsidiaries located in low tax jurisdictions whose share capital is at least 20 percent, either directly or indirectly owned by a Korean parent.

South Korea 251 The Korean tax law provides for some exemptions to CFC rules, where a subsidiary carries on bona fide operations in the low tax jurisdictions through a fixed place of business, such as an office, sales outlet or factory. However, even if these conditions are met, CFC rules still apply to wholesaling, retailing, repair of consumer goods, transport, warehousing, communications, banking, insurance and real estate leasing or services. There have also been some major revisions of the foreign investment regime in Korea. The Korean government recognizes the fact that foreign direct investments play an important role in the economy and plans to designate special economic zones to facilitate an inflow of foreign investments. In designated special economic zones, qualified foreign investments on a large scale shall be granted the same benefits as in foreign investment zones. In the foreign investment zones, companies are exempt from tax for seven years and enjoy 50 percent reduction for the next three years. In designated special economic zones, qualified foreign investments on a medium scale shall be granted the same benefits as in Jeju Free International City. In Jeju, companies are exempt from tax for three years and enjoy 50 percent reduction for the next two years. What is noteworthy about new measures to promote foreign direct investments is that the proposed measures will expand tax incentives to advanced technologies. Suggestions for further improvements One basic goal of mid- to long-term tax reforms is securing tax base. The basic thrust of income taxation reform must ensure an adequate and equitable distribution of tax burden through a continuous broadening of tax base. The reintroduction of the Comprehensive Financial Income Taxation (CFIT) in the fiscal year 2001 will probably contribute to improvement of equity of tax burden in the long run. However, in order to improve effectiveness of this policy, some complementary measures should be prepared. First of all, it is necessary to extend the scope of financial income subject to this policy. Under the current system, a wide range of tax-exempt and tax-favored savings is still taxed separately and their withholding tax rate is 10 percent, which is lower than the rate allowed under the system of CFIT (Jae Jin Kim 2002) According to Chun and Lee (1999), the proportion of the stock of taxexempt and tax-favored savings of total financial assets was over 25 percent as of 1998. In addition, there are other means to avoid CFIT. Long-term treasury bonds and public bonds are not subject to CFIT. Allowance for tax-exemption, favored taxation and separate taxation will restrict the effectiveness of CFIT in improving the equity of tax burden. Individual trading on the stockmarket has increased significantly, but taxation in this area is still in an early stage of development. Therefore,

252 Lidia Ceriani for a more equitable tax burden, capital gains taxation on securities has to be steadily expanded. To accomplish this, small shareholders must bear the tax burden on the capital gains from listed stocks and it is necessary gradually to expand capital gains taxation to derivatives and bonds (Dae Hee Song 2002). At the root of the problems in Korea’s tax system is the wide variation in average effective taxation across various sources of income. While this is partly due to the statutory design of the system, it also reflects an inequitable administrative enforcement of taxation. A number of measures have been implemented since the mid-1990s to enhance tax administration. These include the introduction of the real name system for ownership of financial assets, the adoption of self-assessment as a ruling principle and the implementation of an automated taxpayer database system called the tax integrated system. Although these changes are steps in the right direction, a number of deficiencies are still in place in tax administration and require a comprehensive intervention to improve the effectiveness and transparency of the system. Transparency is one of the most important goals in tax administration, since Korean administration does not depend on evidence-based taxation. Effectiveness is another important goal, since tax administration should be organized with economic judgment. Transparent tax administration can be achieved by evidence-based taxation, which refers to, for example, book-keeping, invoices, etc. Effective tax administration can be organized by reducing the work burden in tax information processing, and by emphasizing tax inspection and audits by tax officers (Jin Kwon Hyun 2002).

Notes 1 The main source is OECD Economic Surveys – Korea (2003). 2 The lower the Gini coefficient, the more equal is the income distribution.

References Beom-Gyo Hong (2002) “An overview of Korean taxation,” Working Paper, Korea Institute of Public Finance. Chun, Y. J. and Lee, C. (1999) “The economic effect of taxation on interest income,” Policy Report 99-06, Korea Institute of Public Finance. Dae Hee Song (2002) “Korean tax reform for the global era,” Working Paper, Korea Institute of Public Finance. Dalsgaard, T. (2000) “The tax system in Korea: more fairness and less complexity required,” Economic Department Working Paper 271, Paris: OECD. Jae Jin Kim (2002) “Personal Income tax in Korea,” Working Paper, Korea Institute of Public Finance. Jin Kwon Hyun (2002) “Tax Administration in Korea,” Working Paper, Korea Institute of Public Finance.

South Korea 253 OECD (2000) Economic Surveys, Paris: OECD. —— (2002) Economic Surveys, Paris: OECD. —— (2003) OECD Economic Surveys – Korea, Vol. 5, Paris: OECD. —— (2004a) Economic Survey of Korea, Paris: OECD.

Websites http://www.imf.org – International Monetary Fund. http://english.korcham.net/ – Korea Chamber of Commerce and Industry. http://www.kipf.re.kr – Korea Institute of Public Finance. http://www.nso.go.kr – Korea National Statistical Office. http://english.mofe.go.kr – Korean Economic Portal. http://english.mofe.go.kr – Korean Ministry of Finance and Economy. http://www.nta.go.kr – Korean National Tax Service. http://www.oecd.org – OECD. http://www.unpan.org – UNPAN: United Nations Online Network in Public Administration and Finance.

11 Thailand Marco Bartolich

Introduction, contents and main conclusions The aim of this chapter is the discussion of the main features of Thai fiscal system, its development during the 1990s and the recent and underway reforms. The second section describes the structure of the tax system and its development during the last decade. Fiscal pressure is low if compared with industrialized countries and also with some less developed ones: it had an increasing trend until the financial crisis of 1997, when it fell by 2.2 percentage points of GDP. After the economic recovery, it started to increase again, but it has not yet reached the pre-crisis level. Indirect taxation is the main source of receipts: it amounts to more than 60 percent of tax revenue, although recently the gap between direct and indirect taxes has somewhat decreased. Personal income tax has limited room as a source of receipts, because of some structural weaknesses, while capital taxation remains at a low value, which explains the inflow of foreign investments during the last decades (together with the low cost of labor). In the third section we examine the main taxes levied at central and local levels. Among direct taxes, personal income tax has many allowances and deductions which reduce the tax base. The brackets of income are eight, while the minimum and maximum rates are 5 and 37 percent, respectively, and the first Baht80,000 are exempt. Corporate income taxation is about at the same level as in the more developed countries. VAT is the second main indirect tax after the excises, because it has a relatively low rate (7 percent). Thailand has a highly centralized fiscal system: tax receipts of local governments are very low, despite the recent reforms, which are leading to a major fiscal decentralization. Hence central government’s grants are the main financial resource for local authorities. The fourth section focuses on the distribution of tax charge by analyzing the evolution of fiscal burden on the main components of national income (consumption, labor, and capital and business) by economic func-

Thailand 255 tion and implicit tax rates during the last decade. Consumption is the most heavily hit (at an average of 10.4 percent of GDP), while the taxation of labor and capital is comparatively low. This is usually indicated as one of the explanatory factors of the rapid economic growth of the last decades. In the last section we describe the fiscal measures taken in response to the financial crisis of 1997 and the main features of the 1999 public sector management reform program, which had, among its objectives, the improvement of revenue collection and distribution of tax burden as well as a progressive increase of fiscal decentralization. In the first phase of the crisis, budget cuts prevailed together with an increase of taxation, especially through indirect taxes, while in the second phase, tax policy became more expansionary in order to reduce the social impact of the economic fall. Among the aims of the public sector management reform program were the enforcement of tax collection, an increase in taxpayer compliance, the improvement of information technology, a devolution of expenditure functions and a greater degree of tax autonomy for the benefit of local governments. Many improvements should be taken, especially for the main taxes. VAT refunds should be redefined in order to reduce their costs, PIT allowances and deductions must be simplified, while for corporate income tax the IMF suggests avoiding fiscal competition with other developing countries by lowering the tax rate and reinforcing the controls over corporations with international businesses, which very often try to repatriate dividends without paying taxes. Finally, some further improvements in fiscal decentralization should be taken in order to take advantage of all its benefits.

A broad view of the tax system and its development from the early 1990s A short reminder of Thailand’s economy and public sector outline In 2002 general government expenditure was 25.3 percent of GDP (without considering intergovernmental transfers), with an increase of 4.5 percentage points with respect to 2001. Central government expenditure represents 90.3 percent of the total: this value indicates that the Thai fiscal system is still highly centralized. Local government expenditure concerns mainly transport, housing and community amenities and general public services. Central government expenditure reflects Thailand’s development priorities: public expense focuses on the sectors that are fundamental for the country’s development and growth. These sectors are agriculture, forestry, fishing and hunting (1.7 percent of GDP), transport (1.4 percent), health

256 Marco Bartolich (1.8 percent, mainly for hospital services), education (4.2 percent, of which 2.8 for pre-primary and primary education), social protection (1.6 percent); less is spent on social security and welfare and defense. Public debt transactions represent 1.3 percent of GDP. These results can be explained by considering that after the 1997 crisis structural reforms became the priorities in order to create a solid ground for the economic recovery: education, health care and regional development have been considered the key sectors that would have led to an improvement of competitiveness, employment and economic growth. Current spending represents 60 percent and capital expenses 40 percent. Among the former, wages and salaries constitute 24.3 percent, a relatively high value in international comparison, while in the second category capital transfers to public enterprises and households have strongly increased during the last year and have caused the increase of expenditure and the deterioration of the general government overall balance. In 2002 per capita GDP was US$1,983 (not PPP corrected), while the labor force was 54 percent of the total population (IMF 2003). In the UN’s 2004 Human Development Report, Thailand occupies the 76th place in the high human development rank, with a life expectancy at birth of 69.1 years, an adult literacy rate of 92.6 percent and a total fertility rate of 1.9 births per woman in the period 2000–5. Infant mortality is still high (24 per 1,000 live births), despite a sharp decrease since 1970. For the period 2002–15 the estimated annual population growth rate is 0.9 percent (it was 1.5 percent from 1975 to 2002). Aids is a serious problem in Thailand: the HIV prevalence is 1.5 percent of the population between 15 and 49 years old. The tax system The current structure and its developments In 2002 general government revenues stayed at 19.2 percent of GDP. Among them, direct taxes amount to 28.9 percent of total receipts, indirect taxes to 54.3 percent, other taxes to 0.4 percent, social contributions to 2.9 percent and non-tax revenues to 13.5 percent. In 2002, the fiscal pressure (i.e. tax revenues plus social contributions) was 16.6 percent of GDP: tax revenues were 16 percent, social contributions 0.6 percent. Direct taxes represent a lower fraction of tax revenues, 34.6 percent, while indirect taxes reach 65 percent. Table 11.1 present the fiscal pressure’s evolution from 1992 to 2002 as percentage of GDP. During the past few years the quantitative spread between direct and indirect taxes has decreased, and the composition of the general revenues has slightly moved towards direct taxes, as the economy has passed from the early stages of modernization to a more sophisticated structure. Among direct taxes, corporation income tax constitutes 18.3 percent of tax revenues, while personal income tax represents

17.2 16.1 1.1

Fiscal pressure Central government Local government

17.5 16.3 1.2

0.2 0.1 0.1

11.4 3.3 4.2 3.3 0.1 17.3

5.8 1.8 3.4 0.6

1993

18.0 16.7 1.3

0.2 0.1 0.1

11.4 3.6 4.2 3.2 0.1 17.8

6.3 1.8 3.8 0.6

1994

Source: own elaboration on Thailand Ministry of Finance data.

0.2 0.1 0.1

11.2 3.9 3.9 3.0 0.1 17.0

Indirect taxes, of which: VAT excises import duties other taxes Tax revenues

Social contributions, of which: employees employers

5.6 1.8 2.9 0.3

Direct taxes, of which: PIT CIT on property

1992

18.3 17.0 1.3

0.2 0.1 0.1

11.4 3.1 4.1 3.1 0.1 18.1

6.6 2.1 3.8 0.6

1995

18.5 17.2 1.3

0.3 0.1 0.2

11.5 3.7 4.0 2.8 0.1 18.2

6.6 2.3 3.8 0.5

1996

17.5 15.8 1.7

0.3 0.1 0.2

10.7 3.2 4.1 2.2 0.1 17.2

6.2 2.3 3.3 0.4

1997

15.3 13.7 1.6

0.2 0.1 0.1

10.2 3.9 3.6 1.4 0.1 15.1

4.9 2.5 1.9 0.3

1998

Table 11.1 Structure and development of fiscal revenues in Thailand, 1992–2002 (% of GDP)

15.0 13.4 1.6

0.4 0.2 0.2

9.6 3.2 3.9 1.4 0.1 14.6

4.9 2.2 2.2 0.3

1999

15.4 13.5 1.9

0.6 0.3 0.3

9.5 3.2 3.7 1.8 0.1 14.8

5.2 1.8 2.8 0.3

2000

15.8 13.7 2.1

0.6 0.3 0.3

9.8 3.2 4.0 1.8 0.1 15.2

5.4 1.9 2.8 0.3

2001

16.6 14.5 2.1

0.6 0.3 0.3

10.4 3.0 4.4 1.8 0.1 16.0

5.5 1.9 2.9 0.3

2002

258 Marco Bartolich only 12 percent, a low value by international comparison. This is mainly due to some structural weaknesses, like a narrow tax base (due to extensive exemptions, income deductions, allowances and tax relieves) and, more generally, to a non-transparent and unfair tax system (IMF 1998). Social contributions amount to 0.6 percent of GDP: employees pay 46.2 percent of them, employers 52.6 percent, while self-employed or nonemployed contributions represent 1.2 percent. Among indirect taxes, VAT plays an important role: it constitutes 18.5 percent of total tax revenue, despite the low VAT standard rate (7 percent). Excises are another important item of indirect taxes: in 2002 they totaled 27.2 percent of tax revenue. The most important were on petrol products, motor cars, tobacco and spirits. Finally, the third main category of indirect taxes is import duties (11.1 percent of tax revenues). Thailand’s fiscal system is highly centralized, despite recent reforms aimed to increase fiscal decentralization: central government collects 92 percent of tax revenues and pays 90.6 percent of general expenditure (including grants from central to local government units). Among central government expenses, grants to local governments represent 7.4 percent (1.6 percent of GDP). These grants constitute 53.1 percent of local government total receipts; this high value emphasizes the important role of central government in the intergovernmental fiscal system. The development of the system from 1992 to 2002 During the last decade, fiscal pressure shows an increasing path until 1996, the year before the financial crisis, when tax revenues and social contributions were 18.5 percent of GDP, and a deep fall until 1999 (15 percent). During the last three years fiscal pressure has grown again, but the 2002 level is lower than the average of the pre-crisis period. Social contributions have been increasing over this period, while tax revenues, which have been strongly hit by the financial crisis, show the same path as the total fiscal pressure. Social contributions’ weight in the fiscal pressure has constantly increased over this period: in 1992 it was 1.1 percent, in 2002 it was 3.4 percent. Among tax revenues, indirect taxes have always played the main role: they have totaled on average 66 percent of tax receipts, even during the crisis, albeit import duties decreased sharply because of the devaluation of the Baht and consumption shift from imports towards domestic goods. In fact, the first fiscal measures of the central government during the financial crisis concerned indirect taxes (especially VAT) and caused a slight increase of tax revenues at the end of 1997 because of the growth of indirect taxation pressure, while direct taxation had not been modified at that time. VAT has had an increasing weight during the last decade: despite the low rate, its receipts amounted to an average of 20.7 percent of tax revenues during the 1990s, with a peak of 25.6 percent in 1998, after the

Thailand 259 1997 VAT rate increase from 7 to 10 percent, a fiscal measure which was introduced to accommodate the expected revenue shortfall. Finally, excises have been increasing during the decade (receipts have doubled in ten years). They had a negative peak in 1998, when central government implemented an expansionary policy (one of the measures was the reduction of some important excise duties). Corporate income tax constitutes the main direct tax, with an average value of 18.7 percent of tax revenue, despite the sharp fall in 1998 (only 12.9 percent), when receipts decreased because of the high buoyancy of corporate income tax with respect to GDP (on average 1.5 percent from 1965 to 1996) and the devaluation of the Baht, which led to a strong increase of the passivities of many corporations because of their indebtedness to foreign creditors. In the 1997–8 period, GDP decreased by 12 percent, which had highly negative consequences on corporate income tax receipts. Personal income tax always had a secondary weight in the mix of total tax revenue (about 12.6 percent as average value) because of the already mentioned structural weaknesses. This tax has not been hit by the 1997 crisis as much as corporate income tax, because of its low buoyancy with respect to GDP: in 1998, personal income tax receipts were 0.6 percentage points higher than corporate income tax revenues as a fraction of GDP. After the fiscal stimulus package measures, its weight relative to total revenues has decreased. In the last decade Thailand’s fiscal system has presented few changes in its tax mix. Tax receipts have always been dominated by indirect taxation. Economic growth has slightly shifted the composition of fiscal receipts away from indirect towards direct taxes, but direct taxation, especially personal income tax, still represents a low fraction of tax revenues. Finally, as to fiscal decentralization, the central government’s share of total tax receipts has been high in the 1990s (an average 93.3 percent), despite a slight but constant decrease (94.7 percent in 1992, 92 percent in 2002). We can see the same evolution on the expenditure side, where the central government’s share of total expenses (including grants from central to local government units) has moved from 91.4 percent in 1992 (but 92.6 percent in 1993 and 92.7 percent in 1994) to 90.6 percent in 2002 (an average 91.5 percent). An international comparison In this section we compare the Thai distribution of fiscal revenues (as percentage of GDP) with that of some selected countries: Vietnam, Japan, Hungary, Germany, Mexico and the United States (Table 11.2). These countries have been chosen in order to see if, and how, Thailand’s fiscal system is similar to other economic realities with different levels of development: Vietnam, Mexico and Hungary can be considered as still-developing countries and hence they can present an analogous situation with

15.8 15.2 0.6 1.9 2.8 0.3 9.8

n.a. 16.6 n.a. 0.4 5.6 0.9 9.7

Vietnam 27.3 17.0 10.3 5.5 3.5 2.8 5.2

Japan 39.0 27.5 11.6 7.6 2.4 0.7 15.1

Hungary

Sources: own elaboration on Thailand Ministry of Finance data, IMF (2003), OECD (2001).

Fiscal revenues Tax revenues Social contributions PIT CIT Tax on property Tax on goods and services

Thailand

Table 11.2 Structure of fiscal revenues in selected countries, 2001 (% of GDP)

36.8 22.2 14.6 10.0 0.6 0.8 10.6

Germany

18.9 15.7 3.2 n.a. n.a. 0.3 9.7

Mexico

28.9 21.8 7.1 12.2 1.9 3.1 4.6

United States

Thailand 261 Thailand (although Hungary has a higher level of economic growth and a more consolidated fiscal structure because of its recent EU membership), while, by analyzing the cases of Japan, Germany and the United States, we can compare the Thai fiscal system with those of some industrialized countries. Fiscal revenues as percentage of GDP in Thailand are similar to those of Vietnam and Mexico but much lower than in Hungary and the industrialized countries. These differences are due to both tax revenue and social contributions. By considering tax revenue, developing countries have a similar situation: indirect taxation is the main source of receipts, at an average of twothirds of the total, although Hungary is moving towards an imposition mainly based on direct taxes. Among direct taxes, individual taxation is much lower in Thailand than in industrialized countries: because of its structural weaknesses, Thai personal income tax is a marginal fiscal revenue, while it is the first source of receipts in Japan and the United States and the second in Germany. Indeed, the social contribution percentage of GDP indicates that the social security system in developing countries is still underdeveloped (Hungary excepted, where social contributions have a relevant weight). These two features lead to the low fiscal burden on labor in Thailand. Finally, the value of capital income taxation can be considered as one of the main causes of the considerable foreign investments which have come to Thailand during the last decades (together with the low cost of labor) and have promoted one of the fastest economic growths among developing countries, especially in East Asia.

Some quantitative and institutional features of the main taxes Direct taxes Personal income tax (PIT) Personal income tax (PIT) is payable by resident and non-resident individuals. An individual is resident if s/he stays in Thailand for more than 180 days during the calendar year. Tax is levied on income from Thai sources and, for income from foreign sources, only on the portion that is brought to Thailand. A non-resident is subject to tax only on income from sources inside Thailand. Taxable income (called “assessable income”) is divided into the following eight categories: wages and salaries; income from jobs, positions or services rendered; income from liberal professions; income from construction and other contracts of work; dividends, interest, shares of profits and capital gains; income from leasing property; income from goodwill,

262 Marco Bartolich copyright, franchise and other rights; other income from business, commerce, agriculture, industry, transport and other activities. The main exemptions are: insurance policies, gifts and bequests, workers’ compensation, insurance claims, medical benefits, proceeds from sales of immovable property acquired by bequest (these proceeds are included only if the sale is made for a commercial purpose). For income from employment, a 40 percent deduction (but not exceeding Baht60,000) is allowed. Other deductions are applied to the different kinds of income. The main tax allowances are: personal and spouse allowance (Baht30,000), child allowance (Baht15,000, a maximum of Baht45,000 for family), education, life insurance premiums, provident funds, estate, social insurance and charitable contributions. Interest income and dividends can be excluded from the assessable income: in the first case, the taxpayer can opt for withholding tax at a rate of 15 percent (certain forms of interest income are exempted from this tax, such as interest on bonds or debentures issued by a government organization). Dividends can be subject to a withholding tax at a rate of 10 percent; alternatively, the taxpayer can include them in her/his assessable income and claim for a tax credit, which is calculated according to the formula: dividends ⫻ corporate income tax rate ᎏᎏᎏᎏ 1 ⫺ corporate income tax rate After the deductions and allowances, income is subject to a progressive tax rate scale (the first Baht80,000 are exempted) (Table 11.3). For certain categories of income, the payer of the income has to withhold tax at source, file a tax return and submit the amount of tax withheld to the district revenue office. The tax withheld will then be credited against the tax liability of a taxpayer at the time of filing the PIT return.

Table 11.3 PIT rates in Thailand Taxable income (Baht)

Tax rate (%)

0–80,000 80,001–100,000 100,001–500,000 500,001–1,000,000 1,000,001–4,000,000 4,000,001 and over

0 5 10 20 30 37

Source: Thailand Ministry of Finance, Revenue Department.

Thailand 263 Corporate income tax (CIT) Corporate income tax (CIT) is levied on Thai and foreign companies. In the first case, a company is incorporated under the law of Thailand and its net worldwide profits are subject to tax at the end of each accounting period. In the second case, a company is considered foreign if it is incorporated under foreign law and only its net profits arising from a Thai source, or in consequence of business carried out in Thailand, are subject to CIT. A foreign company engaged in international transport is subject to tax on its gross receipts. Incomes received by foreign companies, such as service fees, interests, dividends, rents and professional fees, are subject to CIT on the gross amount received; the payer of the income pays a withholding tax which depends on the type of income and the status of the recipient. The withheld tax is credited against the final tax liability of the taxpayer. In the calculation of CIT, 50 percent of inter-corporate dividends are exempted; a full exemption is applied if the recipient company is listed in the Stock Exchange of Thailand, but only if the shares are held three months prior to and after the receipts of dividends. The main deductions concern ordinary and necessary expenses (R&D, job training and equipment for the disabled are allowed special deductions, 200, 150 and 200 percent, respectively), interest, taxes (except for CIT and Thai VAT), provident fund contributions, donations and depreciation (rates depend on the kind of assets). An initial depreciation of 40 percent is allowed for the purpose of encouraging investment. Net losses can be carried forward for five consecutive years. The ordinary tax rate is 30 percent, but for small companies, with paidup capital of less than Baht5 million, a progressive tax rate scale is applied (Table 11.4). For foreign companies many proportional tax rates are applied, which depend on the kind of business. Indirect taxes Value added tax (VAT) Valued added tax (VAT) was introduced in 1992 to replace business tax. It is charged on any person or entity that regularly supplies goods or services

Table 11.4 CIT rates in Thailand Net profits (Baht)

Tax rate (%)

0–1,000,000 1,000,001–3,000,000 more than 3,000,000

20 25 30

Source: Thailand Ministry of Finance, Revenue Department.

264 Marco Bartolich in Thailand and has an annual turnover exceeding Baht1.2 million. Service is considered to be provided in Thailand if it is performed in Thailand regardless of where it is utilized or, if it is performed in a foreign country, it is considered to be utilized in Thailand. VAT is imposed on the value added at each stage of production and distribution: the tax liability is the difference between the “output tax” (i.e. the tax collected when goods or services are supplied) and the “input tax” (the tax collected on any purchase of goods or services; it includes any tax charged on imported goods). The main exemptions are: unprocessed agricultural products and related goods; base services, such as transportation, healthcare, educational and professional services; renting of immovable properties; services in the nature of employment of labor; research and technical services and services of public entertainers; sales and import of newspapers, magazines and textbooks. Certain businesses are excluded and are subject to specific business tax. The tax base includes, for imported goods, the CIF price, import duties and any other taxes, if any, such as excise tax and fees; for exported goods, it includes the FOB price and the eventual taxes (excise tax and fees). Tax rates are: •

•

0 percent rate: export of goods, services rendered in Thailand and utilized in a foreign country, goods and services supplied under international programs and agreements; 7 percent: any other goods and services subject to VAT.

Initially the ordinary tax rate was 10 percent; subsequently it was reduced to 7 percent to reduce the fiscal burden during the 1997 financial crisis. Local governments receive 10 percent of the total VAT revenue; the VAT receipts are collected for the revenue department by the customs department and allocated to all levels of local government. Excise taxes Excise duties are collected by the excise department: collection is primarily by means of excise stamps. Tax is imposed on specific goods and services including liquors, tobacco, petroleum products, playing cards, non-alcoholic beverages, telephone and other services (entertainment places, golf, horse racing). In some cases, exemptions are allowed: for example, goods which are exported, have deteriorated or are damaged, those which are donated to charitable organizations and services which donate the receipts to public charities. Tax rates may be: ad valorem (entertainment and telephone services, motor vehicles), specific (some

Thailand 265 petroleum products, playing cards) or whichever is greater (spirits, tobacco, non-alcoholic beverages). Customs duties Customs duties are collected by the customs department, which collects excise tax for the excise department and VAT for the revenue department too. Tariff duties are levied on an ad valorem or specific rate base; in some cases, however, the tariff which gives the most revenue is applied. Tax rates change frequently: the majority of goods imported by businesses are subject to rates ranging from 0 to 80 percent on the CIF price. Specific duties are levied on certain commodities in the form of ad valorem (on the import duty) or specific tax rates. In order to promote exports, most exported goods are exempted: only a few items are subject to duty, including rice, rubber, raw hides and wood. On the other hand, exemption is granted for imported goods covered by privileges according to international agreements, laws or treaties, for re-exported goods that have been imported within the preceding two years and have not undergone a change in character or form and for crude oil, fertilizers, jewelry and munitions of war, while duties on inputs used in the production of exports are refunded. The current tariff structure is undergoing reform, which includes lower tariff barriers and a reduced number of tax rates, as shown in Table 11.5. Local taxes Shared taxes are the most important tax receipts for local governments in Thailand. They include VAT and selected excises. The surcharge is collected by the revenue department for VAT and by the excise department for selected excises and allocated to local governments. The surcharge is 10 percent of the total VAT and excise taxes. The two most important taxes levied by local governments are the land and building tax and the local development tax. The central government Table 11.5 Tax rates of the customs duties undergoing reform in Thailand Goods

Tax rate (%)

Raw materials Intermediate goods Finished goods Protected goods Input cannot be produced in Thailand

1 10 10 20 1

Source: IMF (2003).

266 Marco Bartolich determines the rate and the base of these taxes, while the local authorities collect them. The land and building tax is levied annually on buildings rented or used for other commercial purposes. Owner-occupied dwellings and buildings used by government, agencies, public hospitals, religious and educational institutions and buildings unoccupied for 12 months or longer are exempted. The tax rate is 12.5 percent of the annual rent received during the previous year. Inadequate databases, underreporting and evasion cause a low level of tax collection. Indeed the tax is disproportionately paid by low income households (unable to buy their own home) because often the owners of the rented dwellings pass on the tax to their tenants by increasing the rent or charging an annual payment in addition to the rent. There is a horizontal equity violation too, because owner-occupied property is not taxed, while property occupied by the owner’s immediate family is taxed. The local development tax is levied annually on the value of unimproved land not subject to the land and building tax. Land occupied by the owner or used for annual crops or owned by government agencies, public hospitals, schools, public utilities and religious organizations are exempted. A total of 34 different rates are applied, ranging from Baht 0.50 per rai with an assessed value of under Baht200, to Baht70 per rai with an assessed value of over Baht30,000 and Baht25 per rai for each additional Baht10,000 (one rai is equivalent to 0.16 hectare). This tax presents some equity problems. First, public enterprises and owner-occupiers receive many local services, but they do not pay for them because they are exempted from the local and development tax. Second, the tax is regressive because the rates decrease as the value of the property increases. Another important local tax is the real estate transfer tax. Transfer of real estate (by sale, gift or succession at death) is taxed on the basis of the assessed value of the property at the tax rate of 2 percent (0.5 percent if the transfer is made to parents, spouses or children). The same tax is applied by the central government (in this case some exemptions and deductions are allowed).

The distribution of taxation charge Analysis of the distribution of tax charge by economic function can be used in order to see how the main direct and indirect taxes are levied on productive factors and main employment of the economic system and, hence, to analyze the impact of taxation on resource allocation, because the tax burden has to be considered as part of the relative factor prices. Usually, study is focused on three main items: consumption, labor, and capital and business. The three indicators are obtained by using, as numerator, indirect taxes (general and specific), social contributions and

Thailand 267 direct taxes paid by employees, and taxes on capital (corporation income tax included), respectively, and, as denominator, GDP. In this way one can understand the weight of taxation on the three main economic components of national income as a percentage of the gross domestic product. During the last decade, consumption taxation shows an increasing trend until 1997, when the fall of tax receipts began because of the overall crisis. Indeed, consumption taxation registered a decrease due to the reduction of the VAT rate in 1999 (from 10 to 7 percent), one of the temporary measures taken in order to stimulate the economic recovery. Only in 2001 did the weight of consumption taxation start to rise again. Labor taxation represents just a small percentage of GDP because of two main factors: the structural weaknesses of personal income tax and the poor development of the social security system, which causes the low weight of social contributions on GDP (on average 0.35 percent). Capital and business taxation was strongly hit by the 1997 crisis, which exploded in the financial sector. Because of its high buoyancy with respect to GDP, corporate income tax fell from 1997 to 1998 by 43 percent, but taxation on financial and capital transactions had a steep decrease too (⫺46 percent). Another important indicator of the distribution of tax charge is the implicit tax rates. There are many definitions and methods of computation of these rates, which can lead to different values. As a rule, the three main indicators (implicit tax rates on consumption, labor, and capital and business) use, as numerator, the same values for the distribution of tax charge by economic function, and, as denominator, three economic aggregates taken from the national account statistics: private consumption, compensation of employees, and gross (or net) operating surplus.1 These rates can be considered as macroeconomic indicators of factor tax loads. The consumption tax burden is very high in comparison with that levied on labor, despite a gradual reduction from 1992 to 2002 (⫺2.1 percent) due exclusively to the decrease of tax receipts (private consumption has yearly increased in this period, 1997–8 biennium excepted). This result indicates that the Thai economy is still mainly based on indirect taxation (a common feature of developing countries). The implicit tax rate for labor was relatively higher during the crisis then before and after it. This result is due to PIT, which reaches a peak in 1998, while social contributions have a slightly increasing trend. Table 11.6 shows the two groups of indicators for the period 1992–2002.

20.2 5.8

Implicit tax rate Consumption Labor

20.3 5.4

11.1 1.9 3.7

1993

20.2 5.9

11.0 2.0 4.1

1994

Source: own elaboration on Thailand Ministry of Finance data.

10.7 2.0 3.5

Economic functions Consumption Labor Capital and business

1992

20.5 6.2

11.0 2.2 4.2

1995

20.8 7.1

11.2 2.5 4.0

1996

19.7 7.2

10.8 2.6 3.7

1997

18.6 7.2

10.2 2.8 2.2

1998

16.7 6.7

9.4 2.5 2.5

1999

Table 11.6 Structure and development of taxation by function and by implicit tax rates in Thailand, 1992–2002

16.6 6.2

9.3 2.3 3.2

2000

17.0 6.5

9.7 2.4 3.2

2001

18.1 6.7

10.2 2.5 3.4

2002

Thailand 269

Tax reforms A quick glance at the macroeconomic and budget outlook In 2002 the GDP growth rate of Thailand was 5.2 percent, the best performance after the 1997 crisis. Private consumption and investment provided the major contribution to this result: thanks to growing consumer confidence and credits, high increases of farm income, low interest rates and a supportive government policy, private consumption grew by 4.7 percent, especially in the sectors of automobiles, mobile phones, telecommunication services and televisions, while private investment rose by 13.3 percent, with a decreasing inflow of foreign direct investments, although their level was still high. Residential construction, purchase of commercial vehicles and office equipment registered the highest inflow of new investments. Exports grew by 5.4 percent in value and by 14 percent in volume, with China and ASEAN (Association of Southeast Asian Nations) countries as the main trade partners. Manufactured products (semi-conductors, radio and television receivers and parts, video recording, automobiles and parts, steel, plastic and chemical products) have been the primary source of the export volume growth, while natural rubber was the main agricultural export product. Inflation decreased from 1.7 percent in 2001 to 0.6 percent in 2002 and the unemployment rate continued to fall, from 3.2 to 2.4 percent, following the trend begun in 1998. Agriculture’s share in GDP was 9.9 percent, while manufacturing (including construction) and services (commerce, transport and communication, financial and other services) were 39.6 and 50.5 percent of GDP, respectively. The labor force by occupation presented the following composition: agriculture 41.6 percent, manufacturing 23.1 percent and services 35.3 percent. In 2002 Thailand’s fiscal system registered a general government overall balance in deficit for an amount equal to 6.6 percent of GDP, a value in line with the negative trend that began in 1997, the financial crisis year. This negative result was mainly due to the sharp increase of expenditures (⫹30.6 percent); the ratio between general revenues and GDP decreased because GDP grew more than total revenues. In comparison with 2001, the expenditure as a proportion of GDP has increased by 5.1 percent: while the main expense categories have slightly increased (salaries and wages, goods and services, interest payment, grants) or have decreased (social benefits), capital transfers to public enterprises and households have shown a strong increase, which has caused the 5.1 percent increment of the incidence of expenditure on GDP. Public debt was equal to 55.1 percent of GDP, the lowest value after the 1997 crisis; by comparison with 2001, it has decreased by 2.8 percent (as a proportion of GDP), and this reduction is mainly due to the decrease of non-financial public enterprise debt (⫺2.2 percent as fraction of GDP).

270 Marco Bartolich Thanks to increasing export growth and the relatively weak US dollar, Thailand registered a current account surplus, in line with the post-crisis trend, and the external debt stock continued to decrease, as a consequence of the government’s efforts to reduce its multilateral loans (World Bank, IMF, Asian Development Bank). Tax reforms of recent years, under way and planned In order to face the financial crisis after the Baht devaluation, Thai authorities implemented fiscal measures necessary to reduce the effects of the economic decrease and to promote the recovery. From July 1997 to February 1998 budget cuts and revenue measures have been taken to accommodate an expected revenue shortfall, to reduce the current account deficit and to produce a budget surplus of 1 percent of GDP. The FY98 budget was cut from Baht982 to 800 billion, while VAT rate was raised from 7 to 10 percent and selected excise taxes and import duties were increased, primarily on luxury consumption goods. In a second phase, fiscal measures were taken to reduce the social impact of the crisis and to stimulate the economy. Because of some difficulties and constraints on the expenditure side, the government began an expansionary policy on the revenue side. The VAT rate was changed from 10 to 7 percent and the 1.5 percent VAT on gross revenues for small enterprises with sales between Baht600,000 to Baht1,200,000 was eliminated. Personal income tax relief was provided to all taxpayers, but particularly low-income earners, by exempting from tax the first Baht50,000 of net income (Baht80,000 since 2003). Further measures to reduce the tax burden were introduced in August 1999 through selected excise tax and import tariff reductions and accelerated depreciation of fixed assets, targeted to provide investment incentives, and the postponement of corporate income tax payments and remittances from state-owned enterprises, in order to alleviate liquidity constraints in private and public sectors. Under the influence of the 1997 new constitution, promulgated after the financial crisis, in May 1999 the government launched the three-year public sector management reform program. Among the main objectives of this reform were an improvement of revenue collection, a more equitable distribution of the tax burden and an increasing fiscal decentralization. To achieve the first two objectives, the tools of the revenue department were: strengthening collection enforcement, increasing taxpayer compliance and improvement in information technology. During the crisis, tax debts increased strongly (from 0.7 percent of GDP in 1996 to 2 percent in 1999): as a consequence, in 1999 collectible debt was 9.3 percent of total tax collections, but thanks to the public sector reform

Thailand 271 loan program (a US$400 million loan from the IMF) this percentage was reduced to 6 percent in 2001. A debt collection management division has been created with the following functions: preparing a three-year debt reduction plan, setting collection targets for each office of the revenue department, developing new enforcement procedures and monitoring debt collection performance. Indeed, more resources have to be allocated by the revenue department for debt collection activities. Another consequence of the crisis was an increase in non-compliance, both in formal sectors, where registered taxpayers have many incentives to understate their income and over-claim expenses, and in informal sectors, where a weak business registration permits many street vendors to evade tax. The main actions (implemented or underway) to improve taxpayer compliance were: a tightening of the rules regarding VAT refunds, the development of automated systems for VAT refunds and for audit case selection, the reduction of taxpayer compliance costs, the simplification of excise tax and VAT forms, support to the large business tax administration office in managing the largest taxpayers, tighter control over marketplace and street vendors, a strengthening of the requirement for business registration and an increase of registered VAT taxpayers. Finally, the 1997 crisis evidenced the difficulties of the institutions in formulating and implementing tax policy. These difficulties were due especially to the lack of information, revenue data and other statistics that were necessary for the revenue department to formulate forecasting models and predictions of the impact of tax changes. Recently, the responsibility for determining tax policy has been transferred from the revenue, customs and excise departments to the fiscal policy office. The main tasks underway are the creation of a functional and integrated computer system for the revenue department, the implementation of online taxpayer registration and automated audit case selection systems, the development of modules for debt management and the creation of data entry, processing and reporting schedules for all taxes. Increasing fiscal decentralization was another main scope of the 1999 reform. The 1999 National Decentralization Act introduced changes in the local and central government roles and in their fiscal and administrative relations. First, it specified a gradual four-year devolution of administrative competences in order to define the expenditure functions of local and central governments and to avoid overlapping functions. Second, it aimed to increase local authority revenues to 20 percent of total government revenues by 2001 and to 35 percent by 2006. Some national tax bases have been shifted to local governments, such as mineral resource tax, land registration fees, gambling tax, underground water fees and bird net tax, but the receipts have registered only a small increase. With the introduction of two local shared taxes, VAT and excise taxes, local tax revenues have increased sharply in the period 2001–2. Indeed, local grants have increased in the same period (⫹75 percent in 2001 and ⫹170

272 Marco Bartolich percent in 2002 from 2000). Most of these grants were specific grants allocated through national government departments, in contrast with the principles of fiscal decentralization: in fact, local authorities do not have full discretion in using these funds. An objective of the reform was a redefinition of intergovernmental transfers, by changing from specific to general-purpose grants allocated through a more transparent system. Finally, reforms underway concern the promotion of responsible local borrowing (the regional urban development fund has recently been created in order to channel credit to projects of creditworthy local governments) and the enhancement of local accountability. Suggestions for further improvements Since 1997 the IMF has suggested some measures necessary to improve Thailand’s fiscal policy and tax administration. These measures particularly concern revenue collection and fiscal decentralization. An increase of tax receipts is indispensable to reduce the budget deficit, especially because a rising proportion of the budget will go towards mandated expenses, such as wages and salaries, payment of public debt and specific grants allocated to local governments. Some changes have been suggested for the VAT, excise taxes, personal income tax and corporate income tax. For the indirect taxes, VAT rate could be raised to 10 percent, while the excise taxes should be increased, especially on beer and tobacco, petroleum products and motor cars. Indeed, VAT refund represents a strong inefficiency in VAT collection. The monthly refunds require significant resources spent on this task and a proportional number of staff. In 1998 and 1999, 60 and 42 percent of audits, respectively, were for VAT refunds. Legal procedures and measures that reduce these administrative actions and the risk of fake invoicing should be taken: among the options, VAT refund rights could be limited or the tax administration could shift the VAT subject from the seller to the buyer, by forcing a partial or total withholding pre-payment in certain transactions where tax evaders are involved, or VAT pre-payment could be applied for sales of certain goods to risky retailers in order to shift the VAT subject and prevent tax evasion. Finally, a minimum amount for tax refund could be introduced. A general redefinition of allowances and deductions should be introduced in order to improve the horizontal equity of PIT and to simplify the withholding tax rate computation. Targeted exemptions, specific sectorbased allowances and the lack of automatic adjustments of brackets and monetary allowances necessary to avoid inflation effects are the cause of the low progressivity of the rate structure. Some deductions, such as that of 40 percent for income from employment against that of only 30 percent for income from professional income, appear illogical, while the many tax allowances should be simplified and redefined for appropriate-

Thailand 273 ness. Finally, the need for employees in many cases to file tax returns should be eliminated, in order to make the PIT a withholding system. The two major problems concerning corporations are the tax rate and the possibility for corporations with international business to avoid taxes on dividends. In the first case, the IMF suggests that the tax rates should not be too low in order to attract foreign investments, because there could be a net transfer of capital from poor countries to rich ones. For the second problem, it suggests a reinforcement of the controls over the corporations with international business, because they try very often to repatriate dividends without paying taxes. The mechanisms used are thin capitalization, overpayments of royalties and intangible assets, and over-invoicing of imports. Against them, a reduction of options for hidden equity capitalization, anti-avoidance rules and high penalties associated with a strong coordination between revenue and customs offices should be applied. In order to increase tax administration efficiency, some improvements have recently been introduced, such as the “large taxpayers office” and the computerization of the tax-paying system, but others are still necessary. The priorities concern a more effective coordination between the different departments of the Ministry of Finance and third-party obligations for providing information and collection of taxes. The other field where the IMF has made some suggestions is fiscal decentralization. According to the 1999 National Decentralization Act, in 2006 local authority revenues will rise up to 35 percent of the total revenues. In order to take advantage of all the benefits of fiscal decentralization, local governments have to increase their financial management, to improve the technical skills indispensable to decentralized functions, to strengthen accounting and financial reporting and to plan efficient programs for the use of the resources. Collaboration between local governments in providing some public services should be supported when a single small-size government cannot do it efficiently. Finally, an increasing cooperation between central and local governments is necessary to minimize and avoid duplication of functions.

Note 1 Because of the lack of detailed data, the computed indicators have to be considered as approximated measures of tax distribution by implicit tax rate. Indeed, due to the lack of data on the operating surplus, the implicit tax rate for capital and business could not be created.

Bibliography Asher, M. G. and Rajan, R. S. (1999) “Globalization and tax systems: implications for developing countries with particular reference to Southeast Asia,” 23 CIES, Centre for International Economic Studies.

274 Marco Bartolich International Monetary Fund (1998) Thailand: Statistical Appendix, Washington, DC: IMF. —— (2000a) Thailand: Selected Issues, Washington, DC: IMF. —— (2000b) Thailand: Statistical Appendix, Washington, DC: IMF. —— (2001a) Thailand: Selected Issues, Washington, DC: IMF. —— (2001b) Thailand: Statistical Appendix, Washington, DC: IMF. —— (2002) Thailand: Selected Issues and Statistical Appendix, Washington, DC: IMF. —— (2003) Thailand: Statistical Appendix, Washington, DC: IMF. —— (2004) Thailand: Selected Issues, Washington, DC: IMF. OECD (2001) Revenue statistics, Paris: OECD. —— (2003) Revenue Statistics 1965–2002, Paris: OECD. The World Bank Group (1999) Thailand Economic Monitor, October, Washington, DC: The World Bank. —— (2000a) Thailand Economic Monitor, December, Washington, DC: The World Bank. —— (2000b) Thailand Economic Monitor, February, Washington, DC: The World Bank. —— (2001) Thailand Economic Monitor, July, Washington, DC: The World Bank. —— (2004) Thailand Economic Monitor, April, Washington, DC: The World Bank. United Nations (2004) Human Development Report 2004.

Websites http://www.adb.org – Asian Development Bank. http://www.bot.or.th – Bank of Thailand. http://www.imf.org – International Monetary Fund. http://www2.mof.go.th – Ministry of Finance Thailand. http://www.nesdb.go.th/ – National Economic and Social Development Board (NESDB). http://www.nso.go.th – National Statistical Office Thailand. http://www.oecd.org – OECD. http://www.worldbank.org – The World Bank Group. http://www.un.org – United Nations.

Index

ability to pay xv, 21 Acemoglu, D. 101–2 Acharya, S. 114 aging population xxiii, 5–6, 97, 193–4, 198, 209–10, 213–14, 234, 236, 249 agriculture tax (in China) 147–8, 158–9 Aguirre, C. A. 41 Ahmad, E. 70, 159–60 allowances 5, 21–3, 116, 122, 147, 175, 182, 210, 215, 226, 232, 242, 254–5, 258, 262, 272 ASEAN Free Trade Area (AFTA) 4, 16 “ASEAN plus Three” cooperation 16 Asher, M. G. 20 Asian Development Bank 270 Asian financial crisis (1997) xvi–xviii, xxii, 6, 12, 20, 27, 92, 122, 137, 140, 231, 239, 248–9, 258–9, 271 Association of Southeast Asian Nations (ASEAN) 4–5, 16, 269; see also under ASEAN audits 51–4 Australia 44, 60 authoritarian regimes 99–105 autocracy 94, 106 Bangladesh 53 Barro, R. J. 101, 105 Baunsgaard, T. 18 Belgium 54 Boix, C. 99–100, 103–6 Brazil 58–9 Brooks, N. 39, 59–60 budget xix, xxii, 7, 9, 30, 50, 59, 60, 81, 84, 140, 158, 173, 180, 212, 220–1, 249, 255, 270, 272 budget constraints, hard and soft xxv, 84, 98

Cambodia 17 capital gains, tax treatment of 121, 150, 202, 215–16, 226–7, 240 Chalk, N. A. 122 Chelliah, R. J. 173, 185 Chicago school of political economy 103 Chidambaram, P. 59 China xvii–xxvi, 3–19, 28–31, 51–2, 58, 63–5, 92–4, 109–10, 113–14, 118–22, 126–30, 137–60, 248, 269; compared with India 65–7, 80–4; corporate taxation 148–50, 158; customs duties 143, 153–4; economic development and democratic transition 105–6; efficiency and equity of tax system 154–6; intergovernmental fiscal relations 68–72, 80–4; pension provision 96–8; personal income tax 138, 145–7, 160; property taxes 150–1; tax reform 156–60; value added tax 138–40, 144, 151–2, 158–60 Chun, Y. J. 251 civil society xxv, 93, 102 clustering of economies and tax regimes 9, 15 colonialism 101 competitiveness 9, 115, 128, 180, 219, 232, 239, 249, 256 computerization xiv, 35–8, 49–51, 54–8, 140, 273 consumption taxes 6, 18, 21, 26, 37, 154–5, 165, 181–2, 238, 245 controlled foreign corporation (CFC) rules 250–1 corporate taxation xxvi, 3, 6–7, 12–15, 25–8, 109, 118–29, 202–3; in China 148–50, 158; in Malaysia 225–6; in

276 Index corporate taxation continued Thailand 259; interaction with FDI 114–18 corruption 38–9 cost-benefit analysis 130 credit cards xix, 241 customs duties xxi, 3, 15–17, 21; in China 153–4; in India 178–9; in Thailand 265 Czech Republic 20 Dalsgaard, T. 211, 248 data on political characteristics 94 data on taxation xxi–xxii De Marco, D. V xvi De Mooji, R. A. 117 decentralization, fiscal 80–4, 254, 273 Dekke, R. 214 DeLong, J. B. 167 democratization xxv–xxvi, 93–5, 98; in China 105–6; economic determinants of 99–102; and public policy 103–5 Deng Xiaoping 66, 139 depreciation 120, 202 direct taxes 6, 12, 15, 169, 173, 185, 197–8, 220–1, 228, 236, 254, 256, 259, 261, 267 dividend payments 121 domino effects 9, 16 dual economies xx, xxiii earmarking of tax revenues xix, xxii–xxiii, 238 East Asia Free Trade Area proposal 16 education 101–2 efficiency-seeking investment 129–30 employment income 22, 146, 160, 200–1, 272 Enste, D. H. 41 equity and inequity xiv–xvii, xxi–xxvii, 4, 6, 14, 22, 26, 29–30, 43, 61, 239; horizontal and vertical 207–10; in Chinese tax system 154–6; in Japanese tax system 205 erosion of tax base 122 ethics xxi Everdeen, S. 117 exchange rate revaluation 9 excise duties 3, 7, 15, 29, 144, 165, 169, 173–4, 178, 180, 183, 227, 239, 259, 264 export promotion 112

fiscal deficits xvi, xviii, 157, 164, 212, 272 fiscal federalism xix–xx, xxiii–xxvii, 63–84 fiscal pressure 10–15 foreign companies and individuals, taxation of 26, 138–9, 145, 243 foreign direct investment (FDI) xxiii, xxvi–xxvii, 4–5, 12–16, 20, 109–14, 137–8, 251; and corporate taxation 114–18; types of 129 France 54 free trade agreements xxiii, 5, 16–20, 131 Freedom House political rights index 102 fringe benefits 59–61 Furutani, I. 208–9, 215 Gandhi, Rajiv 167 Germany 54 Giavazzi, F. 101, 106 gift tax (in Korea) 242 Gini coefficient 205–6 globalization xiv, 36, 40, 51, 74–5, 102, 128 gross domestic product (GDP): estimation of 44; ratio of taxes to xvi, xix–xx, 139 growth, economic xvi–xxii, xxvi–xxvii, 3, 12, 38–9, 61, 66, 100, 129, 167 Henley, J. S. 114 Highfield, R. 44 Hines, J. R. 19 Hong Kong 44, 52 horizontal equity 6, 26, 29, 192, 208, 239, 266 Hungary 20 implicit tax rates (ITRs) xxii, 14, 229, 245, 267–8 incentives for tax officials xxiv, 37, 44; see also tax incentives income distribution 100, 156, 205–7; see also redistributive effects of taxation income per capita 9–12, 105 income taxes xv–xvii, xx–xxiii, 3–6, 10, 14, 18–29, 45, 174–6, 192, 200–2; in China 138, 145–7; in India 174–6; in Malaysia 224–5; in South Korea 240–1; in Thailand 261–3 India xvi, xix–xxvi, 3–16, 27–30, 36, 44, 49–50, 54, 57–61, 63–5, 94, 109–10, 114, 118–26, 130, 164–88; compared

Index 277 with China 65–7, 80–4; critical issues for tax system 181–3; economic reforms 166; income tax 174–6; intergovernmental fiscal relationships 72–84; sales tax 180–1; service taxation 180; tax reform 164–6, 173, 183–8; value added tax 166, 183, 187; wealth tax 176–7 indirect taxation xvii, xx–xxiii, 3, 6–7, 12, 15–21, 26–9, 50; in China 138–9; in India 181–3 Indonesia 20 industrial policy 129–30 informal sector of the economy 19, 36–40, 271 informers, use of 52 inheritance tax (in Korea) 241–2 International Monetary Fund (IMF) xxii, 17, 139, 270–3 international taxation 51 investment tax allowances 128 Ishi, H. 203 Japan xv–xxvi, 3–25, 44, 63–5, 92–4, 109–12, 118–21, 127–9, 192–216; equity and efficiency in tax system 205–11; income tax 192, 200–2; tax reform 198, 212–16; typical features of tax system 196–7; value added tax 203–4 Jha, R. 29–30 Kawagoe, M. 211 Keen, M. 17–18 keiretsu 112 Kelkar, Vijay 166, 186–7 Keynesianism 212 Korea see South Korea labor supply elasticities 210 Laos 17 Lawrence, R. 112 Lee, C. 251 liberalization, economic and political xxvi, 4, 93, 101, 105–6, 114 Ligthart, J. E. 17 Lipset, S. M. 99, 101 local taxes (in Thailand) 265–6 MacArthur, Douglas xv Malaysia xvi, xxi, xxiii, xxvi, 3, 6–15, 25–7, 53, 63–5, 94, 109–13, 118–31, 218–32; compared with other developing countries 223–4;

corporate taxation 225–6; economic prospects 232; income taxes 224–5; tax reform 219, 229–32 market failure 115 Mittal, S. 29 modernization theory 99–101, 104–6 money laundering 59 moral hazard 98 Morinobu, S. 209–10 Mulligan, C. B. 103–6 multinational companies 51–4, 110–11, 114–17, 128, 131 Musgrave, R. A. 24–5, 27, 80, 173 Myanmar 17 neo-liberal economics 167 New Zealand 60 Oates, W. 80 Oman, C. 131 openness, economic 110–11, 143 optimal taxation structure 27 Organization for Economic Cooperation and Development (OECD): economic surveys 14, 248; standards and guidelines of xxvii, 6, 52–3; statistics xxi–xxii, 17, 207, 209, 249 Pan Euro Life 54 pension provision 5–6, 23, 92–3, 106, 209–10, 214, 221; in China 96–8, 106 personal income tax (PIT) xx, xxii, 3, 5, 6, 10, 12, 15, 21, 63, 69, 139, 143, 145, 155, 158, 160, 165, 181, 185–6, 192, 198, 208, 224, 229, 234, 254, 256, 259, 261, 267, 272 Peru 58 petroleum tax 25 Philippines, the 44 “pioneer company” status 126–8, 231 Poland 20 political economy: of democracies 93, 98 Polity dataset 94 population growth (in Korea) 236 PriceWaterhouseCoopers 51 principal-agent problems 98 “prisoner’s dilemma” situations 131 property taxes (in China) 150–1 Purfield, C. 74 “race to the bottom” 7, 29 Rajan, R. S. 20

278 Index Rao, M. G. 83 redistributive effects of taxation xvi, xxiii, 22, 24, 29, 99–100, 103–4, 139, 205–7 reform of tax systems: in China 156–60; in India 164–6, 173, 183–8; in Japan 198, 212–16; in Malaysia 219, 229–32; in South Korea 235, 238–9, 248–52; in Thailand 255, 269–73 religion 101 research and development (R&D) expenditure 128 “residence” approach to taxation 118 Robinson, J. A. 102 sales taxes xxiii, 7, 18, 26, 29–30, 180–1 Sanchez, I. 43 Schneider, F. 41 self-employment 22 Seneca xxi service tax (in India) 180 Shome, P. 41 Shoup, Carl S. xv–xvi, 197, 207 Simons, Henry xv Singapore 19–20, 44, 53 small enterprises, taxation of 204 Sobel, J. 43 social protection 92–3 social security contributions 6, 23, 141, 196, 214, 238, 245 South Asian Association for Regional Cooperation (SAARC) 4, 16 South Asian Free Trade Area (SAFTA) 16, 20 South Korea xvii–xxvi, 3–17, 22–4, 52, 63–5, 92–4, 109–12, 118, 121, 126–9, 234–52; Comprehensive Financial Income Taxation 251; fiscal burden 245–8; income tax 240–1; inheritance tax 241–2; tax reform 235, 238–9, 248–52; value added tax 238–9, 243–5 spillover effects 115, 129–30 structural economic changes xviii, xxii Tabellini, G. 101, 106 Tachibanaki, T. 211 Tajika, E. 208–9, 215 Tanzi, V. 20, 154–5 tariff reductions 4–5, 16–18 tax administration xxiv–xv, xxvii, 4–7, 15–16, 19, 27–8, 35–43, 252, 273; cost and design of 44–54; taxation policy in support of 58–61

tax allowances 21–2, 25, 28, 174–5, 226, 241, 272–3; see also investment tax allowances tax avoidance 38 tax burden 7, 21, 26, 28–9, 52, 114–17, 122, 129, 155, 187, 192, 197–8, 208–9, 214, 234–6, 248, 251–2, 266, 270 tax competition xxvi–xxvii, 4–5, 14–16, 19–20, 131 tax deduction at source 49–50, 240 tax elasticity of investment 117–18, 130 tax evasion xiv, xxvii, 35–45, 54, 58; effects of 43–4; estimation of 40–1; growth in scale of 41–2; resources required to deal with 45; sources and causes of 39–40 tax gap 40–1 tax holidays 115–16, 122–7, 149 tax incentives xviii–xix, xxii, xxv–xxvi, 5–7, 12, 19–20, 27–9, 115–31, 149, 210–11 tax policy issues xiv, xxiii, xxvii, 4–5, 14–20; and tax administration 58–61; for China 28–31; for India 27–30; for Japan 21–5; for Malaysia 25–7; for South Korea 22–4; for Thailand 25–7; linked to FDI 109, 129–31 tax returns 49–50, 273 tax wedge on labor xix, xxii–xxiii, xxvii, 3, 14, 109, 211 taxation change, distribution of 266–8 territorial principle of tax-ation 52, 118 Thailand xvi–xvii, xxi, xxiii, xxvi, 3, 6–17, 20, 25–7, 44, 63–5, 92–4, 109–10, 113, 118–31, 254–73; compared with selected countries 259–61; corporate taxation 259; customs duties 265; income tax 261–3; local taxes 265–6; public sector management reform program 255, 270–1; tax reform 255, 269–73; value added tax 263–4, 271–2 thin capitalization 250, 273 trade taxes 4, 16–19 transfer pricing 37, 48, 51–3, 250 transparency in tax administration 252 Transparency International 39 Tullock, G. 104 underground economy see informal sector unemployment rates xviii uniformity of tax systems 7–15 United Kingdom 44

Index 279 value added tax (VAT) xvii, xx–xxiii, 3, 6–7, 15–19, 23–31, 35–6, 41–3, 50–1, 57–8; double system of 183; in China 138–40, 144, 151–2, 158–60; in India 166, 183, 187; in Japan 203–4; in Korea 238–9, 243–5; in Thailand 263–4, 271–2 Venkatachaliah Commission 77–8 vertical equity 21–2, 192, 208 Vietnam 17

welfare spending xxv–xxvi, 92, 105–6, 140 withholding tax 121 Wittman, D. 103 World Bank 17–18, 39, 92, 160, 270 World Customs Organization 178 World Tax Organization proposal 14 World Trade Organization 42, 118, 137–8, 143 Xu, J. 97–8

Wagner’s law xxiii, 3, 10, 12 Wang, X. 98 wealth tax (in India) 176–7 welfare effects of taxation xxiii, 4, 7, 14, 17, 19

Zee, H. 17–18, 20, 154–5 Zhao, Y. 97–8