THE JAPANESE TAX SYSTEM
This page intentionally left blank
The Japanese Tax System Third Edition
HIROMITSU ISHI
OXFORD UNIVERSITY PRESS
This book has been printed digitally and produced in a standard specification in order to ensure its continuing availability
OXFORD UNIVERSITY PRESS
Great Clarendon Street, Oxford 0X2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Bangkok Buenos Aires Cape Town Chennai Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi Sao Paulo Shanghai Taipei Tokyo Toronto Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York © Hiromitsu Ishi 1989, 1993, 2001 The moral rights of the author have been asserted Database right Oxford University Press (maker) Reprinted 2004 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover And you must impose this same condition on any acquirer ISBN 0-19-924256-9
For the late Dr Carl S. Shoup
This page intentionally left blank
Foreword
One of the most difficult but rewarding undertakings in the field of tax theory and practice is that of tracing the development of a given country's tax system over a period of decades. Japan's tax system is particularly interesting for such a study, since it has changed notably through periods of war, post-war reconstruction, high-level economic development, and moderated economic growth. Those of us, including myself, whose inability to read Japanese shuts us off from a considerable literature that might inform us on this subject will be especially appreciative of this treatise written in English by Professor Hiromitsu Ishi of Hitotsubashi University. Professor Ishi's thorough and perceptive analysis of the development of Japan's tax system over time, coupled with his detailed description and critique of that system as it stands today provides the reader with an unusual opportunity to view a tax system as a virtually living, organic whole, in enough detail to allow comparisons of generally accepted tax theory and tax principles with the real world of an existing tax system. Professor Ishi brings special qualifications to the writing of this book. He has composed a volume on taxation that is part of a Postwar Fiscal History of Japan undertaken by the Ministry of Finance. In the course of this study Professor Ishi has had access to first-hand information, including confidential documents. For many years he has served as a member of the Tax Advisory Commission, and is now involved in drawing up plans for current tax reform. Through his activities in scholarly international tax organizations, including the International Institute of Public Finance, and through several research and teaching posts abroad, notably in Australia and the United States, Professor Ishi has developed a comprehensive outlook with which to appraise the tax system of his own country, as shown, for example, by the numerous comparative tables in this volume. This outlook assures that many of his appraisals and conclusions will be widely applicable. Among the significant developments in the Japanese tax system since the adoption in 1949-50 of the Tax Mission recommendations, none has been more important than the movement, first, away from equity, towards tax preferences designed to achieve certain economic goals and facilitate certain political compromises, followed by a turnaround, now gathering momentum in Japan as elsewhere, toward fairness among taxpayers similarly circumstanced, the economic goals to be achieved by other means. Professor Ishi's detailed account of these phases is among the most important of his contributions. January 1989 C A R L s. S H O U P
Preface to the First Edition Japan's image in the eyes of the world has changed greatly in recent years; Japan is now seen as a great power, comparable with the USA in terms of economic and competitive strength. The story of the economic miracle of the Japanese economy has attracted the world-wide attention of academics and non-academics alike. Reflecting this long-standing interest in Japan, a large literature on the Japanese economy already exists, much of which is now available in English. It seems to me, however, that very little material is available in English regarding the Japanese tax system and tax policy issues. In fact, no book has yet been written to explain the actual performance of tax policies from an academic point of view specifically for English readers. Europeans, Americans, and others interested in understanding the Japanese tax system must continue to have difficulty gaining access to reliable literature. The main aim of this book is to help fill this gap. There is a great need for a systematic description of how the Japanese tax system has developed during or since the postwar economic growth period, and how it has contributed to the superlative performance of the Japanese economy. In addition, it is important to assess Japan's present and future tax reforms because the most sweeping reforms of the Japanese tax system have been debated in the past several years. Many see a need to reform the tax system in order to adapt to ongoing changes in the economy. The tax debate will be with us for years to come. In this sense, it is an appropriate time to publish this book. Generally speaking, this book is non-technical and is written not only for academics but also for the broader tax community (bureaucrats, businessmen, tax accountants and lawyers, etc.). It should also be of great help to general readers who are not tax specialists but are interested in Japanese economic policy in general. Instead of using sophisticated theoretical and econometric models, the discussion in this book will be developed in terms of what general insights are to be derived or what general lessons are to be learned from this course of study, and will include simple statistical procedures. Many people and circumstances have helped me in this undertaking. I am fortunate to be a member of the government Tax Advisory Commission and the Postwar Financial History Project at the Ministry of Finance. These valuable experiences have had a major influence on my ideas about tax issues and have enabled me to have easier access to the resources needed to pursue my research. Members of staff of the Research Division, Tax Bureau at the Ministry of Finance read parts of earlier drafts and saved me from some major errors. My deep thanks should be expressed to Dr Carl S. Shoup for writing the foreword in this book. In addition, I have accumulated many debts during my career to overseas friends who have contributed to my intellectual development and research
Preface to the First Edition
ix
through discussions and the exchange of papers: the late Joseph A. Pechman, Henry Aaron, Oliver Oldman, Richard M. Bird, Mervyn King, and Sijbren Cnossen, among others, have greatly assisted me in the writing of such a book in English. Final publication of the book would not have been possible without the warmhearted support from Andrew Schuller at Oxford University Press, who first invited me to write it. I also owe a special debt to Gillian Bromley and Sue Hughes for their kind help in preparing the manuscript. The Nissan Institute of Japanese Studies at Oxford University very kindly provided me with office space a couple of times for my writing; I wish to express my gratitude to Director Arthur Stockwin, and Jenny Corbett. Finally, I am greatly indebted to Hiroko Sugimoto at Hitotsubashi University, without whose dedicated typing and other assistance the manuscript could not have been completed, to David Gross, who assisted tremendously in editing the final drafts in English, and to Sinji Yamashige, who provided assistance in computing the empirical analyses. I would like to thank all of them very much.
H.I. Tokyo, December 1988
Preface to the Second Edition Four years have passed since I wrote the first edition of this book. At that time, the Takeshita Cabinet was making a strenuous effort to get the Tax Bill approved, including the introduction of the consumption tax (Japan's VAT), in the Diet. It was during the height of the movement among the opposition parties against the consumption tax. In preparing the first edition, I was unfortunately unable to witness and review the whole process of the Takeshita tax reforms. Since that time, the Japanese economy has changed considerably, generating the 'bubble' phenomenon, and there were two important tax reforms at the beginning of the 1990s. I think we need to study these changes in taxation and in the Japanese economy. Two tax reforms in particular are important: (1) amendments to the consumption tax, and (2) land tax reform, both of which I think must be considered in detail to understand recent developments in the tax system. This is the main reason why I am motivated to revise the earlier edition of the book. Another reason is that almost all the data cited in the first edition are now out of date; a new edition has made it possible to make use of the latest data available to us. All the equations have been recalculated to include the more recent period and the data in figures and tables have been completely revised. In the present edition, three chapters—Chapters 4, 16, and 17—have been added, while one chapter has been dropped. At the same time, Part IV has been virtually rewritten to cover recent tax developments, while changes have been made in each chapter. Fortunately, I have been deeply involved in the deliberations on recent tax reforms as a member of the Tax Advisory Commission (in particular, as chairman of the Subcommittee on Land Taxation in 1990). I am greatly obligated to those persons who have helped me in revising this book. (A number of book reviews have come out in relevant journals and literature.) I should like to express my gratitude in particular to Richard M. Bird, Peter Jackson, Simon James, Sylvain Plasschaert, William Vickrey, and Richard Woodworth for their constructive comments and suggestions. I owe much to both Andrew Schuller for the publication of this book and Tracy Mawson for her kind help with the manuscript. For the opportunity to work on the revision, I wish to express my indebtedness to the Nissan Institute of Japanese Studies at Oxford University, and in particular to Director Arthur Stockwin and Jenny Corbett who provided me with pleasant surroundings for my work. I also wish specially to thank the Daiwa Anglo-Japan Fund for financial assistance. As I was at the time of writing the first edition, I am again greatly indebted to Hiroko Sugimoto for her endless aid in the production of the manuscript and
Preface to the Second Edition
xi
to Yoshihiro Kaneko for computational calculations. Others who have assisted in one way or another include the staffs at the Ministries of Finance and Home Affairs, and at the National Tax Administration in collecting data or computing empirical results. I would like to express my sincere gratitude to all of them. H.I. Oxford, October 1992
Preface to the Third Edition Since the second edition of this book was published in 1993, the Japanese economy has completely changed from a success story of good performance to the recent agonies of prolonged recession after the collapse of the bubble phenomenon in 1991. After nearly zero real growth rates continued for the three years 1992-94, the stagnant state of the economy continued for a long period with little sign of recovery, although there was a slight upward movement during the two years 1995—96. What was worse was that once again the economy experienced minus growth rates during the years of 1997-98, as a result of increasing the consumption tax rate, the Asian economic crises, and successive bankruptcies of big banks in 1997. A great deal of pessimism has spread all over the country. Many people, in and out, of Japan even believe that Japan's successful era is over and that it will never return to the success of previous days. To combat the uncertain state of the economy, the Japanese government has repeatedly acted to stimulate aggregate demand by using a Keynesian type of fiscal expansionary policy with increased public investment and tax reductions. But the government has also occasionally tried to increase tax revenues to make up for revenue losses due to successive tax cuts. As a consequence, the Japanese tax system has been affected by a round of tax changes in its macroeconomic policies. In particular, the individual and corporate income taxes, and the consumption tax (Japan's value-added tax) experienced big changes in the 1990s. The main aim of this revised edition is to clarify the process and the effect of another tax reform following the two previous tax reforms under the Nakasone and Takeshita Cabinets in the late 1980s which were treated in previous editions. For this purpose, it has been necessary to trace back the tax reform process institutionally, to update all the data used in the second edition, and to revise all figures and tables. In this edition, the contents have been substantially altered, particularly in the latter part. One chapter—Chapter 16—is added to discuss environmental taxes which have become important and is included in Part IV, an independent Part, different from the second edition. At the same time, Part III has been rewritten to include land tax reform (Chapter 13) and public finance moved to Chapter 18 in Part V. In contrast, Parts I and II basically remain unchanged apart from the updating of the relevant figures and tables. As was in the case with previous editions, I am greatly obliged to those who have assisted me in revising this book. My special thanks to Andrew Schuller at Oxford University Press for his kind support, to Hiroko Sugimoto at Hitotsubashi University for her warmhearted assistance, and Toshiya Hatano at Chiba University for computational calculations. Equally my deep thanks to the staffs of the Tax Bureaux at the Ministry of Finance and the Ministry of Home Affairs who always
Preface to the Third Edition
xiii
helped me in collecting necessary data and computing empirical results. To all of them I wish to express my warm thanks. While I was revising my work, I was informed that Dr Carl S. Shoup had passed away at the age of 97.1 have been indebted to his scholarship for a long time, and he was kind enough to write the Foreword for the first edition of this book in 1989. To Dr Shoup I dedicate this book as a token of gratitude.
H.I. Tokyo, April 2000
This page intentionally left blank
Contents
List of Figures
xvii
List of Tables
xix
List of Abbreviations
xxiii
Introduction
hxx I. Overview
1
1. General Characteristics of the Tax System 2. Background of the Japanese Tax System
3 16
3. Two Strategies of Postwar Tax Policy 4. Tax Administration and Tax Equity
36 53
II. Individual Income Tax 5. Basic Structures 6. The Erosion of Individual Income Tax
73 75 94
7. Inflation Adjustments 8. The Taxation of Investment Income and Savings
110 130
9. Effects of Taxation on the Distribution of Income
147
III. Corporate Taxation and Taxes on Capital
165
10. Principles of Corporate Taxation
167
11. Corporate Tax Levels and Tax Incentives 12. Inheritance and Gift Taxes
187 204
13. Land Tax Reform
219 IV. Indirect Tax System
14. The Traditional Framework of Indirect Taxes 15. The Value Added Tax 16. Design of Environmental Taxes
249 251 268 300
xvi
Contents V. Recent Tax Developments
317
17. Rebuilding the Tax System
319
18. Local Taxation and Intergovernmental Fiscal Relations
349
19. Appraisal and Further Reform
383
References
397
Name Index
409
Subject Index
411
List of Figures 1.1 1.2 1.3 2.1 3.1 3.2 4.1 4.2 4.3 4.4 4.5 5.1
5.2 5.3 5.4 5.5 6.1 6.2 A6.1 7.1 7.2 7.3 8.1 9.1 9.2 9.3 9.4 10.1 11.1 11.2 11.3 11.4 12.1
Trends in the relative shares of direct and indirect taxes, 1934-1998 Trends in tax collection costs, 1965-1998 The taxation process under Japanese and US-Western systems Changes in the level and composition of Japanese tax revenues, 1885-1995 Actual and anticipated rates of economic growth, 1955-2000 The gap between government expenditures and tax revenues, 1960-1998 Administrative cost as a percentage of tax revenue, selected countries, 1960-1995 Comparison between per-staff tax revenue (T/N) and administrative cost (ON) Relative share of withheld income tax, 1960-1998 Tax delinquency as a percentage of revenue, 1960-1996 Tax gap among three income sources, 1970-1990 Marginal rates of income tax (excluding local taxes) up to an income of ¥10 million for a family of four with a single wage-earner; selected countries, 1998 An exposition of Japan's income tax schedule Variations of progressivity of income tax, 1951-1997 Tax progressivity by income class: self-assessed income tax, 1970 Tax progressivity by income class: taxes on wage and salary incomes, 1970 Income tax erosion by income class, 1996 The ratio of tax erosion, selected years Steps in creating a comprehensive tax base, 1996 Changes in effective tax rates, actual and after inflation adjustment, 1960-1975 and 1975-1990 Tax thresholds for a working family of four, 1970-1992 Tax thresholds and inflation, 1970-1992 Movements of personal savings rates in six major countries, 1965-1995 The Lorenz curve, before and after taxes The movement of income distribution in terms of the Gini coefficient, 1951-1997 Redistributive effects of income tax, 1951-1997 Redistributive effects of self-assessed income taxes, by types of income, 1951-1997 Trends in self-financing by corporate firms, 1968-1998 Corporate tax revenue losses from special tax measures as a percentage of corporate tax revenue, 1965-1997 Effective corporate tax rates before and after adjustment for special tax measures, by size of corporation, 1973-1996 Nominal tax rates of corporate taxes, Japan and USA, 1970-1993 Average effective tax rates of corporate taxes, Japan and USA, 1970-1993 Computation of the Japanese tax base
10 11 13 19 49 50 55 59 61 64 68
83 85 87 89 90 102 104 108 123 128 129 137 150 151 152 155 183 189 196 199 203 207
xviii 12.2 13.1 13.2 13.3 13.4 13.5 14.1 14.2 15.1 16.1 16.2 16.3 17.1 17.2 17.3 17.4 17.5 17.6 18.1 18.2 18.3
List of Figures Computation of taxes due Long-term trends of land prices (nationwide, residential land), 1956-1998 Movement of land tax revenue and land price Long-term gains taxes of individuals for land sales after 1989 Corporate capital gains taxes for the sales of land after 1982 A schematic chart of property tax assessment Alcohol tax as a percentage of alcohol consumption and national tax revenues, selected fiscal years Total tax burdens of car-related taxes on the standard passenger car: international comparisons, 1997 Mechanism of the local consumption tax Alternatives of the environmental tax Tax rates, revenues, and earmarking of energy taxes, 1998 Patterns of household environment-related expenditures as a percentage of annual income, 1992 Percentage changes of non-entitlement expenditures over the preceding year, 1970-2000 Trends in the bond dependency ratio, 1970-2000 Package scheme of tax reductions and increases, 1994-1998 Trends of major national taxes, 1983-1999 Individual income tax structure among major countries, 1999 Effective corporate tax rates among major countries, 1999 Tax shares between central and local governments, fiscal 2000 Trends in the ratio of local taxes to total annual revenues at local level, 1955-1998 Relative shares of four major taxes, 1955-1998
208 220 242 244 245 246 256 263 298 306 307 312 322 323 340 342 343 346 353 355 358
List of Tables 1.1 Source of tax revenues in Japan and seven major countries, 1995 1.2 Tax levels in OECD countries: tax revenues as a percentage of GDP, selected years 1.3 Sources of tax revenues, by source, selected years 1.4 Tax revenue as a percentage of national income, by level of government, selected years 2.1 Correlation of per capita real income (yIN), openness (M/Y, (M+X)/Y), and agricultural products' share (Ag/Y) with tax shares (T/Y), 1885-1997 2.2 Regression equations for tax sources as a function of economic development variables, 1885-1944 and 1953-1997 A2.1 Shoup Mission recommendations and their modifications 3.1 Comparison of estimated revenue loss from special tax measures and total individual and corporate income tax revenue, 1958-1998 3.2 Percentage distribution of the estimated revenue loss from special tax measures, by type of incentive, 1958-1997 3.3 Estimated annual tax changes, fiscal years 1950-1998 3.4 Year-to-year income elasticity of the national tax system, 1965-1998 4.1 Tax collection methods by taxable income in 1996 4.2 Conceptual adjustment between TS-base and NIS-base incomes 5.1 Minimum taxable level of the individual income tax, selected countries, 1998 5.2 Different income tax liabilities among one-earner and two-earner couples with a combined income of ¥10 million, 1998 5.3 Income tax liability and income-splitting, L998 5.4 .Statutory rates of income taxes at all levels of government, 1999 5.5 Major changes of tax rates, national income tax, 1949-1999 5.6 Taxes on personal income as a percentage of GDP and at the income level of an average production worker (APW), selected countries 6.1 Comparison of actual and comprehensive income tax base and yield, 1996 6.2 Comparison between actual and comprehensive income tax rates, 1996 7.1 Effect of 10 per cent inflation on the tax liability of a family of four with an annual income of ¥5 million, 1975 7.2 Ten per cent inflation, statutory tax rate, and the deduction for employment income, 1975 7.3 Ten per cent inflationary effects on income tax liability for a family of four, selected income levels, 1975 7.4 Inflation and the self-assessed income tax liability, 1960-1990 7.5 Inflation and withheld income tax on wages and salaries, 1960-1990 7.6 Income taxes and fiscal dividend, 1960-1975 8.1 Personal saving with non-taxable interest income, 1987 8.2 Total principal outstanding of personal saving, by source, as of March 1986
5 7 9 10 21 23 33 39 41 43 47 60 67 76
78 79 81 82 93 100 101 112 113 114 118 120 126 131 132
xx 8.3 8.4 8.5 8.6 9.1 9.2 9.3 A9.1 A9.2 10.1 10.2 10.3 10.4 10.5 10.6 10.7 11.1 11.2 11.3 11.4 12.1 12.2 12.3 12.4 12.5 13.1 13.2 13.3 13.4 13.5 13.6 14.1 14.2 14.3
List of Tables Special tax measures for interest income: reduced tax rates in the case of separate taxation at source Proportion of families using the privileged saving system, by income class, 1986 Proportions of portfolio investments in the household sector, selected years Equations estimating personal savings, 1965-1998 Percentage distribution of taxpayer, taxable income, and tax yield under the self-assessed income tax, 1955-1995 Redistributive effects of taxation, by household, 1972 Redistributive effects of taxation, by household, 1984 Redistributive effects of the self-assessed income tax, 1951-1997 Redistributive effects of the withheld income tax on wage-salary incomes, 1951-1997 Numbers of corporations, by the scale of paid-in capital, 1996 Nominal tax rates among major advanced countries, 1998 Historical patterns of the corporate tax rate, 1950-1999 The ratio of tax credits Types of company taxation in OECD member countries, 1987 Adjustment of corporate and individual income taxes, 1974 Comparison of the total tax burden of corporation and non-corporation in 1961(before tax revision) Revenue losses of special tax measures of corporations, by type, 1977, 1987, and 1998 Effective corporate tax rates before and after adjusting for special tax measures, by size of corporation, 1973-1996 An international comparison of corporate tax rates, 1984 Taxable and economic incomes of non-financial corporations, Japan and USA, 1970-1993 Inheritance and gift tax rates, 1975-1987 and 1998 Trends of tax payments in the inheritance tax, 1958-1996 Inheritance tax revenues, 1950-1998 The inheritance tax burden, 1998 Redistributive effects of the inheritance tax, 1958-1996 Outline of land taxation, 1990 Effective tax burden of the land-holding tax, 1970-1995 The ratios of property tax assessment to official valuation of land price, selected years Outline of land transfer taxation, 1990 The property tax on agricultural land in Urbanization Promotion Areas, 1971-1991 The burden ratio of land value tax Percentage distribution of national indirect taxes, selected fiscal years Alcohol tax rates as a percentage of retail prices, 1987, 1992-2000 Tobacco tax burden on retail prices, 1987 and 1998
133 138 140 144 156 159 159 162 163 170 171 173 176 177 179 181 191
193 198 201 210 211 213 214 216 224 225 226 228 231 243 253 257 259
List of Tables Proportion of cigarette taxes in retail prices, selected countries, 1985, 1991, and 1997 14.5 Car-related taxes, 1998 14.6 Proportion of taxes on petrol in the retail price, selected countries, 1985-1986 14.7 Commodity tax revenues collected from the top ten major items, 1986 14.8 Rate structures of the commodity tax, 1986 14.9 The commodity tax as a percentage of retail price, major items, 1987 15.1 Past trends of VAT among OECD countries 15.2 History of tax reforms in Japan before 1989 15.3 Results of public poll for or against the introduction of the consumption tax by mass-media 15.4 Various tax rates applicable to different sales—the case of non-wholesale firms, 1989 15.5 Tax variations in the national and local tax system under the Takeshita tax reform 15.6 Trends of the consumer price index (CPI) in a nationwide area 15.7 Effects of VAT on prices of major commodities: results of the sixth price monitor 15.8 Extent of forward-shifting, May 1989 15.9 Extent of forward-shifting by firm size at the retail stage, fiscal 1989 15.10 Full-, over-, and under-shifting at the retail stage by business type, April 1989 15.11 Small and medium traders applying to the special simplified rule as a percentage of total taxpayers, 1989 15.12 Value added ratio by type of industry under the application of the special simplified scheme 15.13 Chronology of events after implementing the consumption tax 16.1 CO2 emissions by major countries, 1991 16.2 Tax burden per ton of CO2 emissions, 1992 16.3 Hypothetical types of carbon tax and carbon/energy tax, 1992 16.4 Implicit carbon taxes, 1988 17.1 Tax increases of national corporate income taxes, 1980-1986 17.2 Tax increases of national domestic indirect taxes, 1980-1984 17.3 Reform package of tax increases and reductions: tax bill proposed in July 1988 17.4 Revenue changes of income taxes, full-year basis 17.5 A modified flat rate structure of the income tax under the proposed tax reform plan, 1988 17.6 Basic rate on retentions % 17.7 Removal of reduced tax rates on dividends % 17.8 Changes of tax rates, national income tax, 1993-1999 17.9 Changes of major exemption and deductions, national income tax, 1993-1999 17.10 Changes of the national corporate tax rates, 1990-1999 18.1 Total annual revenues of local government, fiscal 1998
xxi
14.4
260 262 262 265 266 266 269 275 280 283 284 286 287 288 289 290 293 293 295 302 308 311 314 324 325 329 331 334 337 337 345 345 347 354
xxii 18.2 18.3 18.4 18.5 18.6 18.7 18.8 18.9
List of Tables Local tax collection, by source, 1999 International comparison of taxes on real estate, selected years Local transfer taxes, fiscal 1998 Fiscal equalization at the prefectural level per capita, fiscal 1998 Trends of the tax-sharing ratio, 1954-2000 Local allocation tax: receiving and non-receiving local governments, fiscal 1998 Specific-purpose grants from the national government, fiscal 1998 Proportion of support coming from special-purpose grants, selected items, 1984-1999
356 363 364 368 370 371 377 378
List of Abbreviations EPA
Economic Planning Agency
LDP
Liberal Democratic Party (the present ruling political party)
LPFP
Local Public Finance Programme
MITI
Ministry of International Trade and Industry
MOC
Ministry of Construction
MOF
Ministry of Finance
MOHA
Ministry of Home Affairs
MOHW Ministry of Health and Welfare NLA
National Land Agency
NTA
National Tax Administration
SCAP
Supreme Commander for the Allied Powers
TIN
Tax Identification Number
VAT
The value-added tax
This page intentionally left blank
Introduction
The Japanese tax system has undergone drastic changes during the postwar period. Immediately after the end of the Second World War, it was completely overhauled under the American influence, particularly by tax recommendations of the Shoup Mission in 1949. Thereafter, however, the process of building a Japanese-style tax system began by the government initiative in the 1950s and 1960s, when it was reorganized to promote the growth of the Japanese economy. Since the oil price shocks of the 1970s, there have been some further attempts to reform the tax system as a whole in view of the changed circumstances experienced by the Japanese economy. The Japanese tax system is characterized by several unique features in comparison with those of other major industralized countries. The most distinctive feature is that the tax burden is among the lower third in this group: in the ratio of total tax revenues of GDP, Japan comes third-to-last of the 24 OECD countries, the lowest ranking apart from Turkey and the USA. A second feature is the heavier reliance on direct taxes: nearly 70 per cent of total taxes is collected from individual and corporate income taxes. Third, there had been no broad-based consumption tax during the postwar period until 1989, and indirect taxes came solely from selective excise taxes. Japan was the only country that did not rely to some extent on revenues from the value added tax. A fourth feature is the centralized collection system of the tax structures under a unitary nation. More than 60 per cent of total taxes is collected by the national government, and as a consequence the workings of intergovernmental fiscal transfers become important. A principal aim of this book is to describe systematically these aspects of the developing Japanese tax system. Based on a great deal of empirical evidence, the main features of tax policy issues and reforms are clarified. Major issues are divided among four parts, each having three, four or five chapters. Part I provides an overview of Japan's tax development, dating back to the prewar period. Chapter 1 begins this review with a brief explanation of general characteristics that are thought to be of basic importance to an understanding of the Japanese tax system. In Chapter 2, the historical background is investigated in relation to an empirical analysis of tax structure development and the impact of the Shoup tax proposals. Chapter 3 sheds light on the strategies the Japanese government has adopted concerning tax policy during the postwar period. Chapter 4 was added in the second edition to attempt an in-depth analysis on major issues of tax administration in view of tax equity. Part II is devoted to an analysis of the fundamental framework and economic effects of the individual income tax. After a brief outline of basic structures is given in Chapter 5, four important aspects of policy issues relevant to the individual income tax are considered in Chapters 6-9 by using empirical analyses with statistical data: the phenomenon of tax erosion (Chapter 6), adjustment for inflation
xxvi
Introduction
(Chapter 7), the tax treatment of savings and investment (Chapter 8), and the effects on income distribution (Chapter 9). These studies constitute the primary part of the book. In Part III the main features of corporate income tax and capital taxes are explored. Basic principles of corporate taxation and its economic effects are dealt with in Chapters 10-11, focusing on differences between Japan and other major countries. Chapter 12 looks at unique aspects of Japan's inheritance and gift taxes with some statistical procedures. Chapter 13 is devoted to exploring the movement of both land issues and land tax reform, in which a new tax named the land value tax was created. Part IV considers several aspects of the indirect tax system. Chapter 14 treats past trends and the framework of the excise taxes which had important influences on past tax reforms. The process of introducing value added tax (VAT) under both the Nakasone and Takeshita administrations and its aftermath are carefully examined in Chapter 15, which covers the latest debates and modifications. Chapter 16 is newly added in this edition, and analyses the basic nature of environmental taxes with certain empirical results of a hypothetical carbon tax. Part V has been substantially rewritten for the third edition to clarify all issues relating to tax developments in the past several years. Chapter 17 begins with an explanation of the general background against which the tax system has been rebuilt with special reference to fiscal deficits and administrative reforms since the late 1970s. Problems of local taxation are described in detail with intergovernmental fiscal relations in Chapter 18. This chapter covers both sides of local government budgets in a broader scope. Finally, in Chapter 19 the book concludes with an appraisal of recent tax reforms, and of further reforms which might be worth pursuing in years to come. Observing the past trend of Japan's tax developments since 1993 when the second edition of this book was published, it can be seen that the Japanese tax system has come to a turning-point in its search for a better form. As a result of tax reforms undertaken in recent years, successive rounds of income tax cuts in relation to fiscal stimuli have substantially changed the tax structure. While reliance on direct taxes has been reduced, indirect taxes have begun to increase their relative share in total revenues. This trend might be continued with population ageing in the new century. Needless to say, there are substantial parts of the present tax system that do not work satisfactorily and could not be made to do so even with great difficulty. From a practical point of view, switching from direct to indirect taxes could represent the first of a series of steps towards a better tax system. In this respect, I am sure that this book will be of great help. In writing the first edition, I basically relied on information that was correct as of December 1988, when the Takeshita tax bill was hotly disputed between the LDP and opposition parties in the legislative process of the Diet. The second edition was written by using data available to me as of December 1992. For this edition I have used data as of March 1999.
Parti
Overview
This page intentionally left blank
1
General Characteristics of the Tax System The purpose of this chapter is to provide a brief overview of the tax system in Japan. In general, a tax system presents certain significant features which have a close bearing on the workings of the economy in which it operates. What, then, are the general characteristics of the Japanese system? To answer this question, three important areas must be covered before beginning to consider major issues. First, a brief description is given of the evolution of the prewar and postwar tax systems. Second, three main features in tax levels and structure are explored. Lastly, several notable aspects of tax process and administration are examined as a preliminary step for further analysis.
HISTORICAL BACKGROUND
Modernizing the Tax System The origin of a modern tax system in Japan can be dated to 1887, when the national government instituted an income tax. Although this took a truly modern form only in 1940, Japan is thus counted as a pioneer in the use of the income tax. However, since the Japanese economy was at that time still underdeveloped, the income tax played only a minor role in total national tax revenues (say, 1.5 per cent in 1888). Before the income tax became predominant in the tax system as a whole, the main revenues for the national government were raised first from a land tax, and then from indirect taxes. In fact, the land tax accounted for the largest share of national taxes until 1908, after which time the primary source was revenues from indirect taxes, mainly on alcoholic beverages and tobacco. Only after 1935 did income tax on individuals and corporations became the most important single source of total revenues. In short, before the Second World War the Japanese government relied mainly on indirect taxes, deriving more than two-thirds of its total revenue from them. In 1940 an overall tax reform was carried out to prepare for the wartime economy. The whole tax system was thoroughly overhauled, resulting in the modern tax system, based mainly on direct taxes. Although individual and corporate incomes had been taxed together by a single form of income tax, in 1940 separate taxes were imposed for each type of income. Since then, individual and corporate income taxes have coexisted in the tax system. The individual income tax was a schedular tax, under which different sources of income were levied by different tax rates. It was supplemented by a progressive
4
Overview
comprehensive tax which applied to an individual's aggregated income above a specific amount. On the other hand, the corporate income tax was imposed on corporate income at a flat rate of 18 per cent. Moreover, the commodity tax was introduced in 1937 mainly to collect revenue for wartime expenses, and the tax on alcoholic beverages was also simplified in 1940. The relative share of indirect taxes, however, began to decline as a result of the 1940 tax reform. Evolution of the Postwar Tax System The process of developing the tax system in postwar Japan was initiated by the USA. In 1947, several important reforms were undertaken under the influence of the US occupation authorities. The schedular tax on individual income was replaced by a unified tax on an aggregate basis with graduated tax rates. Furthermore, a turnover tax, which was levied on the basis of the sales amount at every stage of transaction at the rate of 1 per cent, was enacted in 1948 to collect necessary revenues. In 1949 a tax mission headed by Carl S. Shoup came to Japan with the task of reorganizing the tax system as a whole. The Shoup Mission recommended a tax plan intended to achieve a complete overhaul of the Japanese system. Essentially, the Shoup recommendations placed more importance on direct taxes, mainly income taxes on individuals and corporations. Thus, the entire tax system was fully reconstructed producing epoch-making change. However, the ideal tax system achieved by the initiative of US influences was of temporary duration: many of the taxes were abolished or modified soon after their enactment. Hence it is necessary briefly to describe the post-Shoup evolution of four major taxes.1 First, and most importantly, comprehensive income taxation has been replaced by a combination of a comprehensive tax and a schedular tax. This hybrid system was produced as a result of modifying the global income tax approach of the Shoup proposals. For example, instead of aggregating most incomes with progressive tax rates, some incomes (e.g. capital gains or interest income) are not now subject to global income taxation but are taxed at reduced flat rates, separate from other incomes. This special treatment is due to a number of tax concessions intended to stimulate saving and investment and to improve the welfare level among specific taxpayers. Second, the corporate income tax was a split-rate system until March 1990, but it has now become a uniform one in which a single rate is imposed on a whole corporate income. The old system was quite similar to that used in West Germany, in which retained profits and dividends are taxed at different rates. Also, numerous special tax measures have made the corporate income tax extraordinarily complicated. Third, an accession tax on transfer of wealth at death proposed by the Shoup Mission was replaced by a combination of inheritance and gift taxes. Fourth, in contrast to the general trend of modifying direct taxes, the whole system of indirect 1 For a general discussion, see e.g. Pechman and Kaizuka (1976), MO1; Tax Bureau (199)), and Aoki (1986).
Table 1.1
Source of tax revenues in Japan and seven major countries, 1995 (%) Japan
Income Individuals Corporate Social security contributions" Goods & services General Specific5 Death and gift Property Other Total
USA
UK
Canada
France
Italy
Sweden
36.6 21.4 15.2
45.8 36.3
36.9 27.4
45.9C 37.3
17.6 13.9
30.1 27.3
35.1 26.2
41.4 35.3
9.4
9.5
8.1
3.7
2.8
8.7
6.1
36.3 15.1
25.1 17.9
5.2 9.9 1.9 9.7 0.4
8.0 9.9 1.0
17.7 34.7 19.0 15.7
16.8 25.5 15.2 10.3
45.7 27.3 17.4
32.0 27.3 13.9 13.4
31.3 24.3 15.1
0.0 10.5
0.0
0.6 9.9 0.2
1.3
9.9 0.8 4.4 4.2
39.4 27.8 17.3 10.5
0.3 2.5 —
0.2 5.5 —
9.2 0.2 2.7 0.1
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
10.2
* Include taxes on payroll, if any. Include taxes on specific goods and services and those on use of goods and performance of activities. 0 Unallocable items are omitted. Notes: Tax revenues include social security contributions, as well as national and local taxes. Source: Calculated from OECD, Revenue Statistics ofOECD Member Countries, 1965-1996.
b
Germany
6
Overview
taxes has remained unchanged for a long time since the time of the Shoup recommendations. There had been no general consumption tax in Japan until April 1989, and indirect taxation depended mainly on selective excise taxes. As a result of the evolving postwar tax system, what significant characteristics can we observe from the present tax system of Japan as compared with those of other major countries? As seen in Table 1.1, the Japanese system has several features different from those of other countries.2 Japan depends more on revenues from taxes on income, and places greater emphasis on the corporate income tax, which is the highest among OECD countries. Social security contributions (equivalent to the payroll tax) play an equally important role in raising revenues, in spite of heavier reliance on income taxes. The relative share of social security contributions has increased very sharply during the past decade or so. Among major advanced countries, Japan is the only one that did not impose a general consumption tax until 1989, and so its relative share was at the lowest level in 1995 even after the adoption of VAT. Thus, taxes on goods and services derive from specific excise taxes (e.g. alcohol, tobacco, or petrol taxes). Minor sources of revenue are obtained from death and gift taxes, while the property tax, a main source of local government revenue, occupies a relatively higher share in the totals. In 1995, of total tax revenues collected in Japan, 36.6 per cent came from individual and corporate income taxes, 36.3 per cent from social security contributions, 15.1 per cent from taxes on goods and services, 9.7 per cent from the property tax, and 1.9 per cent from inheritance and gift taxes. The USA raised total tax revenues in relatively similar proportions from each source. Japan's tax structure and that of the USA, with their heavier reliance on income taxes and social security contributions, should be compared with those of many European countries where a much larger share of revenues is obtained from taxes on goods and services.
MAIN FEATURES OF TAX LEVEL AND STRUCTURE
Low Tax Burden It is difficult to judge whether taxpayers in any one country are burdened to a greater or lesser degree than those in other countries. Since many taxpayers believe that they are personally overburdened and overtaxed, we need some objective yardstick. Perhaps the best tool is an international comparison, which is frequently used by economists. The relevant question is, Is the Japanese tax system required to raise more revenue than are tax systems in similar countries overseas? 2 OECD (1987, 62) classified OECD member countries into three categories geographically, by using the percentage distribution in question: those in (1) northern and central Europe, which rely on general consumption taxes; (2) OECD non-Europe, which rely on income (individual and corporate) taxes; and (3) Mediterranean Europe, which rely on social security contributions and consumption taxes. In accordance with this classification, Japan and the USA belong to category (2).
General Characteristics of the Tax System
1
One measure of the tax burden is provided by tax levels i.e. the ratio of tax revenues (including social security contributions) to GDP. Table 1.2 lists this ratio for 24 OECD countries for the years 1970, 1980, 1990, and 1995. Although the level and timing of increases change from country to country, the highest tax levels are to be found, as expected, in the Scandinavian countries, followed by other European countries, such as Belgium, France, and Luxemburg. In this comparison, Japan came third to last of 24 countries in 1970. In fact, apart from Spain and Turkey, Japan was ranked lowest among the major advanced countries. The same holds for 1980. However, this ratio in Japan rose, exceeding that of the USA and Australia in 1990, but it was still ranked among the lower group around the 30 per cent level. In 1995, it had declined again to the 20 per cent level, reflecting successive tax reductions, although its rank remains the same.
Table 1.2 Tax levels in OECD countries: tax revenues" as a percentage of GDP, selected years
Denmark Sweden Belgium Finland France Luxemburg Netherlands Austria Norway Greece Italy Germany New Zealand Canada UK Spain Switzerland Ireland Portugal Iceland Australia Japan USA Turkey
1995
1990
1980
1970
51.3 49.7 46.5 46.5 44.5 44.0 44.0 42.4 41.5 41.4 41.3 39.2 38.2 37.2 35.3 34.0 33.9 33.8 33.8 31.2 30.9 28.5 27.9 22.5
48.7 55.6 44.4 45.4 43.7 43.4 44.6 41.0 41.8 36.5 39.2 36.7 38.1 36.5 36.4 34.4 31.5 34.8 31.0 31.4 30.8 31.3 26.7 20.0
45.5 48.8 44.4 36.9 41.7 42.0 45.2 40.3 42.7 29.4 30.4 38.2 33.0 31.6 35.3 24.1 30.8 33.8 25.2 29.2 28.4 25.4 26.9 17.9
40.4 39.8 35.7 32.5 35.1 28.0 37.1 35.7 39.3 25.3 26.1 32.9 27.4 31.3 36.9 16.9 23.8 31.0 20.3 27.0 24.2 19.7 27.4 12.5
" Social security contributions are included. Source: As Table 1.1.
8
Overview
As is evident from such an international comparison, the Japanese tax system imposes a lower burden on taxpayers, as does that of the USA, among major industrialized countries.3 Perhaps this is one of its most salient features, which deserves wider attention.
Heavier Reliance on Direct Taxes Tax structure is another important element of the tax system of a country. Past trends in the variation of tax structure are seen in Table 1.3, in which we compare the percentage distribution of tax revenues (excluding social security contributions4) by source at both national and local government levels. The dominance of income taxes since 1950 was maintained and even accelerated until fiscal 1990. In 1990 approximately 70.7 per cent of total national taxes was collected from individual and corporate income taxes, and 63.7 per cent of local taxes came from the same sources. In contrast, the relative importance of taxes on consumption has continued to decline for the past decade or so. Taxes on wealth, which are used more prevalently by the local governments, increased their relative share until 1990. However, substantial changes have recently taken place in both national and local taxes, switching from income taxes to taxes on consumption. The Japanese often use the ratio between direct and indirect taxes3 as a crude measure of investigating a change in tax structure (simply referred to as the 'direct-indirect taxes ratio') in Japan. In Figure 1.1 this ratio is depicted for national and local governments, respectively for selected years over a 60-year span. Obviously, the relative weight of direct taxes moved upward until 1990. In particular, the shift from indirect to direct national taxes in the 1980s progressed markedly since the prewar period. It was widely acknowledged in Japan that the reliance on direct taxes was excessive. By definition, taxpayers felt the burden of direct taxes more strongly than other taxes and were inclined to avoid and evade them. Consequently, many Japanese taxpayers had complaints about the fairness and reliability of the tax system. Such an excessive reliance on direct taxes was partly responsible for the tax reforms, and as a result the relative share in direct taxes began to fall in the 1990s.
3
Kay and King (1986, 224) also attempted to classify selected OECD countries into three group: (1) Holland and the Scandinavian countries, (2) other Western countries, and (3) a somewhat heterogeneous group including Japan and the USA. 4 Revenues from social security contributions are generally not included when studying tax policy issues in japan. In fact, tax data do not include social security contributions, mainly for two reasons: (1) social security contributions are not treated as part of the general account of the national government, and (2) they are administered mainly by the Ministry of Health and Welfare, not the Ministry of Finance. Thus, we are likely to neglect them in analysing tax issues because of the lack of combined data. 5 At the national level, direct taxes are composed of individual and corporate income taxes and inheritance and gift taxes, and the remainder are all classified as indirect taxes including miscellaneous ones The same holds true for local taxation.
Table 1.3 Sources of tax revenues, by source, selected years Fiscal year
National taxes Taxes on income Individual Corporate Taxes on consumption Taxes on wealth Total Local taxes Taxes on income Individual Corporate Taxes on consumption Taxes on wealth Total a
1985
1990
1995
2000a
69.5 38.1 31.5 25.2 5.3
70.1 39.4 30.7 21.9 8.0
70.7 41.4 29.3 22.0 7.3
60.5 35.5 25.0 29.4 10.1
56.5 36.9 19.6 36.9 6.6
100.0
100.0
100.0
100.0
100.0
100.0
55.6 20.1 35.5 24.5 19.9
54.8 26.3 28.5 20.0 25.1
57.1 27.6 29.6 19.2 23.7
58.1 28.7 29.4 17.1 24.7
63.7 31.1 32.6 12.3 24.0
52.7 30.5 22.2 14.7 32.6
45.9 28.5 17.4 21.7 32.4
100.0
100.0
100.0
100.0
100.0
100.0
100.0
1950
1960
1970
1975
54.4 38.8 15.6 43.4 2.2
53.6 21.8 31.9 42.2 4.2
64.3 31.2 33.0 30.9 4.9
67.2 37.8 29.4 26.8 6.0
100.0
100.0
100.0
44.9 38.9 6.0 18.1 36.9
49.6 15.4 34.1 23.3 27.1
100.0
100.0
1980
Preliminary figures. Note: The classification of taxes in this table is rearranged according to the OECD criterion, but social security contributions are excluded. Source: Tax Advisory Commission, Tax Report on a Sweeping Tax Reform (in Japanese), Oct. 1986, pp. 18-19; and MOP, Primary Statistics of Taxation (Zeisei Shuyo Sanko Shiryoshu), Feb. 2000.
10
Overview
Fig. 1.1 Trends in the relative shares of direct and indirect taxes, 1934-1998 Source: As Table 1.3.
Centralized Tax Collections Attention should also be paid to the relative proportion of national and local taxes. Japan is centralized to a degree comparable to that of the UK, and local governments are given only limited fiscal responsibility and powers over their fiscal activities. Under a centralized fiscal system, unitary taxation is generally prevalent throughout the country (see Chapter 18). This is quite different from the decentralized systems of federal nations such as the USA, Canada, and Australia. Table 1.4 summarizes the relative share of total tax revenues of the national and local governments for selected years since 1950. In 1998 the total tax ratio (not including social security contributions) amounted to 24.5 per cent, of which 14.5 and 9.6 per cent were allocated respectively to the national and local governments. More than 60 per cent of total taxes is collected by the national government, but its relative share has steadily declined for the last three decades. This ratio of tax share is of course not applicable to the allocation of public expenditures to different levels of government, in accordance with the workings of intergovernmental fiscal transfers. It is also interesting to compare collection costs (i.e. administrative costs in the public sector) at two different levels of the government (see Chapter 4 for an international comparison). As seen in Figure 1.2, collection costs of national taxes constantly moved at a much lower level than those of local taxes during the period 1965-98, although two lines show the declining trends over the long run until fiscal 1990. It is thus conjectured that tax administration at the national level has been performed more efficiently by enlarging automation and computerized procedures with an unchanged number of staff. The problem lies in the costs of collecting local taxes by prefectural and municipal governments. The collection of local taxes costs over twice as much as that of
General Characteristics of the Tax System
11
Table 1.4 Tax revenues as a percentage of national income, by level of government, selected years Fiscal year
1950 1955 1960 1965 1970 1975 1980 1983 1984 1985 1990 1995 1998
Level of gover nment National
Local
16.9(75.1) 12.8(71.1) 13.6 (70.8) 12.4 (67.9) 12.7(67.5) 11.7(64.0) 14.2 (64.1) 15.0 (63.3) 15.3(63.1) 15.1 (62.7) 18.2 (65.2) 14.5 (62.0) 14.9 (60.7)
5.6 (24.9) 5.2 (28.9) 5.6 (29.2) 5.9 (32.1) 6.1 (32.5) 6.6 (36.0) 8.0 (35.9) 8.7 (36.7) 9.0 (36.9) 9.0 (37.3) 9.7 (34.8) 8.9 (38.0) 9.6 (39.3)
Total
22.5(100.0) 18.0(100.0) 19.2 (100.0) 18.3 (100.0) 18.8(100.0) 18.3 (100.0) 22.2 (100.0) 23.7(100.0) 24.3 (100.0) 24.1 (100.0) 27.9(100.0) 23.3 (100.0) 24.5 (100.0)
Notes: Social security contributions are excluded. Figures in parentheses are percentage distributions. Source: As Table 1.3.
Fig. 1.2 Trends in tax collection costs, 1965-1998 Note: Costs are calculated per ¥100. Source: National Tax Administration, Annual Reports (Kokuzeicho Tokei Nenpo).
national taxes. Why are costs so much higher in the collection of local taxes? One reason often mentioned is that tax offices of local governments perform less efficiently; tax revenues cannot be collected in proportion to the larger numbers of tax officials. Given this situation, it is often proposed that the two administrative
12
Overview
mechanisms be merged to improve the efficiency level of tax administration and to lower total costs of collecting taxes as a whole. In summary, the most salient features of the Japanese tax system are that (1) the tax burden is relatively lower among major industrialized countries; (2) heavier reliance is placed on direct taxes; and (3) tax collection is centralized.
TAX PROCESS The Formulation of Tax Policy The Japanese have 'generated a unique system to create a general consensus on tax policy. For convenience, we distinguish the Japanese type from the US-Western type in formulating tax policy. In Figure 1.3, the two types are drawn as upper and lower pathways. Each process starts from a government inquiry which is initiated when the government admits the necessity for promoting any tax policy or reform. In the US-Western type, a task force or committee is usually established by the president, prime minister, or other authorities; several tax experts are appointed as members, and they are headed by an influential chairman.6 The task force or commission of inquiry generally undertakes a review of the whole tax system for five or six years, and then publishes a report to recommend a plan of structural tax reform for specific goals of tax policy. This report is virtually independent of political powers, at least while it is in preparation. In this sense, it can be considered a public report of an independent task force or commission. After the report is published, the government begins the legislative process in the congress or parliament. It should be emphasized, however, that there is a low degree of implementation of inquiry recommendations by the promoting government (see Kay 1987,67-9). Needless to say, the proposals contained within the report tend to define both gainers and losers, which causes a great deal of controversy by a variety of vested interests. In particular, it is very difficult to win public support for potentially unpopular policies, including tax increases. In general, governments lose their inclination to tackle the political problems of structural reform, and in turn the initial objective is likely to be frustrated. Therefore, under a tax formulation of the US-Western type, the possibility of implementing the tax reports is much poorer than might be expected. In contrast, the Japanese government uses the tax commission on a more formal basis; that is, as a Tax Advisory Commission of the prime minister. After the prime minister submits the inquiry to the Commission, it starts the discussion with a review, public hearings, and analyses. In the case of long-range tax reform proposals, after an intensive study for a relatively shorter period, say less than three years, the 6 Typical examples include the Carter Commission in Canada in 1962, the Ashrey Committee in Australia in 1972, the Irish Commission in 1980, or the US Treasury case in 1984 (see Kay 1987). The Shoup Mission can be regarded as an example of this category.
General Characteristics of the Tax System
13
Fig. 1.3 The taxation process under Japanese and US-Western systems * Liberal Democratic Party (in Japan).
Commission presents its recommendation to the prime minister, not to the general public. What is of primary importance in the Japanese method is that tax recommendations differ from those undertaken in the US—Western type. As will be explained shortly, the Tax Advisory Commission contains representatives from many vested interests. Despite serious conflicts between each group, reconciliation is conventionally made before the Commission recommends a tax programme, since political pressure groups can resolve differences. The Commission does encounter considerable difficulty and resistance in making reconciliations, and as a result a great number of political compromises are involved in the final tax proposals. This makes the contents of proposals difficult to understand readily. The greatest benefit from the Japanese method is that almost all inquiry recommendations can be relatively easily approved by the Diet, chiefly because political problems have already been resolved by the major interests. In short, the extent of implementation is much higher, in spite of many defects in the theoretical level of tax proposals. This process, however, has changed with the intervention of the tax committee of the ruling party, the Liberal Democratic Party (LDP), as is shown in Figure 1.3. The LDP has its own tax committee, which makes tax recommendations every year, and it has become much more influential politically. The Tax Advisory Commission's recommendations, which were adopted virtually intact two decades ago, have often been altered by the different views expressed within the LDP. Thus, the Tax Advisory Commission is inclined to exclude engineering features of tax proposals (e.g. percentage of tax rates or the level of deductions) in order to avoid any possible conflicts with the LDP Tax Committee. Usually, its recommendations are merely made as a fundamental argument for tax reforms without supplying any concrete figures necessary to construct tax structures. The detailed framework of tax changes is politically decided by the LDP, not the Tax Advisory Commission. A typical example will be seen in the case of the land holding tax (see Chapter 13).
14
Overview The Role of the Tax Advisory Commission
The Tax Advisory Commission has long played a central role in assisting the formulation of tax policy and reform, although it is often criticized that its role has tended to be less important. It is impossible to understand the Japanese tax process without a detail knowledge of the Commission itself. The Tax Advisory Commission was established by the government in 19537 to review the whole tax system and to formulate annual tax changes as well as long-run tax policy. One of Japan's most prestigious committees, it consists of 30 regular members and about the same number of supporting members, all of whom are appointed by the prime minister. Usually, tenure for members is for three years. Members of the Commission are selected from a variety of people, including academics, tax experts, tax lawyers, journalists, former government officials of each ministry, representatives of big and small businesses, labour union leaders, and so on. Because of this diverse membership, the Commission can arbitrate conflicting issues among the vested interests who greatly affect tax policy. The Commission's major objectives include the formulation of both short-run and long-range tax proposals. Annual tax revisions are prepared from a short-run standpoint by the Commission when the budget is compiled. Likewise, the Commission is responsible for the formulation of long-run policy on the structure of the tax system. Both types of tax proposals have had a substantial impact on general attitudes towards changes in the tax structure. High-level staff of the tax bureaux in both the Ministry of Finance (MOP) and the Ministry of Home Affairs (MOHA) are deeply involved in the decision-making process of the Commission as its secretariat. In fact, they appear to have much influence on the Commission's recommendations in two respects: first, they are responsible for the recruitment of members to assist the Commission and the government; second, the necessary data are for the most part limited to that provided by the tax bureaux. Therefore, the role of the staff of the tax bureaux is of great importance to the functioning of the Tax Advisory Commission. These staff members are the most enlightened and imaginative of government officials, and if necessary they could formulate tax proposals by themselves without the aid of any outsiders. In fact, before recommendations are actually made by the Commission, tax proposals are frequently discussed with the MOF and other ministries. Why do the government and tax bureau staffs seek the assistance of the Tax Advisory Commission? Generally speaking, a consensus is reached in the movement between the government and the Commission. While the Commission is sometimes seen as 'hiding behind the government',8 the government's views also have been 7 This was not a permanent establishment, but a temporary committee whose term lasted for a couple of years. After 1960 it was replaced by the long-standing committee that still operates today. 8 We use the term 'Kakuremino' in Japanese. The government sometimes uses the commission as a screen in order to divert attention from its real motives.
General Characteristics of the Tax System
15
altered as a result of conferring with the Commission's members. Thus, tax recommendations prepared by the Commission, which signify the government's approval, are instrumental in smoothing the way for the enactment of tax reform. Without them, the government would encounter obstacles in having any tax proposals implemented, chiefly because it is difficult to obtain public support of tax changes. Strictly speaking, the Commission has not played a crucial role in short-run tax revisions for the past two decades or so. Attention should also be paid to the more influential role of the LDP tax committee, as argued before, which usually attempts to present tax proposals in the annual tax process. Around the end of the calendar year, the recommendations of the Tax Advisory Commission are transmitted to the prime minister for budgetary preparations; those of the LDP tax committee follow almost simultaneously. The final decisions are made by the cabinet based on these two recommendations. In the case of opposed views, it became the general tendency for the LDP Tax Committee to predominate over the Tax Advisory Committee in the tax process. Indeed, the powerful LDP committee often succeeded in significantly modifying the proposals of the Tax Advisory Commission. Consequently, as noted above, the Tax Advisory Commission has tended to limit its proposals to the basic arguments in favour of the desirable direction for the longterm tax system and not to raise politically subtle issues of tax policy (e.g. any judgement regarding the value added tax rate). Also, it is of great importance to note that deliberation in the Commission has become open to the press since autumn 1998. This is a result of the recent movement to make official information open to the public.
2
Background of the Japanese Tax System In order to study the main features of a nation's tax system, it is important to explore its basic background from an historical perspective. The main issues in this chapter are considered from two distinct viewpoints. The first approaches tax issues in terms of the long-run trend of tax structure development. The second examines the starting point of the postwar tax system, with particular attention given to the Shoup tax reform.
A LONG-TERM
VIEW OF TAX S T R U C T U R A L
DEVELOPMENT
Generalizations of Change If we take a long and broad view back into time, our attention is drawn to the salient features of the Japanese tax structure during the process of economic development. In other words, we are led to consider how tax structures appear to change during the transition from a traditional society to a modern one, and whether there is any theory to tie together common threads among tax systems in varying stages of development. Deriving an answer to these questions would be important to achieving an understanding of the basic framework of the present tax system and to elucidating long-term changes in the size and composition of tax revenues. In fact, many studies to date have attempted to investigate tax structure change from a similar point of view. Based on broad empirical findings with special emphasis on the size and structure of tax revenues over time, in past studies generalizations have made an attempt to incorporate these findings into a consistent framework (see e.g. Hinrichs 1966). The purpose of generalizations is to determine whether there is some economic law which, as Engels's law confirmed for consumption spending, reveals a relationship between tax revenues and the development process.1 The basic nature of this study is to examine whether or not such generalizations can be applied to the Japanese experience. What is of greater significance to this study is how the tax structure changes at different stages of economic development. Consequently, an ideal approach would be to examine the same countries at different levels of development, using time-series data rather than cross-sectional data.2 However, this approach presents This chapter is based partially upon Ishi (1978,1987). 1
The necessity of constructing such a law is stressed by Thorn (1967, 19—20). Most of the published studies, however, have focused on tax structure development in developed and developing countries, using cross-sectional data. This approach is the only one feasible in many 2
Background of the Japanese Tax System
17
several difficulties. For one thing, data for many countries are not available for the entire period of their economic development; for another, few countries have completed all phases of economic development. Nevertheless, there are some eases (e.g. the UK, the USA, Germany) for which one can analyse the entire process of tax structure development from a long-term standpoint.3 The present study represents an effort to extend past analyses of developed and developing countries through an examination of Japan's experience during the 113-year period of 1885-1997. Japan's experience is notable for two reasons. First, Japan is the only non-Western nation to have succeeded in attaining the level of economic development enjoyed by Western nations. This means that Japan has passed through all the levels of development within the past 100 years. Second, in the process of economic development, Japan has not oriented itself too closely to the European pattern. It can be argued that the economic development of Japan was different from that of Western countries in many respects. Thus, Japan should prove an illuminating case-study of tax structure change.4 We must now consider what empirical evidence in Japan's case supports the generalizations of tax structure development. There are two generalizations presented in past studies which will be investigated here: 1. that the size and composition of tax revenues tend to change over time, reflecting structural changes in the economy; 2. that forces (e.g. social, political, and cultural) other than changes in economic structure also govern the determination of tax shares. Here we seek the similarities or dissimilarities of Japan's experiences in terms of these generalizations.5 Fortunately, such an investigation is now feasible, through use of the long-term statistical data prepared by the Hitotsubashi University group.6 A Model of Tax Structure Change Based upon empirical analyses, past studies have developed a general theory to explain and predict the size and composition of tax revenues in the process of cases, chiefly because there is a lack of reliable historical series data on GNP, its components, price levels, and other related data in most countries. Yet a cross-sectional approach is necessarily very rough and appears to have several defects. See e.g. Oshima (1957), Martin and Lewis (1956), Williamson (1961), Lewis (1963). 3 Some useful hypotheses emanated from such works as Peacock and Wiseman (1961), and Musgrave (1969). 4 For past studies of Japan, see Hinrichs (1966, 49), Musgrave (1969, 137). Recently Chelliah (1986) has attempted to compare the Indian experience with that of Japan with special reference to Ishi (1978). 5 More detailed analysis has already been attempted in Ishi (1978). For a general discussion on the Japanese economy, see Minami (1986). 6 The Hitotsubashi group has been engaged in the lengthy project of estimating economic statistics of Japan from 1868; see Ohkawa et al. (1965-88). Fourteen volumes, including National Product, Capital Formation, and Government Expenditure, have finally been published. In particular, these three volumes
18
Overview
economic development. Generally speaking, it is difficult to generalize the development of tax structures in different countries and time periods because tax structure change at first sight appears as a multi-coloured fabric containing numerous patterns. However, a general theoretical pattern of tax structure change emerges from the empirical and historical observations. This pattern was presented in the form of an 'heuristic model' by Hinrichs (1966, ch. 6), which he derived from a crosssectional analysis of countries at different levels of development. Hinrichs uses an heuristic device to establish a typology and an average picture to which individual cases can be compared. It is helpful for the purpose of our investigation to compare the tax structure development of Japan with such a general pattern. According to Hinrichs's model, tax structure generally develops through the shifting of the relative weights of land taxes, indirect taxes (divided into taxes on foreign and internal sectors), and income taxes as the economy advances through each phase of development (traditional, transitional, modern). Figure 2.1 illustrates the pattern of tax structure development in Japan for comparison with Hinrichs's heuristic model (1966, 99).7 Let us first pay attention to the lower part of the figure. Here, three characteristics of Japan's experience are illustrated. 1. A shift from land taxes to indirect taxes, and further to income taxes, can be seen over time. This is almost identical to Hinrichs's ideal type of tax structure change. 2. However, taxes on foreign trade played no major role in the initial stage of Japan's development—a sharp contrast to the experience of many other countries. The chief reason for this is that tariff autonomy was not achieved until 1899, and even after that date only partial revision of tariff structure was undertaken (see Yamazawa 1975, 41). 3. There is one more dissimilarity: the trend of internal indirect taxes has been to decline, not rise, until recent years in postwar Japan. This reflects the sharp growth of individual and, until recent years, corporate income taxes in the rapidly growing economy. Next we turn to the ratio of taxes to GNP (T/Y) at all levels of government; this curve is depicted in the upper part of Figure 2.1. The level of tax share provides an important indicator of the role of government and taxation during development. It represents the fiscal capability of the government to meet the increased need for public services. Also, it measures citizens' power or capacity to bear the burden of taxation. (For more extensive discussion, see Bird 1964, 303-4.) The tax-GNP ratio reveals six major swings, although four of them are not perfect. The first swing, in 1885-95, shows a large tax share of around 8-12 per cent, are essential to my analysis. For other major statistical materials, see Bank of Japan (1966), Economic Planning Agency (1992). 7 In his model, both the expenditures line and the expenditure-revenue gap are depicted clearly, in addition to each line of individual taxes as a percentage of GNP.
Background of the Japanese Tax System
Fig. 2.1
19
Changes in the level and composition of Japanese tax revenues, 1885-1995 (in
terms of GNP) Note: While tax shares cover all level of government, four ratios of each tax to GNP are limited to the scope of national government.
the same level as that reached in developing countries during the postwar period.8 The principal taxes were those collected on land and alcoholic beverages, and these were used to meet extensive government needs while the country was still at a low level of national income.9 The second swing peaks around 1910 and bottoms in 1918. New or increased taxes on income, alcoholic beverages, tobacco, sugar, textiles, and beverages, as well as custom duties, all contributed to the upswing. The third swing, falling between 1918 and 1932, is a plateau at a fairly low level, reflecting the depression in the 1920s. The main revenue sources at that time were taxes on alcoholic beverages and income. In the 1930s, the tax revenue share rapidly increased in the fourth swing with no downswing following. Direct taxes on individual and corporate incomes, including an excess profits tax instituted in 1935, rose tremendously during this period. Fifth, the highest peak emerges immediately after the end of the Second World War, followed by a relatively stable level of T/Y after 1952-3. After postwar reconstruction was complete, the economy's rapid growth and the government's decision to stress private-sector growth (partly through a tax reduction policy) lowered the tax share, and it remained between 14 and 17 per cent until about 1970. Lastly, from that time the tax share began to rise rapidly, apart from a sharp fall in the mid-1970s. This reflects the expansion of fiscal deficits without a tax cut during the 1980s, reaching its highest in 1990; subsequently it began to 8 9
For data on the developing countries, see Musgrave (1969, appendix table 6). Oshima (1965, 386—7) notes the high level of Japan's tax burden in international comparison.
20
Overview
decline as a result of the recent slump. Until the tax share moves upward in the future the six swing (trough-peak-trough) will continue. Tax Structure Change and Economic Development The major question to be answered here is, How have the size and composition of tax revenues been determined? More specifically, what kinds of factors are most important in explaining the variation of both tax level and structure? Various studies have shown that the ratio of tax revenues to GNP increases with economic development and that the structural change of taxation reflects different levels of development between developed and developing countries. In what follows, attention is given to tax structure change during development. Tax structure, of course, is greatly affected by institutional, economic, and socio-political factors.10 Indeed, it can be regarded as a product of the historical interaction between such forces. Although changes in tax structure can, in principle, influence these economic, social, and political forces over time, such changes have generally tended to be more a determined than a determining factor. If one were to emphasize the passive nature of the evolving tax system, one could even say that the major determinant of tax structure change is the structural change in the economy itself during the process of economic development. There are several specific variables which have been employed in past studies to explain the size and structure of tax systems. Among them, the following three factors are pertinent to the case of Japan: 1. y/N
2. Ag/Y
3. M/For (M+X)/Y
= per capita real GNP (y is real GNP in the prewar period at 1934—6 prices and in the postwar period at 1960 prices, and Nis total population); = agricultural products' share in GNP (Y is nominal GNP, and Ag is output of the primary sector, i.e. agriculture, forestry, and fishery); = openness of the economy (M and X stand for imports and exports, respectively).
Factors (1) and (2) are viewed as indices of economic development, while (3) measures the size of the foreign trade sector, which is used as an alternative to (1) and (2). All these indicators have been found to be significant variables in explaining variations in 777 and the composition of tax revenues, although a fuller explanation could be made by introducing additional factors. Table 2.1 shows the correlation of these three variables with tax shares. The simple correlation coefficients during the prewar period are as high as would be expected from past empirical observation of developing countries on a cross-sectional basis, and we find that the estimated results of the postwar era are reasonable, too. 10
For the socio-political aspect of the problem, see Deutsch (1961).
Background of the Japanese Tax System
21
Table 2.1 Correlation of per capita real income (y/N), openness (M/Y, (M+X)/Y), and agricultural products' share (Ag/Y) with tax shares (J7Y), 1885-1997 Period
Sample size
Correlaticin coefficient y/N
M/Y
(M + X)/Y
Ag/Y
Prewar (1) 1885-1944 (2) 1885-1909
60 25
0.281* 0.337
-0.030 0.458*
-0.081 0.563**
-0.307* -0.377
Postwar (3) 1951-1997
47
0.674**
0.283
0.504**
-0.465**
* Significant at the 5% level. ** Significant at the 1% level.
This, therefore constitutes a rough sketch of the interdependence between the variables. The results may be summarized as follows. 1. There is a significant correlation between y/N, Ag/Y, and T/Y, with reasonable positive or negative signs in the prewar period of 1885-1944. 2. Negative correlation exists between Ag/Y and T/Y for the postwar period of 1951-97, and y/Nand T/Y are still significantly related. 3. Openness provides a better index than the other two variables mentioned above for the first 25 years of the prewar period. In addition, it becomes more significant in explaining the variation of T/Y for the postwar period. These findings are similar to the results of cross-sectional analysis that have been obtained from regression or correlation between T/Y and various development variables." Another approach can be taken to pursue the same analytical objective. In addition to the investigation of T/Y in Table 2.1, an equally important question to ask might be, Is there any systematic relation between the source of tax revenues and the level of income, or is tax structure influenced by institutional factors relatively independently of economic development? In this question, the focus shifts to changes in the composition of the tax structure. Tax revenues Tn are disaggregated into three sources: 1. land taxes—Te; 2. indirect taxes, including profits of government monopolies12—T,-; 3. income taxes on individual and corporate income—T y . 11 Another way of explaining the variation of T/Yuses the hypothesis that the existence of the E-R gap (i.e. the discrepancy between government expenditure and revenue) necessarily stimulates a concomitant increase in tax burdens. This hypothesis is tested by using statistical procedures in Ishi (1975,213-16). 12 If the share of foreign trade taxes is assumed to be high (which is not the case in Japan), it should be treated as one of the dependent variables, distinguished from domestic indirect taxes as in Lewis (1963).
22
Overview
These figures, however, are limited to national government tax revenues because the data for classifying local taxes in such a manner are lacking.13 Therefore, for the dependent variables, we let TtITn stand for the relative share of land taxes, TjTn for the indirect tax share, and Ty/Tn for the income tax share, respectively. (For the same procedure, see Williamson 1961, 51-2.) There are three reasons for stressing the relative importance of each tax share. First, it appears that the effects of economic development upon the tax structure are more a function of institutional change than they are inherently an economic matter. Thus, greater emphasis should be placed on the various sources of tax revenues, as their change reflects the institutional setting of the tax system, which is relatively independent of changes in economic structure. Our 'share-approach' seems to capture the effect of institutional factors. Second, there is a high correlation between T/Y and each share component of total national government taxes during the time period in which each constitutes the principal share of the total (e.g. land taxes for 1885-98, indirect taxes for 1899-1935, and income taxes for 1936-90).14 Third, the relatively poor results of Table 2.1 must be reconsidered; they may be the result of aggregating all the taxes and using GNP as the divisor in tax revenues. A simple regression model was constructed with regard to the development of tax structure. As shown in Table 2.2, almost all the coefficients of determination are statistically significant with a few exceptions. The regression coefficients of all the independent variables in all the equations are also statistically significant. The conclusions from these estimations are as follows. 1. As y/N (an index of development) increases, TtITn and Tj/Tn decrease. The relative importance of these two taxes tends to decline over time, a result found in other studies. 2. The declining trend of Tt/Tn is influenced by the decreasing share of AgfY. 3. 'Openness' (M/For M+X)/Yor can explain the variation in T,-/Tn in the prewar period,15 while it is not significant in the postwar era. 4. Ty/Tn is dominantly affected by y/N. Obviously, the relative share of income taxes tends to rise in the course of development.
13 In general, national government taxes dominated the tax composition of local taxes in the prewar period, since the latter was levied virtually as a surtax on national government taxes. A more complete coverage of tax revenues would not alter the conclusions presented here. This is inferred from data available for the prefectural level, not including the lower levels of local government (i.e. city and town). 14 The correlation coefficients between T/Y and each relative share of national government taxes are: 0.662 of land taxes for 1885-98, 0.512 of indirect taxes for 1899-1935, and 0.492 of income taxes for 1936-90. All the coefficients are significant at the 1 % level. 15 As is evident from the positive coefficient of the indirect taxes equations in the prewar period, the 'openness' of the economy expanded the tax base for indirect taxation, through spillover effects which stimulated consumption, commercialism, transportation, etc. On the other baud, the sign of the 'openness' coefficient in the postwar period is insignificant. It appears that 'openness' was no longer effective in increasing the indirect tax base at this level of economic development.
Table 2.2 Regression equations for tax sources as a function of economic development variables, 1885-1944 and 1953-1997 Prewar period (1885-1944)
Postwar period (1953-1997)
Land tax
Indirect tax
Income tax
The generalized least square (GLS) method was used to generate all these equations. R2 is the coefficient of determination adjusted for degrees of freedom, ** and * indicate significance at the 1 and 5% levels, respectively, DW is the Durbin-Watson statistic, and the values in parentheses are f-statistics.
24
Overview
Ay 5. Real growth rate — is added only in the postwar case. It is significant to explain y
the changes in Tf/Tn and Ty/Tn with opposite signs each. Negative interrelation Ay between TyITn and — appears to reflect the large scale of income tax reductio y in a growing economy. The purpose of the present study has been to examine the development of the tax structure in Japan during the period 1885-1997 in the context of generalizations made in previous published studies. The evidence of Japan presented here seems to be in full support of past generations. Since time-series analyses of tax structure change are relatively rare in the literature, Japan's case-study is especially important if it provides support for these empirical generalizations. Although the findings indicate some divergence from those patterns, many similarities can be found in Japan's experience. In particular, great emphasis should be put on the fact that Hinrichs's heuristic model fits Japan's case with only minor exceptions. Other Determinants of Tax Structure Development In addition to the economic factors underlying tax structure development, there is another key factor in determining the growing ratio of tax revenue to GNP and the 'proper' tax structure composition. In view of the results of various studies, attention should be directed towards the cultural-political preferences for adopting a specific size and composition of the tax system. When a country has reached a high income level and a large government sector share of GNP (say, between 20 and 40 per cent), these preferences appear to become much more important than at lower income levels. For instance, the level at which the government sector share settles between 20 and 40 per cent is likely to be determined by differing commitments towards 'security and defence' and/or 'welfare policy', rather than by change in economic structure (see Hinrichs and Bird 1963, 433). In postwar Japan, such ideological commitments are probably among the determinants of tax structure development. What is of great interest is the low level of the tax share in the postwar period. As we have seen in Table 1.2, Japan has the lower level of tax burden among major advanced countries during four selected years. Japan's low ranking remains unchanged today, even though the tax share has rapidly increased to reduce the gap with other countries. This feature, peculiar to postwar Japan, needs to be explained. Among explanatory factors, of most importance is the difference in the level of military expenses between the prewar and postwar years. If military expenses are shown relative to GNP (or GDP) during selected fiscal years (the relevant table is omitted), the ratio rises drastically during the war period (e.g., 8.44 per cent in 1894-5 (the Sino-Japanese War), 22.97 per cent in 1904-5 (the Russo-Japanese War), and 27.98 per cent in 1941-4 (the Second World War)). Attention should, however, be directed towards the very low figures for the postwar period in comparison with those of prewar Japan: the percentage of postwar military expenditures as a part of has been less than 1.00 per cent. The low level of military spending i
Background of the Japanese Tax System
25
also apparent in international comparisons of the military expenses-GNP ratio: between 1971 and 1984 this ratio averaged 6.3 per cent for the USA, 5.1 percent for the UK, 3.5 percent for West Germany, and 3.8 per cent for France. The 1985-91 averages showed 5.7 per cent for the USA, 4.5 per cent for the UK 2.5 per cent for West Germany, and 3.1 per cent for France, while Japan's figure was only 0.97 per cent. During 1992-96 the ratios averaged 4.8 per cent for the USA, 4.5 per cent for the UK, 2.0 per cent for Germany, 3.6 per cent for France, and 1.2 per cent for Japan. Thus, the gap between Japan and other major countries had shrunk substantially. In addition to the absence of military expenses, reference may also be made to the low level of Japan's welfare commitments. The percentages of social welfare transfers to GDP in 1970 in various countries were: 16.8 per cent in France, 12.7 per cent in West Germany, 8.6 per cent in the UK, 7.6 per cent in the USA, and 4.7 per cent in Japan. These ratios tended to increase in each country: by 1990 the US ratio had increased to 11.1 per cent, the UK's to 11.6 per cent, Germany's to 15.2 per cent, and France's to 21.1 per cent, in comparison with Japan's 10.8 per cent. Japan still had the lowest ratio, although this showed a markedly faster rise during the previous decade. In 1996, Japan's ratio rose to 13.5 per cent, exceeding 12.9 per cent in the USA, but both still remained lower than those of major Western countries: 14.2 per cent in France, 18.5 per cent in Germany, and 23.4 per cent in the UK.16 Cultural-political factors also appear to have some influence on the pattern of tax composition in postwar Japan. It is widely acknowledged that there are typically two tax styles in the world, reflecting the cultural determinants of any tax system: one is the direct tax style (or at least, an even split between direct and indirect taxes), and the other is the indirect tax style. Japan has preserved the former style in the postwar era, to a large extent because of the American influence during the occupation period, i.e. the tax recommendations of the Shoup Mission in 1949. Had there been no Shoup Mission, Japan's tax system might have moved towards a different type of system with a greater share of indirect taxation. THE DAWN OF THE POSTWAR TAX SYSTEM:
THE
SHOUP TAX REFORM
The Shoup Mission As described above, cultural, political, and social forces are equally as important as economic ones in determining the nature of a nation's tax system. In this regard, we must stress the significant role of the 1949 Shoup Mission in shaping the style of the tax system in postwar Japan. (For a more expanded discussion, see Ishi 1987, Shoup 1989.) The mission, headed by Professor Carl S. Shoup, visited Japan in April 1949 at the request of the Supreme Commander for Allied Powers (SCAP).'7 They stayed 16
These figures are derived from OECD estimates. In regards to the process of inviting the Shoup Mission, see Moss (1948), MOF Tax Bureau (1977a). The Mission was composed of seven members, including S.S. Surrey, W.S. Vickrey, J.B. Cohen, H.R. Bowen, R.R. Hatfield, and W.C. Warren. 17
26
Overview
in Japan for about four months, investigating the Japanese tax system as well as its economic and social background. As a result of intensive studies, they presented to SCAP the Report on Japanese Taxation by the Shoup Mission in August 1949. The Japanese government then attempted to reorganize the country's entire system of national and local taxes in accordance with Shoup's recommendations, and the new system went into force with the next supplementary budget in 1949. What were the main reasons for the visit of the Shoup Mission? Two points should be stressed. First, conditions of the postwar Japanese economy were chaotic. Rampant inflation was wreaking havoc among business accounts, tax assessments, and revenue collection. The tax system was truly in a mess. Second, there was mutual understanding between the USA and Japan as to the necessity of overhauling the tax structure and its administration after implementing the 'Dodge Line'.18 The Shoup tax reform was not the first reform of the Japanese tax system during the occupation period, but earlier tax reforms had proved far from satisfactory (see Shavell 1948a,fo; Cohen 1949). Consequently, an authoritative plan was absolutely required to revise the tax system in order to attain greater equity and efficiency. It was not simply a political expedient. In the earlier years of the occupation, the tax system had developed certain defects which the Shoup Mission was asked to remedy. The most important defects were as follows. 1. Individual income taxes had become higher and more progressive, with low exemption and broad coverage. Heavier tax collection overwhelmed tax administration and weakened tax morale. Revenues were collected by the 'goal system', in which each tax office was assigned a goal or quota. 2. Corporate income and excess profits tax rates were high, and businesses were not permitted to adjust depreciation allowances to allow for drastic price hikes. 3. Tax sources remained concentrated at the national government, although a great volume of public functions was allocated to local governments (prefectures and municipalities) in the name of strengthening 'local autonomy'. In addition to trying to remedy these defects in the tax system, the Japanese government negotiated with SCAP to achieve a substantial tax reduction (especially of individual income tax) on behalf of taxpayers. SCAP, however, did not want to verify the necessity of such tax cuts.19 Thus, tax reduction became a crucial issue before the Shoup Report was published. Fortunately, economic conditions, which had been improving in the period before the beginning of the Shoup Mission, favoured its recommendations; inflation 18 The 'Dodge Line' is the name of the anti-inflationary programme conceived by Joseph Dodge, an American banker, who was invited by General MacArthur in 1948 to evolve a formula for arresting runaway inflation. Dodge aimed at stabilizing the yen value by establishing a true balance in the consolidated budget and eliminating the subsidies that had been the prime cause for the continuing growth of fiscal deficits. It was very successful in halting inflation, although a great depression ensued. See Cohen (1950), Yamamura (1967). 19 For a Japanese view, see Economic Stabilization Board (1949). However, Dodge completely disagreed with the request for tax reductions from the Japanese side; see SCAP (1949).
Background of the Japanese Tax System
27
had been halted as a result of Dodge stabilization policy of 1948. Nevertheless, there were still a number of difficulties in the wake of the Shoup Mission. For instance, in view of anti-tax and anti-inflation sentiments in Japan, it was necessary for the Mission to include tax cuts without unbalancing the budget.
Basic Framework of the Shoup Report Like the Carter Report in Canada and the Meade Report in the UK, the Shoup Report has been highly evaluated by many tax experts and has received considerable attention over a long period of time, especially for its theoretical and logical consistency (see e.g. Hicks 1951). The Shoup Report had epoch-making significance in the history of Japanese taxation. In contrast to the tax reports noted above, most of the Shoup recommendations were put into practice in Japan, although a movement towards modifying them began very soon after implementation. The contribution of the Shoup recommendations to Japanese taxation should not be underestimated; throughout the postwar period, the report has served as the benchmark of a well designed tax system whenever Japan discusses tax reform. The Shoup Report is generally considered to contain new and advanced views long cherished by Shoup, Surrey, Vickrey, and other tax experts. The Shoup Mission attempted to reconstruct the Japanese tax system along lines generally familiar to American tax experts, and a number of novel features were designed to make the Japanese tax system 'the best tax system in the world' (Shoup Mission 1949, vol. 1, p. ii) in a so-called experimental manner. Three points characterize the Shoup Report as a whole. First, the fundamental aim of the report was to establish a permanent and stable tax system in Japan over the long term.20 Needless to say, the goal of the Shoup Mission was to create a modern tax system based on direct taxation. Alternatively, it would have been possible to choose another form of taxation, based upon the indirect tax. Indeed, the Japanese government preferred the indirect tax system, mainly because it had favoured indirect taxation during the prewar period.21 Second, the tax proposal was intended as a single integrated plan. The Shoup Mission felt very strongly that the whole plan would be destroyed if any parts were 20 At the time, the Shoup Mission seems to have thought that their proposed tax system should be preserved for more than ten years. Shoup referred to this point in retrospect when he came to Japan in 1972, see MOP Tax Bureau (1972). Furthermore, long-term tax reform was possible because there was not an immediate need for revenue. Accordingly, the Mission recommended a tax plan that would bear fruit only in the long run. 21 The Shoup Mission gave two reasons for not recommending an indirect tax system. (1) Such a system could raise the required revenue, but it would perpetuate gross inequities among taxpayers, dull the sense of civic responsibility, keep the local governmental units in uneasy financial dependence on the national government, and give rise to undesired economic effects on production and distribution. (2) Moreover, the difficulties in obtaining fair and efficient administration of the tax laws, and a high degree of compliance by the taxpayer in Japan, should not be seen as inevitable.
28
Overview
eliminated. On this point, the Shoup Report argued very strongly that 'What we are recommending here is a tax system, not a number of isolated measures having no connection with one another' (Shoup Mission 1949, vol. 1, p. ii). By referring to 'a tax system', they placed strong emphasis on the interlinkage among individual taxes. The Mission carefully considered the effect of individual taxes on one another. For example, individual income and net worth taxes, and succession and real estate taxes (the land and house tax), were investigated jointly. Furthermore, the principle of full inclusion of capital gains and losses in income taxes was closely related to the inter-relationship between individual and corporate income taxes. It is evident from the basic idea of the Shoup Report that the whole income tax structure would be seriously weakened without full inclusion of capital gains and losses. Third, among various tax criteria, most importance was placed on tax equity throughout the whole report. In the press interviews immediately after the Shoup Mission came to Japan (19 May 1949), Shoup himself emphasized the importance of restoring fairness in the Japanese tax system as one of five objectives for his tax reform. Thus, the basic philosophy in support of tax equity is repeatedly argued: A tax system can be successful only if it is equitable, and the taxpayers must realize that it is equitable.... We have often encountered surprise at the emphasis we place on the search for equity. But no one remains in the tax field for long without realizing that nothing he recommends will stand up unless it meets the test of fairness in the distribution of the tax burden. (Shoup Mission 1949, vol. 1, p. 16.)
Turning to major parts of the recommendations, several points are worth noting. First, of greatest importance is the fact that the progressive and broadly based individual income tax was retained as the mainstay of the Japanese tax system. In retrospect, the individual income tax proposed by the Shoup Mission really provided an ideal form of comprehensive tax base with a single 'progressive' rate system in the true sense of the term. Of course, this was the first time that anything of the kind had been attempted in an Asian country, although Japan had experienced schedular income taxes since 1887 on a smaller scale.22 Second, a major concern of the Shoup Mission's recommendations was the improvement of tax administration, especially of the income tax. For instance, withholding taxes at sources from wage and salary incomes and the self-assessment with universal filing of returns were recommended. Furthermore, the use of the 'blue form' for tax returns was especially suited to encourage the proper keeping of accounts, particularly in the case of small businesses. Obviously, these efforts to improve assessment and administration were indispensable for an effective and equitable tax system.23 22
U.K. Hicks argued that the income tax was introduced under rather primitive conditions: 'It is not the sort of economy in which one might, on a priori grounds, expect to be able to recommend a very large sphere for income and profits taxes' (Hicks 1951,200-1). See also Kimura (1952). 23 Tn addition, the Shoup Mission attached great importance to the provision of regular training for tax assessors and collectors, the establishment of training colleges, and the improvement of pay and conditions of the tax administration.
Background of the Japanese Tax Syste
2
Third, a general revaluation of all assets (i.e. land and fixed capital) was recommended as a prerequisite for the adoption of the Shoup tax plan. Since the value of the yen had depreciated on the order of 200- or 300-fold since the prewar period, such a process of revaluation would stimulate private capital accumulation. In addition, it was proposed that a tax of 6 per cent be imposed on the appreciation in the written value of all assets, although it would consist almost wholly of paper gains in terms of book value. Fourth, great emphasis was placed on the reform of local finance in order to educate the Japanese in democratic citizenship. The provision of a fiscal framework for 'local autonomy' was an important element of the Shoup proposal. The general recommendation of the Shoup Mission was that local powers and duties should be substantially increased; in particular, priority should be given to the lowest of the three levels of government (i.e. the municipalities). For this purpose, local governments were given new tax resources (e.g. property tax and value added tax), and at the same time intergovernmental transfers were overhauled to implement a new scheme for the equalization of local budgets called the Equalization Grant Scheme. Finally, the Shoup tax plan contained several novel fiscal experiments. The Mission suggested these experiments for Japan to try without the benefit of any large-scale applications in other countries. Special attention was given to three of these—the net worth tax, the accession tax, and the value added tax—although they were minor in size. The Aftermath of the Shoup Tax Reform It is rare in history that a tax report is enforced in practice. The Shoup Report was almost wholly enacted in both the 1949 supplementary budget and the 1950 budget. The Shoup tax reform is interesting to tax reform experts as a case-study of the accomplishments of a tax mission in a short period under ideal conditions. In seeking the necessary conditions for a successful tax reform, there seem to be a number of relevant factors to learn from the impact of the Shoup Mission. From the very beginning, however, some of the Shoup tax plans were criticized as being too theoretical to be carried out, given the state of socio-economic development in postwar Japan. No doubt, the Mission thought of tax reform primarily in terms of US practice and experience. This was apparent in such matters as the treatment of capital gains taxation or the emphasis on 'local autonomy'. Accordingly, modifications to the Shoup tax system were implemented shortly after 1950. Two tendencies emerged from these modifications. One was the revival of the old system: equity was sacrificed for the convenience of incentives and administration. The other was the reduction of the tax burden of firms, especially big businesses. The goal of this trend was to give priority to the restoration of the postwar economy and the promotion of capital accumulation. Tax equity, on which the Shoup Mission put utmost priority, began to be replaced by incentives as the criterion o
30
Overview
taxation.24 As time has passed, essential features of the Shoup plan have been 'eroded' or 'patched and tattered' by the later tax reforms of the Japanese government. (See Table A2.1 in the Appendix to this chapter.) The most symbolic modification of the Shoup system occurred with the repeal of full taxation on capital gains from sales of securities in 1953. It has often been pointed out that the Shoup Mission was well aware of the shortcomings of the US tax system in respect of capital gains taxation, and that they tried to introduce a better treatment of capital gains as an experiment in the Japanese situation. As noted earlier, the Mission repeatedly insisted on the need for capital gains taxation. In spite of their strong appeals, the actual capital gains tax mostly disregarded proceeds from security sales since 1953, partly because the difficulties of administration were great, and partly because the promotion of capital accumulation became a national goal. In addition, the innovative tax devices of the Shoup Report disappeared from the Japanese tax system after brief or no trials. The net worth and accession taxes were abolished in 1953 because of inadequate revenues. The value added tax was not even brought into operation; its enactment date was postponed twice, and it was finally repealed in 1954 (see Ito 1950; Bronfenbrenner 1950). When the Japanese government departed from the Shoup system, its departure was not in the direction of further experimentation, but towards a return to prewar traditions and practices which it considered particularly suitable to the Japanese economic situation.25 Thus, the tax innovations advocated in the Shoup Report were disregarded. Necessary Conditions for a Successful Tax Reform The modifications by the Japanese government of the Shoup proposals were drastic in their later consequences. Consequently, it may be argued that the Shoup reforms achieved only a partial success, largely because the modifications gradually made the tax system more inequitable and complicated. In spite of these drawbacks, the Mission's contribution in reconstructing the postwar tax system in Japan was considerable. Throughout the postwar period, the Japanese system has retained substantial features of the Shoup framework. Thus, the Shoup tax reform can be considered one of the most successful tax reforms in the world. 24 In general, these modifications of the Shoup tax reform were accepted as inevitable by the Japanese. For instance, Hanya Ito commented on this point: 'However, it is to be observed that the tax system in practice is a product of historical development depending on the social, economic, and political conditions of time and place. It would not be wise to condemn such a course of events merely from the standpoint of abstract theory'(Ito 1953,382). 25 See Bronfenbrenner and Kogiku (1957, part 1, 241). Ito (1953) also argues that, 'Judging from the development of tax reforms these last three years, Japanese taxation is showing a tendency to restore the old system which was in effect before the Shoup recommendation' (p. 358).
Background of the Japanese Tax System
31
To conclude this discussion, we shall seek to explore the necessary conditions for successful tax reform, a rare event in history. Particular attention should be paid to the following three points. The first and most important point is that the foundations on which the Shoup Mission was to erect the new tax structure in Japan had experienced a complete break with the past by the events of the Second World War and postwar inflation. Prewar values had become irrelevant as a basis of postwar taxation, and any former injustices could be disregarded in view of the sweeping change that had affected all values. Furthermore, the changes recommended by the Mission were far less drastic than the overhaul of values that the Japanese had experienced during the war. These circumstances facilitated the work of the Mission and encouraged them to experiment with innovative tax reform. To use Feldstein's terms, a 'tax design', rather than 'tax reform', was implemented (Feldstein 1976, 77). The second point is that circumstances greatly favoured the Shoup Mission. Seldom has any advisory mission received instructions as broadly defined as in the case of the Shoup Mission.26 In addition, the Mission's arrival coincided with the introduction of a national programme for the redirection of the entire economy, which had become feasible since economically chaotic conditions had settled down to a considerable extent. Thus, the Mission was implicitly given leeway for wide and sweeping changes in forming the tax plan. Of utmost importance was the fact that the Mission was supported by SCAP and General MacArthur. Reflecting this support from the highest authorities, the recommendations of the Shoup Mission received high-priority consideration from the Japanese government. These circumstances put the Mission in an exceptionally favourable position as to the enactment of recommendations. The Japanese government and Diet acted with vigour in accepting nearly all of the proposals. Third, from a professional point of view, the Shoup Report itself has been rated as one of the best tax reports, and of the highest quality. The academic specialists of the Mission first followed the basic principles of taxation as developed in textbooks. Thereafter, they tried to link these theoretical considerations with the institutions of Japan, although they were handicapped by unfamiliarity with the Japanese tax system. It is often pointed out that the Shoup proposals not only are logical and well balanced in theory, but also can stand the test of practicability to some extent. In the case of tax missions to other countries, when results generally fall short of expectations, the host countries often tend to refuse to enforce the proposals wholeheartedly. As far as the Shoup proposals are concerned, this tendency was at a minimum in Japan. The Japanese people would not have accepted the proposals if the Shoup Mission had prepared a set of recommendations doomed to failure. Instead, the
26
Shoup himself mentioned in retrospect that General MacArthur did not interfere with the work of the Shoup Mission, and refrained from making any special requests or issuing any orders when they were writing the Report; see MOF Tax Bureau (1972).
32
Overview
Japanese understood that they were to benefit from some of the best thinking on tax issues which the Shoup Mission provided.27 It is obvious from the above discussion that the conditions under which the Mission created a 'tax design' for Japan were exceptional. They would never reappear in the future. 27 For the discussion of the relationship between the Shoup Mission and Japanese taxation, see e.g. Sundelson (1950); Bronfenbrenner and Kogiku (1957a, 1957, pt. 2); Dodge (1949); Moss (1949).
Background of the Japanese Tax System
33
APPENDIX
Table A2.1 Shoup Mission recommendations and their modifications (major items only) Shoup recommendation I. Individual income tax (1) Type of tax: to be single on an aggregation basis, not schedular
(2) Top bracket rate, to be lowered to 55% from 85% with 8 income brackets (3) Exemption, deduction and credit personal exemption, dependants, and earned income deductions to be changed (4) Capital gains and losses: to be included in or deducted fully from income, and treated as form of fluctuating income with averaging system (5) Interest income: source collection (separately from other income) to be abolished
Japanese legislation (1949-56) Carried out Subsequent Japanese moves towards schedular income tax, e.g. special treatment of bank interest, dividend, retirement income, etc. Carried out Top bracket rate raised to 65% (1953) when net worth tax repealed Carried out Medical deduction (1950), life insurance payment deduction (1951), and social insurance payment deduction (1953) Carried out Both gains and losses from security sales disregarded (1953); instead, security transfer tax introduced Carried out, but old system revived at 50% rate (1951); rate cut to 10% (1953); bank interest income made tax free (1955)
II. Corporate income tax (and asset revaluation) (1) Corporate income tax rate: not to Carried out be increased above 35%; no Raised to 42% (1952); lowered to progression to be imposed 35% on first ¥500 000 of income, 40% on the remainder (1955) (2) Excess profit tax. to be replaced Carried out (3) Revaluation procedures: land and Four revaluations carried out depreciable assets to be revalued (1950, 1951, 1953, 1954) as of 1 July 1949 Last revaluation made compulsory for depreciable assets of large-scale corporations; farm land not to be revalued until sold (4) Tax on revaluation gain: to be set at Carried out 6% of gain; payable in instalments over Repealed in connection with 1954 3 years for depreciable assets; payable for revaluation non-depreciable property at time of sale III. National indirect taxes (1) Turnover tax. to be repealed as soon as revenues permit (2) Textile consumption tax: to be repealed
Repealed as of 1 Jan. 1950 Carried out
34
Overview
Table A2.1
cont'd
Shoup recommendation (3) Alcoholic beverage excises: (a) Rates to be raised to pre-May 1949 level, with further increases as local alcohol taxes are repealed (b) Alcohol consumption tax to be repealed
(4) Tobacco taxes (monopoly profits) prices of cheapest (rationed) cigarettes and cut tobacco to be reduced (5) Commodity taxes: rates to be reduced (6) Minor excises to be repealed: Soft drinks Travelling (3rd class) Registration and stamp taxes
Japanese legislation (1949-56) Partially carried out
Carried out
n Tobacco prices increased by local tobacco consumption excises (1954) Substantially carried out Never carried out Soft drinks included in items subject to commodity tax Carried out Never carried out
IV. Local taxes and intergovernmental fiscal relations (1) Value added tax: enterprise tax to Never carried out become an income-type VAT Effective rate postponed annually exclusively at prefectural level through 1953; tax repealed (1954). Enterprise tax remained in effect (2) Inhabitants' tax (local income tax): (a) Allocation: to be reserved for Carried out municipalities Made partially prefectural (1954) (b) Variable element: to be based Carried out on income alone, not on National income tax used as property or social status as standard, takes form of 18% surtax formerly. Base may be (i) income tax, (ii) taxable revenue, or (iii) the difference (ii) — (i) (c) Corporation to be exempted Carried out Corporations made taxable (1951), tax taking form of 15% surtax on national corporate income tax; surtax lowered to 12.5% (1952) (3) Property tax: (a) Allocation: to be reserved for Carried out municipalities Made partially prefectual (1954) (b) Coverage: to be extended to Carried out depreciable assets as well as real property, but not to inventories
Background of the Japanese Tax System Table A2.1
35
cont'd
Shoup recommendation (c) Assessment: to be based on capital rather than on rental values of property
(4) Equalization Grant. (a) To be established as replacement for shared taxes and partial subsidies, but approximately double the total amount of former (b) Distribution among local units to be based on needs for major activities (c) Distribution element of income taxes to be eliminated
(5) National subsidy of local activities: Methods to be changed: (a) 100% subsidies to be replaced by national government performance (b) Partial subsidies to be replaced by Equalization Grant (except for promotional purposes)
Japanese legislation (1949-56) Carried out, with capital values determined by selling prices and volume of business Rate originally set at 1.6% of base, subsequently lowered twice, in 1955 to 1.4% Carried out Equalization Grant system abolished (1954); instead, new tax shared programme introduced Carried out for 90% of grants (1951-4) Carried out 20% of personal income, corporate income, and alcohol taxes distributed to local governments (1954); percentage raised to 22% (1955)
Partially carried out (in some cases, subsidy reduced instead) Never carried out.
Note: This table is partially based on information provided by Bronfenbrenner and Kogiku (1975).
3
Two Strategies of Postwar Tax Policy
The postwar Japanese tax system emerged from a blueprint proposed by the Shoup Mission, but the blueprint was promptly altered by the Japanese government. In the 1950s and 1960s, when the Japanese economy saw very rapid and sustained growth (see Patrick 1970; Patrick and Rosovsky 1976), a Japanese brand of tax system gradually developed as part of postwar growth policy. As a result many of the Shoup tax proposals were replaced by Japan's own tax measures. What is of particular interest to outside observers is how significant the contribution of postwar tax policy was to the remarkable record of economic growth. This chapter is devoted to the study of two aspects of the performance of tax policy since economic reconstruction: (1) tax incentive policies and (2) annual tax reductions. In addition, attention is paid to a renewed movement aimed at restoring tax equity and neutrality in the second half of the 1980s, instead of relying on the older growth policies. The Japanese experience during the postwar period provides an interesting and important case-study, since it was different from those of many other advanced countries.
TAX INCENTIVE POLICIES
Two Strategies
Soon after the new tax system initiated by the Shoup Mission was enacted, modifications began to be proposed, mainly for two reasons. First, the new tax system proved too idealistic and far-reaching for the relatively conservative government. Second, the government wished to enable Japan to revive its postwar economy more rapidly; thus, it behaved in a more growth-oriented manner and used the tax system more effectively to promote private economic growth. It is not easy to describe briefly the special character of Japan's tax policy in the 1950s and 1960s.1 Special conditions have certainly existed in Japan for which the government has been able to create unique tax policies. In this connection, there are 1 Note that 'some of its particular features seem strange to a visitor from a country in which economic conditions and social attitudes are quite different. Japan's high rate of growth and moderate government expenditures (when compared with other developed countries) permit the Japanese to adopt tax policies that can well be envied elsewhere' (Pechman and Kaizuka 1976, 323).
Two Strategies of Postwar Tax Policy
37
two strains of postwar tax policy that are of interest to the outsider: 1. an incentive tax policy to achieve specific policy goals; and 2. a successive tax-cutting policy to maintain a lower level of tax burden. These two strategies were gradually adopted by the government to complement the actual performance of the economy and were sustained at least until the outbreak of the oil crises in the 1970s. If we extend our view to cover government expenditures, a third strategy of sound public finance should be added to clarify the whole image of government fiscal activity (see Ishi 1973, 59-60). Special Tax Measures Let us begin with the first strategy of creating tax incentives. Especially before the outbreak of the first oil shock in 1973, there was 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 saving and investment, housing, technological development, environmental quality, etc. These usually included tax exemption tax-free reserves, and accelerated depreciation for individuals and corporations. Linked with Japanese-type industrial policies, such measures deserve attention as contributing factors to the rapid economic growth in the 1950s and 1960s. It is, however, very difficult to ascertain the effectiveness of using tax incentives for specific policy purposes. In fact, the evaluation of these policy effects has been controversial, subject to multiple difficulties. These difficulties are due mainly to the scarcity of quantitative studies concerning tax incentives on economic activities. In some cases, it is impossible to quantify the effect of these policies. We shall begin our discussion by considering the performance of tax incentive policies that the Japanese government adopted in the postwar period. Although there are a number of notable contributions in this field (see Komiya \966b; Pechman and Kaizuka 1976), it is worth reviewing the development of tax incentives in the system, covering a more recent period (see Kaizuka 1984). In the postwar period, numerous tax measures were introduced to stimulate economic growth and other related policy targets that had a high national priority.2 Such measures are included in the Special Tax Measures Law, which is separate from the ordinary income tax laws. This special law was formulated to prepare a list of most (though not all) of the incentive provisions applying mainly to individual and 2 The origin of special tax measures dated back to the prewar period when the Temporary Tax Measures Law was enacted in 1937. During the war these measures were substantially expanded in co-ordination with the war effort, and were continued in the form of the new Special Tax Measures Law after the war. Naturally, the number of special measures was sharply curtailed when the Shoup tax reform was carried out in 1950. Only a few measures survived at that time, including a tax-free reserve for bad debts and tax-exempt interest income of specific types.
38
Overview
corporate income taxes. However, it must be noted that the same type of incentive measures are often found in the ordinary income tax laws as well. In fact, at present, a number of significant tax benefits are contained in the ordinary income tax laws, rather than in the Special Tax Measures Law. Thus, it is rather difficult to distinguish those tax provisions that are considered special from those that are not. In what follows, I shall use the term 'tax incentives' chiefly to refer to those taxes that the Ministry of Finance (MOP) calls 'special tax measures'. These can also be termed 'tax preferences'. Each item included in the 'special tax measures' is drawn from both the Special Tax Measures Law and the ordinary income tax laws, in accordance with the specific definition of policy incentives adopted by the MOF.3 There is no comprehensive list of 'tax expenditures' for Japan comparable to that contained in the US budget. The scope of'special tax measures' seems to be narrower than that of'tax expenditures' in the USA in terms of tax incentive policy (see Office of Management and Budget 1986). One major aspect of the special tax measures was their proliferation, coming as a result of the political bargaining that plays an important role in the process of Japanese decision-making. Once a particular incentive was approved, people with similar needs requested similar tax preferences for themselves. This process was repeated many times. For instance, the introduction of a deduction on life insurance premiums emerged initially from the needs of life insurance companies. Once they received special treatment, marine and fire insurance companies also demanded a special deduction for fire and other casualty insurance premiums to maintain equal positions in the market. In this way, special tax measures proliferated in the 1950s and 1960s. (For a more detailed discussion, see Ministry of Finance 1991.) Significance of the Tax Incentives Is there any quantitative factor that can indicate how significant the tax incentive measures have been so far? For this purpose, a series of revenue losses caused by the special tax measures, which are compiled annually by the MOF, are of great use. Table 3.1 shows the trend of estimated revenue losses from the special tax measures in individual and corporate income taxes. From these estimates, it is obvious that the revenue loss resulting from the special tax measures was 13.2 per cent of total income tax revenues in the late 1950s, fell to 10.3 per cent in 1960, rose to 12.0 per cent in 1965, and then declined to 4.0-5.0 per cent in the 1980s. In the 1990s, however, it rose to 5.0-6.0 per cent once again. In a word, a long-term declining trend can be observed clearly throughout the postwar era. This trend reflects the 3 The definition of special tax measures was officially announced when the Tax Advisory Commission presented their tax report in 1976. In both the Special Tax Measures Law and the ordinary income tax law, a great deal of effort was taken to draw the distinction between incentive measures relating to specific policy purposes and the ordinary tax structure. The tax staff at the MOF assert that the distinction between the two is clear-cut, given their professional knowledge and understanding of the income tax law (see Tax Advisory Commission 1976, 3-4).
39
Two Strategies of Postwar Tax Policy Table 3.1 Comparison of estimated revenue loss from special tax measures and total individual and corporate income tax revenue, 1958-1998 Fiscal year
1958 1960 1965 1967 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997b 1998°
Total individual and corporate tax revenue (¥b.)
586.3 964.0 1 897.5 2597.6 4014.3 4995.3 5445.6 6718.3 9850.2 11 166.6 9610.2 11004.5 12144.6 15665.9 16657.9 19722.3 20802.8 21980.1 23 467.3 25404.0 27455.7 29917.8 33 247.9 36391.9 40374.8 44379.1 43344.4 36945.0 35824.4 32780.6 33250.5 33448.2 34288.0 35829.0
Revenue losses from special tax measures" Amount (¥b.)
% of income tax revenue
77.6 99.1 228.2 255.4 357.7 434.5 532.7 580.4 645.0 727.0 796.0 759.0 840.0 933.0 959.0 1 026.0 1 121.0 1149.0 1 193.0 1281.0 1525.0 1 562.0 1 528.0 1 694.0 1 836.0 2121.0 2355.0 2183.0 2097.0 2034.0 1 923.0 1 799.0 1 994.0 1 864.0
13.2 10.3 12.0 9.8 8.9 8.7 9.8 8.6 6.5 6.5 8.3 6.9 6.9 6.0 5.8 5.2 5.4 5.2 5.1 5.0 5.6 5.2 4.6 4.7 4.5 4.8 5.4 5.9 5.9 6.2 5.8 5.4 5.8 5.2
a
Includes only the items listed as special tax measures by the Tax Bureau, excluding revenue gains from the curtailment of corporate special and entertainment expenses. Preliminary figures. c Estimated on the basis of budget data. b
Source: tax revenues, from Ministry of Finance Tax Bureau (1998); revenue loss from special tax measures, from data presented to the Budget Committee, National Diet by the Tax Bureau, MOF.
40
Overview
fact that many of the special tax measures for specific purposes have been eliminated in the past decade to make the tax system more neutral and equitable. Objectives of the Tax Incentives The MOP classifies the special tax measures of individual and corporate income taxes into six different objectives: (1) promotion of saving; (2) promotion of environmental quality and regional development; (3) promotion of natural resources.; (4) promotion of technological development and modernization of industrial equipment; (5) strengthening the financial position of firms; and (6) other incentives. This classification, however, is not useful for economic analysis. Some attempt must be made to rearrange these into more meaningful categories. Table 3.2 summarizes the percentage distribution of revenue losses from the special tax measures, dating back to 1958, for which the relevant data are available. The first category, the promotion of individual saving and housing, is tied exclusively to the individual income tax. The other categories are related mainly to the corporate income tax. Throughout every time period, the first category has maintained the highest share. In addition, it has recently shown a trend towards an increasing relative share, because the other tax incentives in the corporate category have been substantially eliminated. The second-largest share of revenue loss arising from tax incentives is related to the promotion of business saving and investment. The tax devices used to promote these activities include tax exemptions an credits, accelerated depreciation, and tax-free reserves. These devices are generally used in particular industries or specific activities, and in many cases two or three measures are employed to promote the same objective. Third, the promotion of export and foreign investment has greatly declined in importance, although in earlier periods it was a major consideration. In the 1950s and 1960s, devices to promote export and foreign investments were prevalent as special deductions of export income from taxation and accelerated depreciation for export-oriented firms. These tax provisions were finally eliminated by the early 1970s. Today, remaining provisions for exports and overseas investment are very minor. Consequently, their relative share in Table 3.2 has continued to fall over the long term. The fourth category includes measures concerned with the promotion of environmental quality. In the early 1970s, the Japanese economy began to experience the side-effects of rapid economic growth, such as rampant inflation, urban congestion, and environmental pollution. Popular attention was focused on the prevention of air and water pollution in order to preserve the quality of the environment. Since that time, the share of tax devices relating to pollution control sharply increased, reaching a peak of 9.2 per cent in 1975. Typical measures include special additional depreciation and tax-free reserves for pollution control activities.
41
Two Strategies of Postwar Tax Policy
Table 3.2 Percentage distribution of the estimated revenue loss from special tax measures, by type of incentive, 1958-1997 (%) Fiscal year
1958 1960 1965 1967 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
Promotion of individual saving a housing (1) 49.5 45.6 59.9 61.0 54.6 47.0 44.5 44.7 49.6 49.4 44.6 47.0 48.2 48.1 49.7 53.6 57.4 58.0 57.3 54.6 59.0 59.8 57.4 59.3 58.2 60.3 61.8 61.2 61.7 61.8 66.0 62.3 63.1
Promotion of business s investment (2) 25.8 35.9 18.7 20.9 18.2 18.1 19.9 27.5 25.1 21.4 22.2 18.8 18.9 18.3 22.4 20.6 20.2 21.1 23.4 26.0 24.2 24.2 29.1 26.4 28.7 22.8 21.2 22.0 22.4 19.7 18.5 18.3 18.2
Promotion of export and f investment (3)
Promotion of environment q
Others
(4)
(5)
18.0 12.6 12.9 10.9 14.8 19.3 17.0 4.7 3.8 5.5 6.5 5.5 4.2 2.8 2.1 1.9 1.7 2.3 1.6 2.4 2.2 1.7 1.9 1.4 1.0 0.8 0.6 0.4 0.6 0.4 0.8 0.9 0.7
0 0 0.3 0.8 1.5 1.6 5.8 6.9 7.3 8.3 9.2 6.7 4.6 4.4 5.5 3.6 4.2 4.3 4.1 4.2 4.1 4.1 2.9 3.2 2.8 3.5 4.0 3.9 3.9 3.5 3.5 3.2 3.2
6.7 5.9 8.2 6.3 10.9 14.0 12.8 16.2 14.2 15.4 17.5 21.9 24.0 26.4 20.2 20.4 16.5 14.4 13.7 12.7 10.6 10.2 8.7 9.7 9.3 12.7 12.4 12.5 11.4 14.6 11.2 15.3 14.8
Note: The official classification was rearranged into five items as listed above. Source: As Table 3.1.
42
Overview
Finally, other miscellaneous measures have accounted for 10-20 per cent of revenue loss since the 1970s. Mostly, these consist of measures such as the special deduction of physicians' fees, which are not related to the corporate income tax. ANNUAL
TAX REDUCTIONS
Past Trends Turning to the second strategy, let us focus on the annual tax-cutting policy. The growing economy of postwar Japan automatically generated large increases in annual tax revenues. Given the fact that nominal GNP rose by an average of 15 per cent a year in the 1950s and 1960s, tax revenues were growing annually at a rate of more than 15 per cent (because the income elasticity of total taxes usually exceeds 1.0). If the tax system had remained unchanged, the government would have obtained fiscal resource increases annually in excess of 20 per cent. In Japan, these abundant revenues were never used to expand the size of the government sector as they have been in many other countries. In fact, the Japanese government adopted a tax-cutting policy to keep the ratio of total tax revenues (including both national and local taxes) to national income constant (i.e. at 20 per cent), rather than simply to increase public expenditures. This guideline was followed especially strictly during the period between 1955 and 1965, and as a result the ratio of total taxes to GNP remained relatively constant during those years. The government therefore had the luxury of providing successive annual tax reductions. It is widely believed that, if annual tax reductions had not been implemented, the individual income tax would have unduly overburdened taxpayers. Thus, tax reductions were very popular with Japanese taxpayers during the high-growth period, incurring envy in many other countries. Table 3.3 summarizes estimated annual tax changes since 1950. These estimates of tax decreases or increases are calculated as the effects of tax actions, using official economic forecasts, that were enacted at the beginning of each year. Before 1975, total tax revenues were reduced annually at both national and local government level with only minor exceptions. This reflects the fact that until the late 1970s a strong sentiment prevailed not to increase national and local taxes even as potential revenues increased. Interestingly enough, annual tax policy since 1976 has adopted the opposite stance towards tax increases in accordance with the new direction of fiscal consolidation. The trend of tax increases, however, was ended in fiscal 1987, and thereafter tax reduction policy was initiated once again by using increased revenues due to the 'bubble boom' until 1991. In the 1990s after the collapse of the bubble, fiscal stimuli packages led to successive rounds of large tax reductions to buoy up the depressed economy. Cuts in individual income taxes in a growing economy prior to the early 1970s occupied the lion's share of annual tax reductions. In comparison with the continuous long-term reductions in individual income tax, both the corporate income tax
43
Two Strategies of Postwar Tax Policy Table 3.3 Estimated annual tax changes, fiscal years 1950-1998 (¥bn.) Fiscal year
National taxes Total
1950 1952 1954 1956 1958 1960 1962 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 a
Local taxes
-206.8 -89.5 -16.9 -1.5 -37.3 6.6 -116.4 -118.2 -115.1 -310.6 -94.0 -68.0 -181.0 -205.5 -108.5 -4.6
-378.1 -1 115.0 -372.0 383.0 81.0 463.0 634.0 367.0 1531.0 308.0 7.0 157.0 187.0 131.0 -599.0 -1725.0 -298.0 -306.0 -4.0 473.0 101.0 -4386.0 36.0 -1 586.0 -164.0 -2092.0"
Individual income tax -135.8 -112.7 -31.4 -22.6 -6.3 0.0 -50.3 -74.5 -65.4 -158.3 -92.5 -125.1 -183.0 -288.7 -206.9 -32.2 -375.2 -1783.0 -186.0 0.0 -141.0 -12.0 80.0 22.0 -13.0 24.0
-1.0 -770.0 24.0 -30.0 -464.0 -1838.0 -30.0 -173.0 0.0 -7.0 -32.0 -3845.0 2.0 - 1 400. -82.0 -1 494.0
Corporate tax
-24.4 19.1 2.6 14.4 21.5 0.0 -1.3 -58.6 56.6 98.7 30.3 0.0 2.4 75.2 12.1 30.6 26.8 352.0 -6.0 115.0 110.0 49.0 218.0 371.0 640.0 324.0 36.0 414.0
194.0 51.0 -512.0 —1 112.0 -39.0 -83.0 1.0 29.0 38.0 88.0 42.0 26.0 16.0 259.03
Other direct taxes
13.9 -2.4 -2.9 0.0 -3.3 0.0 -2.4 -4.5 -0.5 -15.0 -3.1 0.0 0.0 0.0 -6.6 -12.1 -39.7 0.0 -298.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -6.0 0.0 -2.0 0.0 -584.0 0.0 0.0 0.0 371.0 -150.0 -3350.0 -4.0 -182.0 0.0 -134.0
provisional figures
Source: MOF, Primary Statistics of Taxation (Zeisei Shuyo Shiryoshu), Feb. 1992.
Indirect taxes
-60.5 6.5
20.0 6.7 -6.3 6.6 -62.4 19.4 7.4 -38.6 31.9 57.1 -0.4 8.0 92.9 9.1 10.0 316.0 118.0 268.0 112.0 426.0 336.0 -26.0 904.0 -40.0 -28.0 519.0 -31.0 112.0 377.0 1 809.0 -229.0 -50.0 -5.0 80.0 110.0 21.0 -4.0 -30.0 -98.0 -205.0a
n.a. -27.7 -26.3 12.3 20.0 -11.8 -39.9 -55.7 -8.7 -53.4 -19.3 -20.7 -94.7 -96.6 -87.0 -98.1 -145.7 -113.4 -466.8 34.7 -53.4 77.7 181.5 213.2 153.7 54.8 56.0
31.4 89.4 42.3 117.5 -2214.9 -31.8 -107.8 -722.3 4.8 263.4 - 1 564. 19.2 -640.5 2.6 -806.3
44
Overview
and indirect taxes show varying changes over the years. Corporate income tax was increased more frequently, with successive tax increases beginning as early as 1969. Indirect taxes were also often lifted until 1970, chiefly because most of them were levied on a specific basis and periodically needed to be adjusted for the rise in commodity prices. This tendency, however, has been reversed since mid-1970. Other direct taxes, including inheritance and gift taxes, showed no significant variation as a whole except for 1975. Increases in revenue from corporate and indirect taxes were small relative to income tax reductions in the 1950s and 1960s. Consequently, tax policy during the high-growth period was characterized by net annual tax reductions, owing to the major impact of individual income tax reductions in total tax collections.
High-income Elasticity of Tax Revenues As mentioned, one of the most notable features of Japan's tax policy in the 1950s and 1960s was the government's annual tax reductions. Even though a substantial amount of tax revenues was cut annually, however, this did not reduce the size of the government sector. This fortunate circumstance can be explained largely by the rapid growth in revenue, which financed both tax cuts and expenditure increases during the postwar period. At this point, therefore, we must determine whether there are any significant features of the Japanese tax system that render its revenues especially responsive to income growth. The tremendous growth of Japanese tax revenues has been achieved automatically, owing to the highly elastic income taxes and the high rate of nominal GNP. Thus, frequent tax reductions, as shown in Table 3.3, were required periodically to avoid overburdening taxpayers by large revenue increases. To make this mechanism clear, it is necessary to investigate how responsive the tax system has been to income growth. In general, the income elasticity of tax revenues is used to measure the responsiveness of tax revenues to the growth of nominal GNP. The formula for elasticity
where Y= nominal GNP, and T = tax revenues. From the elasticity formula (3.1), we obtain the growth rate of tax revenues:
This rate is indispensable for estimating tax amounts each year in the process of compiling the annual budget. The necessary amount of taxes can be calculated by multiplying ET, if this value is definitely determined by the specific rate of A17Y, which is based on economic forecasts by the Economic Planning Agency.
Two Strategies of Postwar Tax Policy
45
The response of taxes to a change in nominal GNP logically takes place during two stages: (1) the response of the particular component of national income—i.e. taxable income—on which the tax is based to changes in GNP, and (2) the response of the tax yield to a change in the tax base. In order to observe these factors more closely, let us divide (3.1) into the two stages as follows. Thus, we obtain
as the new expression for ET, where B stands for the tax base,
The elasticity of the tax yield ET can be represented as the product of the elasticities of (1) the tax yield with regard to changes in the tax base Et and (2) the tax base with respect to changes in total income Ef,. Thus, Er varies directly with the elastici
t and Eb. The first 'tax-rate elasticity' relies heavily on the statutory-rate for mula of a progressive schedule of income tax rates. The second 'tax-base elasticity', however, is determined primarily by the way in which the private sector works during boom and slump periods. The value of ET tends to vary considerably from one tax to another. In the case of individual and corporate income taxes, it is generally considered to be high, depend ing on the rates and bases of the taxes. The terms for Et and Eb given in expression (3.3) could be the major reason for the high value of Ej- in the direct taxes. A progressive individual income tax should produce proportionately greater swings in tax revenues than in income. This is a typical characteristic of progressive rates, and consequently Et>\. On the other hand, the high ET of the corporate income tax seems to indicate that £;&> 1, since it shows the volatile changes in the level of corporate profits over the cycle. As regards the indirect taxes, it appears that the values of Ej, are lower with few exceptions. Neither Et nor Eb can rise to any significant degree because of their proportional or regressive rates and stable tax bases (see e.g. Musgrave 1959k, 506-7; Lewis 1962, 28; Maxwell 1955, chs. 12-14). Since the concept of elasticity used above deals with the automatic aspects of tax response, the calculation of ET requires estimates of that part of the change, in actual recorded data, which results from automatic rather than discretionary actions. Therefore, the practical measurement of ET raises some serious difficulties The most serious is that the data on tax collections in postwar Japan reflect not only an automatic response to changing income, but also tax reductions enacted in the tax code almost every year. This does not allow for a separation of the effects arising from changes in the tax code. In general, the tax revenue T may be assumed to depend on income Y, the minimum level of exemptions and deductions e, and statutory tax rates t. In other
46
Overview
words, T can be shown to function in the following way:
In measuring T for ET, T should be calculated on the assumption that e and f are fixed by the institutional setting. This entails using the tax revenue as reflected in income changes arising from the 'fixed tax system'. However, it is very hard to assume the tenability of a fixed tax system in the long run. As mentioned earlier, many tax changes have been enacted in postwar Japan, and these make such an assumption unreasonable. Under such circumstances, we cannot hope to measure any elasticity in the strict sense of the term. Actual tax revenues must be adjusted for any changes in tax laws. An attempt has been made by the Tax Bureau in the Ministry of Finance to develop a method for adjusting revenues for tax changes. It is a bold and crude attempt, but it is the only one that has yet been made and whose details are presently available in Japan. The Tax Bureau adjusts tax revenues for statutory changes that are made in the tax code each year. Actual taxes are converted to estimated tax accruals assuming the absence of tax code changes; i.e. tax rates, exemptions, and deductions remain unchanged. In calculating the value AT in (3.1), estimated tax accruals in the current year are compared with actual taxes in the preceding year. In other words, the calculations are based upon estimated tax receipts before the annual tax reductions.4 These estimates provide measures of the year-to-year elasticities of national taxes to changes in nominal GNP (or GDP) on the basis of the annual figures. Table 3.4 lists them for each category of major national taxes during the period 1953-98. Despite their uncertain accuracy,5 these elasticities are very important because the Japanese government has been using them as an indicator for forecasting tax amounts in the budget process every year. As can be seen from each column in Table 3.4, the elasticities show a high degree of variability from year to year, reflecting mainly volatile, short-run factors. The approximate magnitude of the long-run elasticities can be inferred by averaging the annual figures. To observe the long-run tendency of elasticities, it is informative to distinguish between the periods before and after the first oil shock.6 Over the period 1965-74, the elasticity of the whole tax system averaged 1.32, while the same figure fell to 0.85 for the period 1975-85. However, it jumped up to 1.94 for the second half of the 1980s, due to the outbreak of the 'bubble economy'. Once again it declined to 0.28 for 1991-8 after the bubble boom was terminated. 1 Lewis (1962, 29-31) made the adjustments simply by multiplying actual recorded tax accruals by the ratio of pre-change to actual tax rates. Pechman (1956,144) also made the adjustment for tax revenues in terms of an 'index of tax rates', which was based on estimates prepared by the Treasury Department. 5 There has been some criticism of the accuracy of the method used by the Tax Bureau series. Ishi (1968) attempted to recalculate relevant elasticities, selecting specific periods of'no tax change' on a quarterly and annual data basis. 6 Pechman and Kaizuka (1976, 349) tried to average year-to-year elasticities for a five-year period by using a specific formula in order to remove the erratic fluctuations in the annual data.
Two Strategies of Postwar Tax Policy
47
Table 3.4 Year-to-year income elasticity of the national tax system, 1965-1998 All taxes
1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 Ave. 1965-74 Ave. 1975-85 Ave. 1986-90 Ave. 1991-98 a
0.67 1.19 1.51 1.27 1.41 1.58 1.35 1.38 1.93 0.92 -0.34 0.99 1.13 1.01 1.80 1.25 0.56 0.82 1.37 1.10 0.95 1.91 3.33 1.96 1.48 1.03 0.04 -5.21 -1.30 5.75 1.09 0.14 0.67 1.08 1.32 0.85 1.94 0.28
Individual income tax
1.75 1.48 1.76 1.85 1.94 2.25 3.29 1.96 2.52 1.55 1.64 0.99 1.41 1.23 2.14 1.84 1.97 1.39 1.67 1.25 1.44 1.78 2.69 1.57 2.80 2.59 0.63 -7.05 1.50 5.50 -1.82 -1.11 1.67 1.21 2.04 1.54 2.29 0.07
Corporate income tax
-0.19 1.17 1.82 1.37 1.41 1.59 -0.15 0.91 2.37 1.10 -2.96 1.22 1.20 0.83 1.71 1.70 -1.29 0.08 1.37 1.46 0.51 1.70 4.98 2.88 0.59 0.15 -1.48 -9.00 -12.40 8.25 5.91 1.64 2.22 2.17 1.14 0.53 2.06 -0.34
Indirect taxes
0.22 0.89 1.01 0.67 0.87 0.83 0.71 0.96 0.87 -0.28 1.00 0.69 0.71 0.88 1.55 0.01 0.51 0.51 0.60 0.34 0.46 1.87 2.02 1.04 0.79 0.06 0.11 1.26 3.30 8.25 1.41 1.18 -1.00 0.33 0.68 0.66 1.16 1.85
Alcohol tax
-0.09 0.73 0.57 0.23 0.49 0.63 0.44 0.73 0.71 -0.11 1.21 -0.11 0.82 0.40 1.09 -0.29 -0.20 0.96 0.53 -1.81 0.42 0.50 1.15 0.85 0.55 0.58 0.36 0.26 -0.40 5.75 -1.64 0.00 0.78 0.08 0.43 0.27 0.73 0.65
Indicates the average of the 1957-74 period.
Note: Income elasticity is the ratio of the percentage of tax revenues, based on estimated tax liabilities before the annual tax reductions, to that of GNP (1953-85) and of GDP (1986-98). Source: Unpublished data from the MOF Tax Bureau.
48
Overview
The responsiveness of Japanese tax revenues to income growth is due primarily to the high elasticity of the individual income tax, showing 2.04, 1.54, and 2.29, respectively, for the three periods defined above. On the other hand, the elasticity of the corporate income tax reveals more volatile figures. In fact it conspicuously fell to 0.53 in the second period, while it greatly increased to over 2.06 in the third period. Corporate income tax elasticity responds closely to sharp drops in business conditions and even shows negative figures during periods of slump, such as in 1965, 1971, 1975, and 1981. Indirect taxes and the tax on alcoholic beverages are less responsive to income fluctuation, and their elasticities are less than unity, as was pointed out in the theoretical discussion above. While income growth slowed down during 1991-8 due to the collapse of the bubble boom, the elasticities of individual and corporate income taxes have indicated substantial drops to 0.07 and —0.34 from the higher values of 2.29 and 2.06 in the second half of the 1980s. These facts imply how deeply income tax revenues can be affected by economic growth. In contrast, both indirect taxes and alcohol tax show relatively stable values of elasticity. Biased Estimates of Tax Revenues We must now turn our attention to the important concept of the 'natural increase' in tax revenues. Taxes are usually estimated at the time when the budget is being drawn up at the end of each calendar year. During each fiscal year in the 1950s and 1960s large 'natural' tax increases were realized, providing a substantial amount of new financial resources for the current and following year's budget. As noted above, part of these tax increases were appropriated to the financing of new government programmes, while the remainder were devoted to the financing of tax reductions. Therefore, this phenomenon of 'natural' tax increases bore a close relation to the size of the annual tax reductions. The natural increase in tax revenues seems to have been created intentionally, since the government consistently underestimated the tax amounts necessary to finance government expenditures at the first stage of budget planning. Natural tax increases, therefore, occurred during the fiscal year as a result of biased estimates of tax revenues. Some tax revenues, such as individual income taxes, naturally register increases as the tax base expands with the growth of the economy, even if there are no changes in the tax rate or exemptions. The higher the rate of economic growth, the larger is the amount of natural increase in tax revenues that can be expected. A question is raised about the estimation of the natural tax increase. This is based largely on the anticipated rate of economic growth, which is usually computed five or six months before the beginning of fiscal year. As an illustration, let us suppose that the government believes that GNP will expand 12 per cent in the following year. Based upon this anticipated rate, the MOF usually estimates projected natural tax increases. When doing so, some non-economic biased factors can easily be introduced into the calculation of the anticipated rate of economic growth. Frequently, the GNP growth
Two Strategies of Postwar Tax Policy
49
Fig. 3.1 Actual and anticipated rates of economic growth, 1955-2000 Note: In calculating the actual rate, the SNA data are used after 1965. The GNP series is published at three different stages, depending on the time of estimation: Y" stands for preliminary GNP, Y' for interim GNP, and Y for realized GNP. Anticipated rates of economic growth in n periods are defined as
n-rn i)/i"n i, while actual rates are as (Yn-Yn i)/Yn__l. Source: National Budget (MOF), Annual Report on National Accounts (EPA).
rate is intentionally underestimated in order to decrease the expected amount o natural tax increase used as a financial resource at this stage of budgetary preparations. Thus, since at the end of each fiscal year the realized rate of growth is always much higher than the anticipated rate, an enormous natural increase in tax revenue materializes after the implementation of the new budget. Figure 3.1 illustrates both actual and anticipated rates of economic growth from 1955 to 2000. Actual rates of economic growth markedly exceeded anticipated rates before 1974, but this relationship was reversed for about one decade after 1975. Hence, the gap between the anticipated and actual rates of economic growth substantially contributed to the production of large national tax increases in the 1950s and 1960s. In contrast, the government has occasionally suffered from a slowdown in revenue growth for a decade since the late 1970s. The large-scale revenue shortages in 1981 and 1982 mainly reflected the slowdown of income growth (see Ish 1982). Once again, anticipated rates have surpassed actual ones, generating a number of revenue losses in the 1990s.
TOWARDS TAX E Q U I T Y AND NEUTRALITY
The Emergence of Fiscal Deficits
To a considerable degree since the late 1970s, the basic strategy of tax policy has shifted from tax incentives to tax equity and neutrality. The main reason for this
50
Overview
Fig. 3.2 The gap between government expenditures and tax revenues, 1960-1998 Notes: Tax revenues include non-tax revenues, and all figures are related to the scope of general account in the national govenment. Source: MOF Budget Bureau, Fiscal Statistics (Zaisei Tokei).
is that fiscal deficits have begun to grow, reflecting a conspicuous slowdown in economic growth caused by the two oil crises (see Ishi 1986a; Noguchi 1986, 1987a; Lincoln 1988). To curtail the growth of fiscal deficits, the Japanese government terminated the annual reductions in individual income tax, and gradually began to increase corporate taxes and indirect taxes (see Table 3.3). As a result, the tax incentive policy necessarily came to a close. Traditionally, the Japanese government has followed a balanced budget policy. The roots of this policy lie in a conservative view of'sound' finance which developed in reaction to the extravagant government spending and inflationary pressures during the wartime period and immediately afterwards. A balanced budget was maintained in the true sense of the term until 1965, when national bonds were first issued in the postwar period. Although national bonds have been floated periodically since then, fiscal deficits posed no serious problems for a while.7 But they did become serious after the mid-1970s. Figure 3.2 depicts government expenditures and tax revenues during the period 1964-98. The gap between the two lines, which corresponds roughly to fiscal deficits, began to expand rapidly at the outbreak of the 7 Needless to say, the concept of fiscal deficits differs in accordance with the scope of the government (the 'general account' of either the national government or the 'general government', which includes all levels). In this regard, general account deficits are generally used when focusing on policy issues in lapan. For more detailed discussion, see Ishi (1986fo).
Two Strategies of Postwar Tax Policy
51
first oil shock in 1973. The situation has worsened since then. Since the collapse of the bubble boom in 1991, the gap has continued to expand to a great extent, reflecting revenue shortages in recession. What are the main causes for the sharp rise of fiscal deficits since 1973? Many factors can be put forth, but two are the most important. First, there was a conspicuous slowdown of Japanese economic growth caused by the two oil crises. Indeed, the real growth rate declined sharply, from 10 to 5 per cent in 1973, and to 3 per cent in 1979 when the second oil shock struck the Japanese economy. These experiences resulted in a sharp reduction of revenue growth, which in turn contributed greatly to the increase in fiscal deficits during the early 1980s. The second factor concerns the expanded role of government in income maintenance, medical care, public pensions, etc. In the early 1970s, important institutional reforms were undertaken in the social security system with a slogan, 'Construct the welfare state in Japan'. The target was to catch up with the Western level of social welfare programmes. Previously, it was often pointed out that Japan had lagged behind Western countries in the development of social welfare policies. Ironically, it was in 1973 that new social welfare programmes were launched and expanded to the same size as those in Western countries. At the time, it had been expected that the high rate of economic growth would continue and would thus generate the additional resources needed to finance higher public expenditures. Unfortunately, in the decade that followed, the rate of growth fell considerably in Japan as in most industrial countries. Growing Emphasis on Equity and Neutrality Because of the huge debt accumulation, the Japanese government decided that a substantial tax increase could not be avoided in the future in order to curtail fiscal deficits. At first, the government intended to resolve the fiscal deficits problem by instituting tax increases, rather than by making expenditure cuts. The introduction of a value added tax (VAT) was attempted in 1979, but resulted in a complete failure. As a consequence, it became politically difficult to introduce tax increases sufficient to reduce the level of national debt, and therefore the government changed its policy stance from advocating tax increases to promoting expenditure cuts.8 At the same time, the Japanese people demanded that the government should take the initiative in revising the inequitable burden of income taxation as a prerequisite to future tax increases. Because of these requirements, postwar tax policy was forced to shift towards putting greater emphasis on tax equity and neutrality. These trends gave rise to two important issues. One concerns the manner of reducing the number of special tax measures for tax incentives; the other concerns the perception of increasing unfairness of income taxation. As for the former, there 8 As regards the introduction of VAT and subsequent fiscal performances, see, for an expanded discussion, Ch. 15.
52
Overview
has been a growing realization that special incentives have eroded the fairness and neutrality of the tax system, and have made it increasingly complex. As noted earlier, Japan introduced a tax system after the Second World War which had relatively few special exemptions, credits, or similar provisions, under the influence of the Shoup Mission. Gradually, however, an interventionist approach to taxation had resulted in the enactment of various special tax measures designed to promote specific policy goals. The appearance of a renewed emphasis on tax equity and neutrality, with restricted use of special provisions, indicates that there has been a reversal of the interventionist trend in recent years. This trend is illustrated in Table 3.1 by a sharp decline in revenue losses from special tax measures in the 1980s. Concerning the latter issue, the Japanese government has been under increased pressure to address perceived inequities in the tax system. Complaints from taxpayers have centred round three issues and have concerned mainly the individual income taxes paid by salaried workers. First, there have been widespread complaints about the problem of 'bracket creep', which refers to the phenomenon by which salaried workers are pushed into higher brackets when they receive wage hikes for inflationary adjustment. Second, the problem of bracket creep is exacerbated by the fact that a large number of the self-employed and farmers have been able to escape their tax burden. In general, non-wage-earners, such as farmers and the selfemployed, can manipulate their declared taxable income to reduce their tax burden, whereas salaried workers cannot because their income tax is withheld at the source; as a result, they are considered to be subject to the utmost taxation by the tax office. Third, tax inequities among various income groups have been aggravated by the income-splitting practices available only to business owners and the self-employed. These groups are allowed to split their income by paying salaries to family members and even themselves, thus lightening their tax burden. Furthermore, they are allowed to deduct their business expenses to a greater extent than salaried workers, who are permitted only a standard deduction. As can be seen, salaried workers have major complaints about the present income tax system in terms of fairness.
4
Tax Administration and Tax Equity The aims of this chapter are twofold. One is to analyse how efficiently the Japanese tax system is administered from an international perspective. A major concern is with the well-established withholding tax system which may be the main factor enhancing the efficiency of tax administration. The other is to shed light on the reverse side of the withholding system and to explore the views of different taxpayers on the unfairness of the burden from a standpoint of equity. 'Tax gap' phenomena in Japan are also estimated using the available data to provide some empirical evidence. The plan of Chapter 4 is as follows. First, empirical findings which support the efficient aspects of Japanese tax administration are presented in comparison with those of the other major countries in the second section. Secondly, the third section analyses the well-established framework of the withholding system in Japan in institutional settings. Thirdly, empirical evidence of the 'tax gap' phenomenon between different sources of income with a discussion on policy is presented in the fourth section. Lastly, the discussion is concluded by proposing a desirable tax structure.
E F F I C I E N T TAX C O L L E C T I O N : A N I N T E R N A T I O N A L COMPARISON
Two Types of Tax Collection
Generally speaking, tax collection established two principles which have remained fundamental to tax administration in the leading industrialized countries: (a) withholding at source, and (b) self-assessment on a tax return basis. Income is divided into two categories: wages, salaries, interest, and dividends withheld at source; and other incomes required to file a return. It is widely acknowledged that the withholding tax system improves efficiency in collecting tax revenues by saving costs. Payers of income subject to taxes withheld at source are responsible for collecting the tax and paying it over to the government on behalf of each taxpayer. By contrast, under the self-assessment system each taxpayer is required personally to assess taxable income, file a return, and pay the tax due to the tax office. From an administrative point of view, tax collection is much more expensive under the self-assessment system because most of the collection work must be done by the revenue staff. This chapter is based on Ishi (1992fc).
54
Overview
Under these two different methods, taxpayers with income withheld have no freedom of manipulating taxable income, while self-assessed income-earners have a substantial margin to manipulate their income for tax purposes. Given the current state of tax collection, the former must feel that the basic rule of horizontal equity, that equals should be treated equally, is not being observed. For example, income tax of employees is administered under the withholding system, and unless the employer co-operates in evasion it is difficult for an employee to pay less than the tax due. However, in any country, avoidance and evasion on self-employment income including agricultural income are much more common; with the exception of a few abnormal cases, only a minority of taxpayers making tax returns are subject to tax inspection. There seems to be a general belief that income-earners who file their own taxes pay less tax than those whose taxes are withheld at source. Thus, different collection systems tend to induce a potential conflict between efficiency and the fairness of tax administration. This is obviously true in Japan. The main cause behind this fact is that the Japanese tax system has developed the withholding system to a greater extent over a broad area of primary taxes. Administrative Costs It is very difficult to investigate in quantitative terms how efficient tax administration is. To some extent, this analysis may be pursued by the measurement of collection costs, although this measure is far from satisfactory. Needless to say, costs of various kinds arise from the collection of individual taxes and the existence of the tax system itself. In what follows, particular attention will be paid to public sector costs (i.e. administrative costs), chiefly because necessary data are not available to cover compliance costs. To explore efficient aspects of tax administration, account must be taken that compliance costs in the private sector play an equally important role in collecting taxes. Compliance costs are less transparent than administrative costs, but they are significant in the process of administering the tax system as a whole. In particular, since there is a substantial degree of transferability from administrative to compliance costs, the government often tends to cut the former at the expense of the latter. Therefore, consideration of compliance costs should not be neglected for the purpose of our argument, but unfortunately there are no reliable data at hand.1 Our empirical analysis will therefore be based on only one aspect of collection costs. Let me begin with a preliminary discussion on the relative size of administrative costs compared to tax revenue. In general, administrative costs are officially calculated by the revenue department of a country, such as the Internal Revenue Service (IRS) in
1 Japan's tax authority is still cautious in exploring the actual situation of compliance costs. Thus we cannot obtain any relevant data in Japan at a similar level to the attempt by Sandford et al. (1989).
Tax Administration and Tax Equity
Fig. 4.1
55
Administrative cost as a percentage of tax revenue, selected countries, 1960-1995
Note: Each figure is calculated as a percentage of costs per $100, £100, or ¥100. Sources: USA: Internal Revenue Service, Annual Report, 1989 and 1990. Canada: Revenue Canada Taxation, Inside Taxation, 1975 and 1989; Supply and Service Canada, Report of the Department of National Revenue Customs Excise and Taxation, 1977-9 and 1981. UK: Board of Inland Revenue, Report for the Year, 1970,1972-5,1977-85, and 1987-90; Customs and Excise, Report of the Commissioners of Her Majesty's Customs and Excise, each year. Japan: National Tax Administration, Annual Report of Statistics, 1970, 1985, and 1990.
the USA. This includes wages and salaries of staff, accommodation costs, travel, postage and telephone, computing, and equipment costs. Therefore, based on official data, an approximate international comparison becomes feasible in terms of administrative costs as a percentage of tax revenue. Figure 4.1 shows the movements of such a ratio at the central government level in the USA, the UK, Canada, and Japan mainly during the period 1960-97. Unfortunately, it is not possible to obtain a detailed breakdown of data by individual taxes except for the UK where direct and indirect taxes are separated. In view of the possible different coverage of costs in each country, a strict comparison is not possible, but two interesting points are worth noting in the case of Japan. First, the ratio of costs to revenue until the early 1970s was higher in Japan than in the other countries. In particular, it was at a much higher level than that of the USA where the ratio has been kept very low. On this point, it seems that Japan's tax system was administered less efficiently in terms of the costs-revenue ratio. Second, however, the ratio in Japan declined sharply in the mid-1970s, and was lower than that of Canada. Thus, Japan now seems to lie between the USA and Canada. This comparison is a common procedure as the first step to present a measure of the efficiency in administering the tax system or of the relationship between input and output (i.e. 'productivity') in tax offices as a whole. It is, however, necessary to interpret the results derived from the data used in Figure 4.1 with care. Particular
56
Overview
attention should be paid to (a) nominal income growth and (b) changes in the tax system.2 The growth of nominal GNP generates additional tax revenue which tends to diminish the cost-revenue ratio. This was true until the early 1970s during which rapid economic growth was still prevalent in the Japanese economy. By contrast, the peak of the ratio around the mid-1970s implies that recession caused by the oil shock tended to decrease tax revenues and in turn led to an increase in the costrevenue ratio. If we emphasize the effects of income growth on changes in the ratio, the investigation of relative ratio would tell us nothing about the efficiency of tax administration. Tax changes are another and more important factor in exploring administrative efficiency in any time-series data. Tax reductions had been repeated almost every year before the outbreak of the oil shocks, producing a substantially decreased revenue, but towards the 1980s a deliberate tax-cutting policy was terminated to secure finance to make up for debt accumulation (see Ishi 1986«). Thus, changing the tax system led to automatic increases of revenue in the economy which resulted in a sharp decline in the cost-revenue ratio in the 1980s, as seen in Figure 4.1. However, it began to rise after 1990, reflecting the decrease in tax revenues. Another Measurement The next step is to make a more significant comparison by using per-staff basis data. Tax statistics in four countries provide us with three series of data: (1) the number of personnel in tax administration (N), (2) administrative cost (C), and (3) tax revenues (T), all of which are similar in the scope of coverage. Particular attention is paid to the gap between T/N and C/N, which implies a proxy of output and input per staff, respectively. If T/N is larger than C/N, the tax system is administered more efficiently and vice versa. Figure 4.2 shows the trends of these related data in four countries in terms of the base year= 100. Although the period covered and the scale of the index on a vertical axis substantially vary from one country to another, it seems that certain interesting facts can be derived from the comparison. Most importantly, great stress should be placed on the unique phenomenon of Japan's case in which T/N is consistently greater than C/N. On the other hand, the cases of the USA, Canada, and the UK turn out to be quite the opposite; the movement of C/N is in general over and above that of T/N, although the discrepancy between the two lines sometimes gets wider or narrower. In Japan, T/N has increased
2 See, for an expanded discussion, Sandford et al. 1989, pp. 19-20. In addition to these two points, the scope of revenues is also important. For instance, one of the main reasons why the US cost—revenue ratio has been kept much lower must be due to simultaneous collection of social security taxes with other taxes. By contrast, in Japan, the social security contribution is not included in tax revenues because it is collected by the Ministry of Health and Welfare.
Tax Administration and Tax Equity
57
more rapidly than C/N since the late 1970s, expanding the gap to a great extent. As stated above, the gap may be considered as evidence of the 'productivity' of tax staff. This being the case, the increased 'productivity' may indicate that the efficiency of tax administration has been improved in Japan as compared with that of other countries. One reason behind this is that the total number of national tax staff in Japan is much smaller. In fact, in 1996 the total was 55 108 in Japan, while the corresponding figures were 111 543 for the USA, and 93 138 for the UK in 1990. In addition, tax staff in Japan have increased by only a small margin: from about 52 000 in the 1960s and 1970s to about 55000 in recent years.3 In contrast to such a relatively stable number, tax revenues have enormously expanded, say, by more than 100 times during the same period. Needless to say, computer technology has enabled tax authorities to collect increased tax revenues with the very limited number of staff.
MAIN
F E A T U R E S OF A W E L L - E S T A B L I S H E D WITHHOLDING TAX SYSTEM
Current State of Tax Collection One of the distinctive factors explaining the efficient tax administration in Japan is obviously the withholding tax system which is firmly built into the basic structure of individual income tax. Historically, individual income tax was first introduced in 1887, and then the withholding system was adopted for interest income in 1899. Income withheld at source was widely expanded to cover employment income and dividends in 1940 when a sweeping tax reform was attempted by the government. Since the Shoup Mission recommended the overall reform package in 1949, the Japanese tax system has placed increasing importance on withheld income tax. In principle, the individual income tax is paid on a self-assessment basis, in which taxpayers themselves compute their income tax liability on the basis of their annual income and file a final return to the district tax office. As noted below, however, a major proportion of taxable income is now subject to being withheld at source. Taxes on such incomes as wages and salaries, interest and dividends are calculated by payers, deducted from the relevant source of income, and transferred to the tax offices on behalf of income-earners. At present, taxable income in the individual income tax is divided into ten categories, which are in turn classified into four types, depending upon tax collection methods. The relationship between taxable income and different types of collection are summarized in Table 4.1. Employment and retirement incomes (Type 1) are collected from payers on the basis of the withholding system on behalf of taxpayers—in its strict form. Such income is first computed comprehensively and progressive tax rates are strictly 3 The recent increase in staff reflects the fact that both the consumption tax (Japan's VAT) and the land value tax were introduced in 1989 and 1992.
58
Overview
Tax Administration and Tax Equity
59
Fig. 4.2 Comparison between per-staff tax revenue (TIN) and administrative cost (C/N) Notes: Temporary employees are included, and tax revenue excludes reimbursement. b Staff do not include temporary employees. Tax revenue is calculated by excluding revenues of the Canada pension plan and unemployment plan from total collections. c Casual staff are not included after 1968. '' Casual staff are included. 0 Okinawa regional tax offices are added after 1972. a
Source: As Fig. 4.1.
applied. Types 2 and 3 are kinds of interim measures, so far as the basic nature of comprehensive income taxation is concerned. In Japan, however, separate taxation at source has gradually become the practice in the case of specific incomes, mainly because of administrative considerations.
60 Table 4.1
Overview Tax collection methods by taxable income in 1996 (¥bn.)
Income
Withholding (Type 1)
Employment Retirement Interest Dividends Capital gains Stock etc. Others Real estate Timber Business Agricultural and self-employed Others Occasional Miscellaneous
262 660 11379
Subtotals
274039 (85.9)
Total
Separate taxation at source without final returns (Type 2)
with final returns (Type 3)
Filing returns (Type 4)
1 1 808 255
4003
1 010 6488
437 30 6345
28 7720 3407
297 3 136 13073 (4.11)
17034
14857
(5.3)
(4.7)
319003 (100.0)
Note: Figures in parentheses are percentage distribution. Interest and dividends contain some amounts that corporations have earned as well as individuals. Source: Calculated from NTA (1997).
Interest, dividends (some portion), and capital gains on the sale of stocks (Type 2) are separated from other incomes and tax is withheld at source by applying a flat rate of 20 or 35 per cent.4 Moreover, these incomes are excluded from the tax base in filing a tax return. Another type of taxation is applied to certain other dividends, capital gains on the sale of land and building, some business income, and miscellaneous income. Once tax on these incomes is withheld separately at a lower rate of tax,5 the incomes are required to be filed later as taxable income. The necessity of 4 All interest income and capital gains on the sale of stocks are withheld at source at the tax rate of 20 per cent (including 5 per cent of local tax) (see Ch. 8) while dividends attract 35 per cent. 3 In Type 3, dividends may alternatively be subject to the rate of 20 per cent at the taxpayer's option, but he/she must be required to file a final return. Capital gains on the sale of land and buildings are taxed separately on a self-assessed basis, depending upon the holding period of the relevant assets (see Ch. 17). Some portions of business income and miscellaneous income, which are mainly composed of fees, royalties, and remuneration paid to professionals, are withheld, usually at the rate of 10 per cent, as an advance taxation.
Tax Administration and Tax Equity
Fig. 4.3
61
Relative share of withheld income tax, 1960-1998
Note: Calculated from MOF, Primary Statistics of Taxation (Zeisei Sanko Shiryoshu), 1970, 1980, 1991, and 1998.
final returns distinguishes one type of separate taxation (Type 2) from another (Type 3). The amount of tax withheld under Type 3 is regarded as an advance payment of tax due, with final adjustment for the tax payable being made by incomeearners themselves as taxpayers. The remaining income (Type 4), from other categories of capital gains (e.g. the sale of valuable assets, paintings, or jewels), real estate, timber, and occasional, agricultural, or self-employed incomes, are levied under a self-assessment method on a calendar year basis. A tax return is required for annual income, to be filed not later than 15 March of the following year and be paid to the tax office at the same time. The relative share of taxable income collected under the present withholding system in 1996 is calculated in Table 4.1. The pure method of withholding occupies about three-quarters of total taxable income, while the share of filing returns accounts for only 4.7 per cent. Since separate taxation can be grouped into a withholding category, the importance in the tax system of non-filing returns increases further. It is difficult to estimate corresponding figures for international comparison, but the coverage of withheld tax collection in Japan seems to be much broader than in any other country.6 As argued earlier, individual income tax is paid in two ways: withholding or filing returns. In order to make clear the trend of relative weight in withholding taxation, Figure 4.3 shows two ratios of withheld income tax relative to national taxes and the 6
Rough information can be derived from OECD 1990 (Table 4, pp. 30-1). As far as this table is concerned, it seems that the source of income to which withholding tax is applied in Japan was the largest among the OECD countries.
62
Overview
individual income tax for 1960-98. Particular attention is paid to the upward movement of relative share in national taxes from 16.3 per cent in 1960 to 27.2 per cent in 1998. Meanwhile, the highest level of 33.1 per cent was reached in 1993. A sharp fall can be observed for the years 1986-90, but that was due to the abnormal state of the 'bubble economy'7 in which stock and land price hikes generated a great amount of tax revenues in the form of non-withholding taxes, such as the corporate tax, the security transaction tax, registration, and licence tax. No doubt, this induced a drop in the relative share of withheld income tax, A similar pattern is, more or less, shown in the upper line of Figure 4.3, although the increasing trend is less sharp. These facts imply that the withholding tax system has been entrenched in the framework of income taxation. How is Employment Income Taxed? There are two points to be emphasized about the significant role of the withholding tax system unique to Japan with particular reference to employment income (Type 1). First of all, the withholding system plays an important role in reducing the number of 'direct taxpayers' to pay taxes due. The number of taxpayers who are paying withheld income tax on employment income in 1996 is estimated at 52009683. By contrast, the number of withholding agents (i.e. 'direct taxpayers'), who have the obligation to withhold the tax at the time of income payment, accounts for 3 893 483: merely one-thirteenth of employment income-earners. It is easy to understand the efficient mechanism of tax collection through withholding agents if we compare the small number of withholding agents with the large number of self-assessed income taxpayers for filing returns: i.e. 8 239 8588 in 1996 (see NTA 1997). Second, great stress should be placed on how to withhold employment income at source. For this purpose, both the withholding tax table and the year-end adjustment are prepared to calculate the tax payable and enable it to be paid to the tax offices almost simultaneously. The amount of withheld tax on employment income is estimated every month, based on elaborate withholding tax tables (see MOP 1997, p. 66). These tables are prepared by the National Tax Administration to take account of many factors, such as progressive tax rates and a variety of exemptions and deductions. Since the Japanese companies generally pay salaries monthly, employers easily withhold the tax due on income, based on a 'withholding tax table for monthly salary payments'. Similarly, tax on a bonus, which is equivalent to a few months' salary in summer and winter in accordance with the Japanese wage custom, is also calculated by using a 'withholding tax table for bonuses'. 7 One of the most remarkable phenomena in the second half of the 1980s was a 'bubble economy', mainly caused by easy monetary policy. Excess money risked in stock and land markets, produced the abnormal hike of asset prices. During the period 1985-90, nominal rate of GNP growth accounted for 6.0 per cent. 8 This figure contains some taxpayers, subject to withholding tax, whose income exceeds ¥15m. per year or is earned from more than two sources. These taxpayers are obliged to file a final return in addition to an advance payment of withholding tax.
Tax Administration and Tax Equity
63
However, the use of such withholding tables provides merely a provisional calculation of the tax liability. Thus, at the end of the year, employers are requested to calculate annual income and the tax due as a whole, and to adjust for the difference between the annual tax liability and the tax amount already withheld. This is a 'year-end adjustment'. When such an adjustment is made every December, certain special deductions not considered in the monthly withholding table are added to recalculate total tax liability.9 As a consequence, the year-end adjustment plays an equal role in filing a final return. Since most employees usually have no other income of Type 4 (see Table 4.1) or no income higher than ¥15 million, they do not need to get direct access to the tax offices.10 Given the two factors mentioned above, the withholding tax system on employment income is characteristic of the following three points (see Ozaki 1991). 1. A great number of taxpayers who are employment income-earners have no relation with the tax offices. Indeed, only 3.5 million people out of 46 585 000 taxpayers paid income tax on employment income by tax returns in 1991. 2. Withholding agents perform the same job as the district tax offices. If they fail to collect the tax at source and to pay it to the tax offices, unpaid taxes will be collected directly from the agents, not income earners. Even in the case of tax delinquency, additional taxes and interest on them are paid through the agents. 3. Taxpayers are requested to present their personal data for tax purposes to their employers, not the tax offices. Therefore, taxpayers need to follow minor procedures which require a much lighter work load than that for self-assessed taxpayers. Likewise, other incomes under separate taxation, such as interest and dividends (Type 2), are equally withheld at source. Since such capital income is paid very widely to an unspecified number of people by financial institutions, withholding taxation is absolutely necessary to collect the tax on such items of income adequately. In particular, because there are no Tax Identification Numbers (TIN) at present, it is impossible to tax interest, dividends, and capital gains on the sale of stocks adequately in the strict sense of the term. Additional evidence can be derived from Figure 4.4 to show the efficient aspect of withholding taxation. Data on tax delinquency is available by categories of national taxes, and the ratios of tax delinquency to relevant tax revenue are depicted for 1960-96 for four selective taxes. Among them, the withheld income tax has kept the lowest level of the tax-delinquency-revenue ratio as compared with other taxes and the average of total national taxes. Obviously, the withholding tax system has been very effective in securing necessary revenues, leaving behind no possible delinquency of taxation. 9 For instance, deductions for accidents, medical expenses, life insurance premiums, etc., are taken into consideration for a year-end adjustment. 10 The method of withholding taxes on employment income in Japan is substantially similar to the UK cumulative PAYE system applied to Schedule E.
64
Fig. 4.4
Overview
Tax deliquency as a percentage of revenue, 1960-1996
Source: Calculated from National Tax Administration, Annual Report of Statistics (Kokuzeicho Tokei Nenposho), 1991.
What is of great interest is the sharp rises of each ratio after 1991 when the Japanese economy plunged into prolonged recession. In particular, self-assessed income tax began to increase the tax-delinquency ratio sharply, reflecting the bad time of business and the accumulation of bad loans. Even withheld income tax tended to lift the relevant ratio.
EMPIRICAL EVIDENCE OF THE TAX GAP A TEST OF THE 'KU-RO-YON' PHENOMENON 'Ku-ro-yon' Ratios Turning to the other side of administrative efficiency, we shall focus on the unfair tax burden among different income sources in relation to horizontal equity. Horizontal equity is often related to administrative practices. In principle, horizontal equity is frequently impaired when administrative arrangements are not satisfactory. This is the case in Japan. As noted earlier, it is widely believed, especially among salaried workers, that there are large divergences in the identification of taxable income among different classes of taxpayers. Since salaried workers are taxed at the source of income under the withholding system, their income is almost fully (90 per cent) identified by tax authorities. On the other hand, the self-employed (including practising doctors and solicitors/barristers) and farmers file their own income returns. They are not taxed fully at source and can easily dodge tax liability by underreporting their income. Reputedly,
Tax Administration and Tax Equity
65
only 60 per cent of the incomes of the self-employed and 40 per cent of farmers' incomes are caught by the tax office. These percentages (90-60-40) are used so often in describing the present unfair situation in the Japanese tax system that a special term, 'Ku-ro-yon', has been coined. 'Ku-ro-yon' is a portmanteau word of Japanese numbers—9 (ku), 6 (ro), and 4 (yon). This term is, by and large, used in the same way as the 'tax gap' among different tax sources is referred to in the USA. It is very difficult to test the 'Ku-ro-yon' ratio statistically. One possible method, which I have tentatively attempted, is to compare the scope of taxable income quoted in tax statistics (TS) with that which appears in national income statistics (NIS). The most difficult task is to make the necessary adjustments for obtaining a common base against which comparisons can be made. In Table 4.2, differences in concept definitions are enumerated in employment, self-employed and agricultural incomes. Y stands for income recorded in national income statistics (i.e. NIS-based income), while y denotes taxable income in tax statistics (i.e. TS-based income). The relation between the two income concepts is basically expressed as follows:
The term 'a is considered as an unreported or underreported portion of taxable income, presumably due to tax evasion and avoidance (for general discussion, see Goode 1981; Roth et al 1989; Webley et al. 1991). If we regard NIS-based income as a reference level11 for comparison with taxable income recorded in tax statistics, the tax gap ratio (8) can be defined as follows. Both sides of (4.1) are divided by Y,
8 indirectly indicates the magnitude of unreported or underreported income which may be considered to be another interesting measure. The tax gap ratio is calculated for each income source,12 and 81( 82, and 83 are linked with that of employment, self-employed, and agricultural incomes, respectively. 11 Of course, this assumption would not be plausible. Not all the value of economic activity is recorded in national income accounts and some is presumably untaxed because of the so-called underground economy. However, since there would be no other reliable alternative, we rely upon NIS-based income as a reference to which reported taxable income is matched. 12 The tax gap is defined by income sources, not income earners, because income earners tend to have other incomes apart from income from their major occupations. In Japan, for example, farmers frequently earn both agricultural and employment incomes by working at nearby factories. Thus, it is more accurate to define the tax gap in terms of income sources.
66
Overview Empirical Results
The matching of TS-based income to NIS based income can be justified by the fact that the two incomes are obtained statistically from two different data sources. For instance, employment income in NIS is estimated by the Economic Planning Agency, mainly based on the Monthly Report of Working and Wage and other related statistics of the Department of Labour, while the corresponding figure in TS practically is collected by the National Tax Administration for tax purposes. The same holds for both self-employed and agricultural incomes.13 Such a comparison may make sense to obtain evidence of the tax gap. Procedures for tax gap estimates are rather complicated, based upon data processing by a number of different statistics.14 Thus, it should be emphasized that the estimates are subject to large potential errors. In particular, this is true in the case o self-employment income, because income below minimum taxable level unrecorded in tax returns must be estimated with bold assumptions. The matching for employment income is the easiest procedure among the three cases, producing the most reliable result. Basic tax statistics13 provide us with almost all the necessary data pertinent to taxable income in TS, including both b and c in Table 4.2. Therefore, it is not necessary to attempt troublesome estimates of these two items, and only minor adjustments need to be made for trivial differences in concept definitions; say, bonus for company executives which is originally included in company profits. On the other hand, both self-employed income and agricultural income require some additions to taxable income in TS. As is seen from the notes in Table 4.2, the minimum taxable level under the income tax law is first set at a specific income level each year, for a standard household (couple and two children), and then income below such a threshold is estimated by using data on the income distribution by income classes from Employment Status Survey (Statistics Bureau, Management and Coordination Agency).16 Procedures for these estimates must essentially become crude, reflecting less reliable data sources. Furthermore, special deduction for wages paid to family employees should be included in tax returns. The whole process of estimation concerning the cases of self-employment income is done by attempting
13 NIS-based self-employed income is estimated on the basis of Statistics of Non-corporated Business Offices (Jigyosho Tokei Chosa) and Economic Survey of Non-corporated Business Offices (Kojin-kigyo Keizai Chosa) (Statistics Bureau, Management and Coordination Agency). Similarly, NIS-based agricultural income employs farmers Economic Survey (Nohka Keizai Chosa) and Agricultural Census (Nogyo Chosa) (Ministry of Agriculture, Fishery and Forestry). 14 For example, in order to obtain detailed classification of incomes in question, I must use worksheet-level data at the Economic Planning Agency which are fortunately available to me. 13 Basically, Annual Report of the National Tax Administration and Statistics on Private Wages and Salaries (NTA) is employed. 16 Employment Status Survey (Stingyo Kozo Kihon Chosa) has not been published every year. In the past, data were only available in 1974, 1979, 1982, and 1987 for the period to cover our estimates. Thus, any single-year data were extended to make multiple-year estimates before or after the limited years.
Table 4.2 Conceptual adjustment between TS-base and NIS-base incomes3 Taxable income subject to tax code
Unreported or underreported income
Income below minimum taxable level*3
y
a
b
Non-taxable income due to additional deductions0 c
income in TS
?
including in
including in
y\
y\
b2
C
Employment income
y\
Self-employed income
the same Vi
Agricultural income
the same
yi
?
?
b3
2
C3
Income excluded in TS, but included inNIS d
Income Difference included in TS, in coverage but excluded inNIS e /
special deduction for wages paid to family employees d2
capital gains in inventory
the same d3
the same
e
2
bonus for company executives /i livestock farming and fishery incomes h
«3
Notes: TS—Tax Statistics, NIS—National Income Statistics. Minimum taxable level is calculated in the case of a standard household (couple and two children, including basic exemption, exemption for dependants, exemption for spouse, deduction for social insurance premiums, and special deduction for blue/white return). c Deduction for medical expenses, deduction for life insurance premiums, deduction for fire and other casualty insurance premiums, and deduction for small-scale enterprise mutual aid premiums are added to five items of minimum taxable level. a
b
68
Overview
Fig. 4.5
Tax gap among three income sources, 1970-1990
separate procedures for blue and white tax returns (see below).17 Certain adjustment is also needed to change the coverage of income in the case of self-employed income (i.e. the term'/' in Table 4.2). Final empirical results for 1970—90 are depicted in Figure 4.5 where we can find two kinds of line for each category of income. One line is the tax gap ratio itself;1;8
2, and 83, and the other is a matching ratio of original income data before concep tual adjustment: y\IY\, y2/Y2, and y^/Y3. Since a substantial number of estimat procedures are included to obtain the final tax gap ratio, original matching might be of use to ascertain the accuracy of the tax gap in question. Major fact findings are shown in the following four points. First of all, Sj, and y\IY\ in the case of employment income substantially remained stable in the band of 90-100 per cent during the period 1970-90. It can be conjectured that there would be almost no difference between NIS-based and TS-based incomes, and that the 'ku' ratio is justifiable by empirical data. The incomes of wage and salary earners are fully captured by the tax offices under the withholding system. Second, self-employed income varied the level of 82 to some extent, but it can be pointed out that the 'ro' ratio did not come into existence until about 1985. As will be argued shortly, it is necessary to explain the upward movement of 82 in the late 1980s. The other line of y2/y2 moved more smoothly with a regular margin agains
2, and the adequacy of our estimates would be reinforced by the original matching ratio. Third, the growth of 83 and y^Y3 in the case of agricultural income is differen from the other two cases. 83 moved below the 40 per cent level for 1970-84, but it 1 Statistics on the Self-assessed Income Tax (NTA) is employed to obtain necessary information.
Tax Administration and Tax Equity
69
turned upward sharply after 1985. Also, y^Y3 followed a similar pattern to 83 durin the whole period. The 'yon' ratio may not exactly be justified, but it is noted that 83 and 73/^3 were kept at the lowest level with a minor exception as compared with corresponding figures. Fourth, judging from our estimates, it seems that the tax gap among different income sources in Japan may be empirically tested to some extent in conjunction with the 'Ku-ro-yon' phenomenon. Particularly, this would be the case for 1970-85. Other studies of the tax gap also support my empirical evidence.18 Recent Development of Tax Gap The tax gap, as stressed previously, has been produced by different tax collection between the withholding system and tax returns. Given the existence of the tax gap found above, the well-developed withholding system has obviously strengthened taxpayers' perception of the unfair tax burden of wage and salary earners, while it has made tax administration more efficient. As a consequence, the conflict between efficiency and equity of tax administration in the Japanese tax system has been induced to a great extent by developing the withholding collection. It seems, however, as seen in Figure 4.5 that the tax gap narrowed up to 1990. As far as this evidence is concerned, the conflict between the two objectives of tax administration begins to be mitigated to some extent. This being the case, it would be necessary to explore the factors necessary for increasing the 'ro' and 'yon' ratios of self-employed and agricultural incomes. Most importantly, great emphasis should be placed on the improvement of tax return methods. The self-assessment on a tax return basis is generally divided into two systems, i.e. (a) blue returns and (b) white returns. When the Shoup Mission proposed tax recommendations in 1949, the blue return system was introduced to improve book-keeping and to promote honest self-assessment by taxpayers subject to income tax returns. Since then, the National Tax Administration has considered the 'blue return' as the fundamental requirement for efficient tax administration. The main aim of the blue return is to encourage small- and medium-sized business to keep a minimum set of accounting records, but in addition certain significant advantages are offered to individuals and corporations by the tax offices (see MOF 1991,63-4). The major advantage is that taxpayers filing a blue return are not subject to reassessment as long as errors cannot be found in their accounting books and records. Moreover, they are allowed to deduct reasonable amounts for wages paid to family members working in the same companies and to use special tax-free reserves 18 Following my own estimate in Ishi 1983fc, three studies have so far tried to testify to the 'Ku-ro-yon' ratio: i.e. Homma et al. 1984; Hayashi 1990; Okuno et al. 1991. These estimates lead, more or less, to the same empirical results, although statistics and procedures used are quite different in each. For example, Hayashi estimates 101.3-52.5-13.3 in 1979, 99.4-58.6-14.3 in 1982, and 101.4-61.7-20.7 in 1987 as the 'Ku-ro-yon' ratio. Likewise, Okuno etal. find 104.8-60.4-27.6 in 1985 (all figures are percentages).
70
Overview
(e.g. reserves for bad debts, losses due to price fluctuations, etc.). By contrast, taxpayers, who are not filing a blue return, are not given these advantages for tax purposes, but they are not obliged to keep books and records. This case is usually called the 'white return' system. Traditionally, most farmers do not file blue returns, and the tax offices estimate their income on the basis of their crops. This is considered to be a major factor leading to understatement of agricultural income for tax purposes. In fact, blue returns as a percentage of total filing returns were very low for farmers until the early 1980s; 2.5 per cent in 1970, 7.1 per cent in 1975 and 10.2 per cent in 1980. However, the blue-return ratio has sharply increased to higher values since then; 17.8 per cent in 1985 and 32.0 per cent in 1990. The sharp rise in using blue returns is evidently thought of as the most important factor to explain the narrower tax gap in agricultural income after 1985. The same reasoning may be applied by and large to the case of self-employed income, but it is not so clear-cut as the agricultural case. The corresponding percentage using blue returns merely increases from 48 per cent in the early 1970s to 51-3 per cent in the 1980s. More importance should be put on increasing exemptions and deductions applicable to self-employed taxpayers.
SOME POLICY IMPLICATIONS
In this chapter, the 'Ku-ro-yon' ratio does indeed seem to be approximated by these statistical procedures, although the results are far from satisfactory. A tax gap between the three income sources would probably arise from both evasion and avoidance. Obviously, it would be impossible to draw borderline distinctions statistically. It is necessary to observe with care changes in the tax gap, as seen in Figure 4.5 and to identify whether or not these phenomena would be temporary in the future. At the moment, however, let us assume that the 'Ku-ro-yon' ratio is still prevalent in tax administration. A high proportion of popular complaints about the present tax system emerges from inequities of this kind. These complaints have arisen among a majority of wage and salary earners. The national atmosphere vis-a-vis inequitable income taxation has attracted wide attention among the general public. Without doubt, this is one of the inherent features built into the Japanese income tax system. It is important to note that this atmosphere among the general public potentially supports the need to increase reliance on indirect taxes in the tax system. In the past, two sweeping tax reforms by both the Nakasone and Takeshita cabinets attempted to introduce the value added tax in the Japanese tax system and were finally successful (see Chapter 17). No doubt, this success was primarily due to the support of wage and salary earners who had major complaints about unfair tax burdens because of the 'Ku-ro-yon' phenomenon.
Tax Administration and Tax Equity
71
It is still uncertain whether recent tax reforms will solve the conflict between efficiency and equity of tax administration. It seems, however, that efficient administrative practices have essentially led to the adoption of VAT in Japan in order to mitigate inequities of tax burdens caused by the well-developed withholding tax system. The argument concerning the 'Ku-ro-yon' phenomenon was most popular in the 1980s in relation to the introduction of VAT. It seems, however, that the atmosphere against it virtually cooled down towards the 1990s for two reasons. For one thing, the long-term decline in the agricultural sector has structurally been remarkable in Japan, reflecting the steady decrease of farming families from 5.3m. in 1970 to 3.6m. in 1994. The other is that smaller-size farms have tended to be merged into large ones with modernized business management.
This page intentionally left blank
Part II
Individual Income Tax
This page intentionally left blank
5 Basic Structures
Similar to practically all other developed countries, Japan depends upon the individual income tax for a significant portion of its tax revenues. Throughout the postwar period, the income tax has reigned supreme in the Japanese tax system. In more recent years, tax experts have begun to pay attention to certain of its defects, particularly from the standpoint of those favouring an expenditure tax. In my view, however, the income tax will continue to be widely regarded as one of the best methods of taxation yet devised. In Japan, the tax on individual income is used simultaneously by the national, prefectural, and municipal governments. Our major concern here is with the national income tax; local income taxes will be discussed in Chapter 18. In this chapter we consider the basic structures of the individual income tax at the time of writing (April 1999) from three angles. First, it is necessary to define a few important terms used in computing the income tax base, such as 'income for tax purposes', 'exemptions and deductions', the 'tax unit', and the 'treatment of fringe benefits'. Second, the structure of progressive tax rates is considered with special reference to the application of separate tax rates on certain incomes. Lastly, attention is focused on some salient features of the present income tax, including a hybrid of an income tax and an expenditure tax.
THE
TAX BASE
Computation of Taxable Income 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. So long as this primary principle is followed, it is relatively easy to compute the final tax liability for each taxpayer. Each incomeearner 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 may use certain tax credits to arrive at the final amount due. Taxable income is classified into ten categories: (1) interest income, (2) dividends, (3) real estate income, (4) business income, (5) employment income, (6) retirement income, (7) timber income, (8) capital gains, (9) occasional income and (10) miscellaneous income. Most of these categories permit specific exemptions and deductions,
76
Individual Income Tax
Table 5.1 Minimum taxable level of the individual income tax (wage-earners), selected countries, 1998 Japan a (¥1 000)
USA ($)
National Local
Federal
Single
1 107
1 053
Couple
2095
1 857
Family of 3
2698
2380
Family of 4
3616
3031
9800 (1166) 12500 (1488) 17869 (2126) 20569 (2448)
UK (£)
Germany France (DM) (F)
4045 (789) 5418 (1056) 5418 (1056) 5418 (1056)
17714 (1187) 32996 (2211) 45416 (3043) 55784 (3738)
Stateb 7500 (893) 13000 (1547) 14000 (1666) 15000 (1785)
72771 (1455) 116442 (2329) 138268 (2765) 160112 (3202)
Note: Figures in parentheses are those expressed in terms of ¥1000, using 1998 exchange rates: $1 =¥119, £1 =¥195, DM1 =¥67, and Fl =¥20. a b
Includes the deduction for social insurance premiums. New York State income tax.
Source: MOF, Tax Bureau, Primary Statistics of Taxation (Zeisei Shuyo Sanko Shiryoshu), Dec. 1998.
except for interest and dividends. Fundamental personal exemptions for average households consist of the basic exemption, the exemption for spouse, and the exemption for dependants: this came to a flat ¥350 000 each for the taxpayer himself, his spouse, and other dependants in 1992.' Besides these personal exemptions, wage and salary workers are permitted to subtract two special deductions for earned income and social insurance premiums from their employment income. The MOF includes these five exemptions and deductions in calculating the minimum taxable level (the tax threshold) of each household. Table 5.1 shows the minimum taxable level of wage-earners among five selected countries. It is very difficult to obtain comparative figures since the income tax law admits different structures of exemptions and deductions in each country. Japan reveals relatively higher levels for the tax threshold, as do Germany and France. This implies that the actual tax paid by taxpayers at various income levels is rather modest, particularly in comparison with the UK and the USA. Allowable Deductions and Credits In addition to personal exemptions, which are widely applied to taxable income as a whole, numerous deductions from different income sources are allowed for
1 In addition, a special exemption for spouse (¥350000) was allowed from Sept. 1987 in the case of one-earner couples to balance the tax burden between wage-earners and others.
Basic Structures
77
specific reasons. Some of the major deductions are explained in the following paragraphs. Employment income, as already noted, is given a special deduction instead of being entitled to deductions of actual amounts of personal expenses.2 The main aim of this deduction is to put wage and salary workers, the only categories to whom this deduction is available, on a more equal level with the self-employed, who can treat their personal expenditure as business expenses. This special deduction for employment income is considered to be rather generous, although it can be used only on an across-the-board basis. In fact, the ratio of the amount used for this deduction to average earnings per worker accounted for 34.1 per cent in 1985, 31.6 per cent in 1990, and 31.7 per cent in 1996. A variety of special treatments is also involved in the determination of business income. Generally speaking, net income is subject to taxation after all necessary expenses are subtracted from gross receipts. Taxpayers filing a blue return were permitted to deduct the cost of preparing the tax returns up to a maximum of ¥100 000 until 1993. A special deduction for wages paid to family employees is also allowed in lieu of deducting them as necessary business expenses from the proprietor's business income. Even the proprietor's own remuneration is deductible from his business income, and this can also be applied to the special deduction for employment income, if the firm elects to be treated as a quasi-corporation for tax purposes (i.e. as a 'deemed corporation'). As a result of tax reform in 1992, this 'deemed corporation' system was abolished from 1993. Instead of this, however, the special deduction for filing a blue return was increased from ¥100000 to ¥350000 on condition that good record-keeping is observed, and it was raised to ¥450 000 in 1998. Retirement income, timber income, and capital gains are each granted special deductions. For retirement income, the deduction is ¥400000 per year for up to 20 years of employment or ¥700 000 per year for anything over 20 years of employment in excess of the amount accumulated during the first 20 years. For timber income, a special deduction of ¥500000 is deductible from gross receipts minus necessary expenses. Capital gains on the sale of securities and real estate are taxed separately from ordinary income tax after allowing for special deductions. Besides the special deductions permitted to certain taxable incomes, there are nearly 20 deductions involved in the present structure of individual income taxes.3 To compute final tax liability, several tax credits may also be taken, such as credit for dividends, credit for foreign taxes, credit for incremental research and experimental expenditure, credit for acquisition of a house. Some of these are linked to the promotion of specific policy goals. 2 'Employment income' includes not only wages and salaries, but bonuses, pensions, and other allowances of a similar nature. In 1999 the deduction was 40 per cent of the first ¥1,80m., 30 per cent of the next VI.80m., 20 per cent of the next ¥3.Om., 10 per cent of the next 3.7m., and 5 per cent for amounts in excess of ¥10m., but the minimum deductible level is set for ¥6.5m. 3 Typical categories of such deductions include deductions for casualty losses, medical expenses, donations, physical handicap, etc. For a comprehensive list, see MOF Tax Bureau (1998).
Individual Income Tax
78
The Tax Unit In prewar Japan, the family unit was adopted as the basic tax unit, reflecting the belief that the head of the household (usually the husband) was responsible for filing a tax return of joint income on the grounds that the wife and children were dependent on him. The Shoup Mission recommended that the family be replaced by the individual as the primary tax unit. Since then, the Japanese tax system has maintained a system in which the individual is the unit of income taxation, although some exceptions have been allowed. According to this system, two-earner couples are taxed as separate individuals. For one-earner couples with several dependants, the head of the household is treated as a single taxpayer. The tax unit adopted by the Japanese is similar to that of Australia and New Zealand in which the individual is still predominant. It is in sharp contrast, however, with the USA, the UK, and France. (For a more detailed discussion, see Kay 1987.) The USA uses the family unit as the basis of taxation with distinctive tax rates for single persons and couples under an income-splitting approach. Similarly, the UK employs a unit basis of assessment in which the income of a working wife is aggregated with that of her husband after relief is taken for the wife's earned income. France has a unique quotient system that splits household income among the members of family. Given such diversity of existing practices regarding the tax unit, the Japanese tax unit does not seem to raise any serious problems. Accordingly, married couples in Japan with the same total income pay different taxes, depending on whether there are two wage-earners or one. Table 5.2 shows Table 5.2 Different income tax liabilities among one-earner and two-earner couples with a combined income of ¥10 million, 1998 Tax liability (¥000) Husband
Tax burden rate (%) Wife
Total
One earner Husband Wife
¥10 m. 0
720
0
720
7.2
Two earners Husband Wife
¥8m. ¥2m.
520
70
590
5.9
Husband Wife
¥6.5 m. ¥3.5 m.
294
165
459
4.6
Husband Wife
¥5m. ¥5m.
177
273
450
4.5
Note: Tax liabilities arc calculated on the 1998 tax code, assuming that each couple is a wage and salary worker with two dependants, excluding temporary tax cuts in 1998.
Basic Structures
79
Table 5.3 Income tax liability and income-splitting 1998 Income scale ¥m. 3 4 5 8 10 20 30 50
Present system
Income-splitting
Tax (¥000)
Tax burden rate (%)
Tax (¥000)
Tax burden rate (%)
0 28 101 368 720 3309 6822 15785
0.0 0.7 2.0 4.6 7.2 16.5 22.7 31.6
0 28 101 349 525 2366 4929 11392
0.0 0.7 2.0 4.4 5.3 11.8 16.4 22.8
Note: Tax liability under income-splitting is simply computed as ((taxable income X 1/2) X present tax rate} X 2.
the tax liabilities of various married couples with the same combined income of ¥10 million. Married couples with only one wage-earner pay higher income taxes than those with two wage-earners. To the extent that total income can be earned more equally by couples, they tend to pay less tax. Given the present structure of the individual income tax, the Japanese tax system benefits two-earner couples with evenly split income, although an exemption for a spouse is allowed for a one-earner couple. What happens if the income-splitting approach is introduced to one-earner couples in Japan? In Table 5.3, the tax liabilities of married couples with one wage-earner are computed as if they were two single taxpayers with their total income divided evenly between them. The effect of income-splitting would evidently lower the tax burden at higher-income classes. For instance, at the top income of ¥50 million, the tax burden would be reduced from the present rate of 31.6 per cent to 22.8 per cent. If children are recognized as independent earners, they may be taxed separately on their own income. However, few children have incomes large enough to be subject to taxation. Their financial reliance on their parents enables the family to take exemptions for them as dependants. Despite the fact that the tax unit is based upon the separate individual, the investment income of the spouse and other dependants used to be taxed as if it were received by the head of the household. The main purpose of aggregating income from investment was to prevent tax avoidance by allocating such income arbitrarily among family members.4 ^ The aggregation tax method of investment income was established as an exception to the individual tax unit in 1950, but it was repealed in 1951. In 1957 this method was re-established and was maintained until 1988. It has now been repealed.
80
Individual Income Tax
The problem of the tax unit in Japan has some bearing on the special treatment of business incomes. As noted earlier, business proprietors are allowed to split their incomes by paying wages and salaries both to other members of their families and to themselves. Obviously, this income-splitting lightens their income tax burden, and as a result this practice is greatly criticized by wage-earners as being inequitable. Fringe Benefits In Japan, tax advantages for fringe benefits have not received any more attention than in other major countries. Apparently, the existing tax treatment of fringe benefits is more favourable than that accorded to ordinary income. Since this affects mostly employment income, it would be unfair to omit this aspect of taxation in relation to the definition of taxable income. Under the individual income tax schedule in Japan, cash income is taxable, but non-cash benefits given to the wage-earner and the business executive either are not taxable at all or are only partially taxable. Although no clear boundary can be demarcated between ordinary income and fringe benefits, there are several kinds of tax preferences for fringe benefits worth noting here. The most important of these benefits is the exclusion of the value of subsidized housing from taxable income. Most large companies, as well as the government, provide housing to many of their employees at much cheaper rents than market prices for comparable facilities. Companies also subsidize loans for the purchase or construction of private residences, so that employees can obtain mortgages at a very low rate of interest over ten or more years. The subsidized portion of such rents and interest payments is excluded from taxable income. Furthermore, the system of employee compensation includes free or cheap provision of recreation facilities and other welfare benefits. Such payment in kind is non-taxable, and greatly benefits Japanese wage-earners, particularly those working for large companies. For the majority of business executives, a compensation package is provided by their companies: a car, a large residence, an expense account for entertainment, etc. It is not common, however, for corporate executives in Japan to hold stock in their own companies, because of the lack of stock option plans. Of greatest importance is perhaps the expense account, which is by and large used for inviting business customers to expensive restaurants, clubs, and golf facilities. Thus, many people find it easy to see why their employers are generous in paying out large sums of money for entertainment purposes: all of these benefits go untaxed in their personal incomes.5 Needless to say, the tax advantage of fringe benefits aggravates the inequities between the treatment of taxpayers who receive only a cash income and those who 5 At present, total expenditures for entertainment expenses are not deductible from corporate income. However, small firms with ¥10-50m. of capital are permitted to deduct ¥3m. a year (but 20 per cent of the portion if less than ¥3m.), and those with less than ¥10m. are allowed ¥4m. (likewise, 20 per cent below ¥4m.). In the past decade or so, such expenses have been taxed more strictly.
Basic Structures
81
receive fringe benefits as well. Traditionally, major tax reports written around the world have reached a clear agreement on the appropriate taxation of fringe benefits (see Kay 1987, 74-5).
TAX RATES
Progressive Rates
All taxable income in excess of the minimum taxable level is subject to tax at progressive income tax rates. Table 5.4 illustrates the schedule of statutory tax rates at all levels of government in 1999. National income tax rates start at 10 per cent of taxable income up to ¥2.0 mil lion, rising to a top rate of 37 per cent above ¥1.8 million. By contrast, local income tax rates, consisting of both prefectural and municipal rates, are relatively simple. Prefectural income tax rates are 2 per cent on the first ¥7.0 million of taxable income, and 3 per cent above it. The rates for the standard municipal income tax begin at 3 per cent and rise to 10 per cent on taxable income above ¥7.0 million. Local income taxes cannot be deducted from the national income tax base. Thus, the combined marginal tax rate seems to have a significant effect on the attitude of taxpayers. At the top income bracket, it reaches 50 per cent which was a punitive level of 65 per cent until 1998.6 Table 5.4
Statutory rates of income taxes at all levels of government, 1999
Taxable income (¥m.)
under 2.0 2.0-3.3 3.3-7.0 7.0-9.0 9.0-18.0 18.0 and over
Tax rates (%) National
Prefectural
Municipal
Total
10 10 20 20 30 37
2 2 2 3 3 3
3 8 8 10 10 10
15 20 30 33 43 50
Note: Since the exemption levels are different for all three income taxes, the taxable income for any specific taxpayer is not exactly the same in each bracket. Source: As Table 5.1.
6 When the top rates were 70 per cent in the national income tax and 18 per cent in the local income taxes (i.e. a combined rate of 88 per cent), in 1986, a maximum amount of income tax was set to mitigate the total tax burden at the effective rate of 78 per cent. The adjustment was made in the prefectural and municipal income taxes to maintain that level. This method, however, was repealed after Sept. 1987. In addition to the income tax based on the progressive rates, there are per capita taxes levied by local governments. For a detailed discussion, see Ch. 18.
Table 5.5 Major changes of tax rates, national income tax, 1949-1999 1949 Tax rates %
1962
1950 Taxable income classes (¥000)
Tax rates %
Taxable income classes (¥000)
Tax rates %
1970 Taxable income classes (¥000)
8 10
100 300
Tax rates %
10
600
12 14
600 900
16
1 200
18
25
20
40
20
25
50
80
70
30
100
30
1800
35
100
35
120
35
2500
40
150
40
150
40
4000
200
45
200
500 50 50 250 55 500 55 300 60 500 65 700 70 1000 75 2000 5000 80 86 5000 No. of income brackets 14 8
45 50 55 60 65 70 75
1988
Taxable income classes (¥000)
Tax
rates %
Taxable income classes (¥000)
10.5 12 14
500 1 200 2000
10.5 12
1500 2000
16
3000
17
3000 20
5000
Tax rates %
1989
1995
Taxable income classes (¥000)
Tax rates %
Taxable income classes (¥000)
21 24
2000 2500
27 30 34
3000 3500 4000
38
5000
42
6000
46 50 55 60 65 70 75
8000 10000 20000 40000 60000 80000 80000
6000 10000 20 000 30000 45000 60000 60 000
15
10
3000
10
3000
1 500
1 200
30
45
Tax rates %
800
20
25
1987
Tax rates %
1999 Taxable income classes (¥000)
Tax rates %
Taxable income classes (¥000)
500
15
20
Taxable income classes (¥000)
1984-86
19
21
4000
25
6 000
25
6000
30
8000
30
8000
35
10000
35
10000
40
12000
40
12000
45
15000
45
15000
50 55 60 65 70
20000 30000 50000 80000 80 000
15
50 55 60
30000 50000 50000
12
20
6000
20
6000
30
10000
30
10000
40 50
20000 50 000
40 50
20 000 20000
60
50000
6
Source: MOF, Primary Statistics of laxation (Zeisei Shuyo Sanko Shiryoshu), Feb. 1998 and MOF, Outline of the 1999 Tax Reform.
5
10
3300
10
3300
20
9000
20
9000
30
18000
30 37
18000 18000
40 50
30000 30000
b
4
Basic Structures
83
Tax rate structures have varied greatly in Japan throughout the postwar periods. Major changes in national income tax rates are tabulated in Table 5.5. In 1950, a drastic change was enforced to make the income tax less progressive than it was under the influence of the Shoup proposals. The mitigation of progressiveness was carried out in conjunction with the adoption of the 'net worth tax' supplementary to a comprehensive income tax. After the repeal of the net worth tax in 1953, however, the rate structure again became more progressive with an increased number of income brackets. This trend continued until 1984, when the reverse movement began to emerge from general support for tax structures that have a smaller number of rate bands and lower marginal tax rates, particularly at higher-income classes. As a result of flattening tax rates in line with the movement of world tax reform, Japan's rate structure was simplified with five brackets from 1989 and further reduced to four brackets in 1999. Figure 5.1 shows the marginal tax rates implied by the statutory tax schedules in five major countries in 1998. The diagram depicts the rates that apply to a wage earner with a family of four and an income level of up to ¥10 million. Clearly, rate structures up to ¥10 million are by and large the same in the UK, the USA, and Japan in comparison with Germany. In particular, the long, basic rate band in the UK is conspicuous, so that the basic marginal rate applies to most taxpayers. The USA and Japan show similar patterns, although the basic bands are much narrower. In contrast, in Germany a non-linear rate begins to rise sharply after the basic rate band. The flatter rate structures in the USA and Japan were introduced by tax reforms. As a matter of fact, in 1986 the initial tax rate was lower and started rising more or
Fig. 5.1 Marginal rates of income tax (excluding local taxes) up to an income of ¥10 million for a family of four with a single wage-earner, selected countries, 1998 The US figure includes the state income tax for New York (but excludes the New York City tax). Exchange rates used for the conversion of yen are those for 1998: $1 =¥130, £1 =¥215, and DM1 =¥72. Source: MOF Tax Bureau, data presented to the Tax Advisory Commission, with some modifications.
84
Individual Income Tax
less continuously. The main features of Japan's rate structure included a higher level of minimum taxable income, and an excessive number of marginal tax rates with narrow rate bands, but the progressive tax structure has recently been flattened to a great extent. Application of Separate Tax Rates So long as we maintain a global tax system of comprehensive income taxation, it is not entirely necessary to separate certain income sources from others for tax purposes: all incomes should be aggregated and taxed at progressive rates. This was actually realized under the aggregation approach proposed by the Shoup Mission. Today, however, many incomes are taxed separately at specially reduced rates. These separate taxation methods have been developed mainly to promote policy objectives. The majority of the ten taxable incomes described above are subject to forms of separate taxation. The most important of these is the application of a separate flat tax rate on interest derived from taxable personal savings. We shall consider these points in greater detail in Chapter 8. A similar procedure applies to dividends. Retirement income and timber income are also taxed separately under progressive tax rates. Only half of retirement income, after the special deduction for retirement is applied, is subject to taxation, and this is separated from other incomes. Timber income is handled by a sophisticated procedure of averaging; after allowance for the special deduction, taxable income is first divided by five and taxed at the rate applicable for that one-fifth; the tax amount thus computed is multiplied by five to obtain the total tax due. Likewise, capital gains on the sale of real estate are taxed at flat reduced rates favourable to all taxable long-run gains. Fluctuating and extraordinary income, which are usually classified as 'occasional income', are eligible for the 'averaging taxation' method. The Income Tax Schedule The progressiveness of the individual income tax is the most important means of securing vertical equity in the income tax burden distributed between individuals whose circumstances are different. This is also concerned with the redistributive implications of income taxation among the rich and the poor. How should the tax burden be distributed among people in different economic circumstances? To answer this question, it is important to formulate a progressive income tax schedule which relates tax liability to income received. In general, progressive income taxation is defined as taxation in which the proportion of income expropriated as tax rises as income increases. This implies that the average tax rate must increase with income, although the marginal rate of tax need not also increase with income: a flat marginal rate with a basic allowance can produce substantial progressivity (see Kay and King 1986, 211).
Basic Structures
Fig. 5.2
85
An exposition of Japan's income tax schedule
Figure 5.2 is an expository diagram of the income tax schedule that has been adopted in Japan during the postwar period, as derived from Table 5.5. The marginal rate of tax tm rises sharply at lower income levels, and begins to increase more gradually as it approaches the top rate t at income OA. At incomes above OA, tm maintains a constant level of f . The average rate of tax ta also shows a gradual upward movement, but it consistently maintains a lower level of tax rate than tm. Therefore, the income tax in Japan is progressive on the grounds that tm is higher than ta. Effective Progressivity The progressive income tax schedule illustrated in Figure 5.2 is based on the statutory tax rate schedule used thus far. In actual practice we may see a different progressivity of income taxation from that suggested by a statutory rate, because the income tax liability is not determined only by a statutory schedule of progressive tax rates. Now an attempt is made to seek empirical evidence of the effective progressivity involved in the income tax system. As a point of departure, let us construct a simple cross-section model. Tax statistics yield taxable income Y, number of taxpayers N, and taxes T by income class. Let us postulate a relation among 7;, N,, and T, for income class i:
86
Individual Income Tax
A priori, (3 > 1 is hypothesized in (5.1), because the present income tax system is progressive. Expression (5.1) shows that taxes per capita are related to income per capita in a progressive tax function. Now, setting 11= TIN and y= Y/N, (5.1) is simplified to
The value of |3 is an indicator of tax progressivity.7 The required data, Y, N, and T, are available from two sources prepared by the National Tax Administration; one is the self-assessed income tax on the selfemployed, agricultural, and professional incomes, and the other, the withholding income tax on wage and salary incomes. (For a more detailed explanation, see the appendix to Chapter 6.) I estimated (5.2) for each of two types of the income tax for the period of 1951-96, depending upon two different tax sources. My empirical results are highly significant in statistical terms. (They are omitted because of space limitations.) In Figure 5.3, the values of p show the level and variation of income tax progressivity over three decades.8 As hypothesized above, (3 > 1 is verified. It should be noted first of all that the values of (3 are not at all stable. In other words, the tax function itself has changed. It is probable that these values would have been much more stable had the tax system been fixed in the postwar period.9 The tax function has been volatile, partly because of changes in the income distribution and other factors, but mostly because of the influences of annual tax changes. In other words, these changes in the tax function measure the effects of tax changes on the income tax structure. In the period under study, there is only one year in which no actual tax change was imposed: 1960. This is because the same income tax system was applied in both 1959 and 1960. We should therefore get approximately the same coefficient estimates for these two years. The values for (3 are relatively stable in the years 1959 and 1960 in Figure 5.3; thus, the theoretical expectation that the tax function is stable when the tax system is fixed is substantiated. (3 did vary sightly between the two years, perhaps because of changes in the income distribution and other factors. Except for these two exceptional years, (3 varied considerably from year to year. Since 1951, the income tax on wages and salaries has shown many variations, dropping by about 20 per cent, from 2.09 in 1957 to 1.71 in 1962, and then rising 7 The value of (3 also indicates the income elasticity of tax revenue, as discussed in Ch. 3. From this standpoint, Ishi (1976) calculated p in a similar way. 8 Of course, (3 in (5.2) is not due exclusively to the effect of income tax progressivity: changes in income distribution and other factors also influence (3. If we are to isolate the influence of income tax progressivity, we should use Y and T from the data of the statutory tax schedule. However, here we are interested in measuring the tax structure in actual practice. Thus, 1 have decided to use actual data derived from tax statistics. 9 Factors other than tax changes influence (3. Thus, even without tax changes, (3 would not be stable, in view of the nature of the data used in this study. However, the postwar US tax function, estimated from the same type of data, shows that (3 is virtually unchanged with the fixed tax system; see Snowborger and Kirk (1973, 242-3).
Basic Structures
87
Fig. 5.3 Variations of progressivity of income tax, 1951-1997: (a) self-assessed income tax; (b) income tax on wages and salaries but not returning to the high level of the late 1950s. In the 1970s and 1980s, the level of P moved within a narrow band of 1.9-2.0, registering many ups and downs. However, it began to decline sharply below 1.9 in the 1990s due to successive tax reductions. In contrast, the self-assessed income tax has revealed a somewhat different pattern of change. The values of P stagnated in the late 1950s and the early 1960s, fell briefly, and then quickly rose for a while. This pattern differs somewhat from the variations in income tax on wage and salary incomes since 1970. One important fact that has emerged from my estimates is that the progressiveness of the income tax in Japan has been manifest in terms of the effective progressivity,10 and 10 It is necessary to explain that (3 for the self-assessed income tax is rather stable and its level is generally lower. There are two reasons for this worth noting. For one thing, tax avoidance and evasion can be used more easily to assess income liability in the self-assessed income tax, although it is difficult to prove
88
Individual Income Tax
that the tax progressivity maintained a relatively stable level from the late 1970s until about 1990 without any change in income tax structure. Are the variations in p observed in Figure 5.3 related to the direction of tax rate changes? Looking over the changes in income taxes for 1951-97, no changes occurred in the progressive tax rate schedule in 27 of these years, while tax rates were reduced in 17 years and were increased in 3 years.11 Taxes were reduced mostly in the late 1950s, the early 1960s, and the early 1970s. These variations in the progressive tax rates correspond to the fall in p. On the other hand, tax rates were changed in increasing revenues in 1967, 1968, and 1984, and they remained unchanged in most other years. These rate increases seem to have raised p and elevated the progressivity of income taxes. A flattening tax rate structure in the 1990s seems to have reduced the value of p, particularly in the case of income tax on wages and salaries. However, systemic changes in tax rates and variations in p do not always show exact correspondence. For example, 3 reached a peak in 1957 for the income tax on wage and salary incomes even though that year's tax reduction was one of the largest. This lack of correspondence may be due partly to changes in allowable exemptions and deductions and partly to factors other than tax changes influencing p. Tax Progressivity by Income Class
We have charted the value of p in relation to tax progressivity. Cross-section estimates of p based on Figure 5.2, however, are for all income classes taken together. Since not all income classes have the same P, it is necessary to examine differences in p among income classes. As an illustration, we take 197012 and calculate p for each income class under the self-assessed income tax and the income tax on wage and salary incomes. By definition, p is an indicator of income tax progressivity. Slitor (1948) suggested another indicator, denoted by 8, which is defined as
where ta = average tax rate, tm = marginal tax rate, and Y = average income of an income class. Income tax progressivity increases as the difference between tm and ta increases. 8 can be computed when Y is given an appropriate measurement unit. A comparison of 8 of different income classes enables us to determine their degree of progressivity. Since p is defined as tm/ta, P is closely related to 8. this accurately. For another, the conceptual difference of taxable income between the two types of income taxes appears to explain the different level of pi. For a more expanded discussion, sec Ishi (1976). 11 Tax reductions arising from rate changes were undertaken a number of times, but their amounts were much smaller than those of exemptions and deductions. 12 Other years show more or less similar results.
Basic Structures
Fig. 5.4
89
Tax progressivity by income class: self-assessed income tax, 1970
8 and (3 are shown by income class in Figures 5.4 and 5.5. The shaded area shows tm — ta] when the area widens, tax progressivity increases. The two diagrams lead to some interesting observations. 1. For the self-assessed income tax, neither tm nor ta shows an upward trend, even at high levels of income. Both are already stabilized at an income of ¥13.5 million and even decline slightly towards the highest income class (average income, ¥42.9 million). The distance between tm and ta does not widen with the increase in income. 2. The income tax on wage and salary income, on the other hand, shows rising tm and ta, with the gap between tm and ta widening. At the income level of ¥13 million the upward slope flattens, and the distance between tm and ta starts to narrow. 3. 8 and (3 also decline with increases in income in the case of the income tax on wage and salary incomes. However, they are higher than the comparable figures for the self-assessed income tax. Overall, the low level of (3 and its sharp decline at high-income classes for the self-assessed income tax are quite conspicuous. In particular, at the highest-income
90
Fig. 5.5
Individual Income Tax
Tax progressivity by income class: taxes on wage and salary incomes, 1970
class, which includes individuals with incomes above ¥20 million, (3 is somewhat below 1. The low progressivity at high incomes is due mostly to the particularly favourable tax treatment of high-income individuals, such as special tax measures on interest, dividends, and capital gains.13 13
We note that both average and marginal tax rates are very changeable and are high over the three and four bottom income classes. This holds for other years, too. The changeable and high rates for individuals in these lowest income classes arc not representative of the tax burdens they pay over longer periods of time than a year, because in these classes there is a heavy concentration of people with temporarily low incomes. In order to eliminate this variability in income classes below ¥700 000,1 have computed averages of their Y and T weighted by numbers of taxpayers and then calculated their average
Basic Structures
91
GENERAL FEATURES OF THE PRESENT INCOME TAX
A Hybrid Income Tax: Income v. Expenditure as Tax Bases A great deal of controversy still rages over the choice between income and expenditure as alternative tax bases.14 It seems, however, that past discussion offers no decisive conclusions on either theoretical or practical grounds. As a great deal of literature is available on this debate,15 we need not concern ourselves with the various arguments here. As has been evident from previous chapters, the Japanese tax system has traditionally taken the view that income is the best index of an ability to pay. This bias towards income originated with the introduction of a comprehensive income tax at the inauguration of the postwar tax system. At the initiative of the Shoup Mission, the income tax base was defined to include almost all forms of income, with no major exceptions. In actual practice, Japan adopted a comprehensive tax base that expanded the economic concept of income as widely as possible (see Pechman 1986, chs. 4-6). Furthermore, such broadly based taxable income was taxed strictly at progressive tax rates under a global tax system. During the years following the war, the individual income tax began to be eroded considerably by numerous special provisions. It is interesting to note that the eroding process of the comprehensive income tax base has advanced mainly in the area of income from investment and savings. Most capital gains from the sale of stock were removed from the individual income tax base, and a large portion of interest income became non-taxable under a system of tax-exempt savings. In addition, dividends, capital gains on real estate sales, and taxable interest are taxed favourably at a separate flat tax rate. How should we interpret the present structure of income taxation? Japan seems to have deviated from the comprehensive income tax and phased in a hybrid income tax in which different sources of income are taxed at different rates. It can be concluded that schedular income tax has appeared which is much like the prewar individual income tax of 1940. Moreover, it is possible to justify the basic nature of the present individual income tax in the following way. The distinction between an income tax and an expenditure tax depends upon the treatment of savings. Under an expenditure tax, either savings or interest is excluded from the tax base. As a consequence, savings cannot be subject to taxation. Similar effects of an expenditure tax on consumption over the life-cycle can be achieved by excluding interest from the tax base. This and marginal tax rates; that is, 8 and (3 are computed for an aggregate of the four or five lowest income classes. Musgrave (1959fl, 2223) also stresses this point. 14 As the third tax base, wealth is equally important; but generally, taxes on wealth are rejected as a major tax source, particularly as a substitute for income or consumption taxes. 15 See e.g. Andrews (1974), US Treasury (1977), Institute of Fiscal Affairs (1978), Kay and King (1986), Pechman (1980), Aaron and Galper (1985), Pechman (J986).
92
Individual Income Tax
argument was clearly explained by the US Treasury (1977), which preferred a pre-paid type of expenditure tax (i.e. taxing savings in advance but leaving interest tax-exempt). Either way, an expenditure tax does not discriminate against a pattern of consumption expenditure over a stretch of years, and the 'double taxation of savings' does not take place. Proponents of an expenditure tax often cite this as an advantage of choosing consumption expenditure as the tax base. The treatment of savings or investment income in Japan is almost the same as that which would be applied to all savings under an expenditure tax as described above. In effect, the Japanese individual income tax has been partially transformed into an expenditure tax. It can therefore be considered a hybrid of a comprehensive income tax and an expenditure tax,16 However, this hybrid has developed spontaneously, without any special attempt to avoid the double taxation of savings. Because of its split nature, the Japanese tax system has two distinctive options available for change in the future. One is to return in the direction of a comprehensive income tax by enlarging the tax base. The other is to move in the opposite direction towards some form of expenditure tax. Either direction can be defended theoretically. To be sure, as the third option we are also capable of preserving the current mixture of income and expenditure taxes (see Chapter 19). Income Tax Burden How do Japanese taxpayers bear the burden of the income tax? Table 5.6 provides some comparative information on the income tax burden among selected OECD countries. The highest income tax burdens are observed, not surprisingly, in Denmark, Sweden, and Finland, followed by Belgium and Canada. Both the USA and UK have relatively high ratios of taxes to the income of average taxpayers (APW), while their burdens relative to GDP are not ranked as high. What is of interest is that all these countries fell into two different patterns with respect to the change in the ratio of personal income taxes relative to GDP between 1970 and 1995: tax ratios sharply increased in Denmark, Belgium, and Italy, while they decreased in Sweden, Norway, and the Netherlands. Japan doubled the income tax burden during the quarter-century, like the first groups. Nevertheless, Japan shows the lowest figure as well as France, viewed from either a macro or a micro level of tax burden. Since France is a country in which indirect taxes predominate, it is easy to understand its rank as the lowest among major advanced countries. It is however interesting to learn that income tax burdens in Japan are the lowest by international standards, even though income tax yields Japan's major revenues. Perhaps the most important characteristics of the Japanese income tax system is the light burden that it imposes on the taxpayer. This is certainly true if we look at 16
The British tax system is more or less similar to the Japanese hybrid income tax; see Kay and King (1986, chs. 2-5).
Basic Structures
93
Table 5.6 Taxes on personal income3 as a percentage of GDP and at the income level of an average production worker (APW), selected countries As % of GDP
Denmark Sweden Finland Belgium Canada Australia Italy Norway Germany Switzerland OECD total USA UK Austria Netherlands France Japan
1995
1970
27.7 17.5 16.1 14.5 13.6 12.3 10.8 10.7 10.7 10.6 10.4 10.1 9.8 8.8 8.3 6.2
19.6 19.8 12.8 8.7 10.1 9.0 2.8 12.3 8.8 7.5 8.5 10.0 11.7 7.2 9.9 4.2 4.2
6.1
As % of the APW income 1985
34.3 33.9 25.3 17.7 10.2 16.8 14.0b 15.0 10.9 6.4 15.3 17.9 7.6 8.1 0.4C 2.8
* Includes taxes on income, profits, and capital gains (OECD classification 1100). 1984 figure. c 1983 figure. b
Source: OECD, Revenue Statistics of OECD Member Countries, 1965-96, 1997; Taxation in Developed Countries, 1987.
the low tax burden on the income of average workers. Individual income tax rates in Japan are not so low as those of other countries on a statutory basis (see Figure 5.1), but the actual tax paid by each taxpayer is extremely modest by international comparison. One of the most important reasons for Japan's low tax burden is related to major structural features of the Japanese income tax. In addition to a high minimum taxable level, generous exemptions and deductions for various incomes are allowed, and property income is favourably treated for tax purposes. Attention must now be directed towards the phenomenon of the erosion of the individual income tax in relation to the moderate tax burden.
6
The Erosion of Individual Income Tax For many years, individual income tax in Japan has comprised nearly 40 per cent of total national tax revenues. Because of its progressive character, most Japanese have accepted it as the most equitable means of distributing the tax burden among the people. In recent years, however, it has come to be greatly criticized, ironically from the standpoint of tax equity. This is because the Japanese individual income tax has been eroded by a number of special tax measures. These special provisions have resulted in the erosion of both the tax base and the tax yield, and tax erosion clearly impairs the equity of the existing income tax system. The main objective of this chapter is, first, to estimate the extent of income tax erosion by income class for 1972, 1975, 1983, 1984, 1987, 1991, and 1996 from Japanese tax data, and then, using the results, to choose a standard by which to measure the distributional effects of tax erosion and examine the extent of tax equity in the Japanese income tax system. The major conclusion of this study confirms similar studies made in the USA.
THE RELATIONSHIP BETWEEN INCOME TAX EROSION AND TAX EQUITY
Equity Considerations Tax experts view tax equity or fairness in two dimensions, vertical and horizontal. Vertical equity is concerned with the distribution of the tax burden among different income classes. Progressive taxation of all incomes on an all-inclusive basis is an essential condition in allocating different tax burdens fairly between the rich and the poor. On the other hand, horizontal equity is concerned with equalizing the tax burden among people in similar economic circumstances. This requires that the income tax base be as broad as possible. The Japanese income tax system has violated the principles of both vertical and horizontal equity to a considerable degree. This is due mainly to the fact that many special provisions have been introduced into the income tax law benefiting select groups of taxpayers. For instance, the special application of reduced tax rates to specific types of investment income favours higher-income classes, impairing vertical equity. Likewise, a number of special exclusions, deductions, and tax credits This chapter draws heavily upon Ishi (1979, 1989a).
The Erosion of Individual Income Tax
95
diminish the income tax base and create large differences in tax burdens among people with equal incomes. Obviously, the concept of tax equity and the phenomenon of erosion are closely interrelated. A study of the extent and the distribution of tax erosion would enable us to judge whether or not a specific income tax system can be regarded as equitable. Since tax erosion arising from special tax measures is actually disguised payment to individuals, it is often referred to as 'tax expenditures'. For more than two decades, 'tax erosion' and 'tax expenditures' have been frequently quoted terms in academic discussions on tax issues, and have attracted the attention of tax experts in many countries.1 There seem to be two reasons for this world-wide interest. First, there is a general trend to re-examine present tax systems in the light of tax equity. Second, increasing attention has been paid to comparing the roles of tax expenditure with those of direct subsidies. Eroding Aspects of the Present System
The fundamental structure of the present Japanese tax system was established upon the recommendations of the Shoup Mission in 1949 in the form of a progressive and broadly based income tax. The Shoup proposal dictated that a progressive tax rate structure be applied to a comprehensive definition of'taxable income'. In principle, all types of income were included in the comprehensive tax base, with no exceptions. For instance, capital gains from the sale of securities, or occasional income (e.g. winnings from horse races, prizes won in competitions, etc.), were fully subject to income taxation. Such a broad definition of 'taxable income' rendered the original postwar tax system all-encompassing. The major motivation for altering this system was to stimulate the achievement of specific policy goals, such as capital accumulation or the improvement of social welfare. In so doing, the government has often maintained that the deviation from equity is in the national interest, and that special provisions in the tax law are the most efficient way of achieving specific national goals. Income tax reforms enacted in the postwar period have included special measures such as 1. the full or partial exclusion of certain items of income from the tax base; 2. the separate application of special reduced tax rates to certain items of income; and 3. an increase in the scale and scope of deductions, exemptions, and tax credits. All three categories fall under the special tax measures, which were designed to give preferential treatment to specific sources of income and were motivated by 1 See e.g. Surrey (1973, 1981), Surrey and McDaniel (1985). The 1976 annual congress of the International Fiscal Association dealt with the topic, 'Tax Incentives as an Achievement of Government Goals', and, similarly, the 1977 annual conference of the International Institute of Public Finance considered the issues, 'Subsidies, Tax Reliefs, and Prices'. Also Surrey and Sunley (1976), reporting to the 1976 congress of the IFA, state that the official data published by the Japanese government is incomplete.
96
Individual Income Tax
policy goals. At the time of enactment, these changes seemed appropriate to both the Diet and the government. However, the special tax measures tended to eliminate huge portions of the tax base and set a precedent for enacting still more measures to satisfy special interests. In the postwar period, therefore, the proliferation of special tax measures has eroded the tax base to a substantial degree. Although some of these special provisions have been phased out, a number remain in all three categories. Typical in principle (at the time of writing in 1999) of the first type is the full exclusion of a certain type of interest income for aged groups, although a flat rate on interest income was uniformly introduced in April 1988. As an example of the second type of special tax measures, income from interest, dividends, lump-sum severance pay, and the sale of timber, as well as capital gains from the sale of securities and land, were taxed separately from other incomes at lower tax rates. This provision for separate taxation clearly deviates from the spirit of the Shoup tax system, in which all incomes were aggregated for the purpose of income taxation, regardless of their source. There are numerous existing examples of the third category of special tax measures. While the 1950 tax law permitted only several exemptions, such as the basic exemption, the exemption for dependants, and the deduction for earned income, the present income tax law institutes more than 20 different exemptions, deductions, and tax credits. The introduction of such special tax measures has contributed to the erosion of the tax base and has greatly reduced income tax revenues. Enactment of these measures has also been detrimental to the equity and fairness of income taxation in Japan. In what follows, it is assumed that it is most important to terminate the erosion of the present income tax system and to achieve income tax reform which would promote greater vertical and horizontal equity. Every departure from the comprehensive income tax has narrowed the tax base, and keeps tax rates higher than they need to be. Also, the present system violates the principle of tax equity, and makes the income tax structure more distorted and complex. The major concern of this study is to measure how much the income tax base has been eroded, thereby causing revenue loss.
P R O C E D U R E S FOR THE CALCULATION
OF TAX EROSION
Definition of a Comprehensive Tax Base Estimates of tax erosion measure the difference between the income that is subject to taxation under the current law and the tax base that would result from an elimination of special provisions. Examination of tax erosion thus depends on the calculation of a comprehensive tax base. As a basis for this calculation, it is important to 2 Of course, we can choose a different direction of tax reform if the expenditure tax is preferred to the comprehensive income tax.
The Erosion of Individual Income Tax
97
have a comprehensive definition of income.3 Past studies have traditionally adopted the definition of 'consumption plus the net increase in the value of assets during the year'. This concept of income has been defined to correspond as closely as possible to the economic income concept of Haig and Simons (see Simons 1938). It is impossible, however, to achieve a measure of income that corresponds exactly to this comprehensive concept. Difficulties in measuring such components as accrued capital gains, income in kind, and imputed rent, for example, mean that we cannot compute a truly comprehensive measure. Moreover, available data, particularly those arranged by income class, prevent us from including items which in theory should be counted as income. Thus, while applying the 'economic' income concept, we are forced to modify it in our measure of comprehensive income. In practice, it is conceptually problematic to define how far we should go in eliminating the special tax provisions in broadening the tax base. It is very difficult to decide which special provisions to retain and which to eliminate. My own preference is to approximate the comprehensive tax model as much as possible and to derive a modified concept of comprehensive income from such a model. Eroding Provisions
Erosion of the tax base has resulted from two provisions in the income tax law: those allowing for (1) exclusions, and (2) deductions and exemptions. Amounts excluded in accordance with these two provisions are added to the existing tax base to arrive at a measure of the comprehensive income tax. The specific types of income that came under these categories in 1984 are as follows: 1. Exclusions (a) interest exclusions: interest accruing from postal savings, bank current deposits, and government bonds, each not exceeding ¥3 million in principal or in total face value; interest income or distribution of profits for the formation of employees' assets not exceeding ¥5 million;4 (b) dividend income: non-taxable portions of distribution of profits from securities investment; (c) half-exclusions (or half-taxation) of aggregate capital gains5 and occasional income; (d) special preferential deductions: ¥500000 per taxpayer for timber or occasional income, and aggregate capital gains; ¥1 million for long-term separable capital gains; and (e) special deductions for retirement income (see Chapter 5). 3 For a discussion of this issue, see Royal Commission on Taxation in Canada (1966), Bittker et al. (1968), and Surrey (1973,1981). 4 For a more expanded discussion, see Ch. 8. ' This category of capital gains excludes those gains treated with the separate taxation method under the special tax measures law.
98
Individual Income Tax
2. Ordinary deductions and exemptions. About 20 kinds of ordinary deductions and exemptions are encompassed in the income tax law. Typical examples are the basic exemption, exemptions for spouse and dependants, deductions for medical expenses, and deductions for charity contributions or donations. The latest estimate in 1996 was made by a slightly different category of exclusions, mainly reflecting the fact that taxation on interest income and capital gains on securities was drastically changed in 1988 and 1989 respectively (see Chapter 8 for a detailed discussion). Briefly, the scope of interest exclusions was restricted to groups over 65 years of age and disabled people, leaving the limit applicable to non-taxable treatment unchanged (i.e. ¥3.5 million for bank deposits, post office saving and public bonds or ¥5.5 million for savings for the formation of employees' assets). In connection with these modifications to the current tax base, it is necessary to explain two particular parts of my calculations, because some aspects of my computation of the comprehensive tax base might generate conceptual debates. First, I consider all ordinary deductions and exemptions as tax erosion from the comprehensive tax base. Had I not done so, it would have been necessary to classify all items into 'rational' or 'irrational' sources of erosion (see Pechman 1984). However, such a classification poses the problem of borderline cases in which arbitrary judgement intervenes. To avoid the problem of such an arbitrary classification, I chose a broader definition of tax base erosion (in terms of all deductions and exclusions) than has been utilized in other studies.6 Second, transfer payments are not included as items contributing to the erosion of the tax base for two reasons: (1) there is a lack of reliable data by income class; and (2) most recipients would not be subject to tax. The next step is to calculate the erosion of tax revenues. Although this definition is far from satisfactory, the comprehensive income tax is defined as the actual revenue plus the additional revenues that would result from a larger tax base. There are three sources of additional tax revenues: 1. tax revenues on the additional income items listed above—on exclusions, deductions, and exemptions; 2. tax revenues that would be collected if separate taxation of interest, dividends, retirement income, and capital gains on land7 (including securities in the 1991 and 1996 cases) were replaced by progressive taxation on an aggregate basis; 3. tax revenues raised from the elimination of the following tax credits: credits for dividends, foreign taxes, incremental R&D expenditures, and acquisition of a dwelling house.8 6 In fact, Pechman (1959, 251} excluded personal exemption, which is equivalent to the basic exemption in Japan from the items that promote erosion. 7 During the years of my estimation, four kinds of income were taxed separately from other income sources under the special tax measures. For a detailed explanation, see Ch. 5. 8 A tax credit for savings to purchase a house was admitted until 1982.
The Erosion of Individual Income Tax
99
ESTIMATES OF TAX EROSION BY INCOME CLASS
Two Effective Rates of Tax The procedures described above were used to estimate the comprehensive tax base and yield for Japan in 1972, 1975, 1983, 1984, 1987, 1991, and 1996. Because of space limitations, only the results for 1996 are discussed here.9 Since the data available are scattered and relatively poor, painstaking efforts are necessary to attain a final estimation. (See the appendix to this chapter for data.) Tax erosion is analysed in terms of the effective rate of tax by income class. The 'effective rate of tax' is here defined as the ratio of tax liability to the comprehensive tax base. Comparison is made between the effective rate of tax under a comprehensive income tax (after elimination of all special provisions) and the actual effective rate of tax (the ratio of actual taxes to a comprehensive income). This comparison reveals the extent of erosion caused by the special provisions of the income tax law. Since the effective burden of statutory tax rates is altered by the existence of tax eroding provisions, the usual practice of examining the movement of nominal tax rates by income class does not clarify the true distribution of the tax burden. An examination of the difference between comprehensive and actual effective rates of tax over income classes reveals the distribution of the benefits of tax erosion, and provides a better tool for judging the equity of the tax system. It is generally believed that tax-eroding provisions have greater impact at higher income levels. In particular, it is expected that the exclusion of interest and dividends and of the separate taxation of capital gains will primarily affect taxpayers in the highest brackets. Thus, tax erosion is assumed to benefit higher-income individuals more than taxpayers in the lower brackets, and the entire tax system is thought to be much less progressive than the statutory tax rates imply. Table 6.1 contains the comprehensive tax base and yield for 1996. Using these estimates, two effective rates of tax can be derived. A comparison of these two rates is displayed in Table 6.2. Major Findings The results presented in Figure 6.1 to some extent confirm our expectations. The changes in effective tax rates resulting from the adoption of the comprehensive income tax for taxpayers are illustrated at different income levels. The top line of the graph indicates the effective tax rates that would be paid under the comprehensive income tax, while the bottom line indicates the ratio of actual tax yield to the comprehensive tax base. The various shaded areas between the two lines depict the contribution of the major structural features of the 1996 tax law to tax erosion. 9 The latest year for which necessary data were available to me at the time of writing the third draft of this book was 1996. The results of the six selected years are slightly varied, and the extent of erosion has tended to decrese as the time has progressed. Detailed analysis for the years 1972 and 1975 appeared in Ishi (1979) and that for the year 1984 in Ishi (1989a).
Table 6.1 Comparison of actual and comprehensive income tax base and yield, 1996 (¥bn.) Income class ¥(m.)
under 0.7 0.7-1.0 1.0-1.5 1.5-2.0 2-3 3-4 4-5 5-10 10-15 15-20 20-30 30-50 50 and over All classes
Comprehensive tax base
Tax yield Actual
Increase in yield from:
(1)
(2)
(3)
(4)
(5)
Separate taxation (6)
1 140.6 2 594.4 6 754.7 11280.4 35 962.6 43 206.5 43 056.3 126 862.0 34799.5 11 365.9 6 002.9 5475.6 8 714.3 337215.5
12.6 29.8 121.1 183.7 851.7 1253.6 1 338.5 5425.2 3 185.3 1 600.9 1 145.5 1119.3 1889.8 18157.1
4.2 9.1 17.6 23.1 92.4 148.0 189.0 714.4 250.1 87.3 46.9 44.5 42.6 1 669.2
170.6 298.6 552.3 754.3 2261.5 2779.5 3221.6 10544.2 2582.9 689.4 331.1 248.6 153.6 24588.1
0.0 0.0 0.1 0.3 2.3 5.0 6.6 23.4 8.6 4.9 3.8 4.4 9.2 68.6
2.6 6.3 8.5 10.0 54.1 108.7 158.4 657.3 287.5 126.7 194.6 398.5 1335.5 3348.6
Note: (7) = (2) + (3) + (4) + (5) + (6).
Exclusions
Deductions and exemptions
Tax credit
Comprehensive
190.0 343.7 699.6 971.5 3262.0 4294.7 4914.1 17364.5 6314.5 2509.1 1721.9 1815.4 3430.7 47831.7
(7)
Table 6.2
Comparison between actual and comprehensive income tax rates, 1996
Income class (¥m.)
under 0.7 0.7-1.0 1.0-1.5 1.5-2.0 2-3 3-4 4-5 5-10 10-15 15-20 20-30 30-50 50 and over All classes
Tax erosion (%)
Actual income tax base (%)
Exclusions
(1)
1.10 1.15 1.79 1.63 2.37 2.90 3.11 4.27 9.14
14.06 19.07 20.45 21.93 5.32
Tax credit
Separate taxation
(2)
Deductions and exemptions (3)
(4)
(5)
Comprehensive income tax rate (6)
0.37 0.35 0.26 0.21 0.26 0.34 0.44 0.56 0.72 0.77 0.79 0.83 0.56 0.50
14.96 11.51 8.18 6.69 6.29 6.43 7.48 8.31 7.43 6.09 5.58 4.63 2.02 7.32
0.00 0.00 0.00 0.00 0.01 0.01 0.02 0.02 0.02 0.04 0.06 0.08 0.12 0.02
0.23 0.24 0.13 0.09 0.15 0.25 0.37 0.52 0.83 1.11 3.19 7.13 14.64 0.93
16.66 13.25 10.36 8.61 9.07 9.94 11.41 13.69 18.14 22.07 28.69 33.12 39.28 14.09
Note: Percentages in each column are computed as the ratio of tax yield to the comprehensive tax base in Table 6.1.
102
Individual Income Tax
Overall, effective tax rates under the comprehensive income tax are considerably higher at every income level than actual effective tax rates. The main reason for the disparity, and largest single contributor to the gap, is the elimination of provisions for exemptions and deductions. Indeed, these items comprise a substantial proportion of the total erosion over all income classes. After deductions and exemptions, the next-largest addition to actual effective tax rates results from the elimination of separate taxation. In contrast, the elimination of exclusions and tax credits under the comprehensive tax system adds very little to effective tax rates. As expected, the impact of the comprehensive tax differs for taxpayers at different income levels. As Figure 6.1 shows, under the comprehensive income tax, the effective tax rate rises much more sharply with income level than does the actual effective tax rate, revealing the extent to which tax erosion destroys the vertical equity of the system and reduces the effective progressivity of statutory tax rates. The greatest differences between effective tax rates occur at the two ends of the income scale. An elimination of deductions and exemptions would result in a larger increase in effective rates for taxpayers with incomes under about ¥5 million, while the curtailment of separate taxation would increase tax rates substantially for taxpayers with incomes over ¥20 million. In other words, higher-income taxpayers benefit from the separate taxation provision, while lower-income taxpayers gain from the provisions for exemptions and deductions. On the other hand, the influence of erosion arising from exclusions from the tax base is equivalent over all income classes, partially reflecting the effect of special tax
Fig. 6.1
Income tax erosion by income class, 1996
Note: Figures in parentheses on the income scale are in US dollars, converted at $1 =¥125. The horizontal axis is drawn to a logarithmic scale.
The Erosion of Individual Income Tax
103
measures for the exclusion of certain interest on small deposits and postal savings, which are not heavily concentrated at the higher end of the income distribution. Tax credits, mainly for dividend payments, benefit relatively higher-income taxpayers (over ¥10 million), but these credits have only a small effect on the overall picture. In proposing more equitable income taxation, it would be too controversial to suggest the elimination of all deductions and exemptions. Because the major beneficiaries of an across-the-board elimination of deductions and exemptions would be those in the lower tax brackets, we do not need to place so much importance on this source of tax erosion in order to promote tax equity. Instead, greater attention should be paid to the second-largest area of tax erosion, the separate taxation provisions. It is the separate taxation of certain items of income that is responsible for the largest income tax erosion among the higher-income brackets. This phenomenon is most outstanding among taxpayers with incomes of more than ¥30 million. Our findings on the extent of tax erosion in lapan are similar in a couple of respects to studies carried out in the USA (see Pechman 1986, ch. 5; Pechman and Okner 1972, 28). In both countries, tax erosion is greater at higher-income levels. In the Japanese case, major causes for tax erosion are the provisions for separate taxation of capital gains and certain other types of income. These correspond to the preferential treatment of capital gains and income-splitting provisions which were found to be the greatest sources of tax erosion benefits to upper-income taxpayers in the USA.
IMPROVING THE EQUITY OF INCOME
TAXATION
One Measure of Tax Equity So far, we have devoted our attention to estimates of tax erosion by income class. As is often pointed out, the extent of tax erosion reveals how inequitable the tax burden has been. The gap between the two effective tax rates as depicted in Figure 6.1 is very important in illustrating this fact. Using these tax rates, we can compute the ratio of tax erosion, defined as follows:
where t c =the effective rate of the comprehensive income tax, and f a = the effective rate of actual tax revenues to the comprehensive tax base. As is evident from the formula, the larger the value of 8, the greater the tax erosion is and, in turn, the more inequitable the income tax burden (see Ishi 1989). In Figure 6.2, the values of 8 are depicted for five years, 1972, 1975, 1984, 1991, and 1996. Looking at the movements between each year, we can clearly see that the provisions allowing for tax erosion tend to impair the vertical equity of the tax system, particularly in the 1970s. As we have indicated, high-income taxpayers receive more of the benefit from tax erosion, in terms of greater overall reduction in effective tax rates, than do low-income taxpayers.
104
Individual Income Tax
Fig. 6.2 The ratio of tax erosion, selected years The ratios in 1983 and 1987 are omitted because they are almost the same as those in 1984 and 1991, respectively.
Most conspicuous, however, is the fall in the ratio for the upper-income brackets between 1972 and 1996. This implies that vertical equity has been recovering to a certain degree during these years. What has contributed most to this phenomenon? When we draw the same graphs as in Figure 6.1 for 1972, 1975, 1984 and 1991, the answer becomes clear. We can see that the elimination of separate taxation has played a major role in reducing the tax erosion for higher-income classes. Changes in the income tax law have reduced the scope of special provisions such as capital gains from the level reached in the 1970s. Special Treatment of Capital Gains on Land Of the types of income eligible for separate taxation at reduced rates, short- and long-term capital gains on land are the most important in estimating the phenomenon of tax erosion for selected years. Approximately 60-70 per cent of tax erosion of the top two income brackets in 1972 and 1975 is due to the separate taxation of capital gains on land through special measures in the income tax law, although these ratios have tended to decrease steadily. Special provisions were included in the 1969 tax reform to promote the supply of land and to discourage speculation on land sales. Until 1975, the new tax rates were applied to capital gains on land separately from other incomes: 10 per cent in 1969—71, 15 per cent in 1972-73, and 20 per cent in 1974—75 on long-term capital
The Erosion of Individual Income Tax
105
gains, and 40 per cent on short-term gains.10 This provision of the 1969 tax reform had far-reaching effects, but also, it substantially sacrificed tax equity. The special treatment of capital gains on land was thus a crucial factor contributing to the income tax erosion in the first half of the 1970s. In accordance with the critical view of inequitable taxation, the government in 1976 began to make some modifications towards heavier tax burdens in the income tax law for long-term capital gains on land. In 1983-84, taxable long-term capital gains of ¥40 million or less were taxed at a 20 per cent tax rate as before. In contrast, taxable amounts in excess of ¥40 million (X) were computed as follows: ~(X—¥40m.) X regular tax rate. The regular tax rate is the rate applicable to the taxpayer from the statutory tax schedule, which depends upon the taxpayer's income bracket. Thereafter until recent years, capital gains taxes on land have continued to be levied more heavily to stop land price hikes. Such modifications of the special tax provisions on land sale have contributed significantly to a slowing of the erosion of income taxation. Their success is a major factor in explaining the decrease in tax erosion observed in Figure 6.2. In addition, in 1991 and 1996 a large scale reformation of land tax was enforced, in which capital gains taxes on land were further increased (see Chapter 13 for an expanded argument). This treatment is likely to be of great use in weakening the tendency to erosion of income taxation in higher-income classes. On the other hand, investment income, such as interest and capital gains from the sales of securities, was taxed more heavily in 1988-89, as noted above. In fact, instead of eliminating non-taxable measures, flat tax rates were adopted for such investment income which is separate from other incomes. Therefore, the scope for eroding the tax base by applying separate taxation must have expanded in 1991 and 1996 due to the new treatment of the taxation of investment income in 1988-89. However, as far as the results in Figure 6.2 are concerned, the extent of tax erosion in both years shrunk to a considerable extent, as compared with that in the previous three years. The effects of increasing the burden on capital gains on land is likely perhaps to surpass those of new taxes on investment income. More detailed attention should be paid to the variation of tax erosion in 1996, as compared with 1991; i.e. the increased erosion in lower-income classes while the eroding phenomenon almost disappears in the upper-income classes. We must stress two facts. One is the large scale of income tax reduction in raising deductions and exemptions in 1996 which has contributed to increased erosion among the lower classes. The other is that capital gains tax on land was further strengthened in 1996, reducing the effect of tax erosion in the upper classes. In addition, it is important to note that capital gains on land themselves greatly decreased in 1996, reflecting the sharp drop of land prices after the collapse of the bubble phenomenon. 10 For more detailed information, see MOF Tax Bureau (1977a). The discussion between long-term and short-term capital gains is based on capital gains from property acquired before or after 1 Jan. 1969. More precisely, this type of capital gains, short- or long-term, covers not only land but also building and the right to use land.
106
Individual Income Tax
The preceding analysis has assumed that, in the change from separate taxation to full or aggregate taxation, all special measures for capital gains were eliminated. However, because capital gains on land differ markedly from ordinary income in that they cannot be realized every year, this premiss seems unrealistic. If all capital gains were subject to taxation only in the year realized, the taxpayer would be hit too heavily all at once. Instead, a modified treatment should be introduced for assessing taxes on such irregular income. For this reason, we should consider another case in which the averaging of such income for tax purposes is employed. If I make such an alternative assumption concerning the averaging method of timber income, I can develop a more realistic picture of the possibilities in the real world. (The relevant figure is omitted; for the 1975 estimate, see Ishi 1979.) Certainly, the decrease in the extent of tax erosion at higher-income levels is marked, but the remaining provisions for separate taxation still contribute considerably to eroding the effective tax rates for higher-income-earners. The estimates clearly show that provisions allowing for tax erosion violate the vertical equity of the tax system, although the extent of the erosion has gradually been reduced in the past two decades. High-income taxpayers derive more of the benefit from tax erosion, in terms of greater overall reduction in effective tax rates, than do low-income taxpayers. Moreover, inequity would be even greater if provisions for deductions and exemptions were not classified as tax erosion. Policy Implications Obviously, the elimination of tax erosion in higher-income classes is desirable, both in terms of equity and on fiscal grounds. For the same reasons, a broadly based income tax is more appropriate than the present income tax with its eroded tax base. In order to achieve such a broadly based income tax, the following three tax reforms were suggested in relation to the 1984 income tax. 1. Separate taxation should be reduced as much as possible by aggregating incomes currently taxed separately with other income and taxing the total accordingly. 2. Unnecessary deductions, exemptions, and credits should be eliminated where possible. In addition to such exemptions included in the special tax measures (i.e. the deduction for fire and other casualty insurance premiums, the deduction for life insurance premiums, the credit for acquisition of a dwelling, etc.), other provisions in the law for miscellaneous exemptions also should be eliminated. 3. Income that is currently non-taxable should be fully or partially taxed. This reform would entail the removal of all exclusions for interest income and dividends, and the full or partial taxation of capital gains from the sale of securities. In the 1991 estimate, item 3 listed above was partially enforced, although far from satisfactory. In addition to restoring the vertical equity of the income tax system, two other merits of a broadly based income tax that were suggested earlier should be recapitulated. The first is the increase in tax revenues that would be achieved
The Erosion of Individual Income Tax
107
under a more broadly based tax structure,11 and the second is the benefit derived from tax rates, which could be cut to maintain a constant yield. If the restoration of a comprehensive tax base were co-ordinated with a reduction in tax rates, moving towards the flat-rate tax, the equity, fiscal posture, and compliance rate of the tax system would all be greatly improved. Lastly, the probable effect of one omission from my calculation must be mentioned. I was unable to incorporate the phenomenon of non-reporting and under-reporting into my calculations as a source of tax erosion. If I had been able to take account of this factor, tax erosion in all income classes would have been even more pronounced.
APPENDIX Data Sources
Generally speaking, the quality and quantity of basic data for Japan are much less adequate than those of the MERGE data used by Pechman and Okner (1972) in their estimations.12 In Japan, it is administratively impossible to obtain the necessary information directly from the tax return files. We are forced to employ official tax statistics published by the government and to adjust them to obtain estimates of the desired aggregates. Methods of collecting data and making estimates by income class are even more complicated. Two basic sources for income tax data used in estimating a comprehensive income tax were obtained from the National Tax Administration (NTA). Statistics on the Self-assessed Income Tax (Shinkoku Shotokuzei no Jittai) contains data for self-employed taxpayers, and Statistics on Private Wages and Salaries (Minkan Kyuyo no Jittai) covers data for workers in the private sector falling under the withholding system. In addition, the Annual Report of National Tax Administration (Kokuzeicho Tokei Nenposho) was an indispensable source of data. However, these official statistics alone were not sufficient to complete my estimations, and I was forced to seek unpublished data at the worksheet level from the NTA. Even though I had access to unpublished data from the NTA, my data are still unsatisfactory in several ways: for instance, the basic data have not been corrected 1 ' Needles to say, all estimates in this study are confined to the 'first-order' effects of the revenue loss arising from erosion under the income tax law. No attempt has been made to take into account possible induced changes in before-tax income sources or income received by tax-payers with 'zero-elasticity assumption'. Since I assume that taxpayers try to minimize their tax liability, in practice, these 'second-order' effects would reduce the additional yield below what might be expected on the basis of my estimates. 12 For their basic data, Pechman and Okner (1972) used the 'MERGE file', which combines information on 30000 families and single persons from the Survey of Economic Opportunity (SFO) with the data from 90 000 federal individual income tax returns. Since it contains both data for low-income SEO families who are not in the tax-filing population and income tax information for higher-income individuals, the MERGE file is a much more satisfactory source for a study such as this than the data available for Japan. Furthermore, it is corrected for non-reporting and under-reporting.
108
Individual Income Tax
for non-reporting and under-reporting because no relevant information was available. Moreover, I did not have data on low-income individuals who do not file taxes. In spite of these significant deficiencies, I carried out estimations based on the income tax information available to me, as this was the only possible way of attempting to analyse the extent of tax erosion. Creating a Comprehensive Tax Base The steps in deriving the comprehensive tax base are summarized in Figure A6.1. (Note that I disregard the steps necessary to calculate the tax base by income class.) Since basic tax statistics provide data on taxable income only before deductions and exemptions, the calculations begin with these data (step .1 in the figure). However, not even these data are published directly in the statistics on self-assessed and wage/salary taxpayers. Substantial additional data collection and estimation are
Fig. A6.1
Steps in creating a comprehensive tax base, 1996.
Figures in parentheses show actual amounts in ¥000bn. for 1996.
The Erosion of Individual Income Tax
109
required to separate out the amount of deductions and exemptions (step 2 in the figure). Although deductions and exemptions are partially available from the original data on those falling under the self-assessed income tax, the data from the withholding system contain no such information. Conversely, for wage and salary workers it is necessary to calculate separately the 17 deductions and exemptions provided in the tax law. Finally, the actual tax base (step 3) is estimated by subtracting step 2 from step 1. The amount of exclusions (step 4) has been calculated by estimating from the data collected for the five items in this category. The addition of these exclusions to the tax base in step 1 gives the estimate of the comprehensive tax base (step 5). The difference between the actual and comprehensive tax bases, at the top and bottom of the figure, is the erosion from the tax base. Results of these calculations are provided by income class in each column of Table 6.1. The estimate of the actual total taxable income for 1996 is ¥148000 billion. Under the comprehensive income definition, taxable income would rise to ¥337 200 billion, an increase of 2.27 times. Erosion from the Tax Base and Yield
Estimates of the comprehensive tax yield and tax erosion can be obtained from the comprehensive tax base. Estimates of two types of tax erosion are also presented in Table 6.1. First, erosion of the tax base is computed as the sum of exclusions, deductions, and exemptions. If taxed at 1996 rates, these additions to the tax base would increase tax collection by ¥1669.2 billion and ¥24588.1 billion, respectively (see Table 6.1). Second, increased liability under a comprehensive income tax does not result solely from increases in the taxable income base. Tax credits and separate taxation represent erosion of the tax yield, and an elimination of these provisions would have a substantial impact on the total amount of taxes paid. While elimination of tax credits alone would add only ¥68.6 billion in total to the tax yield, the termination of special provisions for separate taxation would produce a much larger increase of ¥3 348.6 billion.
7
Inflation Adjustments Inflation has probably had the greatest impact on the structure of the Japanese individual income tax system during the postwar period. At least, it was so until the late 1980s. It is widely acknowledged that the major goal of income tax reductions was to adjust for the heavier tax burdens arising from inflation. Inflation causes significant increases in the effective rate of the individual income tax on real income if there are no adjustments for inflation. The Japanese government attempted to reduce the extra income tax liability arising from inflation nearly every year during the postwar period until the late 1970s, but it has ceased such adjustments since then. Although the rate of inflation declined in the 1990s, it produced distorting effects on income tax burdens until then. It is therefore important to identify how effective annual tax reductions were in adjusting the income tax for inflation, and to observe what has happened after inflation adjustment devices were dismantled. The following discussion is divided into three parts. In the first, an attempt will be made to clarify the mechanics of inflation adjustment (i.e. tax indexing), using the 1975 income tax system as a base. The second section examines the impact of the government's decision deliberately to pursue annual tax reductions rather than introduce tax indexing. Finally, a rough estimate is presented of the fiscal dividend that would have been produced by tax cut measures, and this is followed by a brief consideration of policy issues. Needless to say, focus is upon the estimation during the inflation period before 1990. INFLATION AND THE TAX SYSTEM
Distortion of the Individual Income Tax Inflation affects tax liabilities in various ways. Above all, it has a tremendous impact on an individual income tax with a progressive structure of tax rates, exemptions, deductions, and credits. The main aim of this chapter is to explore the effect of inflation on the individual income tax in Japan.1 This chapter is primarily based on Ishi (1981). 1 Of course, it is also important to pay attention to the effect of inflation on the corporate income tax. The difficulty of measuring particular items for tax purposes—e.g. capital gains, business income, interest—complicates the calculation of the tax base of corporate income tax during periods of inflation. Also, indirect taxes require some adjustment for inflation in the case of specific duties; see e.g. Maxwell (1955), Aaron (1976), Kay and King (1986).
Inflation Adjustments
111
In general, inflation induces distorting effects on both the tax base and the rate structure of income tax. As a result, tax authorities attempt to take all steps necessary to adjust for inflation. Income tax laws are generally written for a world of stable prices. Many of their provisions are expressed in fixed nominal amounts, irrespective of variation in price levels. Under the 1992 tax code in Japan, for instance, taxpayers pay 10.0 per cent of the first ¥3 million of taxable income; basic exemptions and deductions for dependants are ¥350 000 each. When prices rise, the real value of those and other quantities is reduced. At the same time, if wage incomes increase proportionally and thus remain unchanged in real terms, income tax liability automatically increases because the rate structures are progressive. Past increases in inflation have raised the question of whether and how the income tax system should be adjusted to cope with the rising trend in prices. Many countries have adopted the practice of indexing their tax structures automatically as price levels change (see Tanzi 1976). Other countries, including Japan, still pursue ad hoc remedies by periodic legislation to deal with the distortion of inflation on tax burdens. It would be interesting, in the case of Japan, to investigate how effective the deliberate measures for tax reductions have been in compensating for inflation. Inflation causes significant increases in income tax liability in two distinctive ways. First, it raises tax liability, because rate bracket boundaries are expressed in terms of money income. When prices rise, say, 10 per cent, a family's money income declines by 10 per cent in real value. Money income is usually elevated just enough to offset inflation (i.e. by 10 per cent) through wage hikes, in order to avoid a decline in real income. So, although real income remains unchanged, the family is thrown into a higher tax bracket, reflecting the illusory increase of money income. As a result, tax liabilities increase faster than inflation and reduce the increase in the family's real income. Second, an increase in tax liabilities is also caused by exemptions and deductions of fixed nominal amounts. The increment of a family's money income is fully included in progressive taxation with the levels of exemption and deduction fixed. Similarly, lower-income families who were below the minimum taxable level before inflation, and therefore held tax-exempt status, are often thrown into the tax rolls. Clearly, the structure of the individual income tax system tends to impose a higher tax burden on taxpayers whose real income stays the same in an inflationary world. Thus, when prices rise sharply over a long period of time, some means of price adjustment becomes necessary in order to reduce the increase of tax liabilities. Mechanism for Indexing Schemes How can the distorting effects of inflation on income tax liabilities be eliminated? One practical way to achieve this is to index the income tax system in order to offset or moderate the distortion of inflation on the calculation of taxation. There are
112
Individual Income Tax
Table 7.1 Effect of 10 per cent inflation on the tax liability of a family of four with an annual income of ¥5 million, 1975 1975 level plus 10% inflation (¥000)
Income before exemptions and deductions (a) less: Basic exemption Exemption for spouse Exemption for dependants Employment income deduction Taxable income Income tax liability (b) Effective tax rate (b/a)%
Case 1: 1975 level
Case 2: No adjustment for inflation
Case 3: After adjustment for inflation
(1)
(2)
(3)
5000
5500
5500
260 260
260 260
286 286
520
520
572
1450 2510 332 6.64
1450 3010 410 7.45
1595 2761 365 6.64
Note: These exemptions and deductions are based on 1975 levels. Deduction for employment income and income tax liability are calculated using Table 7.2. several kinds of indexing schemes.2 Using one of them, we shall move to the question of how and whether the Japanese income tax system can be adjusted for the effects of inflation. Generally speaking, a 10 per cent increase in prices lowers by 10 per cent the real value of the rate brackets, exemptions, and deductions under the income tax law. Thus, a price escalator is seen as a useful scheme for eliminating the distortions caused by fixed nominal amounts in the tax law. Under such an indexing scheme, fixed monetary values are increased annually at a rate equal to that of inflation. For example, assuming 10 per cent inflation now, the basic exemption and the exemption for spouse would be increased from ¥350 000 to ¥385 000 each.3 Table 7.1 shows the results of this type of indexing scheme, applied to a family of four with a ¥5 million wage and salary income in 1975 level.4 It is assumed that 2 Tanzi (1976) argues that there are four different schemes for indexing a tax system, based upon his study of such schemes in a number of countries: (1) all statutory tax rates are lowered proportionally to eliminate the increase in revenue arising from inflation; (2) the increase in income attributable to inflation is deducted from the taxpayer's gross income; (3) taxable income is deflated to a base year; (4) price escalators are introduced into the income tax structure such that income tax rates apply to conslant real incomes rather than constant nominal incomes. As Tanzi correctly pointed out, only the last two methods represent well designed schemes. We shall use these two schemes for our analysis. 3 These figures are based on the 1989-92 tax law. They were ¥290000 in 1980, ¥300000 in 1983, and were further increased up to the previous level of Y330 000 in 1984.1 do not admit the necessity of updating the data for 1975 because it was in the midst of inflation adjustments for tax policy. 4 In constructing Table 7.1,1 am indebted to Sunlcy and Pechman (1976).
113
Inflation Adjustments
Table 7.2 Ten per cent inflation, statutory tax rate, and the deduction for employment income, 1975 Deduction for employment income3
Statutory progressive rate 1975 level (¥m.)
Tax rate (%) (3)
1975 level (¥m.)
(D
After indexation (¥m.) (2)
under 0.6 0.6-1.2 1.2-1.8 1.8-2.4 2.4-3.0 3.0-4.0 4.0 and over
under 0.66 0.66-1.32 1.32-1.98 1.98-2.64 2.64-3.30 3.30-4.40 4.40 and over
10 12 14 16 18 21 24
Deduction rate (%)
(4)
After indexation (¥m.) (5)
under 1.5 1.5-3.0 3.0-6.0 6.0 and over
under 1.65 1.65-3.30 3.30-6.60 6.60 and over
40 30 20 10
(6)
a
Minimum amount deductible, ¥0.5 million. Note: In regard to the statutory tax rate, only the necessary income brackets for our calculation are listed above.
this family claims only four exemptions and deductions; the basic exemption, the exemption for spouse, the exemption for dependants, and the deduction for employment income.5 Before inflation, the family would have to pay a tax of ¥332 000 on ¥5 million, and thus would register an effective rate of 6.64 per cent. Assuming that an inflation rate of 10 per cent has been accompanied by a 10 per cent wage rise, the family income would increase to ¥5.5 million. With no adjustment for inflation in column (2), taxes would rise to ¥410000 with an effective tax rate of 7.45 per cent. This amount represents the inflation-induced tax burden. In contrast, let us assume that the principal fixed monetary parameters are indexed, as shown in column (3). Table 7.2 includes indexing for both the statutory schedules of progressive tax rates and the deduction for employment income. The deduction for employment income allows a specific percentage of deduction for each income bracket. Thus, the boundaries of each income bracket, as well as those of statutory tax rates, must be elevated by 10 per cent. Under indexing mechanics, both income and taxes would increase by the same percentages, and so there would be no rise in the effective tax rate. Undoubtedly, such a method is successful in removing the distorting effects of inflation on income tax liability. If the same family earns a different amount of income, how would the situation be altered? Table 7.3 summarizes the same calculations applied to a family of four at different income levels. Columns (1) and (2) in case 1 of the table show the actual tax amount and the effective tax rate at the 1975 level, while columns (3) and (4) in case 2 present the effect of 10 per cent inflation with no indexing. Unless there is an adjustment for inflation, income tax liability clearly increases at each income level. 5 The minimum taxable level (i.e. tax threshold) is calculated by adding these four to the deduction for social insurance premiums.
Table 7.3 Ten per cent inflationary effects on income tax liability for a family of four, selected income levels, 1975 Income level3
After 10% inflation Case 1: 1975 level
3000 5000 7000 10000 15000 20000 30000 50000 100000
(3 300) (5 500) (7 700) (11000) (16500) (22 000) (33 000) (55 000) (110000)
Case 2: No indexing
Effect of not indexing Case 3: With indexing
% increase in tax
% point increase in effective rate
Amount (¥000) (1)
Effective rate (%) (2)
Amount (¥000) (3)
Effective rate (%) (4)
Amount (¥000) (5)
Effective rate (%) (6)
(7)
(8)
98 332 678 1411 3 130 5213 9959 20650 51 691
3.27 6.64 9.69 14.11 20.86 26.07 33.20 41.30 51.69
132 410 815 1655 3596 5941 11237 23 178 57769
4.00 7.45 10.58 15.05 21.79 27.00 34.05 42.14 52.34
107 365 748 1 554 3445 5737 10957 22717 56862
3.24 6.64 9.71 14.13 20.87 26.08 33.20 41.30 51.69
23.36 12.33 8.96 6.50 4.38 3.56 2.56 2.03 1.60
0.76 0.81 0.87 0.92 0.92 0.92 0.85 0.84 0.65
Notes: Calculation in Table 7.1 is applied to selected income levels. (7) = (3)-(5)-l. (8) = (4)-(6). ' Figures in parentheses are adjusted for inflation.
Inflation Adjustments
115
In contrast, the results for indexing major monetary variables after a 10 per cent inflation are observed in case 3. It is obvious from comparing column (4) with column (6) that indexation eliminates the distorting effects of inflation on income tax liability. The effects of inflation can be judged from the two sets of indicators in columns (7) and (8). In terms of percentage increases in (7), inflation has the greatest effect on tax liabilities at the lowest end of the income scale. If there is no indexing, lower-income classes encounter greater damage owing to the increase in nominal tax burdens caused by inflation. For instance, the tax payment for a ¥3 million income increases by 23 per cent when prices rise by 10 per cent, while that for a ¥30-50 million income rises only 2.0-2.5 per cent under the same conditions. However, as far as the change in tax burdens is concerned, it is more accurate to observe it in terms of effective tax rates. The effect of inflation on effective rates is much more uniform by income classes, as is seen in column (8). On the basis of these figures, the impact of inflation increases somewhat from the lowest income levels to about the ¥10—20 million income level, and declines thereafter. An explanation for this movement seems to be related to the progressive nature of the statutory tax rate. The effect of inflation is most markedly observed at the income level having the maximum progression of statutory effective rates.6
AN OFFSETTING EFFECT OF TAX REDUCTIONS
Estimates and Data The postwar Japanese economy has undergone very rapid economic growth, which has brought about a continuous rise in nominal income. Any rise in nominal income from economic growth increases income tax liability, and inflation accentuates this pattern. Unless there is an adjustment for the rise in income, tax burdens automatically increase, as we saw from the results of Tables 7.1 and 7.2. Consequently, every nation must take steps to arrest the automatic increase in the income tax burden. In Japan, the government has not provided automatic adjustments through indexing mechanics to offset the impact of inflation on income tax liabilities. Instead, it reduced taxes nearly every postwar year by undertaking deliberate tax cut measures until the mid-1970s, and only in 1960 and 1976 were there no reductions 6 The extent of rate progression is calculated by using the concept of the income elasticity of tax yield ET in the statutory tax schedule. As in Figure 5.2, ET is defined as the quotient of marginal rate tm by average rate ta: E 7 -=f m /t 0 . At the lowest income level, ET is equal to 1 since tm = (0, following the statutory schedule. As income rises, however, the value of Er increases, reflecting the fact that tm increases faster than la. These values reach a maximum near the middle range of income and decline thereafter. At the highest income level, ET returns to 1 because ta becomes 100 per cent and tm cannot exceed 100 per cent. An income of ¥10-20 million shows the greatest disparity between marginal and average rates under the progressive pattern of the statutory tax schedule.
116
Individual Income Tax
in income taxes. From 1977 there were no major income tax cuts until the late 1980s, because the huge fiscal deficit prevented the government from reducing income tax liabilities (see Table 3.3 above). Once again in the 1990s, substantial amounts of cuts in income tax were enforced, a policy carried out to encourage the depressed state of the Japanese economy, not to adjust for inflationary bias. Thus, our major concern is with the effects of inflation adjustments for the inflationary period of 1960-90. Hence, we must question whether these annual tax cut policies have successfully corrected for the distorting effects of inflation on the individual income tax system and if so, in what way. In the past, some argued that the tax reductions proved insufficient to offset the impact of inflation on income tax liabilities.7 However, I doubt this. It seems to me that the reductions in taxes have more than offset the increases in taxes caused by inflation. It is necessary to investigate the offsetting effect of deliberate tax cuts enacted by periodic legislation. For this purpose, creating an indexing scheme with 'deflated taxes' would be of great value.8 According to such a scheme, taxable income is first deflated to a base year. Next, after the tax system is also fixed to a base year, it is assumed that tax revenues increase with the growth of real income each year. The result is deflated taxes keyed to a base year, which are thought of as tax revenues after offsetting the inflation-induced tax burdens. The concept of deflated taxes is of central importance. Deflated taxes Tdn in year n are defined according to the Sunley-Pechman formula (Sunley and Pechman 1976, 165):
where T*, Y* = taxes on taxable income, fi = the elasticity of tax liabilities to real taxable income, in a base year, respectively, and Y%= deflected taxable income in year n (« = 0,1, 2, ...), assuming automatic inflation adjustment. Necessary data can be obtained from two kinds of tax statistics collected by the National Tax Administration (NTA).9 In order to investigate the effect of inflationary adjustments, actual reductions in tax liabilities are compared with the tax liabilities that would have applied under an indexed system for two sub-periods (i.e. 1960-75, and 1975-90). We need to set out two base years, i.e. 1960 and 1975, when 7
For example, see Wada (1974, ch. 2), Nakagiri (1978); for another view, see Ihori (1984). This scheme is the third one proposed by Vito Tarm: see n. 3. Two basic sources for income tax data are obtained from the NTA data which have already been employed in Chapter 6: Statistics on the Self-Assessed Income Tax (Shinkoku Shotokuzei no Jittai), and Statistics on Private Wages and Salaries (Minkan Kyuyo no Jiltai). 8
9
Inflation Adjustments
117
both the tax system and the price level are fixed. Values of (3 are calculated by using cross-sectional data in 1960 and 1975 compiled on the basis of the two tax data sources. We have two values of (3, 1.477 and 1.819 in 1960, and 1.503 and 2.040 in 1975, for the self-assessed income tax and the income tax on wages and salaries, respectively.10 Tables 7.4 and 7.5 summarize the estimates of income tax liabilities and effective rates that would have applied to taxable income if the individual income tax had been indexed for inflation from 1960 (case 1) and 1975 (case 2). Columns (1) and (4) show the actual income and tax yields in current prices, and column (7) presents the percentage ratio of (4) to (1). In column (5) we have deflated taxes to the 1960 and 1975 price levels, calculated by the above formula. Furthermore, in column (-6) they are converted into current prices multiplied by the GNP deflator.11 The most interesting observations can be derived from the comparison between actual taxes and those assessing automatic inflation adjustment (i.e. deflated taxes) in current prices. Deflated taxes are considered as theoretical standards fully adjusted for inflation, since they are calculated according to the growth of real income and the income elasticity of tax yields, having fixed the price level and tax system to base years 1960 and 1975. In contrast, actual taxes have varied with the periodic changes in tax law imposed for the purpose of inflation adjustment.12 If the government succeeded in achieving its initial objective of inflation adjustment through deliberate tax reductions, actual taxes must be nearly equal to deflated taxes. Let us compare column (4) with column (6) in Tables 7.4 and 7.5. In case 1 of self-assessed income taxes (Table 7.4), actual tax liabilities rose from ¥95 billion in 1960 to ¥1413 billion, a factor of 15 times over the 16-year period. On the other 10 Values of (3 in 1960 were each estimated, according to the following regression equations. Selfassessed income tax:
logl 1 = 0.066+1.477 logy (20.191) R 2 = 0.926,
n=ll
Withheld income tax on wage and salary: logll = 0.049+1.819 logy (12.670) R 2 = 0.925, n=14 where 11 = per capita income tax liability, y = per capita income, R 2 = coefficient of determination adjusted for degree of freedom, n = number of observations, and figures in parentheses are t-values. The results for 1975 were calculated in a similar manner. 11 The consumer price index or the cost-of-living index might be a better measure than the GNP deflator because the tax burden of an individual is closely related to his ability to purchase consumption goods. However, because there are no data that cover the entire period using the same base year (i.e. I960), I am obliged to use the GNP deflator. There is little difference between the two indices when comparing them for overlapping years (i.e. 1967-73). 12 Tax reductions were not undertaken only for the purpose of inflation adjustment. Nevertheless, the main object of the periodic tax changes was to adjust for price hikes. Thus, it is reasonable to treat all tax reductions as if they were intended for inflation adjustment.
Table 7.4 Inflation and the self-assessed income tax liability, 1960-1990 Income (¥bn.) Current price
Base-year price
Deflator (Base year = 100)
Tax liabilities (¥bn.)
(2)
Case 1: under the 1960 tax system 1282 1282 1960 1375 1484 1961 1641 1835 1962 2252 1930 1963 2644 2167 1964 2818 2198 1965 3257 2423 1966 3943 2800 1967 3210 4718 1968 6386 4158 1969 4902 8044 1970 5912 10127 1971 11658 6487 1972 17258 8616 1973 5380 13010 1974 14339 5521 1975
(3)
(4)
100.0 107.9 111.8 116.7 122.0 128.2 134.4 140.8 147.0 153.6 164.1 171.3 179.7 200.0 241.8 259.7
95 123 144 189 216 230 265 336 420 540 664 812 I 120 1819 1165 1413
Actual
Gap: (8)-(7)
Current price (6)
Assuming automatic inflation adjustment
(7)
(8)
(9)
95 113 150 194 234 251 296 368 450 630 807 1033 1 197 1801 1318 1454
7.41 8.29 7.85 8.39 8.17 8.16 8.14 8.52 8.90 8.46 8.25 8.02 9.61 10.54 8.95 9.85
7.41 7.64 8.17 8.60 8.86 8.92 9.08 9.32 9.53 9.86 10.04 10.20 10.27 10.43 10.13 10.14
0 -0.65 0.32 0.21 0.69 0.76 0.94 0.80 0.63 1.40 1.79 2.18 0.66 -0.11 1.18 0.29
Assuming automatic inflation adjustment
Actual current price
Base-year price (1)
Effective tax rates (%)
(5)
95 105 134 166 192 196 220 261 306 410 492 603 666 899 545 560
Case 2: under the 1975 tax 1975 14339 1976 14387 1977 16107 1978 17910 1979 20625 1980 22652 23904 1981 1982 25816 1983 27535 1984 28712 30440 1985 1986 33302 37750 1987 40887 1988 48993 1989 57550 1990
system 14339 13534 14333 15229 17084 18234 18734 19889 21 100 21065 21994 23635 25803 27852 32749 37787
100.0 106.3 112.4 117.6 120.7 124.2 127.6 129.8 130.5 136.3 138.4 140.9 146.3 146.8 149.6 152.3
1413 1307 1497 1733 2192 2364 2409 2549 2678 2741 2881 3367 4187 4324 5363 6602
1414 1293 1412 1544 1819 1989 2066 2237 2417 2411 2549 2789 3111 3415 4143 4890
1413 1375 1586 1816 2196 2472 2636 2904 3154 3286 3528 3930 4551 5013 6197 7447
9.85 9.08 9.29 9.67 10.62 10.44 10.08 9.87 9.73 9.55 9.46 10.10 11.08 10.58 10.95 11.47
9.85 9.56 9.85 10.14 10.65 10.91 11.03 11.25 11.45 11.46 11.59 11.80 12.06 12.26 12.65 12.90
Note: Income, before exemptions and deductions are applied, is used for this calculation. Effective tax rates are expressed in percentages after dividing taxes by income; i.e. (7) = (4) ^ (1) X 100, (8) = (6H (1)X 100 or (5)^ (2)X100, (6) = (5)X (3)-^ 100. Base year is 1960 in case 1 or 1975 in case 2. Source: Calculated from National Tax Administration (NTA), Statistics on the Self-assessed Income Tax (Shinkoku Shotokuzei no jittai); Economic Planning Agency, Annual Report of National Income Statistics.
0.00 0.48 0.56 0.47 0.03 0.47 0.95 1.38 1.72 1.91 2.13 1.70 0.98 1.68 1.70 1.43
Table 7.5 Inflation and withheld income tax on wages and salaries, 1960-1990 Income (¥bn.) Current price
Base-year price
Deflator Base year = 100
Tax liabilities (¥bn.)
Effective tax rates (%)
Actual current price
Actual
Assuming automatic inflation adjustment
Gap: (8)-(7)
(7)
(8)
(9)
Assuming automatic inflation adjustment Base-year price
(1)
(2)
Case 1: under the 1960 tax system 3516 3516 4094 4417 5362 4796 6425 5506 7523 6166 8704 6789 10625 7459 12264 8710 14604 9935 17865 11631 13887 22788 27992 16341 32858 18285 20584 41229 22512 54435 61559 23704
1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975
(3)
100.0 107.9 111.8 116.7 122.0 128.2 134.4 140.8 147.0 153.6 164.1 171.3 179.7 200.3 241.8 259.7
(4)
174 215 252
332 392 420 441 510 613 794 1020 1 201 1 536 2184 2310 2240
(5)
174 226 289 353 413 469 529 642 752 905 1108 1329 1504
1711 1885 1992
Current price (6) 174 244 323 412 504 601 711 904 1 105 1390 1 818 2277 2703 3427 4558 5173
4.95 4.86 4.70 5.17 5.21 4.83 4.40 4.16 4.20 4.44 4.48 4.29 4.67 5.29 4.24 3.64
4.95 5.52 6.02 6.41 6.70 6.91 7.09 7.37 7.57 7.78 7.98 8.13 8.23 8.31 8.37 8.40
0 0.66 1.32 1.24 1.49 2.08 2.69 3.21 3.37 3.34 3.50 3.84 3.56 3.02 4.13 4.76
Case 2: under the 1975 tax system 1975 61559 61559 1976 71125 66909 1977 76547 68117 83555 1978 71050 1979 90777 75193 1980 98359 79178 104023 81523 1981 1982 108693 83739 1983 114980 88107 120064 1984 88088 129907 98863 1985 135202 1986 95956 1987 140 068 95736 1988 145875 99370 103473 1989 154796 1990 167128 109736
100.0 106.3 112.4 117.6 120.7 124.2 127.6 129.8 130.5 136.3 138.4 140.9 146.3 146.8 149.6 152.3
2240 2961 3139 3747 4493 5250 5780 6245 6630 6838 7661 8158 8107 8196 8442 9735
2240 2637 2726 2944 3252 3547 3716 3880 4204 4204 4634 4789 4749 5015 5318 5779
Note: See Table 7.4. Source: Calculated from NTA, Statistics on Private Wages and Salaries (Minkan Kyuyo no Jittai).
2240 2803 3064 3462 3962 4407 4742 5036 5486 5730 6413 6748 6948 7362 7956 8801
3.64 4.16 4.10 4.48 4.95 5.34 5.55 5.75 5.77 5.70 5.90 6.03 5.78 5.62 5.45 5.82
3.64 3.94 4.00 4.14 4.33 4.48 4.56 4.63 4.77 4.77 4.94 4.99 4.96 5.05 5.14 5.27
0.00 -0.22 -0.10 -0.34 -0.62 -0.86 -0.99 -1.12 -1.00 -0.93 -0.96 -1.04 -0.82 -0.57 -0.31 -0.55
122
Individual Income Tax
hand, deflated tax liabilities began at the same level in I960, and would have reached ¥1 454 billion in 1975, a level ¥41 billion higher than the actual liabilities for that year. Actual taxes showed smaller amounts every year except in 1961 and 1973. This implies that most of the past tax reductions have more than offset the impact of inflation on income tax liabilities. By and large, the same holds for case 2 using the base year of 1975. Next, we shall move to the case of withheld income tax on wages and salaries (Table 7.5). The gaps between actual and deflated revenues in current prices are even more conspicuous than in case 1 of Table 7.4. Actual taxes are smaller than deflated taxes in all years, and they are short by approximately ¥3 000 billion in 1975. Thus, it is conjectured that actual income tax liabilities on wages and salaries were much smaller than they would have been if tax laws had been left unchanged and no inflation occurred. By contrast, in case 2 the opposite phenomenon occurred for the withheld income tax on wages and salaries from 1975 to 1990. Actual taxes exceeded the level of deflated taxes in current prices, hence the amounts necessary to adjust the income tax for inflation were insufficient. Obviously, this is due to the lack of periodic tax cuts and adjustments for inflation since 1977. Tax revenues have automatically risen under a fixed tax system with a growing money income. Let us turn to the changes in effective tax rates in columns (7) and (8). Figure 7.1 illustrates the variations in effective rates for the two types of income taxes under the 1960 and 1975 tax systems, in accordance with Tables 7.4 and 7.5. It is apparent that the dotted lines, which indicate the effective tax rates assuming automatic inflation adjustment, indicate a steady upward increase. This reflects the fact that tax revenues tend to grow in real terms even though the tax system is completely indexed for inflation because of the interaction between the progressive rate structure and growing real incomes. Evident from values of (3 greater than 1, the growth of the tax revenues has been faster than that of real income. In Figure 7.1 (a), under the 1960 tax system, actual effective tax rates vary considerably in a pattern of ups and downs, reflecting the periodic changes in tax law. They tend to decline in the long run in the case of the withheld income tax on wages and salaries, while in the case of the self-assessed tax a slight increasing pattern is indicated, although it is not nearly so apparent. The deviation between the two lines is conspicuously large for the income tax on wages and salaries.13 Again, this strengthens our contention that tax reductions had a very strong offsetting effect on the tax burdens of wage and salary recipients. It appears that the degree of the offsetting effect has a close bearing on the rate of inflation. In the case of the self-assessed income tax, no effective adjustment was made for inflation in 1961 and 1973 under the sharp rise of price levels. Indeed, inflation rates were 7.9 per cent in 1961 and 11.6 per cent in 1973 in terms of 13 The chief reason for the greater reduction in the income tax on wages and salaries is that the deduction for employment income has frequently been raised on a large scale. In fact, the elevation of the deduction for employment income in 1974 reached an all-time high.
Inflation Adjustments
123
Fig. 7.1 Changes in effective tax rates, actual and after inflation adjustment, (a) 1960-1975, under the 1960 tax system; (b) 1975-1990, under the 1975 tax system (i) Self-assessed income tax; (ii) income tax withheld on wages and salaries Source: Tables 7.4 and 7.5.
the GNP deflator, which are enormously high, excluding the unusual year of 1974 just after the occurrence of the oil crisis. In both years, actual effective rates were higher than the effective tax rates after inflation adjustment, and therefore ad hoc remedies by annual tax reductions failed to correct the increased tax burden caused by inflation. The higher the rate of inflation, the more difficult it is to correct the inflation-induced tax burden in the income tax system.
124
Individual Income Tax
How, then, can we explain the case of 1974, when prices rose by over 20 per cent and yet the reduction of the tax burden more than offset the tendency of inflation to push income-earners into higher-rate brackets? We must recall that the largest amount of income tax reductions (i.e. ¥1 783 billion) in the postwar period were enacted in 1974 (see Table 3.3). As far as my estimates are concerned, this tax cut measure not only perfectly eliminated all the distortions caused by the abnormal level of inflation, but in fact offset them. Figure 7.1 (b) depicts the effective tax rates under the 1975 tax system in a similar fashion. As regards the case of the self-assessed income tax, the major findings are the same as those in the previous sub-period. However, in the case of withheld income tax, the actual effective rates outstrip the effective rates assuming automatic inflation adjustment, reflecting the end to annual tax reduction policies. Tax distortions caused by inflation were left uncorrected in the income tax system, and this resulted in inflation-induced tax burdens. The reason why the self-assessed income tax remained the same during the two sub-periods is not clear, but tax avoidance and evasion would probably explain to a considerable extent the fact that tax liabilities in the actual effective rates were lower than those yielded after inflation adjustments.
CERTAIN
ASPECTS
OF
POLICY
ISSUES
Fiscal Drag and Dividend As we have seen, income tax burdens are expected to increase automatically as a result of a progressive rate structure, coupled with an increasing growth in nominal income. Inflation tends to accelerate the tempo of such growth. This function of the individual income tax system is often regarded as a stabilizing effect which compensates counter-cyclically for the fluctuations in national income over changing business conditions. This is referred to as the built-in flexibility of taxation. However, this flexibility does not always have a beneficial effect on the economy. In the long-run process of economic growth, it tends to depress the level of aggregate demand especially during recoveries, and to slow down the potential path of economic growth. Thus, it is generally acknowledged that the depressing effect of progressive income taxes is not desirable from the standpoint of long-term policy objectives. Even if the distortion of inflation on income taxes can fully be offset, tax revenues grow faster than real incomes, and thus effective tax rates increase constantly. Let us return to Figure 7.1. The upward movement of effective tax rates, assuming automatic inflation adjustment (drawn with broken lines), indicates a clear-cut observation of this process. In the USA the phenomenon has been called the 'fiscal drag', drawing attention to the fact that it is one of the shortcomings in the mechanism of
Inflation Adjustments
125
built-in tax stabilizers in a growing economy.14 There are two policy measures that can cure the fiscal drag: (1) tax reductions and (2) increases in government expenditures. These remedies together are frequently called the 'fiscal dividend', in contrast to the concept of fiscal drag. As stressed repeatedly, the most conspicuous characteristic of postwar tax policy in Japan is the successive rounds of annual tax reductions which occurred until the late 1970s, focusing primarily on the individual income tax. This policy seems closely related to the fiscal dividend. The Japanese government selected tax reductions, rather than increases in government expenditures, as the means of effecting the fiscal dividend, mainly because it considered the tendency towards increasing tax burdens undesirable. As a result, this policy option has prevented the level of government expenditures from expanding rapidly in the past years, and has contributed to the construction of a comparatively small government. Now let us examine whether reductions in income taxes were able to function efficiently as the fiscal dividend. It has already been argued that tax reductions in the past have more than offset the increase in tax liabilities arising from inflation during most of the postwar period. This implies that some portion of the tax reductions was appropriated to reduce tax burdens in real terms which more than offset the tax burden caused by inflation. This portion must have some bearing upon the concept of fiscal dividend. It is difficult, however, to estimate what proportion of past tax cuts were meant to eliminate the fiscal drag. In crude terms, I shall attempt to estimate the size of the fiscal dividend. The first step is to calculate tax revenues that are assumed to cause fiscal drag. They must be calculated on the basis that taxes increase after adjusting for inflation because of the interaction between the progressive rate structure and growing real incomes. Until about 1970, the Japanese economy traced a growth path close to its potential output (see Patrick and Rosovsky 1976; Ackley 1976). Therefore, it is assumed that the growth of real income required for calculating hypothetical tax revenues has also been close to potential output. Given such an assumption, our hypothetical taxes can be regarded as those that induce the fiscal drag according to the formulation used in the USA. In this fashion, it is possible to consider certain aspects of the fiscal dividend in the first sub-period. The tax revenues in question have been calculated in columns (5) of Tables 7.4 and 7.5, i.e. as tax revenues assuming automatic inflation adjustment in terms of 1960 prices. In these calculations, it is assumed that the growth rate of real income is
14 The definition of fiscal drag was first developed by the Council of Economic Advisers. They stale: 'As the economy moves along the potential oulput path with reasonably stable prices, the federal tax system generates an increase in revenues of about 6 per cent a year. Unless this revenue growth is offset by reductions in taxes or by increases in expenditures, it acts as a "fiscal drag" by siphoning off income' (see Council of Economic Advisers 1962, 72-3). For reference, see also Heller (1965), Blinder and Solow (1974).
126
Individual Income Tax
14.8 per cent, and that the elasticities of tax liability to real taxable income in two types of income tax are the same as in the preceding analysis.13 Let us further assume that the hypothetical tax revenue based on the 1960 tax law has been realized since 1960. The growth rates of income taxes have become rather high; the average rate of growth is 13.3 per cent for the self-assessed tax and 17.9 per cent for the income tax on wages and salaries. If the government had left tax increases unchanged, the phenomenon of fiscal drag would certainly have occurred in the Japanese economy, as it did in the USA. It was the deliberate tax cut measures that prevented tax increases from resulting in fiscal drag. In Table 7.6, the changes in both actual and hypothetical revenues are indicated in 1960 prices, making partial use of the previous results. The gap between the two can be defined as a sort of fiscal dividend, incurred by annual tax reductions. In this context, tax reductions are assumed to remove the fiscal drag that would occur and depress the economy if the hypothetical tax revenues continued to persist in the long run. Table 7.6 Fiscal year
1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975
Income taxes and fiscal dividend, 1960-1975 (¥bn.) Self-assessed tax (1960 price)
Income tax on wages and salaries (1960 price)
Tax adjusted for inflation (1)
Actual tax (2)
Gap (D-(2) (3)
Tax adjusted for inflation (4)
Actual tax (5)
Gap (4)-(5) (6)
95 105 134 166 192 196 220 261 306 410 492 603 666 899 545 560
95 114 129 162 193 179 197 239 286 352 405 474 623 908 482 544
0 -9 5 4 -1 17 23 22 20 58 87 129 43 -9 63 16
174 226 289 353 413 469 529 642 752 905 1 108 1329 1 504 1 711 1 885 1 992
174 199 225 284 321 328 328 362 417 517 622 701 855 1090 955 863
0 27 64 69 92 141 201 280 335 388 486 628 649 621 930 1 129
Note: Columns (1) and (4) are the same as (5) in Tables 7.4 and 7.5, respectively. (2) and (5) indicate (4) in those tables after converting from current prices to 1960 prices. 15 The CEA assumes a reasonably stable price, say 2-3 per cent, but in my calculation the price level is fixed at the 1960 level.
Inflation Adjustments
127
The amount of fiscal dividend is clearly different for the two types of income taxes. The income tax on wages and salaries produces a steadily larger amount of fiscal dividend than the self-assessed income tax. In particular, it began to increase sharply after 1974 and accounted for over ¥1 000 billion in 1975. The scale of tax reductions seems to be too large around these periods compared with previous trends. In contrast, annual tax reductions do not always result in the fiscal dividend for the self-assessed income tax. As is seen in column (3) of Table 7.6, the gap shows negative values in 1961, 1964, and 1973, when tax cut measures did not perfectly offset automatic increases in tax revenues caused by inflation. In real terms, tax increases were incurred. In other years, the self-assessed income taxes incurred a smaller amount of fiscal dividend, and therefore the annual tax reductions played a minor role in adjusting for inflation. Were Deliberate Measures Successful?
Inflation adjustment for the income tax system poses an important problem, particularly in Japan. The Japanese economy has grown with great speed, and this expansion was accompanied by inflation and a sharp rise in nominal income in the 1960s and 1970s. It is easy to conjecture that inflation caused significant increases in income tax liability with distorting effects on the rate structure and the tax base of income taxes. The government selected one remedy from two possible alternatives to cure the increase in income tax burden caused by inflation. Its policy was to administer annual tax reductions through periodic amendments of income tax law. In contrast with tax indexation, deliberate tax cut measures adjusted for inflationary distortion in crude terms. As my statistical evidence reveals, such measures are far from satisfactory in eliminating the effects of inflation on income tax liability. Since ad hoc remedies were not linked precisely with the variations in both rate structure and tax exemption, only rough corrections could be made. Although tax indexation would be a better scheme to remove inflationary distortions, the Japanese government has so far resisted the adoption of such a system, and it will surely be reluctant to accept it in the future. It seems that Japan's experience has not been as successful as that of other foreign countries that adopted tax indexing systems. For instance, discretionary changes in income tax laws enacted by the Diet have more than offset the automatic increases in tax revenues and have induced tax reductions in real terms. Also, lacking a tax indexing system, it would be impossible to adopt more detailed corrections matched to each type of taxable income, e.g. interest, capital gains, earned income, etc. The inability to undertake such corrections is due to the fact that the whole tax system is adjusted for inflation in crude terms (see Kay and King 1986, ch. 13). After the late 1970s, the Japanese government stopped adjusting the individual income tax for inflation with annual tax reductions. As a consequence, traditional
128
Individual Income Tax
remedies for inflation adjustment were abandoned, and the increase of the income tax burden must have been induced substantially by inflation, despite the smaller increases in the price level. Looking back at this way of reducing income tax liability, major portions of the tax cuts were due to a rise in the minimum taxable level (i.e. tax threshold), rather than to a lowering of the progressive rate structure. In Figure 7.2 the trend of the tax threshold since 1970 is illustrated, in addition to the rise in consumer prices in terms of percentage increases in Figure 7.3. It is apparent that the level of the tax threshold remained almost unchanged between 1977 and 1986 with the exception of 1984. As a result, more income-earners were thrust above the minimal level of income subject to tax as their nominal incomes rose. Since the rate of increase of consumer prices exceeded that of the tax threshold during the decade, the real value of the minimal taxable level was lowered substantially in accordance with the rate of inflation. This level should have been raised to avoid increasing the number of taxpayers. In fact, the ratio of taxpayers among wage and salary workers to the total number of taxpayers increased steadily, from 86.1 per cent in 1977 to 91.7 per cent in 1982 and 89.1 per cent in 1985. During the same period, progressive tax rates had no major changes either, although a small reduction was undertaken in 1982, to correct the distortion caused by inflation. Since measures of inflationary correction were not adopted to enable the government make the proper adjustments in the 1980s, what has happened to the income tax system? There are two points worth noting in this context. First, real tax burdens rise because of bracket creep as nominal income increases. Thus, the
Fig. 7.2 Tax thresholds for a working family of four, 1970-1992 Source: MOH Tax Bureau, Primary Statistics of Taxation (Zeisci Shuyo Sanko Shiryoshu), Feb. 1992.
Inflation Adjustments
129
Fig. 7.3 Tax thresholds and inflation, 1970-1992 The rate of inflation is calculated based on the rate of increase in the consumer price index. Source: See Fig. 7.2.
'fiscal drag' occurs, demanding a higher proportion of the taxpayers' rising money income as tax each year. Second, the government automatically receives increasing tax revenues as a result of inflation. The impact of inflation on the tax system is very clear: individual income tax revenues rise relatively rapidly, while revenues from indirect taxes fall throughout the whole tax system. This represents a switch from indirect to direct taxation, as was seen in Table 1.3, and it should be regarded as an unintended by-product of inflation in the absence of proper adjustment. However, this trend was reversed in the 1990s with income tax reductions in the context of fiscal stimuli. In recent years, action has been taken to prevent further unplanned shifts and to restore the relative proportion of indirect taxes to total tax revenue. These efforts have proven successful to some extent in accordance with introducing the consumption tax (Japan's VAT) in the tax system. Minimum tax thresholds have been raised considerably since 1992. Indeed, the tax threshold for a working family of four was ¥3 277m. in 1993-94, ¥3 539m. in 1995-97, and ¥3 616m. since 1998. Such an upward trend has been induced by income tax reductions as fiscal stimuli, rather than inflationary adjustments.
8 The Taxation of Investment Income and Savings
An important issue relating to the Japanese tax system is the impact of income taxes on saving behaviour. Japan's high savings rate during the postwar period has received world-wide attention in connection with the country's notable economic performance. There has been a general belief that the high rate of personal savings has been stimulated by tax policy adopted by the government. In fact, the most salient feature of these special tax measures is the substantial portion of revenue losses that have been sustained in order to stimulate personal savings (see Table 3.2). Moreover, the country's relatively light tax burden may have helped to increase the savings rate. Despite great interest in such issues, past studies have reached no conclusion on the influence of Japan's tax policy in promoting saving behaviour. Since the subject is a complex one, I do not hope to achieve definitive results. We shall begin with a brief explanation of the way in which the existing income tax system tends to favour income from investment and savings, with special reference to special tax measures prior to fiscal year 1988. Next, some policy issues are discussed relating to Japan's high personal savings rate, mainly related to the late 1980s. Lastly, an attempt is made to obtain some empirical relationships between personal savings and income taxation.
MAIN FEATURES OF TAX P R E F E R E N C E S
Background It is often pointed out that special tax measures were initiated to implement specific policy goals—in particular, capital accumulation, i.e. the promotion of savings and equity investments, during the process of rapid economic growth. Capital accumulation was promoted in two ways: 1. by the full or partial exclusion of certain investment income from the tax base; and 2. by the separate application of special reduced tax rates to certain investment income.
This chapter is partially based on Ishi (1983a).
Investment Income and Savings
131
These measures were designed to give preferential treatment to interest, dividends, capital gains, and other such incomes in order to stimulate personal savings. Although some special provisions were phased out, tax preferences for savings were maintained intact until fiscal year 1987. Needless to say, such special measures applicable to specific sources of income are a clear deviation from a global tax system proposed by the Shoup Mission. There were several forms of savings that received exemption or highly favourable tax treatment. Thus, a major effect of the Japanese tax system has been its impact on saving behaviour, although the nature of its impact is difficult to pin down. The Privileged Savings System until Fiscal Year 1987 Our first concern should be with the tax preferences for specific personal savings.1 According to this system, until March 1987, income earned from interest, up to a specific limit of total principal (or total face value) of personal savings (or public bonds), was exempt from income taxation. These privileged savings consisted of five sources, including the well-known 'Maruyu' system (see note to Table 8.1). In April 1988, these special treatments were replaced by a flat rate of 20 per cent at source. These tax preferences developed gradually over a long time as a variety of government activities were designed to promote savings. For instance, the origin of tax-free postal savings dates back to 1920, and the limit on the principal was altered frequently, although the postal savings system was founded in 1875. Similarly, the idea of not taxing the interest earned on small savings was instituted in 1920 to encourage national saving for the war effort. In contrast, a provision for savings enabling
Table 8.1 Personal saving with non-taxable interest income, 1987 Sources ( 1 ) 'Maruyu' system (2) Postal savings (3) National and local bonds (4) Savings for the formation of employees' assets (5) Postal instalment savings for housing Total
Limit (¥m.) 3 3 3 5 0.5
14.5
Note: the 'Maruyu' system includes deposits at banks, securities companies and other private financial institutions. Savings under this system can exceed ¥3m., the interest of which is subject to taxation. In contrast, the amount of postal saving cannot exceed the threshold by law. Only salaried workers arc eligible for savings for the formation of employees' assets.
1 For an expanded discussion, see e.g. MOF Tax Bureau (1986), and Gomi (1985); see also Suzuki (1987) for the general discussion of financial assets in Japan.
132
Individual Income Tax
the formation of employees' assets was enacted in 1972, following the West German attempt to stimulate employees to creat their own assets. Accordingly, the individual income tax law in Japan permitted each individual to earn tax-exempt interest on personal savings of up to ¥14.5 million (US$96666, where $1 =¥150). In spite of the strict requirements necessary to take advantage of these measures, the amount of savings that was tax-favoured was abnormally high by international standards. Table 8.2 shows the significant role of personal savings under the privileged savings system in accumulating total savings. In March 1986, the total principal outstanding of personal savings, including those with both taxable and non-taxable interest, amounted to ¥503 000 billion. The largest components were deposits and savings held by banks, other financial institutions, and post offices, followed by insurance. Savings, the interest on which was exempted from taxation by the income tax law, made up approximately 60 per cent of total personal savings. Most notable is the large share of the 'Maruyu' and postal savings system, which together account for more than 90 per cent of all tax-free savings.
Table 8.2
Total principal outstanding of personal saving, by source, as of March 1986
Sources
Outstanding (¥000bn.)
Total principal outstanding Deposits and savings Private financing institutions Postal offices Trusts Insurance Security investment Bonds and debentures Investment trusts Total
318 215 103 26 98 61 45 16 503
Principal outstanding of small savings with non-taxable interest 163 Maruyu 153 Private financial institutions 10 Security companies 103 Postal savings 11 National and local bonds Savings for the formation of employees' assets 10 Total Note: Postal Instalment Savings for housing is omitted. Source: Data from the MOF.
287
(%)
(63.2)
(5.2) (19.5) (12.1)
(100.0) (56.7)
(35.9) (3.8) (3.5) (100.0)
Investment Income and Savings
133
Reduced Tax Rates on Interest In addition to the full or partial tax exemption of earned interest under the 'Maruyu' system and the other measures mentioned above, the Japanese tax system developed another type of tax concession on interest. In the case of taxpayers who exceeded the limit on privileged savings, excess interest income was taxed either separately or aggregately in conjunction with other incomes, at the option of the taxpayer. If it was to be aggregated, interest earned on taxable savings was subject to a withholding tax of 20 per cent at source and then was combined with other incomes on the tax return. As an exceptional treatment, interest on ordinary deposits and other similar deposits was taxed up to a 20 per cent flat rate at source, but this income could be excluded as taxable income on the tax return. This meant that, in practice, some interest was taxed only at the preferential rate of 20 per cent. It was also possible to apply reduced tax rates under the separation method to interest exceeding the threshold of tax-favoured savings. Taxpayers who obtained such interest could choose the option of having additional income taxed separately at lower rates. Moreover, they did not need to count such income as taxable income for tax purposes. The past record of reduced tax rates since 1951 is summarized in Table 8.3. Following relatively higher tax rates of 50 per cent in 1951—52, tax rates on interest were lowered to 10 per cent in 1953 when the Shoup proposals were substantially Table 8.3 Special tax measures for interest income: reduced tax rates in the case of separate taxation at source Reduced tax rates (%)
Period
50 10 5 10 0 0 10 10 5 10 15 20 25 30 35
1951-2 1953 (long-term saving) (short-term saving) (short-term saving) (long-term saving) (short-term saving) (long-term saving) (both) (both) (both) (both) (both) (both) (both)
1954
1955-6 1955-8 1957-62 1959-62 1963-4 1965-6 1967-70 1971-2 1973-5 1976-7 1978-87
Source: MOF, Primary Statistics of Taxation (Zeisei Sanko Shiryoshu), February 1988.
134
Individual Income Tax
modified for the first time. Furthermore, no tax was levied on interest income earned on both short- and long-term savings in 1955 and 1956. Since the 1970s, these special rates have begun to be raised in response to criticisms of inequitable income taxation. In addition, it should be noted that gains from discount bonds were subject, at the time of issuance, to a withholding tax rate of 16 per cent at source. Special Treatment of Dividends Dividends as well as interest income are treated under special tax measures. Although they have no counterpart to the 'Maruyu' system, two types of separate reduced tax rates, 20 and 35 per cent, are applied to dividends in a similar fashion. For one thing, dividends are taxed at source at the rate of 20 per cent, but are included in taxable income for filing a return. The other is that they may be levied at the reduced rate of 35 per cent, separate from other incomes, and removed from taxable income when filing a tax return. This is an optional system for taxpayers. Furthermore, recipients of dividends are now allowed to credit 10 per cent of their dividends against their income tax as compensation for the double taxation of dividends. The tax credit is reduced, however, to 5 per cent if the individual's total income is more than ¥10 million. A tax credit of 5 per cent is thus applicable to such fraction of dividends as is equal to the amount of total income minus ¥10 million, and the rest is credited against total income at 10 per cent. Capital Gains Of great interest is the favourable tax treatment given to capital gains on the sale of stocks, compared with taxes on other investment incomes. Non-taxable treatment of capital gains on the sale of stocks had become a symbol of unfairness of the existing income tax. Although capital gains had once been fully taxed (losses fully exempted) in accordance with the Shoup tax proposals, they were exempted from taxation in 1953 because the administration of the tax was too difficult to enforce. Since then, capital gains had in principle been tax-exempt, but those who continuously dealt with stocks in large volumes (thirty transactions a year, involving more than 120000 shares in 1988) had been required to include capital gains in their tax base, which was then subject to aggregate taxation. In response to criticism of the non-taxable status of capital gains, the Takeshita tax reform in 1988 proposed two alternative capital gains taxes, irrespective of the introduction of the Tax Identification Number (TIN). The first was to apply realized capital gains to the self-assessed declaration method, separate from other incomes, at a rate of 20 per cent (plus 6 per cent for the local inhabitants' tax). The second was to impose a 20 per cent tax on capital gains at the sources. Capital gains were deemed as 5 per cent of gross proceeds derived from the stock sales price. Consequently, the taxpayer would be required to pay 1 per cent (0.2 X 0.05) of the
Investment Income and Savings
135
stock sales price as a deemed capital gains tax under the withholding system. This is exactly the same as the existing securities transaction tax. The taxpayer would have the choice of self-assessed or withholding methods. Taxation of capital gains from the sale of stocks became effective from April 1989, but additional requirements were introduced. When capital gains are obtained from the initial public offering (IPO) of stocks in the market, taxpayers cannot opt for separate taxation at source. Such gains must be filed in a return if they are acquired before the IPO and sold within one year, being subject to national and local income taxes in the following formula: Short-term gains (stocks held for 3 years or less prior to the IPO)—the entire capital gains are included in taxable income. Long-term gains (stocks held for more than 3 years)—only half of such gains are included in taxable income. In addition, the taxation of capital gains from the sales of stocks was to be reviewed in the future, the aim being to move towards a unified tax from the current method of separate taxation, on condition that the system of using the Tax Identification Number could be introduced into the tax system. In relation to this shift, the possibility of treating interest income taxation in a similar way was also to be reconsidered. Moreover, capital gains on land, buildings, or the right to use land are taxed separately at low rates, depending on the length of ownership. This category of capital gains taxation has been strengthened substantially in the past decade, as mentioned in Chapter 6. Other Favourable Treatments In addition to the special tax measures applied to interest, dividends, and capital gains, there are certain types of smaller tax concessions allowed on other privileged savings. Particular attention should be paid to the following three types. Insurance is the first kind of savings that attracted tax preferences in this context. Premiums for life insurance and fire and other casualty insurance are non-taxable, with special deductions permitted up to a limit. Second, housing investments as a form of personal savings are given some favourable tax treatment. For practical administrative considerations, imputed rent on owner-occupied housing is not taxed as is normally done in other overseas countries. The interest paid on loans for house purchase and construction is eligible for a tax credit, up to an amount equal to 1 per cent of the total balance of the loan payment. Of course, there are strict requirements for using this credit in terms of the extent of floor space and the level of the borrower's income. Furthermore, the Postal Instalment Savings for housing mentioned above was created as a means for saving for the deposit to be used towards the purchase of a house. Lastly, capital gains obtained on the sale of the taxpayer's residence are treated very generously.
136
Individual Income Tax
The third type of privileged savings is public or private pension funds.2 Pension contributions made by the employer are excluded from taxable income. An employee's pension contributions are also eligible for a special deduction on social insurance premiums in the individual income tax law. On the other hand, payments made out of a pension fund are taxed fully as employment income, but an exemption for employment income and a special exemption for pension contributions for the aged have been adopted to calculate taxable income. This treatment is considered very lenient. In addition, taxes are generally not levied on the accumulated pension funds themselves, with the exception of some private pension plans. Considering the favourable tax treatment on both contributions and payments, savings made through pension funds are regarded as a means of tax-free accumulation of personal savings.
PERSONAL SAVINGS IN JAPAN: SOME POLICY
ISSUES
Significance of the High Savings Rate It is widely acknowledged that Japan's savings rate has been one of the highest among advanced countries. Figure 8.1 illustrates the movement of six major countries including Japan during the period 1965-90. Clearly, Japan maintained the highest level over two decades,3 followed by Germany and France. Indeed, the personal savings rate in Japan reached an historic high of 23.2 per cent in 1974, and then declined steadily to as little as about 16 per cent in 1985. In 1990, Japan still occupied the top rank while the USA and the UK showed personal savings rates of only 6-7 per cent. The lowest savings rate was in Sweden, where it became negative after 1985. It is well known that the high personal savings rate in Japan played a vital role in the achievement of a remarkably high-growth economy between the 1950s and the first oil shock in 1973. It contributed greatly to the financing of private investments in plants and equipment and to the rapid expansion of the productive capacity of the economy. Moreover, since the late 1970s high personal savings have enabled the country to finance the huge fiscal deficit without causing a crowding-out effect in financial markets. This situation should be compared with that in the USA and other countries with lower rates of personal savings. Traditionally, however, Japan's high savings rates have been greatly criticized overseas. The main criticisms are directed towards the fact that the excess of domestic savings over domestic investment induces a massive volume of trade and current
2 For a detailed discussion of the tax treatment of the pension system, see the Tax Advisory Commission (1986), Noguchi (1983). 3 Italy sometimes showed a slightly higher rate of personal saving than Japan.
Investment Income and Savings
Fig. 8.1
137
Movements of personal savings rates in six major countries, 1965-1995
The personal savings rate is defined as the ratio of net saving (= total current receipts - total current disbursements) to final consumption expenditure plus net saving in the account for households and private unincorporated enterprises. Source: Calculated from OECD, National Accounts, vol. II, 1983-95,1973-85 and 1962-79.
account surpluses in Japan, strengthening the country's international competitiveness in the world market. In turn, capital has flowed outward from Japan to the USA and other countries, and hence Japan's high rate of personal savings is now regarded as a primary cause of trade friction and economic disorder in the world. In addition to the macroeconomic criticisms of Japan's high savings rate, many questioned the value of maintaining favourable tax treatment for personal savings. In the late 1980s, this view attained particular prominence with the movement to eliminate inequities from income taxation. As a part of the movement towards tax reforms, there was a growing feeling that the privileged saving system should be revised. Originally this system was established to aid small savers, but it has been more widely used by those with higher incomes. This is evident from Table 8.4: the higher a family's income level, the greater the proportion that it is benefiting from the tax preferences on tax-exempt savings. The percentage of total non-taxable savings used by taxpayers tends to increase with income. Furthermore, it is generally believed that higher-income persons took advantage of this system by opening numerous small savings accounts and using fictitious or false names to evade taxation. The limits on the privileged savings system were applied to individuals, and so families as a whole were eligible for much higher limits.4 Thus, abuses were believed to be common. To the extent that higher-income 4 In general, individuals can open tax-exempt accounts in the name of their spouses, children, and relatives when they have used up their personal tax-free savings. Gift taxes should be paid on the transferred income if the income exceeds their exemption level of ¥600 000 a year. In practice, however, gift taxes were not imposed on such transfers. Thus, interest income derived from virtually unlimited amounts of personal savings is non-taxable.
Individual Income Tax
138 Table 8.4 1986(%)
Proportion of families using the privileged saving system, by income class, Maruyu
% of families using the system Annual income (¥m.) less than 2 2-3 3-4 4-5 5-7 over 7 % of families not using the system Unknown (%) Total (%)
Postal savings
78.9
56.4 71.9 76.9 83.0 88.2 93.0
(19.6) (18.9) (17.3) (19.8) (24.9) (43.1)
54.0
(13.1) (13.9) (11.5) (8.9) (12.0) (29.3)
37.8 49.9 54.0 59.3 59.3 61.8
National or local bonds
13.0
6.4 (9.1) 5.7 (20.5) 10.6 (21.7) 13.6 (16.5) 16.5 (17.4) 28.2 (30.7)
16.3
36.5
69.7
4.8
9.5
17.3
100.0
100.0
100.0
Note: Percentages in parentheses are the proportions of the total number of families that use the full amount of privileged savings. Source: Council for Savings Promotion (Chochiku Zokyo Chuo linkai).
families tend to benefit more from this system, it should be considered unfair or inequitable. Are Tax Benefits
Effective?
Next, we shall consider the effects of large tax benefits on personal savings. There is still controversy about the relation between the high rate of savings and the tax-exempt status afforded by many types of savings. Foreigners in particular often claim that the high savings rate in Japan has been created by these tax concessions. In Japan, opinion is divided among academics, businessmen, and politicians. Some believe that the tax concessions have substantially contributed to the high rate of Japanese savings, while others think they have contributed little or nothing. It seems to me that the special treatment of privileged savings has played no significant role in encouraging greater saving. It is almost impossible to estimate empirically the impact of tax incentives on high personal savings (see Komiya 1966a; Pechman and Kaizuka 1976). A great number of studies have sought possible explanations for Japan's high savings rate,3 5 See e.g. Mizoguchi (1970), Shinohara (1982), Kosai and Ogino (1984), Horioka (1986), Ishikawa (1988). In most of the literature, the relation between personal savings and taxation alone has not been studied.
Investment Income and Savings
139
but it seems that no attempt has so far been successful in determining how much of the high savings rate is attributable to tax policy and how much to other factors. The high rate of personal savings has also been explained by various other factors, such as rapid income growth, the bonus effect, the life cycle hypothesis, the socio-cultural background of savers, the under-developed social security system, etc. (For an excellent survey of the literature, see Horioka 1985.) Many, if not all, of these possible factors help explain Japan's high savings rate, and tax incentives should be regarded merely as one of a number of minor factors. A lot of tax experts, including some members of the Tax Advisory Commission, are sceptical about the effectiveness of special tax measures. They point out that the desirable effects, if any, of the special tax measures have been achieved at the expense of simplicity, neutrality, and equity in the individual income tax system. As already noted in Chapter 6, many of these measures impair the progressiveness of the nominal tax rates as they apply to specific individuals. While the intended effects of tax incentives are ambiguous, their distortion of the structure of individual income taxation is apparently clear-cut. Some reference should also be made to the interest elasticity of savings, because this plays an important role in determining whether the special tax measures can significantly influence the level of personal savings. If the interest elasticity of saving is substantially greater than zero, tax incentives for saving would have a substantial impact on the level of personal savings through the effects on the after-tax real rate of return. As is well known, the sign or magnitude of the interest elasticity of savings is theoretically indeterminable because of two offsetting effects: substitution and income effects. Thus, our concern is with the empirical results of using econometric models. In the USA recent evidence suggests that the interest rate elasticity of saving is positive and large in magnitude (see Boskin 1978; Summers 1981; Kotlikoff 1984; etc.). If the same evidence could be found in Japan, this would imply that various incentive measures for saving have had a significant positive impact on the level of personal savings, because they lowered the effective tax rate on interest and in turn increased the after-tax real rate of return. Unfortunately, I have obtained no convincing evidence in Japan which would suggest a significantly positive value for the estimated elasticities in question. Perhaps such poor results can be attributed to the past regulation of interest rates. Much of the affected savings were thus earned below market-regulated rates because of government intervention. A typical example was seen in the interest earned on postal savings. On this point, the regulation of interest rates can be regarded as a rationale for tax-exempt saving because interest income had already been 'taxed' through the regulatory limits on interest rates. However, these savings accounts have been deregulated, and their interest rates have been adjusted to market levels. Consequently, as time passes, we may see better results for the interest elasticity of saving under deregulation.
140
Individual Income Tax EMPIRICAL RESULTS OF PERSONAL SAVINGS AND TAXATION
The Effect on Portfolios Although scepticism is widespread about the effect of tax incentives on the level of personal savings, it appears that the special tax measures have some effect on saving behaviour. In fact, empirical evidence suggests that tax incentives may have greater influence on the portfolio decision of households (i.e. on the form in which households wish to hold their personal savings) than on the total level of savings. The first three rows of Table 8.5 show the proportion of typical forms of financial investments (e.g. (1) savings deposits, (2) bonds and equity, and (3) life insurance and pensions) to the net acquisition of financial assets in Japan, the USA and the UK for selected years. In comparison with other countries, Japan's portfolio holdings are characterized as follows. First, and most importantly, the proportion of personal holdings of savings deposits occupies more than 60 per cent of total financial assets. It accounted for 70.2 per cent in 1995, and the comparative figures were 2.6 per cent in the USA and 30.1 per cent in UK. This means that Japanese households have given priority to savings deposits as a vehicle for personal savings. Second, life insurance and pension Table 8.5
Proportions of portfolio investments in the household sector,'1 selected years (%) Japan
Proportions of financial investments (1) Saving deposits'1 (2) Bond and equityc (3) Life insurance and pension Financial investments (4) The ratio of netd lending to gross capital accumulation
USA
UK
1975
1985
1995
1975
1985
1992
1975
1985
1987
66.2 12.0 10.8
63.5 2.4 28.3
70.2 -10.3 39.5
49.8 5.5 32.0
23.8 14.6 45.1
2.6 42.8 48.7
44.1 -3.6 43.9
33.5 2.5 44.0
30.1 15.0 37.3
57.4
64.8
59.7
44.7
21.5
11.8C
56.3
36.7
38.5f
Note: Percentages in (1 )-(3) are calculated as the ratio of each portfolio investment to total acquisition of financial assets. Since other items are excluded, totals are not equal to 100%. " Includes private unincorporated enterprises. b Deposits other than transferable deposits. ' Short-term and long-term bonds and corporate equity security. Negative figures in Japan and the UK imply larger values of capital loss in equity. '' Net lending means financial investments after subtracting real investment (i.e. increases in stocks, gross fixed capital formation, purchases of land, etc.) from total capital accumulation. c 1994 figure. 1 1994 figure. Source: OECD, National Accounts, vol. II, 1973-85 and 1983-95.
Investment Income and Savings
141
funds are not so attractive to households in Japan, although the relative shares have accelerated rapidly. By contrast, households in the USA hold over 40 per cent of their financial assets in the form of life insurance and pension plans, and those in the UK hold even higher proportions. Third, portfolio investment, which takes the form of bond and equity securities, was present in a relatively lower proportion in Japan than in the USA and the UK in 1990. The next question concerns the proportions of total assets that have been allotted to financial and to real investments. It is difficult to determine this because of the lack of reliable data, but the ratio of net lending to gross capital accumulation might be of great help here. This ratio roughly indicates the relative share of financial investments in total capital accumulation.6 Accordingly, the higher the value of this ratio, the greater the number of households holding financial assets. In turn, these investments help to finance plant and equipment in the private sector and to make up for fiscal deficits in the government sector. The ratios of net lending to gross accumulation in Japan are much higher than those in the other two countries (see row (4) of Table 8.5). It is of great interest to observe the 1985 figures: Japan showed 64.8 per cent in 1985, as compared with 11.8 per cent in the USA and 36.7 per cent in the UK. Purchases of financial assets in Japan have been maintained at a high level. Given these data, it is often conjectured that households in Japan tended to exhaust the tax-exempt portions of the privileged savings accounts first (the 'Maruyu', postal savings, and national bonds, in that order), and then to direct their income towards the purchase of real assets (e.g. housing and land) and equity investments. In general, however, individuals have limited access to both of these. The Japanese tax system must have had its impact on saving behaviour, mainly reflecting a hybrid of income and expenditure taxes in favour of special treatments on investment income or savings. Three features of a hybrid income tax affect households' portfolio decisions. First, investment income arising from saving deposits, bonds, and equity has greatly benefited from highly favourable tax treatment. As stressed earlier, we have long enjoyed the privileged savings system and the tax exemptions for capital gains on the sale of stock. This has resulted in a relatively larger proportion of savings deposits, bonds, and equity in Japan. Second, special tax measures on the other forms of personal savings, such as insurance and pension funds, have not developed to the same extent as in other countries, say the UK. For instance, premiums for life insurance are given only a small deduction, and private pension funds themselves are specially taxed. Overall, income tax has not played a vital role in establishing forms of personal wealth other than the privileged savings system. 6 Net lending is the difference of real investments (i.e. the sum of increases in stock, gross fixed capital formation, net purchase of land and intangible assets, and capital transfers) and total gross accumulation. It indicates the money amounts appropriated for the purchase of financial assets.
142
Individual Income Tax
Third, the interest paid on loans is usually not deductible under the income tax law, unlike in the USA where households are allowed to deduct 100 per cent of their interest payments. In addition, current laws do not allow taxpayers to deduct their local property tax liability from their national income taxes. Capital gains on the sale of land and buildings are taxed more heavily, although ad hoc tax relief was given during a specific period. It is permitted under the income tax laws to deduct the interest paid on owner-occupied mortgages, but the extent to which home owners are allowed to deduct this interest is much limited. Less favourable tax treatment for the purchase of real assets has resulted in a smaller share of total capital accumulation going towards real estate investment. Determinants of Personal Savings: An Estimated Result In spite of some recent results in explaining Japan's high savings rate,7 it is still very difficult to determine the direct impact of tax incentives on savings from empirical data. The subject is a complex one, and the present discussion is intended more to provide a sense of the argument than to develop it in detail. The light tax burden may be a significant factor in the explanation of the high rate of personal savings in Japan (see Pechman and Kaizuka 1976, 369). Indeed, it has often been argued that the high personal savings rate may be due to the low level of taxes and social security contributions. In particular, this argument seems to be widely accepted within the government.8 We can identify this relation in international comparison since the personal savings rate and the ratio of tax burden (including both direct and indirect taxes) plus social security contributions to personal income tend to show a negative correlation (see Ishi 1983a).9 Although some correlation between savings and taxation can be obtained from an international comparison, we need to investigate Japan's case in greater detail. What factors determine the variation of personal saving in Japan? In this regard, the aggregate savings function seems to be of use to obtain a rough sketch of major determinants. An attempt is made to apply Taylor's type of saving function to the Japanese case. As formulated by Taylor (1971), the reduced-form estimation equation is
where S = personal saving, S__i = personal saving in the preceding period, Yw= labour income, Tr= transfer income (i.e. social security benefits, social assistance
7 In recent studies, the relative importance of new explanatory factors (e.g. the low ratio of the aged to the working-age population, or higher land prices) is added to make the discussion more convincing (see e.g. Horioka 1986; Ishikawa 1988; Yoshikawa and Ohtake 1988). 8 This view was a 'Kasumigaseki' (the name of the district in Tokyo where the main governmental office buildings are located) theory. Tor example, an official statement in support of this theory was published in EPA (1982). 9 Personal income consists of (1) wages and salaries of employees, (2) the self-employed income, and (3) property income in the national account concept.
Investment Income and Savings
143
grants, and unfounded employee welfare benefits), Yp = property income (i.e. interest, dividends, and rent and entrepreneurial income), S/= social security contributions, Tp = personal tax (including non-tax revenues), r=the interest rate on one-year time deposits, and u = an error term. Although the derivation of (8.1) is complicated, the equation itself is rather simple. It requires only the regression of personal savings on its value in the preceding period and the first differences of the components of disposable income. In addition, the estimated model contains one additional variable, the interest rate on one-year time deposits. With the exception of the interest rate, data are obtained from 'Income and Outlay Accounts in Households' in the Annual Report on National Accounts published by the Economic Planning Agency. The quarterly data and nominal values are employed. Observations cover the period 1965 (III)-1991 (I), and the sample period is divided into three sub-periods mainly because tax policy changed substantially in the late 1970s and mid-1980s. (See Chapter 3.) Empirical results are tabulated in Table 8.6, in which four equations have been estimated for each period. Equation (1), based upon the disaggregation of disposable income and the interest rate as an additional variable, is the full model. Equation (2) is estimated with social security contributions and personal tax combined, since the coefficient of personal tax is not statistically significant. Equation (3) is added to (2) with labour and transfer income combined. Finally, equation (4) is intended as a benchmark for comparison, and differs from (2) in that disposable income is not disaggregated. Major findings are summarized as follows. 1. The most important explanatory factor is personal saving itself during the preceding period in all equations. 2. The coefficient of transfer income in the first period is not statistically significant. If it is combined with labour income in equation (3), the result becomes meaningful with a positive sign. 3. Property income has a significant relationship to the variation of personal savings, as was expected, with minor exceptions. 4. Personal contributions to social security have negative coefficients in all equations. Likewise, personal tax has the same result except during the first period. This suggests that in the short run the bulk of the adjustment to a change in social security contributions and personal tax falls on savings rather than consumption. Thus, it seems that households consider social security contributions to be a form of tax. In addition, tax incentives on savings seem to be obtained from our estimate since the late 1970s. 5. The interest rate has no significant impact on the variation of personal savings. This is quite natural, given the fixed interest structure and savers' behaviour during the past years. With regards to tax incentives, a number of reservations are required before reaching more conclusive results. For example, the aggregate savings function
Table 8.6 Equations estimating personal savings, 1965-1998 Equations
Personal saving
St-i
Changesa Labour income
Transfer income
Property income
AY"W
ATr
WP
Personal contributions to social security AS7
R2
DW
0.984
2.547
Personal Interest tax on 1 -year time deposits AT p Ar
(a) Quarterly 1965-7 6b 0.821 (1) (15.266)
1.283 (5.121)
1.668 (0.873)
0.813 (2.015)
(2)
0.857 (19.674)
1.166 (4.955)
0.799 (0.576)
1.025 (2.850)
-1.877 (-2.961)
0.983
2.608
(3)
0.853 (21.519)
1.024 (2.880)
-1.974 (-3.959)
0.984
2.646
-1.993 (-4.085)
0.984
2.670
0.998
2.887
0.998
2.817
(4)
1. 147 (5. 198)
0.856 (22.768)
1.106 (7.336)
(b) Quarterly 1977-86C 0.949 (1) (100.654)
0.561 (10.386)
1.457 (5.656)
0.682 (4.353)
0.949 (104.918)
0.552 (10.674)
1.620 (9.480)
0.682 (4.450)
(2)
-2.572 (-2.411)
-0.131 (-0.089)
-1.034 (-5.010)
-1.211 (-14.623) -1.198
( -14.857)
-400.755 (-1.151)
-97.299 (-0.541)
(3)
0.955 (82.044)
(4)
0.956 (75.661)
0.767 (19.942)
0.224 (1.395) 0.666 (56.451)
(c) Quarterly 1987-1998d 0.949 (1) (55.803)
0.560 (14.293)
1.516 (17.704)
0.715 (3.920)
(2)
0.952 (60.060)
0.557 (14.595)
1.521 (19.549)
0.720 (4.177)
(3)
0.934 (33.197)
(4)
0.942 (31.764)
0.806 (15.791)
-0.007 (-0.025) 0.699 (23.069)
-0.999 (-10.923)
0.997
2.648
-1.213 (-23.306)
0.996
2.596
0.987
3.051
-1.204 (-13.783)
0.988
3.047
-0.973 (-6.471)
0.961
3.124
-1.291 (-14.481)
0.956
3.072
-1.182 (-8.588)
-1.196 (-12.347)
-244.935 (-0.632)
Note: Data are quarterly before seasonal adjustments. R2 is the coefficient of determination adjusted for degrees of freedom. DW is the Durbin - Watson statistic and the values in parentheses are t-statistics. a b c d
First differences. From 1965 (III) to 1976 (IV). From 1977 (I) to 1986(1). From 1986 (II) to 1998 (I).
146
Individual Income Tax
formulated by Taylor (1971) seems to be too crude to obtain any meaningful estimates of tax incentives on savings. Most salient is the fact that the coefficient of A T is an amalgam of two effects: (1) variations arising from changes in tax rates, exemption, and deductions, and (2) changes in the tax base, reflecting those in the general level of economic activity. Unfortunately, it is very difficult to isolate each change into separate components.10 If we could distinguish one factor from another, more conclusive results might be obtained in this estimation. According to my present calculations, tax incentives cannot be interpreted as having been a relevant factor contributing to the high savings rates in Japan for one decade after 1965. More attention should be paid to other possible factors, such as socio-cultural factors, the country's rapid economic growth, etc. Even if no special tax measures had been devised as incentives for saving, the savings rates would still have remained at a high level. While the effects of tax incentives on savings are ambiguous, their deleterious effect in producing tax erosion is very clear. As tax erosion in higher-income classes is considered evidence of an unfair tax system, its elimination would be desirable in terms of equity. At present it would no longer be necessary to use the tax system to stimulate savings while sacrificing tax equity. This will remain the case while personal savings stay at a high level. Greater importance should be attached to eradicating inequities in the present tax system in order to prepare for possible tax increases in the future. On the other hand, however, it is anticipated that the rate of personal savings will fall persistently towards the twenty-first century, as has been observed in Figure 8.1. This would be mainly explained by the coming of an ageing society in Japan. This being the case, it might be possible to revive tax incentives on savings again to promote the higher rate of personal savings. 10 It might be possible to avoid some of the shortcomings of the crude model at the macro level if we estimate the equations of the saving function at the micro level, using the household budget data. In so doing, it is necessary to devise some tax parameters including changes in tax law for savings-induced measures.
9 Effects of Taxation on the Distribution of Income It is widely acknowledged that taxation can reduce the inequality of income distribution or at least slow its growth. In particular, a progressive tax system is expected to play an influential role in determining the size distribution of income. When considering the distributional effects of taxation, emphasis should be placed on the individual income tax, because it is typically equipped with steeply progressive rates. Since a greater tax burden is thereby placed on higher-income classes than on lower-income-earners, it is commonly believed that income tax substantially narrows income inequality. The purpose of this chapter is to correlate information about the redistributional effects of taxation with the distribution of income, and to interpret them as one aspect of policy behaviour in Japan. Given the current state of available data, it is not easy to determine the influence of taxation on the distribution of income. Frequently, sceptics question the quality of the statistics that are obtainable and the results derived from them. The following discussion will be treated in two steps. First, I shall examine the influence of the individual income tax on income inequality, using tax statistics collected by the National Tax Administration (NTA). Based on these data, we can clarify the redistributional effects of taxation on income-earners (e.g. on wages and salaries, self-employment income, property income). Second, I shall use the same procedure on other data sources to ascertain the equalizing effects of income tax on income distribution. Similar calculations are made to assess income distribution more comprehensively by occupation, using non-tax statistics released by the Ministry of Health and Welfare (MHW).
MAIN FEATURES OF THE STUDY
Data Sources
Ideally, the impact of income tax on income equality should be measured by reliable statistics based on a comprehensive definition of income, including accrued or realized income (see Chapter 6). The nature of the data available in Japan, however, leads us to depart from this ideal level to a considerable extent. Before beginning major calculations, let us examine the data sources and their adequacy.
148
Individual Income Tax
Broadly speaking, two kinds of statistics are available for the purpose of our estimations: (1) tax statistics, and (2) other statistics, based mainly on the Annual Household Survey. Whichever type is employed, data on the distribution of income suffer from a lack of comprehensiveness with regard to both population coverage and the definition of income. In this study the tax statistics data are utilized more, for the following reasons. 1. 2. 3. 4.
Information relating to tax revenues is more accurate. Time-series data on individual categories of income-earners are available. Data on higher-income classes are more abundant and reliable. The movement of certain capital gains is captured to some extent.
For our purposes, the accuracy of the information on taxes is most important, because the main aim of this analysis is to focus on the measurement of the redistributional effects of income taxes, rather than on the income distribution itself, as in many other studies.1 Hence, tax statistics provide a better data source than other statistics in that the latter have no specific intention to compile information relating to tax payments. Also, with regard to reason 2, tax statistics permit a more detailed inspection of income-earners for tax purposes, while statistics from the Annual Household Survey are confined to wage and salary-earners. Furthermore, it is extremely important that data on higher-income brackets can be obtained copiously from tax statistics. Under the income tax law pertaining between 1984-86, progressive tax rates began at 10.5 per cent of taxable income up to ¥500 000 and rose to a maximum of 70 per cent above ¥80 million. Since the redistributive effects of tax depend greatly on this progressive rate structure, basic data should cover as many income brackets as possible, especially at the higherincome levels. The highest income bracket found in tax statistics is that above ¥50 million, while the one used in other statistics is usually defined at much lower income levels, at best ¥10 million. Thus, we can get only very limited information for our purposes from non-tax statistics. Since capital gains are not included in the concept of national income, the usual approach for analysing income distribution tends to omit capital gains from the definition of income. They are, however, a very significant item in taxable income, and they play an important role in influencing income distribution. Therefore, capital gains cannot be neglected in this study, although the data on capital gains are very restricted even in tax statistics. On the other hand, tax statistics are not free of problems in collecting the necessary information. For instance, a basic criticism is that the income recorded on tax statistics may be subject to under-reporting and may be less accurate. However, the major proportion of income tax revenues (i.e. more than 80 per cent) is derived from taxes withheld at source from wages and salaries. In contrast, less than 20 per cent of total income is taxed under the self-assessed income tax provisions, whereby 1 There are a number of estimates in Japan that are not directly related to taxation. For example, see Mizoguchi and Takayama (1984) and their reference list.
Effects on Distribution of Income
149
individuals file their own tax returns and are therefore liable to under-report their incomes. Another criticism is based on the argument that tax statistics do not cover those lower income-earners who do not pay taxes. If this study were aiming to consider the redistribution of income through government expenditures (e.g. transfer payments), the lack of data on income-earners below the minimum taxable level would be significant in view of the redistributional effects of budgetary policies. In this study, however, the main focus is on the tax side of the budget, with the assumption that government expenditures will be taken as given so as to isolate the influence of income tax. Therefore, non-reporting persons, who do not need to pay taxes owing to exemptions and deductions, can naturally be omitted.2 A Simple Measure of Income Inequality Next, we shall examine the measurement of income inequality. Several alternative indicators are available, and reliable methods to measure income distribution have developed in recent years.3 Chiefly for the simplicity of calculation, I have utilized the familiar approach of the Lorenz curve and the Gini coefficient. As seen in Figure 9.1, the percentage of incomes cumulated from lowest to highest is plotted on the vertical axis, and the percentage of households or persons cumulated from lowest to highest is plotted on the horizontal axis. Obviously, the more uneven the income distribution, the greater the curvature in the Lorenz curve. Thus, the ratio of the area between the Lorenz curve and the line of equal distribution to the entire area below the line of equal distribution is useful in clarifying questions of income equality. This ratio is called the 'Gini coefficient'. By using before-tax and after-tax data, we can draw two curvature lines on the map of the Lorenz curve and compute two Gini coefficients; one is for the before-tax ratio, Rb, and the other for the after-tax ratio Ra. The effects of taxation on income distribution can be defined a follows:
tp is called the 'equalization coefficient'. The larger
150
Fig. 9.1
Individual Income Tax
The Lorenz curve, before and after taxes
the withheld income taxes on wages and salaries. The former is levied on selfemployed and professional incomes, agricultural incomes, property incomes, etc.; for all of these types of income, the taxpayer must file a tax return. The latter type of tax applies only to wage and salary earnings. In this chapter, analytical results will be shown separately for each type of data source. Since accurate data are available from 1951,4 the estimates are made during the period 1951-97. Careful treatment is absolutely necessary to obtain consistent time-series data (see Ishi 1976, ch. 4).
MAJOR FINDINGS USING TAX STATISTICS
The Extent of Redistributive Effects Before directly investigating the redistributive effects of income taxes, let us outline the long-term variation of before-tax income distribution by using tax statistics. The results are shown in Figure 9.2, where the two types of before-tax income distribution are depicted in terms of the Gini coefficient. (See also Tables A9.1 and A9.2 in the Appendix to this chapter.) The most outstanding feature of the results is that the 'self-employed and other incomes' category can be seen to move towards increasing income inequality, while for wage- and salary-earners such inequality declined markedly, after 1956—57. Particular attention should be paid to the sharp rise of the former during the period of the bubble in the late 1980s when income distribution became markedly unequal 4 As regards the self-assessed income tax, data are also available for the years 1948-50. Since these data were collected on a rudimentary basis, 1 did not use them.
Effects on Distribution of Income
Fig. 9.2
151
The movement of income distribution in terms of the Gini coefficient, 1951-1997
Gini coefficients of before-tax income are depicted in appendix Tables A9.1 and A9.2.
reflecting the sharp rise of capital gains due to asset inflation. However, after the collapsing bubble phenomenon in 1991, this type of income has moved sharply toward more even distribution. Interestingly enough, the results of the wage-salary earnings are nearly the same as those attained with other, non-tax, data sources (see e.g. Economic Planning Agency 1975). The critical point of 1956-57 can be explained with reference to the excess labour force that existed in the first part of the 1950s, combined with the labour shortage of younger workers in the latter part of the decade. The labour shortage tended to raise wages, particularly for the younger workers, which caused income inequality to narrow. Further inspection of this phenomenon is not required for the purpose of this study. Next, let us shift our attention to the equalizing power of the income tax as seen in Figure 9.3. The long-term movements of equalizing coefficient
152
Fig. 9.3
Individual Income Tax
Redistributive effects of income tax, 1951-1997
Notes: Equalization coefficients in appendix tables A9.1 and A9.2 are drawn here.
capital gains. These items of income tend to be concentrated at higher-income brackets; therefore the self-assessed income tax levied on them has greater power to equalize income distribution through its progressive structure. During the period 1969-73 its effect weakened considerably, and this requires some specific explanation. However, the equalizing power of self-assessed income tax was sharply lowered in the 1990s by two factors; first, a great reduction in property income and capital gains caused by the dissipation of the bubble phenomenon, and second, income tax progressivity moderated by adopting a flatter rate structure. Relation to Tax Changes We have so far investigated the simplest aspect of the influence of income taxes on economic inequality in terms of
Effects on Distribution of Income
153
income tax system. The former either reinforces or counteracts the equalizing effects of progressive tax rates. For example, some reference should be made to the following two cases when the equalization coefficient
154
Individual Income Tax
frequently on a large scale in the 1950s; this then explains the sharp decline of
Effects on Distribution of Income
155
Fig. 9.4 Redistributive effects of self-assessed income taxes, by types of income, 1951-1997
clarifying the changes in the redistributive effects of the self-assessed income tax. For example, for several years, starting with 1969, the equalizing power of the self-assessed income tax was lower than that of the withheld income tax on wage and salary earnings. This is rather unusual, since the self-assessed income tax should play a more important role in equalizing the distribution of income. This unexpected result requires further explanation. Incomes subject to the self-assessed income tax are divided into four categories: 1. 2. 3. 4.
the self-employed income; agricultural income; professional income; and capital gains and other related incomes.7
Items 1-3 may be included in a broader category of business income, while item 4 may be defined as non-business income. Let us investigate the redistributional effects of tax on each item of income as classified above. The results are indicated in Figure 9.4. Until the late 1960s, the movement of capital gains and other related incomes differed substantially from that of the three other incomes. Indeed, capital gains and other related incomes strengthened the equalizing power of taxation with increased values of
156
Individual Income Tax
the self-assessed income tax as a whole in the 1950s (see Figure 9.3) is associated with the similar patterns revealed by the taxes on these three incomes. On the other hand, the fact that the equalizing power of the self-assessed income tax declined after 1969 can be explained by the movement of tp in capital gains and other related income during the same period. Since 1969, the pattern of tp in the Table 9.1 Percentage distribution of taxpayer, taxable income, and tax yield under the self-assessed income tax, 1955-1995 (%)
Taxpayer 1955 1960 1965 1970 1975 1980 1985 1990 1995 Taxable income 1955 1960 1965 1970 1975 1980 1985 1990 1995 Tax yield 1955 1960 1965 1970 1975 1980 1985 1990 1995
The self-employed
Farmer
41.5 45.9 40.1 39.4 37.1 35.9 30.9 29.3 26.5
36.8 18.7 8.5 8.1 6.3 2.3 4.3 2.9
4.9 5.2 6.1 6.7 7.3 8.2 9.0 7.8
2.4
39.1
29.7 11.2
37.4 27.0
23.5 20.4
19.8 17.3 14.0 14.9 54.4 40.3 25.0 23.5 14.5
9.4 8.2 10.0 10.5
4.4 3.7 2.7 1.3 2.2 1.4 1.4
18.9 3.8 1.6 1.4 0.9 0.4 0.6 0.6 8.6
Capital gains and other income-earners
Total
7.5
16.8 30.1 45.3 45.7 49.2 53.1 55.8 60.0 63.6
100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
5.8 6.2 7.9 7.8 9.8 11.8 10.3 6.1 7.2
25.4
45.2 60.7 65.0 67.1 67.1 71.2 78.5 76.5
100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Professional
8.0 6.6 11.5 13.1 16.2 21.6 16.9 6.7 9.1
18.7 49.3 61.9 61.9 68.6 68.6 74.3
82.6 71.8
100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0
Source: National Tax Administration, Statistics on the Self-assessed Income Tax (Shinkoku Shotokuzei no Jittai).
Effects on Distribution of Income
157
case of the self-assessed income tax in Figure 9.3 appears to be almost the same as that of the capital gains and other related incomes in Figure 9.4. Particular attention is paid to the declining trends of each item in the 1990s, reflecting the adoption of a flattening, progressive tax structure. Table 9.1 shows the percentage distribution of the four income categories in terms of the numbers of taxpayers, the taxable income, and the tax yield during selected years. No matter what indicator is used, capital gains and other related income-earners continue to increase their relative share, while the other three types of income-earner tend to decline or remain almost constant. Thus, the relative significance of capital gains and other related incomes has tremendously increased its share in the self-assessed income tax since the mid-1960s. Earners of capital gains and other related incomes are likely to be concentrated unevenly at higher-income brackets. Progressive tax rates applicable to such income must therefore function very effectively to redistribute it. This function successively raised the value
RED1STRIBUTIVE EFFECTS OF TAXATION BY HOUSEHOLD UNIT
Household Survey Data The preceding analysis has focused on the income distribution of individual income-earners, using tax statistics. Alternatively, it is possible to employ data on 8 This is quite evident from the estimates of the Gini and equalization coefficients of capital gains: they show far higher values than those of other incomes. For a detailed discussion, seelshi (1980). 9 More exactly, this kind of capital gain covers not only land but also building and the right to use land. For more detailed information, see MOF (1977a). 1(1 Capital gains sharply increased from 1968 to 1969, by 291.5 per cent, while non-capital gains showed only an 18.1% increase.
158
Individual Income Tax
the household unit for the same purposes. The basic objective of using household survey data is to supplement the previous measurements by computing income tax liability on an individual basis. In past studies based upon the household unit, main coverage was restricted to wage and salary-earners whose data were largely derived from the Annual Household Survey. However, in this study analysis is extended to include various types of household units by a cross-sectional approach, rather than by a single household unit on a time-series basis. For this purpose, more comprehensive income distributions, which are based on field surveys carried out by the Ministry of Health and Welfare (MOHW), are available. Information is given for six selected years—1952, 1962, 1967, 1972, 1975, and 1984—but I shall estimate figures for 1972 and 1984. Unfortunately, the necessary data are not available for more recent years because MOHW statistics are no longer comparable because of a change in the way in which information is collected. Surveys cover families of the self-employed, farmers, and those relying on property income, as well as families of wage and salary-earners. Income as defined by this survey includes earnings from employment and self-employment and all property income, but excludes capital gains. Although the MOHW survey places greater weight on the expenditure side of the budget (e.g. the social security programme), necessary information on taxes is also found in the statistics. However, the individual income tax is mixed in with other types of taxes—i.e. inhabitants' tax (local income tax), property tax, and tax on motor vehicles. Thus, it is impossible to single out the income tax from these tax data. However, the relative share of national and local income taxes appears to predominate in the total. The effect of taxation on income distribution can therefore be attributed chiefly to income tax. The Differing Effects on each Household Household units, which differ slightly between 1972 and 1984, are broadly classified in two main categories: (1) non-farm employed and self-employed households, and (2) farm households. Each category has its own sub-categories. Using the same procedure described before, we summarize the final results in Tables 9.2 and 9.3 with the Gini and equalization coefficients. Let us first examine the 1972 case in Table 9.2. Several interesting facts emerge. First, when comparison is made between the two major categories, the non-farm households show a more equalized before-tax distribution of income than the farm households. However, the redistributive effects of taxation are more effective among the non-farm households. Stated differently, tax effects on the farm households are extremely limited. Second, of the three sub-categories of employed households, the most important is the household whose head is working in an enterprise with over 1000 employees or who is employed in the public sector. Families belonging to this group enjoy higher
159
Effects on Distribution of Income Table 9.2
Redistributive effects of taxation, by household, 1972 Gini coefficient
Household
Equalization coerncient W
Before tax
After tax
Rt
Ra
9
Total
0.468
0.453
3.34
Non-farm households Employed households Less than 30 employees 30-999 employees Over 1000 employees and public servants Self-employed households With employees Without employees Other households
0.459 0.419 0.624 0.445 0.226
0.442 0.404 0.613 0.431 0.214
3.64 3.57 1.74 3.22 5.42
0.446 0.246 0.559 0.824
0.430 0.224 0.550 0.822
3.60 8.61 1.72 0.30
Farm households Full-time Part-time
0.518 0.698 0.385
0.510 0.694 0.375
1.60 0.52 2.68
Source: Calculated from figures in Ministry of Health and Welfare, Surveys on Income Redistribution (Shotoku Saibunpai Chosa), 1972.
Table 9.3
Redistributive effects of taxation, by household, 1984
Household
Gini coefficient
Equalization coefficient (%)
Before tax
After tax
Rfc
Ra
Total
0.406
0.387
4.72
Non-farm households Employed households Less than 30 employees 30-999 employees Over 1000 employees and public servants Self-employed households With employees Without employees Other households
0.407 0.322 0.332 0.290 0.243
0.387 0.304 0.322 0.278 0.230
4.87
0.463 0.467 0.410 0.711
0.424 0.416 0.395 0.729
10.81 3.84 -2.64
Farm households
0.398
0.385
3.36
9
5.61 3.05 4.25 5.08 8.44
Source: Calculated from figures in Ministry of Health and Welfare, Surveys on Income Redistribution (Shotoku Saibunpai Chosa), 1972.
160
Individual Income Tax
income levels, and hence the equalizing power of the progressive rate structure imposed on their income is stronger. The opposite applies to households whose heads are employed in firms with 30-999 or with less than 30 employees: their before-tax distribution of income is more unequal, and the equalizing power of taxation is weaker. Third, the same phenomena emerge in the case of self-employed households. Self-employed families whose firms hire employees represent a relatively large scale of enterprise. They earn greater self-employed income, and thus are in higher income brackets where the income tax can perform its redistributive function. Consequently, their value of
are observed in employed households whose head works for a firm with over 1000 employees and in the larger scale of self-employed households. In contrast, the 'Other households' and full-time farmers show the weakest redistributional effect. The results of the 1984 survey are indicated in Table 9.3, and they reveal almost the same findings as in 1972. The ranking of each household in terms of the value of tf remains unchanged, but the redistributive effects of taxation are strengthened in 1984 for every category except 'Other households'. OVERALL
A S S E S S M E N T Of
TAX E F F E C T S ON
INCOME DISTRIBUTION
Our major concern was with the following question: To what extent has the individual income tax played a role in equalizing the distribution of income in Japan? 11 The majority of farmers in Japan have some side occupations. In fact, less than half of Japan's 5 million farming households can be termed full-time farmers: the rest are merely 'part-timers'. This phenomenon is due to the increasing availability of job opportunities in secondary and tertiary industries, coupled with the introduction of labour-saving mechanisms in agriculture. Usually, the head of the household and his sons are engaged in non-agricultural jobs, while farm labour is performed by the women and old people. 12 This is a 'kuroyon' phenomenon—see Ch. 4.
Effects on Distribution of Income
161
Although attention may be directed towards the redistributive impact of inheritance and gift taxes on wealth distribution, they are less significant than the equalizing function of income taxes in mitigating the uneven distribution of income and wealth. Despite the important limitations of available data, several significant facts have been observed concerning the redistributional effects of income taxes. First, the equalizing powers of income tax on income distribution continued to weaken until the mid-1970s. In particular, the sharp decline of the equalization coefficient
162
Individual Income Tax APPENDIX Table A9.1 1951-1997
Redistributive effects of the self-assessed income ta> Gini coefficient
1951 1953 1955 1957 1959 1961 1963 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
Before tax Kb
After tax
0.322 0.314 0.272 0.332 0.373 0.449 0.453 0.439 0.449 0.445 0.443 0.501 0.524 0.565 0.550 0.590 0.515 0.537 0.490 0.492 0.499 0.515 0.524 0.525 0.527 0.521 0.516 0.519 0.532 0.562 0.559 0.587 0.605 0.595 0.518 0.525 0.515 0.515 0.521 0.510
0.285 0.279 0.249 0.312 0.353 0.425 0.427 0.414 0.424 0.418 0.414 0.478 0.503 0.547 0.525 0.565 0.491 0.509 0.464 0.464 0.470 0.482 0.493 0.496 0.500 0.493 0.488 0.492 0.503 0.529 0.528 0.569 0.586 0.579 0.508 0.514 0.503 0.506 0.514 0.505
Ra
Equalization coefficient (%)
Source: Calculated from NTA, Statistics on the Self-assessed Income Tax (Shinkoku Shotoku/.d no Jittai).
163
Effects on Distribution of Income APPENDIX Table A9.2 Redistributive effects of the withheld income tax on wage-salary incomes, 1951-1997 Gini coefficient
Rt
After tax Ra
0.359 0.381 0.392 0.415 0.416 0.402 0.374 0.344 0.339 0.330 0.332 0.320 0.317 0.316 0.314 0.315 0.324 0.301 0.302 0.302 0.307 0.312 0.317 0.319 0.320 0.328 0.329 0.334 0.335 0.336 0.340 0.343 0.344 0.343 0.345 0.344 0.350 0.350 0.348 0.349
0.326 0.348 0.366 0.398 0.401 0.387 0.358 0.325 0.321 0.314 0.314 0.303 0.303 0.303 0.300 0.299 0.314 0.229 0.292 0.292 0.296 0.299 0.303 0.305 0.306 0.313 0.314 0.318 0.320 0.321 0.325 0.329 0.329 0.327 0.328 0.327 0.334 0.337 0.336 0.337
Before tax
1951 1953 1955 1957
1959 1961 1963 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997
Equalization coefficient (%) V
9.18 8.70 6.55 4.09 3.60 3.63 4.12 5.66 5.20 5.12 5.36 5.18 4.48 4.15 4.60 5.20 3.06 3.04 3.31 3.39 3.58 4.00 4.23 4.41 4.47 4.46 4.49 4.53 4.72 4.61 4.29 4.14 4.48 4.78 4.90 4.83 4.68 3.68 3.63 3.60
Source: Calculated from NTA, Statistics on Private Wages and Salaries (Minkan Kyuyo no Jittai).
This page intentionally left blank
Part III Corporate Taxation and
Taxes on Capital
This page intentionally left blank
10
Principles of Corporate Taxation The tax on corporate profits was first introduced in 1899, 12 years after the imposition of the tax on individual income. It was combined with the individual income tax until 1940, when separate taxation for individual and corporate incomes began. During the postwar period, the corporate tax has steadily developed into the largest or second-largest source of tax revenue in the national tax system. Corporate incomes are also taxed at the local government level (i.e. by prefectures and municipalities). In addition to an enterprise tax at the prefectural level, both prefectures and municipalities levy a local corporate income tax known as the 'inhabitants' tax'. As local taxes will be discussed in Chapter 18, here we shall focus mainly on the national corporate tax. The major issues relevant to corporate taxation can be divided into two groups: (a) general features, and (b) incentive policies using special tax measures. The former is treated in this chapter, while the latter is dealt with in the next. There are three points in particular that we shall examine here. First, we shall explore the basic nature of the corporate tax; then we shall proceed to consider integration problems between the corporate and individual income taxes; lastly, we shall investigate other important issues in relation to structural changes, business financing, or the incidence of the tax.
GENERAL CHARACTERISTICS OF THE CORPORATE TAX SYSTEM
Relative Importance of the Corporate Tax Corporate taxes produced more revenue in Japan than the individual income tax in 13 of the 50 years from 1950 to 1999. This phenomenon was particularly conspicuous during the rapid growth period (i.e. 1957-64). In the remaining years the corporate tax was the second-largest revenue after individual income tax, and it still retains a significantly high share in total national tax collections. What is of most importance is that the relative share of the corporate tax in the national tax revenues in Japan is the largest among major advanced countries. In 1997 these proportions were 24.2 per cent in Japan, 18.0 per cent in the USA, 13.4 per cent in the UK, 12.2 per cent in France, and 4.8 per cent in Germany. There are two explanations for the heavier reliance on corporate taxes in Japan. First, there has been a growth of economic activities in the corporate sector, reflecting the high growth rate of the Japanese economy throughout the postwar period. Second,
168
Corporate Taxation and Taxes on Capital
the incorporation of self-employed small businesses has been accelerated by tax inducements. Computation and Payment of Tax The corporate tax is levied on the net income earned by both domestic and foreign corporations in each accounting period or in liquidation. The net income is roughly the excess of gross revenues over expenses incurred doing business. Corporations are required to calculate their net income once during each accounting period (i.e. any one year selected by them), and to file a tax return within two months after the end of the accounting period. Interim returns must be filed and tax payments made for the first six months within eight months after the beginning of the period. The calculation of corporate net income generally accords with the actual practices of the modern business accounting system, although some adjustments must be made in arriving at taxable income. Thus, corporate net income for tax purposes conforms to corporate profits before tax in the profit and loss statement of corporate firms. When corporations terminate going concerns by merger or dissolution, the corporate tax is levied on their liquidation income. In computing the taxable income of corporations under the 1997 corporate tax law, there are several points that deserve special attention (see MOP Tax Bureau 1997). 1. Capital gains of 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 owned by corporations are additionally levied a special tax burden. 2. Dividends received from other corporations are fully excluded from taxable income, but the excess of dividends received over those paid is only partially exempted (i.e. 80 per cent of net dividends received). 3. Inventories are permitted to be valued at cost or market value, whichever is lower, by using any of eight methods of valuation (e.g. the actual cost method, LIFO, FIFO, etc.). 4. Based upon original cost and shorter useful lives, depreciation is calculated by the straight-line method or the declining-balance method. Generous provisions have been allowed for accelerated depreciation, increased initial depreciation, and special tax-free reserves to stimulate particular types of investments. 5. Corporations filing a blue return may carry back net operating losses so as to offset them against the previous year's taxable income1 or may carry losses forward for five years. In effect, this provides a six-year period for offsetting losses against gains. These carry-back and carry-forward systems are required to avoid taxing corporations with fluctuating incomes more heavily than those with relatively stable incomes. 1
The carry-back of losses was temporarily disallowed between 1 April 1 992 and 31 May 2000.
Principles of Corporate Taxation
169
6. If current outlays for research and development (R&D) exceed the largest amount outlayed for such purposes during any accounting period since 1966, 20 per cent of the increase may be credited against the corporate tax in the year it is made. The maximum amount creditable is 10 per cent of the corporate tax liability. In addition, a tax credit of 7 per cent is adopted for specific property used for R&D of basic technologies.
Taxable Corporations Since the corporate tax is levied on corporations, it is necessary to define the scope of corporations subject to corporate taxation. For tax purposes, corporations include not only those in the ordinary sense (e.g. a joint stock company) but also co-operative associations and certain specific organizations. These corporations are grouped into the following five categories. (Both numbers and percentages relate to 1996.) 1. Ordinary corporation (joint-stock companies, medical corporations, mutual insurance companies, etc.) 2. Co-operatives (agricultural co-operatives, consumers' co-operatives, fishermen's co-operatives, etc.) 3. Non-profit organizations (religious and educational organizations, labour unions, private foundations, etc.) 4. Non-juridical organizations with designated representatives 5. Foreign corporations Total
2 694 814 (96.8%) 57 124 (2.1%) 24 075 (0.9%) 6 283 (0.2%) 1 735 (0.1%) 2784031 (100.0%)
It can be seen that ordinary corporations account for more than 95 per cent of total corporations in Japan, followed by co-operatives. The number of firms in the remaining categories is negligible. The Corporate Tax Law exempts a variety of non-profit organizations, including religious, charitable, and educational organizations, labour unions, etc., but the tax does apply to the 'unrelated business income' of these organizations. For this purpose, such non-profit organizations are listed among taxable corporations. 'Unrelated business income' is defined as income from a business that is not directly related to the carrying out of religious, educational, or other tax-exempt functions. This income is now subject to the reduced tax rates noted below. The favourable tax treatment for non-profit organizations has been controversial. Numerous complaints have been made against them for unfair competition with private profit-making businesses. These organizations might not have been subject to corporate taxation had they remained small and stuck to fund-raising activities.
170
Corporate Taxation and Taxes on Capital
Table 10.1 Numbers of corporations by the scale of paid-in capital, 1996 Paid-in capital (¥m.) less than 5 5-10
10-100 100-1000 over 1000 Total
Profit-producing corporations (1)
Deficit-operating corporations (2)
Total
270453 (31.5) 89371 (10.4) 480933 (55.9) 15283 (1.8) 3599 (0.4) 859639 (100.0)
717184 (45.5) 212904 (13.5) 630171 (40.0) 13411 (0.9) 2440 (0.2) 1576110 (100.0)
987637 (40.5) 302275 (12.5) 1 111 104 (45.6) 28694 (1.2) 6039 (0.2) 2435749 (100.0)
(3)
(2)1(3) (%) (4) 72.6 70.4 56.7 46.7 40.4 64.7
Note: Figures are based on a sample survey of corporations, and so the totals are not equal to those mentioned in the text. The target of the survey is limited to ordinary domestic corporations. Source: National Tax Administration, Annual Report, 1997.
In recent years, however, their business-related activities have become spectacular in the grey zone of unrelated business areas' (e.g. fund management), and as a result, activities permitted to them have been spelled out in the tax code. It is often argued that the special tax rates applicable to the unrelated business income should be raised to fill the gap between them and the regular tax rates. A classification of corporations, arranged according to the scale of paid-in capital for profit or deficit operating firms, is depicted in Table 10.1. Two points are worth noting. First, small-scale corporations occupy the dominant share of the total: 40 per cent of all firms are small companies with less than ¥5 million of paid-in capital. Big corporations with more than ¥100 million of paid-in capital occupy only 1 per cent of the total. Obviously, the Japanese 'small corporations' would be classified as self-employed businesses in other countries.2 Second, we note the large share of deficit-operating corporations in the total. The ratio reached nearly 50 per cent (i.e. 48.4 per cent) in 1990, having accelerated from 35.9 per cent in 1965, and 43.0 per cent in 1975, but having decreased from 55.4 per cent in 1985. Furthermore, the smaller the scale of a business is, the more likely that 2 It was relatively easy for self-employed entrepreneurs to incorporate in Japan until 1993. The Commercial Law required them to assemble only five shareholders and ¥350 000 of paid-in capital when they established a joint-stock company. In the case of forming a limited company, requirements were even more generous: two shareholders and ¥10 000 of paid-in capital. Partnerships, limited or unlimited, could be registered as corporations with only two shareholders and no provisions for paid-in capital. However, the Commercial Law was amended in 1993, and minimum paid-in capital of ¥10m. is required to establish a joint-stock company.
171
Principles of Corporate Taxation
it will be a deficit-operating corporation. Indeed, smaller businesses with less than ¥5 million of paid-in capital make up 54.8 per cent of deficit-operating corporations, while the largest companies represent merely 21.4 per cent. It is widely pointed out that such increases of deficit operating corporations of smaller size may be caused by small companies manipulating expenses to avoid tax liabilities. Thus, some argue that corporate taxation in terms of floor space, number of employees, or sales rather than profits might be introduced to tax deficit-operating corporations which have been sustained for a long time. Tax Rates In Japan, the corporate tax is levied at a flat rate on most corporate income, but rate structures have become much more complicated than those under the Shoup tax proposals. Under the 1999 corporate tax law, the structure of corporate tax rates is summarized as follows. 1. Ordinary corporations (a) Basic rate 30.0% (b) Lower rates as a concession to small businesses 22% (These rates are imposed only on taxable incomes of less than ¥8 million in corporations with less than ¥100 million of paid-in capital. The remainder is taxed at basic rates.)
Table 10.2 Nominal tax rates among major advanced countries, 1998
Nominal rates
Japan
USA
UK
Germany
France
34.5%
35%
31%
33-^%
Local tax:
State corporate tax in California: 8.84%
Retention: 45% Dividend: 30% State enterprise tax: 18.5%
31.00%
51.67
3^-%b
Prefectures 1.73%
Municipalities 4.24%
Enterprise tax: 11% Total effective rate
46.36%
40.75%
Note: Total effective rates are calculated on nominal rate structures. * Surtax is levied as 7.5 per cent of corporate tax liability. '' 20 per cent of surtax is imposed in the same way. Source: Data presented in the Tax Advisory Commission.
172
Corporate Taxation and Taxes on Capital
2. Co-operatives 22% 3. Non-profit organizations 26% In addition to these tax rates on corporate profits, certain portions of income in liquidation were taxed at a rate of 30.7 per cent (27.1 per cent after April 1999). Table 10.2 permits an international comparison of corporate tax rates in 1998. Japan's rate structure is unique in that corporate income is levied at all three levels of government. Assuming some ratio of dividends, we can compute these ratios into a single rate in terms of nominal rates, which is then regarded as the total effective rate. This rate in Japan is ranked second-highest, following that of West Germany. As argued later, nominal rates do not necessarily show real tax burdens on corporations, because the effects of special tax measures are neglected.
INTEGRATION
OF THE CORPORATE AND
INDIVIDUAL INCOME
TAXES
Structural Features When the Shoup Mission proposed the tax reform plan in 1949, the most important change in the field of corporate taxation was the introduction of the idea that the corporate tax was an advance payment of the individual income tax of the shareholders. This was based on the concept of the corporation as a mere aggregate of stockholders. The system proposed in order to achieve this aim consisted of four elements (see Shoup Mission 1949,106f.): 1. a tax of 35 per cent on the net income of each corporation; 2. a credit for each individual shareholder against his individual income tax of an amount equal to 25 per cent of the dividends received from corporations subject to the tax imposed in (1), assuming that such dividends are themselves included in the net income of the shareholder; 3. exemption of dividends received by a corporate shareholder from the corporate tax; 4. an interest surcharge on the corporation of 1 per cent for each year of the aggregate retentions accumulated out of the net earnings of fiscal periods beginning after 1 July 1949. The first two items consisted of an approximate withholding tax at a flat rate of the share in the earnings of the corporation, and an approximate allowance for this withholding tax in the return of the individual shareholder. This adjustment was introduced in order to avoid double taxation of dividends, reflecting the Shoup Mission's basic view of the nature of the corporation. Therefore, the corporate tax proposed by the Mission can be regarded as an integrated system of corporate and individual income taxes with a dividend-credit for individuals. The corporate tax proposed by the Shoup Mission of 1949 was regarded as ideal by many economists, but at the time it seemed to be too idealistic to fit with the
173
Principles of Corporate Taxation
reality of the economy. Thus, tax reforms have been undertaken each year since 1951 to readjust the corporate tax system to fulfil the basic needs of economic growth, such as the need to encourage capital accumulation. In Table 10.3, the changing patterns are summarized since the introduction of Shoup taxation in 1950. During the earlier period of corporate tax reform, the abolition of interest surcharges on the retained profits of corporations was initiated in 1951 in order to promote profit retention. On the individual income tax reform, capital gains from securities transactions were made non-taxable in 1953, primarily because the
Table 10.3 Historical patterns of the corporate tax rate, 1950-1999 Accounting period
General tax rate (%) Retentions
Income demarcation
Retention
Dividends
1 April 1950 1 Jan 1952 1 July 1955 1 Oct 1955 1 April 1957 1 April 1958 1 April 1961 1 April 1964 1 April 1965 1 April 1966
38 38 37 35
28 26 26 26
1 May 1970 1 May 1974
36.75 40
26 28
1 May 1975
40
30
1 April 1981
42
32
April 1984 April 1987 April 1989 April 1990 April 1998 April 1999
43.3
33.3 32 35
Reduced tax rate (%) Dividend
35 42
—
—
40
— below ¥0.5m. below ¥lm. below ¥2m. 33 below ¥2m. 33 below ¥3m. below ¥3m. 31 28 below ¥3m. (only applicable to corporations with capital of not more than¥100m.) 28 same as above 28 below ¥6m. (same as above) below ¥7m. 28 (same as above) below ¥8m. 30 (same as above) same as above 31 same as above 30 same as above 29
— 35 35 33
40 40 38
42 40
37.5 34.5 30.0
24 22 22 22
22 22 22 24 25 24 26 28 25 22
Note: Bold-face figures indicate tax rate changes as result of tax reform. Source: MOF Tax Bureau, Primary Statistics of Taxation (Zeisei Shuyo Sanko Shiryoshu), February 1999.
174
Corporate Taxation and Taxes on Capital
taxation of capital gains allegedly hampered the development of the securities market. Moreover, a new exemption system on dividends was adopted for increases of equity capital in 1954, and a reduced tax rate on dividends was introduced at the company level in 1961. In 1955, a reduced tax rate was applied to incomes of not more than ¥0.5 million in order to alleviate the tax burden on small and medium-sized firms. The reduced rate was 35 per cent, in contrast to the general corporate tax rate of 40 per cent. This treatment has been extended frequently, as shown in Table 10.3. However, in 1966 use of the reduced tax rate was limited to corporations with capital of not more than ¥100 million. From the viewpoint of the integrated tax system, the most marked reform of the corporate tax occurred in 1961. In that year, a German-type split-rate system was introduced to help corporations increase equity capital vis-a-vis borrowings. That is to say, the tax rate on distributed profits of corporations was reduced by about one-fourth, from 38 to 28 per cent, while the tax on undistributed profits remained unchanged at 38 per cent. Correspondingly, the dividend credit against income tax was diminished from 20 to 15 per cent of dividends, and the exclusion from taxation allowed for inter-corporate dividends was also reduced, from 100 to 75 per cent of the amount in excess of dividends paid by recipient corporations. This structure had not been changed substantially until two split-rates were merged into one rate in 1990. The reduced tax rate on dividends was eliminated and merged with higher rates on retained profits, mainly because it was generally acknowledged that reduced rates on dividends had had no bearing upon the development of stock markets through increased dividend payments. During the 1960s, corporate tax rates were allowed to decline as much as 35 per cent on retentions and 26 per cent on dividends so as to encourage an increase in retained earnings and to foster capital markets. Since 1970, there has been a trend towards raising tax rates. The shift took place mainly for two reasons: to find replacements for revenue losses arising from individual income tax reductions, and to avoid the heavier reliance on fiscal deficits. In 1981, 1.3 per cent was provisionally added to the regular rates of 42 and 32 per cent in the form of ad hoc measures, but these were terminated in fiscal 1987. Thereafter tax rates continued to fall to as low as 30.0 per cent in fiscal 1999. Recent corporate tax reductions were enforced to buoy up the stagnant state of the economy, as well as individual income tax cuts. Meanwhile, two tax rates on retention and dividends were integrated in one unified rate in 1990, mainly because it was judged by the Tax Advisory Commission that the lower tax rate on dividends had not been so effective in promoting capital markets. Basic Structures of Integration Since the Shoup Mission regarded the corporate tax as an advance payment on the shareholders' income tax, there has been general agreement that the so-called double taxation of distributed corporate earnings should be eliminated or moderated.
Principles of Corporate Taxation
175
Thus, greater interest has been focused on how to make proper adjustments for the relation between corporate and individual income tax. The problem of integrating the two taxes has been the subject of considerable controversy in the process of structural changes in the corporate tax system in conjunction with the promotion of capital accumulation. In general, there are six methods of integrating the corporate and individual income taxes (see Pechman 1987,181-8): 1. 2. 3. 4.
the dividend-received credit for individuals; the dividend-paid deduction for corporations; the split-rate corporate tax; the imputation method (a method in which all or a portion of dividends are regarded as having already paid tax at source); 5. the dividend exclusion from shareholders' income tax base; 6. full integration, or the partnership method. In 1950, integration started from the adoption of the dividend-received credit in accordance with the Shoup proposals. Under this method as was described above, dividend recipients are allowed to deduct 25 per cent of their dividends as a credit against their individual income tax. Basically, this method of integration has been sustained since then, although numerous changes have been added to the original form with particular emphasis on increasing saving and investment. What is most marked was the move to charge a lower rate of corporate tax on distributed profits than on undistributed profits in 1961 under the split-rate system. The main reason for this alternation was not to alleviate the double taxation of dividends in view of integration, but to improve the working of the securities markets and encourage saving and investment. Therefore, the problem of selecting integration methods was combined with the policy goal of stimulating capital accumulation. Particular attention was paid to the treatment of taxation on dividends. Behind such controversy, we note the important discussion of the Tax Advisory Commission. Since 1953, when the Tax Advisory Commission was provisionally established, major tax reforms of each year have been based mainly on the reports submitted by this commission. The major proposals of the Tax Advisory Commission in light of dividend taxation are as follows. In 1960 the Commission stressed the importance of the self-financing of business firms, in particular the development of the securities market (see Tax Advisory Commission 1960). In this connection, it discussed the adjustment of economic double taxation, i.e. alternative methods of the imputation system, full or partial credit of dividend paid at the company level, reduced tax rates on dividends received by shareholders, etc. Ultimately, however, none of these ideas was formally proposed or enacted. Subsequently, the Commission provisionally proposed the adoption of reduced tax rates on dividends paid in 1961. This was put into effect that same year. After the 1964 report, the Commission began to change its attitude towards the basic nature of the corporate tax, noting some contradictory aspects of both reduced tax rates on dividends paid and tax credits for dividends received
176
Corporate Taxation and Taxes on Capital Table 10.4 The ratio of tax credits (%)
1961 1962-70 1971-2 1973-
Standard ratio: income below ¥1 Om.
Additional ratio: income over ¥10m.
20 15 12.5 10
10 7.5 6.25 5
Note: If taxable income including dividends exceeds YlOm., the additional ratio of tax credit is applicable to such fraction of dividend income equal to taxable income minus ¥10m. The standard ratio of tax credit is applied to the remainder of dividends. Source: As Table 10.3.
(Tax Advisory Commission 1964«,fo). Furthermore, it concluded that tax reductions on dividends for the purpose of promoting the increase of equity capital had not been effective. The imputation system was proposed at this time, but was not put into force. In 1968, in order to simplify the complicated system of corporate tax, the Commission discussed the tentative idea of corporate profit tax (Tax Advisory Commission 1968). On this point, the most controversial issue was whether or not dividends paid should be exempted from the tax base as a business expense. Since 1970, the Tax Advisory Commission concluded that the current corporate tax should be regarded as the separate type of tax system (Tax Advisory Commission 1970, 1971, 1973). In accordance with this view, the tax credit on dividends received is seen as a windfall benefit to mitigate the tax burden of shareholders. Thus, the Commission's position is that the ratio of tax credit must be reduced. In fact, this ratio has steadily been reduced since 1961, including the two separate percentages of standard and additional credits (see Table 10.4). A higher percentage of standard credit is applied to shareholders' income below ¥10 million, while a lower ratio is added to an additional portion exceeding ¥10 million. On the whole, both tax credits continued to decline from 1961. Looking over the development of Japan's corporate tax system during the postwar period, it appears to be in the process of gradually shifting from the integrated scheme proposed by the Shoup Mission towards a modified separate system. The main reason for this shift is that the government has grafted a number of special treatments in favour of dividends to promote capital accumulation into the corporate tax system. As a consequence, the Japanese corporate tax has been furnished with special features in an integration scheme that is in contrast to that of other countries. Systems of corporate taxation can be classified according to the varying tax treatment of dividends and retentions. Two points are important to distinguish one
Principles of Corporate Taxation
177
Table 10.5 Types of company taxation in OECD member countries, 1987 Type of taxation
Country
Separate corporate tax (X) Pure classical system Slight alleviation at company level (XI) Slight alleviation at shareholder level (X2)
Australia Sweden (Japan), US A
Partial integration (Y) At company level (split rate) (Yl) At shareholder level (imputation) (Y2) Credit for company tax actually withheld (Y2(a)) Credit for domestic tax deemed to have been paid (Y2(b)) Full assimilation (Z) At company level (Zl) At shareholder level CZ21
Japan, (W. Germany)
UK
Canada Greece W. Germany
Note: Countries listed in parentheses have subsidiary features of the taxation type. Source: OECD (1987, 86).
system from another: (1) whether the corporate tax is considered to be a separate tax or an integrated tax between corporations and individual shareholders, and, (2) when mitigation of the tax due is needed, whether this should be given at company level for dividends or at shareholder level as credit for the corporate tax paid. Table 10.5 illustrates various ways of avoiding the double taxation of dividends in 1987, using the classification of the OECD. Observing the actual methods of integration used by various countries, the main considerations seem to be centred on three broad types of the corporate tax system: the separate system (X), the split rate system (Yl), and the imputation system (Y2). In contrast, the full assimilation (i.e. full integration) is not regarded as a practical method of integration and is used only in West Germany. As is evident from the past experience of corporate tax reforms, Japan is characterized by a hybrid type of integration system combining X2 and Yl. Under the X2 method the shareholder can receive a credit for part of the corporate tax paid on dividends, while the Yl method admits to tax-distributed profits at a lower rate than retained profits. It is often explained that the double taxation of dividends is partially alleviated at both the shareholder and the company level. In some sense, this hybrid method may be thought of as a partial integration scheme. In 1990, however, this hybrid system was partially changed into the present form, i.e. the Yl method was abandoned and the X2 method alone was retained to lighten the double taxation of dividends.
178
Corporate Taxation and Taxes on Capital Partial Alleviation of Double Taxation
Given such a partial integration of the corporate and individual income taxes, to what extent can the double taxation of dividends be eliminated in Japan? An attempt was made to calculate the extent of avoiding double taxation (see MOF Securities Bureau 1975,1-2). The following are the 1975 tax codes relevant to the computation of corporate taxes: 1. corporate tax rate on dividend—30 per cent; 2. corporate tax rate on undistributed profits—40 per cent; 3. surtax rate on corporations levied at the level of prefectures and municipalities (i.e. corporate inhabitants' tax)—17.3 per cent; 4. tax credit for dividends received to be applied against the individual income tax—10 per cent, but the applicable rate on taxable income above ¥10 million is 5 per cent. If we assume that all corporate profits after tax are distributed to shareholders, dividend (D) is calculated by the following formula, if the corporate profit before tax is equal to 100: D+{0.3XD+0.4X (100-£>)}X (1+0.173) = 100 tax on dividends
tax on retentions
surtax rate
corporate tax amounts total tax amounts Solving the equation for D, we obtain D = 60.13. Since we have assumed that all after-tax profits are distributed, tax (T) is therefore equal to 100 —D, or T=39.87. The amount of the tax can be divided into the corporate tax (Tc) and the inhabitant tax. Since the inhabitants' tax is levied as a surtax on Tc, T C X(1 +0.173) = 39.87,
and
Tc = 33.99
Final results can be summarized as follows: Corporate profit before tax (100%) Dividends, D (60.13%) Corporate tax, Tc (33.99%) Inhabitants' tax (5.88%) Table 10.6 has been calculated based on these figures. Column (2) shows the 1975 statutory rate of the individual income tax by taxable income class. Column (8) indicates the degree to which double taxation of dividends can be avoided, given the
Table 10.6 Adjustment of corporate and individual income taxes, 1974 Taxable income class (¥m.)
Marginal rate of personal income
Income tax burden of shareholder
Amount of double taxation
Required tax credit for dividends
T-100f=jR (5)
RlD=r^ (6)
r-i
(3)
Total burdens of personal income and corporate taxes Tc+Ti=T (4)
t
DX t= Tt
(1)
(2)
(7)
(8)
0-0.6 0.6-1.2 1.2-1.8 1.8-2.4 2.4-3.0 3.0-4.0 4.0-5.0 5.0-6.0 6.0-7.0 7.0-8.0 8.0-10.0 10.0-12.0 12.0-15.0 15.0-20.0 20.0-30.0 30.0-40.0 40.0-60.0 60.0-80.0 80.0-over
10 12 14 16 18 21 24 27 30 34 38 42 46 50 55 60 65 70 75
6.01 7.22 8.42 9.62 10.82 12.63 14.43 16.24 18.04 20.44 22.85 25.25 27.66 30.07 33.07 36.08 39.08 42.09 45.10
40.00 41.21 42.41 43.61 44.81 46.62 48.42 50.23 52.03 54.43 56.84 59.24 61.65 64.06 67.06 70.07 73.07 76.08 79.09
30.00 29.21 28.41 27.61 26.81 25.62 24.42 23.23 22.03 20.43 18.84 17.24 15.65 14.06 12.06 10.07 8.07 6.08 4.09
49.89 48.58 47.25 45.92 44.59 42.61 40.61 38.63 36.64 33.98 31.33 28.67 26.03 23.38 20.06 16.75 13.42 10.11 6.80
10 10 10 10 10 10 10 10 10 10 10 5 5 5 5 5 5 5 5
20.0 20.6 21.2 21.8 22.4 23.5 24.6 25.9 27.3 29.4 31.9 17.4 19.2 21.4 24.9 29.9 37.3 49.5 73.5
Source: MOF, Securities Bureau (1975,2).
(%)
Actual tax credit for dividends (%)
-^X100%
180
Corporate Taxation and Taxes on Capital
1975 corporate tax regulations. It indicates that the higher the income class, the greater the success in escaping double taxation. This evidence leads us to conclude that: 1. the total tax burden of corporate and individual income taxes is distributed unevenly in favour of higher income classes. This result still stimulates small firms to incorporate; and 2. the substantial degree of double taxation suggests that the 1975 corporate tax should not be regarded as a fully integrated tax system.
OTHER ASPECTS OF THE CORPORATE TAX
The Difference between Tax Burdens of the Corporate and Non-corporate Sectors Although the Shoup Mission's corporate tax structure included an adjustment mechanism to avoid double taxation, the method was only an approximate one. In actual practice, the problem of double taxation did arise. When all earnings are distributed, given the individual and corporate tax rates and the tax credit for dividends at the shareholder level proposed by the Mission, individual shareholders subject to marginal rates of income tax above 53 per cent will find that they are slightly better off than if they had been owners of an unincorporated business with a profit equal to their share of the corporate profit before tax. Conversely, shareholders subject to marginal rates of tax of 53 per cent or less will find themselves slightly worse off than they would be as individual proprietors (Shoup Mission 1949,107). Although this phenomenon of double taxation seems a minor problem, there has been a marked tendency for small firms to incorporate, and this has been attributed largely to the existence of just such an integration of the corporate tax. One manifestation of this trend was the absolute decline in the amount collected for selfassessed income taxes (mainly, income tax levied on unincorporated business) in the 1950s. The proportion of self-assessed income tax in national tax revenues was 20.2 per cent in 1950, but it fell steadily to 5.4 per cent in 1959, while the relative share of corporate tax increased from 18.4 per cent in 1950 to 31.4 per cent in 1959. It is believed that many unincorporated firms moved towards a corporate form of business organization because corporations benefited more than the unincorporated businesses under the corporate tax. This phenomenon has been described as the emergence of'quasi-corporations'. The government has been trying to cope with the problem of 'quasi-corporations' by introducing a special tax treatment for proprietor and family employees under the individual income tax. As noted before, in 1952 taxpayers filing blue returns were permitted to deduct salaries paid to family employees as business expenses up to a certain limit. The limit was increased several times and was ultimately removed in 1968.
Principles of Corporate Taxation
181
In 1961, the Tax Advisory Commission investigated the relative tax burdens of corporate firms and unincorporated proprietors (Tax Advisory Commission 1961, 67-9). Based on this investigation, the Commission recommended that the allowable limit for deducting salaries to family employees should be raised from ¥80 000 to ¥120000 for blue-return taxpayers. The main purpose of this was to reduce the tax burden of unincorporated businesses in order to eliminate the unnatural phenomenon of burgeoning 'quasi-corporations'. Table 10.7 is a comparison of the total tax burdens of different forms of business organization in 1961. The figures are computed given five family members with a specific age and salary distribution. It is clear that the income class of above ¥1.5 million has a higher tax burden if it is organized as an unincorporated business rather than a corporation. This bias has continued down to the present time. That advantages have occurred to 'quasi-corporations' implies a failure of the system to
Table 10.7 Comparison of the total tax burden of corporation and non-corporation in 1961 (before tax revision) Net income (¥000)
500 800 1000 1500 2000 3000 5000
Tax burden (¥000) Corporate firm
Unincorporated business
32 113 158 313 489 907 1 886
18 86 144 317 527 1005 2089
Notes: Family size and salaries are assumed as follows: (1) Tax burden includes the following items: corporate firms—corporate tax (national), corporate inhabitants' tax (local), and corporate enterprise tax (local); unincorporated business—individual income tax (national), individual inhabitants' tax (local), and individual business tax (local). (2) Family size—five members: Husband (entrepreneur) Wife ( > 25 years old) First son ( > 25 years old) Second son ( > 15 years old) Third son ( < 15 years old) This family composition is assumed to be the same in all income classes in both cases. (3) Salaries—entrepreneur, wife, and first son are assumed to earn the following salaries from the firm (in thousands of yen): Net income 500 800 1000 1500 2000 3000 5000 Entrepreneur 240 300 420 540 600 720 840 Wife and first son (each) 96 120 120 150 180 210 240 (4) Unincorporated businesses are permitted to deduct the salaries of the wife and the first son as a special tax benefit for proprietors and their family employees. Source: Tax Advisory Commission (1961).
182
Corporate Taxation and Taxes on Capital
eliminate double taxation. In practice, the problem of double taxation continues under the current corporate tax. The emergence of'quasi-corporations', however, has declined since the 1970s, and their importance as a group of taxpayers has been greatly reduced. One reason is that special tax treatment of 'deemed corporate income' has been established in favour of unincorporated, self-employed taxpayers since 1974. Taxpayers filing a blue return can elect to be treated as 'corporations' for tax purposes. Under this 'deemed' method, business proprietors can deduct their own salaries, as well as those of family employees, as business expenses. Furthermore, the deduction for employment income is also applicable to their own salaries. The net income computed for business proprietors is considered 'deemed corporate income' and is taxed at the lower rate of dividend income.3 Thus, business proprietors no longer needed to become 'quasi-corporations' to mitigate their tax burdens. As noted before, this 'deemed' system was finally abolished in 1992 to make the income tax burden among taxpayers more equitable. Debt and Equity Financing Taxation on dividends has been discussed in relation to the choice between debt and equity in business financing. During the postwar period, the Japanese government has continually stressed the improvement of self-financing through the stimulation of capital markets. For this purpose, the government has given special treatment to taxes on dividends. However, the extent of self-financing in business enterprises has gradually declined until 1980 (see Figure 10.1). This was generally thought to be an undesirable feature of the Japanese economy. Many have argued that the large amount of debt financing by corporations should be discouraged. Debt certainly makes good business sense if firms have a sufficient margin for paying fixed interest charges. However, firms may be squeezed financially when the economy enters into depression, and at such times their margin would disappear swiftly. Bankruptcies may occur because of defaults on interest and principal payments. It has been maintained that corporations should try to finance a substantial share of their capital requirements through equity capital (mainly retained earnings) to avoid these risks. Obviously, of relevance is the fact that equity financing can be unattractive from a tax point of view compared with borrowing. Corporations are allowed to deduct interest payments on borrowed capital from taxable income, but there is no corresponding deduction for dividends paid out to shareholders. At the 37.5 per cent tax rate, a corporation must earn ¥16000 before tax to pay ¥10000 dividends, but it need earn only ¥10000 to pay ¥10000 interest. Needless to say, this asymmetry 3 Business proprietors electing to be 'deemed corporations' for tax purposes amounted to 135597 in 1992. They represent 3.1 per cent of total taxpayers filing blue returns, which has been reduced from 7.3 per cent in 1986.
Principles of Corporate Taxation
Fig. 10.1
183
Trends in self-financing by corporate firms, 1968-1998
Source: Data presented to the Tax Advisory Commission.
makes the cost of equity finance more expensive for corporations than an equal amount of borrowed capital. Were there no tax considerations, there would be little to choose between alternative methods of business financing. However, when the tax system favours one method rather than another, a corporation, being very sensitive to tax considerations, will look for the cheapest source of finance. This is true in Japan, as it is in many other countries. Throughout the postwar period, the banking community has exerted strong pressure for the reduction or elimination of the tax on interest income. As a consequence of this pressure, complete or partial tax exemption for interest income has been put into effect, and this, in turn, has resulted in excessive reliance upon financing through borrowing in the corporate sector. Once the interest provisions were adopted, securities dealers and representatives of investment companies demanded favourable treatment for dividend incomes in order to maintain a competitive position in relation to the banking industry. Attempts were made to distribute tax benefits more or less equally among the banking and securities industries on the plea of encouraging capital accumulation, in particular, or increasing the ratio of equity capital. For example, a reduced tax rate on dividends and the separate taxation of dividend income as well as interest income at a specific rate were used to mitigate the tax on distributed profits. In addition, the Securities Transaction Council frequently proposed that dividends paid at the company level be made deductible as business costs (Securities Transaction Council 1960). In 1961, when the split rate system was first introduced, the Tax Advisory Commission distributed a questionnaire to investigate the mitigating effects of
184
Corporate Taxation and Taxes on Capital
dividend taxation on the equity financing of businesses (Tax Advisory Commission 1962, 38f). The results indicated that this tax device did not stimulate an increase in the relative share of equity in the supply of corporate funds to any substantial degree. Many Japanese tax experts outside of the government, as well as some members of the Tax Advisory Commission, were sceptical of the effectiveness of special tax measures in correcting the imbalance between debt and equity in corporate finance (see Tax Advisory Commission 1964a, 12). Figure 10.1 depicts the movement of the self-financing ratio and corporate tax rates since 1968. The self-financing ratio shows a long-run decline in the 1960s and the 1970s but turns upward slightly in the 1980s and the 1990s. Unfortunately, it is impossible to observe any significant relationships among the variations of the selffinancing ratio, corporate tax rates, and tax credit ratio. Apart from the issue of business financing, we should pay brief attention to the distribution of corporate dividends. Recently there has been a marked reduction in the percentage of corporate dividends received by individual shareholders. This is in sharp contrast with the enormous increase in the number of individual shareholders during the 1950s. The share of total dividends paid out to individual shareholders was 61.3 per cent in 1950, 33.5 per cent in 1975, 23.2 per cent in 1985 and, 22.5 per cent in 1995, although it rose slightly to 23.6 per cent in 1997. It is widely believed, however, that this phenomenon is not related to the current working of the corporate tax system. The Incidence of the Tax The question of 'who actually pays the corporate tax' is the most controversial issue in tax theory and practice. Although the corporate tax is normally levied on corporate profits, no one can identify who bears the ultimate burden. There are many divergent views among economists and businessmen which hypothesize the shortand long-run incidence of corporate taxation, but no scientific basis has yet been provided by economists to accept or reject the various arguments. Tax specialists have indicated two main groups which probably bear part of the corporate tax burden (see Goode 1951; Harberger 1962; Krzyzaniak and Musgrave 1963; Pechman 1986; etc.). One is the group that supplies capital and/or entrepreneurial skills to corporations. If the corporate tax is not shifted in the short run, the after-tax rates of return on invested capital decrease, and investment, in turn, may be discouraged. Given the fixed supply of capital, the burden of the corporate tax would then fall on the owners of capital. Another group of people who may share the burden of the corporate tax are the consumers who purchase the goods and services produced by corporations. In imperfect markets, business firms set their prices using the full-cost pricing method (i.e. the price is determined at the level that covers the full cost plus a profit margin).
Principles of Corporate Taxation
185
In this case, firms regard tax as a factor of cost and lift their prices to recover the cost. Alternatively, firms might aim at a certain after-tax rate of return on their corporate equity, and so the imposition of a corporate tax would influence the determination of output and prices. To secure the target level of return, the tax must be passed on to consumers. If this forward-shifting takes place, the corporate tax acts as a general sales tax on goods and services that firms produce. Also, part of the burden of the corporate tax may also fall on the workers, but such backward-shifting is not likely to occur so smoothly. We should note that the incidence of the corporate tax is a complicated issue and depends on the answers to empirical questions. However, empirical analyses do not provide a clear determination of the factors affecting corporate profits, prices, and wages. The evidence is generally inconclusive. Empirical analyses of corporate tax-shifting have been carried out less frequently in Japan than in such countries as the USA. Despite the limited research in this area by Japanese economists, there have been some attempts to study quantitative aspects of corporate tax-shifting in Japan. In particular, two articles by Furuta (1965, 1970) are worth noting, and the following is a brief survey using his estimates. In the first place, let us focus on two primitive approaches that have been used in the past: (a) the profit share approach, and (b) the corporate rate of return on capital approach. In the USA, the long-term change in corporate profit share before tax has remained fairly constant, in spite of a rise in the rate of corporation income tax. However, the corporate profit share after tax seems to have decreased with the rise in the tax rate. This fact has been cited as evidence that further tax-shifting has not occurred. On the other hand, looking over the long-term movement in the rate of return on capital in the USA, one notices that the rate of return on capital before tax rose along with the increase of the tax burden, while the rate of return after tax has tended to remain constant. This fact is cited as evidence that the long-term shifting of the corporate tax is nearly complete. After empirically investigating these figures pertinent to pre- and post-war Japan, Furuta concluded that three aspects of the corporate tax—the effective tax rate, the rate of return on capital, and the corporate profit share—show a surprising similarity between Japan and the USA with respect to the shifting of the corporate tax. Next, Furuta tried to apply three models developed for the American situation by (1) Krzyzaniak and Musgrave (1963) (K-M), (2) Kilpatrick (1965) (K), and (3) Gordon (1964) (G), to some industries in Japan. The well-known conclusion of the K-M model is that more than the full amount of the corporate tax burden is shifted forward to the consumer in the short run by US manufacturing firms. In the Japanese manufacturing sector, Furuta also demonstrated an 'over-shifting of the corporate tax'. The degree of shifting estimated by the K model for US manufacturing ranges from 62 to 94 per cent, but this conclusion differs from the K-M and G models. According to Furuta, this result is not appropriate in the case of Japan. In contrast to the 'over-shifting' of the K-M model,
186
Corporate Taxation and Taxes on Capital
Gordon asserts that tax-shifting by US manufacturing is not significantly greater than zero. Furuta finds that the Japanese results obtained by using the G model give a description of tax-shifting in Japan contrary to that of the USA. In sum, the empirical evidence by Furuta suggests that the corporate tax is shifted forward into prices to a considerable extent, in accordance with the K-M model. Furuta's conclusions seem to be accepted by many, although not all, scholars in Japan.
11
Corporate Tax Levels and Tax Incentives
In addition to raising revenues from corporate firms, the Japanese government used the tax system to promote specific policy objectives, such as capital accumulation, export activity, pollution control, and so on. This function persisted at least until the oil shocks of the 1970s. Such tax incentives took the form of numerous special tax measures. Since the late 1970s, however, the basic thrust of tax policy has shifted towards placing greater importance on tax equity and neutrality than on tax incentives. As a consequence, during the past three decades the number of special tax incentives has been curtailed substantially, a move especially affecting the corporate sector. This has tended to increase the tax burden on business firms, which often stimulates debates as to whether or not the corporate tax burden in Japan is too heavy by international standards. The aims of this chapter are twofold. One goal is to give an overview of tax incentive policies, paying much attention to the role played by the special tax measures for corporations. The other is to make clear the tax burden of the corporate sector, with special reference to tax incentive policies, referring to the past debates on the corporate tax burden and its policy implications. GENERAL
FEATURES OF TAX I N C E N T I V E POLICIES
Tax Incentives as an Industrial Policy Japan's industrial policy has attracted world-wide attention, and as a result has become an important target of study for scholars. It is frequently argued that during the postwar period the policies of the Japanese government to stimulate private saving and investment, export, and innovation were effected mainly through tax incentives (see Hollerman 1984). Although it is almost impossible to assess how effective these policies were as an industrial policy in quantitative terms, many people admit the significant role of tax incentives in the policy-making process. There were several steps in the development of tax incentives for the promotion of specific policy goals. First, earlier in the postwar period, both the promotion of exports and the encouragement of certain key industries (i.e. petrochemicals) This chapter draws heavily upon Ishi (1988c), and the empirical estimates of E. Tajika and Y. Yui.
188
Corporate Taxation and Taxes on Capital
formed a major consideration in tax policy. In the main, tax exemptions were used to achieve these objectives. Today, however, these provisions are of very minor significance. Second, a more important goal was to stimulate business saving and investment among targeted industries. Specific targeted industries were authorized to use various types of tax devices, such as special depreciations, tax-free reserves, and tax credits. In particular, we should note the interaction between investment and exports. Tax incentives encouraged the formation of capital in export-related industries as well as in those with export potential. It is argued that this interaction contributed to Japan's rapid economic growth. Third, tax incentives were directed towards developing technological innovation. Tax devices were used to stimulate such innovation indirectly by generating the growth of capital formation in important, selected industries. Major beneficiaries of such special measures were the steel and machinery industries, which may have thereby developed international competitiveness to a great degree. Lastly, as time went by, the benefits of special tax measures were extended from key industries to cover a wide variety of policy objectives such as pollution control and the enhancement of social welfare. The use of the tax incentives was diversified to advance various economic and social policy goals. However, since the late 1970s tax benefits for business investment, corporate savings, and industrial production have begun to be phased out because of the large fiscal deficit. The scope of tax incentive policies has thus become narrower.
Development of Tax Incentives As described in Chapter 3, changes in revenue losses resulting from the special tax measures are a good indicator of the past development of tax incentives. The data relevant to the corporate tax are available from 1965 onwards. Figure 11.1 illustrates the trend of revenue losses from special tax measures as a percentage of corporate tax revenues. In 1966 and 1972, revenue losses reach the top level of 9.0 per cent, and during the intervening years they maintain a relatively high level. Since 1972, however, they declined steadily to as low as 2.2 per cent in 1982, thereafter turning upwards to 3.8 per cent in 1987, and again moving around 3 per cent until 1991. Thereafter, they continued to fall to as low as 2.7 per cent in 1997. The low level of the ratio since the late 1970s mainly reflects the fact that a number of special tax measures have been abolished and have diminished in size. In general, special measures included in the corporate tax system are classified into two types: 1. tax exemptions and credits 2. tax deferrals (a) accelerated depreciation (b) tax-free reserves
Corporate Tax Levels and Tax Incentives
189
Fig. 11.1 Corporate tax revenue losses from special tax measures as a percentage of corporate tax revenue, 1965-1997 Source: Data presented to the Tax Advisory Commission.
Broadly speaking, greater importance was placed on type 1 measures in the 1950s, while type 2 measures became more significant in the 1960s and 1970s (see Ikemoto etal. 1984). The origin of the special tax measures can be traced to the 1938 enactment of the Temporal Tax Measures Law which provided tax incentives for the war effort. Such measures proliferated during the war and immediate postwar period, but were drastically reduced when the Shoup Mission recommended tax reform in 1949. Since then, the list of special measures was considerably expanded again until the late 1970s. During the earlier period (i.e. in the 1950s), the exemption type of tax incentive measures was most prevalent. To give two examples, income from the production of certain products (mostly petrochemical products) was exempted from taxation between 1953 and 1966; and to promote export-oriented industries, export income was exempted from 1953 to 1963. Furthermore, to stimulate corporate savings, income spent on dividends for increased shares went untaxed from 1954 to 1957. Most of the exemption type of measures, however, were phased out in the early 1960s. It is important to notice that these exemptions were not applied to all industries across the board, but only to specific ones, such as infant or exportoriented industries. Instead of tax exemptions, the tax credit device has been used more often in recent years. In particular, the tax credit for experimental and research expenses, which started in 1967, is of special importance. Since 1981, an investment tax credit for energy-saving equipment has been allowed.
190
Corporate Taxation and Taxes on Capital
Tax deferral measures include accelerated depreciation and increased initial depreciation. Both promote business investment. These devices were used often in the 1950s, and some of them have remained in effect during much of the postwar period. However, many accelerated depreciation and increased initial depreciation plans have been replaced. For instance, accelerated depreciation for important industrial equipment was made an option in the form of a three-year, 50 per cent more than normal, rate of depreciation from 1951 to 1961. Also, additional initial depreciation for important industries (i.e. 50 per cent write-off during the first year) was introduced from 1952 to 1961. In the 1960s, depreciation plans expanded to new areas. For instance, accelerated depreciation was granted to small and medium-size enterprises in 1963. The initial depreciation on machinery and equipment used for the prevention of environmental pollution was increased in 1967. As time progressed, the trend towards expanding accelerated depreciation provisions began to slow. In 1972 accelerated depreciation for export industries was abolished, and in 1973 increased initial depreciation for important industries was reduced. Overall, the recent trend has been to reduce the range of tax devices, especially the accelerated depreciation provisions. Another type of tax deferral measure is that of tax-free reserves. Although the term 'tax-free reserves' is used in English, in Japanese two types of tax-free reserves are usually distinguished. The first represents reserves that are set aside for purposes of business accounting (hikiatekin), while the second type is to stimulate involvement in risky activities (junbi-kin). Because thejunbi-kin are those reserves that may not be duly justified by generally accepted accounting principles, these are officially listed among the special tax measures. Both types of tax-free reserves have been allowed since the early 1950s. Those set aside for tax-free accounting permitted reserves for bad debts and special repairs in 1950, reserves for retirement allowance in 1952, and reserves for bonus payments and losses on returned goods in 1965. Special inducements included reserves for price fluctuations in 1952 and reserves for drought in 1952. Numerous other reserves were added in the 1960s and 1970s. Of course, the size of the tax-free reserves and the amount allowed annually are subject to specific limits, but in all cases, the limits are generous in relation to the potential risks. Thus, the tax-free reserves are criticized on the grounds that they conceal profits retained from the amount subject to income taxation. More detailed information on recent special tax measures for corporations in comparison with those in 1977, 1987, and 1998 is given in Table 11.1. Total amount of revenue losses fell from ¥457 bn. in 1987 to ¥356 bn. in 1998 although they doubled from 1977 to 1987. We see three broad types of tax incentives: (1) special depreciation, (2) reserves, and (3) tax credits and allowances. In 1998, the third type occupied the largest share (47.8 per cent), having greatly expanded from 16.2 per cent in 1977. Besides a big increase in credit for research and experimental expenditures, three new items (3(b), 3(c), and 3(e)) to promote specific investments have been added.
Corporate Tax Levels and Tax Incentives Table 11.1 1998
191
Revenue losses of special tax measures of corporations, by type, 1977, 1987, and 1977
¥bn.
1. Special measures for depreciation (a) Special depreciation on qualified equipment (e.g. machinery and equipment for reducing pollution, saving energy, recycling, etc.) (b) Special depreciation on manufacturing machinery used in underdeveloped regions (c) Special depreciation on machinery and equipment acquired by small- and medium-sized firms (d) Others
1998
1987
¥bn.
%
%.
¥bn.
%
128 34
56.1 14.9
129 33
28.2 7.2
99 23
27.8 6.5
14
6.1
23
5.0
11
3.1
43
18.9
43
9.4
24
6.7
37
16.2
30
6.6
41
11.2
2. Reserves (a) Reserve for overseas investment loss (b) Reserve for losses caused by repurchase of computer (c) Others
63 12
27.6 5.3
76 2
16.5 0.4
87
24.4
1
0.4
24
5.2
50
21.9
50
10.9
87
24.4
3. Tax credits and allowances (a) Credit for research and experimental expenditures (b) Tax measures to promote investments for efficient use of energy (c) Tax measures to promote investments by small- and medium-sized firms on equipment utilizing electronic devices (d) Special deduction for income derived from overseas technical service transactions (e) Tax measures to improve managerial fundamentals of small- and medium-sized enterprises (/) Others
37 17
16.2 7.5
252 90
55.1 19.7
170 44
47.8 12.4
0
0.0
55
12.0
0
0.0
52
11.4
41
11.5
12
5.2
27
5.9
5
1.4
0
0.0
18
3.9
8
2.2
8
3.5
10
2.2
72
20.2
228
100.0
457
100.0
356
100.0
Total
Source: iMOF, Tax Bureau (1977, 1987) and NTA 1998.
192
Corporate Taxation and Taxes on Capital
On the other hand, the relative share of the first and second types in 1987 is reduced considerably. Particular attention should be paid to the sharp decline of the tax-free reserves system. Reserves for price fluctuation, for instance, were abolished in 1986, and this helped to recover revenue losses. Use of Tax Incentives What size corporation actually benefits most from the special tax measures? In order to estimate the effect of special tax measures on corporate tax burdens, we need to explore what is meant by the effective rate of tax. Obviously, statutory or nominal tax rates are not representative of relative tax burdens on corporate incomes because the government has adopted numerous devices, such as tax-free reserves, accelerated depreciation allowances, investment credits, and others, which reduce that burden. Tax rates on a nominal basis only imply a weighted average of the multiple-rate structure with split and reduced tax rates, before the adjustment of special tax measures. By contrast, the effective rate of tax paid by corporations on gross profits is estimated after adjusting for special tax measures. Table 11.2 summarizes the two types of tax rates, the nominal and the actually paid, at three different levels of paid-in capital. Of major importance is the fact that we observe a discrepancy in the corporate tax burden by size of corporation. On an actually-paid basis, corporate incomes are broadened to include the major items eroded by the special tax measures (i.e. tax-free reserves, accelerated depreciation, and special deductions), and the effective tax rates are calculated in column (8). These are to be compared with nominal tax rates in column (3). Figure 11.2 depicts the movement of the two tax rates listed in Table 11.2 in a more clearly visible fashion. There are a couple of points worth noting. To begin with, nominal tax rates consistently showed an upward tendency for corporations of all sizes in the first 12 years until 1985. Afterwards, they began to fall up to 1996. This suggests that the corporate tax burden was raised by the government on a statutory basis, but it began to decline substantially after 1986, reflecting the more recent movement towards lowering corporate tax rates (see Table 10.3). Particular attention ought to be paid to the sharp declining trend in the 1990s. In addition, medium-size corporations with ¥0.1-10 billion of paid-in capital incur a heavier tax burden than the other two types of corporations. Small corporations benefit from reduced rates of tax on a larger portion of their corporate income, while large corporations in general could pay out more dividends, taxed at a lower level. By contrast, medium-size corporations seem to have relatively fewer opportunities to make use of such preferential tax rates. To turn our attention to another indicator, the movement of effective tax rates on an actually-paid basis has shown ups and downs over the long run, reflecting the availability of special tax concessions during each period. It is often pointed out that large corporations are given more opportunities to make use of tax-free reserves
Table 11.2 Fiscal year
Effective corporate tax rates before and after adjusting for special tax measures, by size of corporation, 1973-1996 Corporate income (¥bn.)
Corporate tax revenue (¥bn.)
(2)/(D
(1)
(2)
(3)
Taxfree reserve (¥bn.)
Accelerated depreciation (¥bn.)
(4)
(5)
Special deduction for overseas technical service' (¥bn.) (6)
Tax credit for experimental and research expenses (¥bn.) (7)
(8)
1.1 0.5 1.1 0.8 1.7 1.9 0.8 1.3 0.7 1.9 3.3 1.1 5.1 0.6 0.5 0.7 0.5 0.2 1.8 0.4 0.2 0.7 0.0 0.1
1.3 1.8 1.6 1.6 1.4 11.0 23.1 18.1 13.5 6.2 5.9 8.0 15.9
32.4 34.1 35.7 35.0 35.7 35.6 35.6 35.9 37.9 38.2 39.3 38.8 39.8 40.0 39.1 38.5 38.6 37.1 35.0 34.9 35.4 35.5 35.2 35.4
(2)-(7) (l) + (4) + (5) + (6)
Case 1: less than ¥0.1 billion of paid-in capital
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
5 994.2 7331.1 5 626.5 5 700.0 5 652.6
6 395.2 8 168.7 8 960.2 9098.2 9 026.2 9 009.5 9 584.3 10436.7 10412.6 12 624.4 14490.0 17831.2 19 176.5 19 347.4 17208.0 14359.8 13071.1 12253.2 14050.5
2036.9 2578.7 1 992.2 2021.9 2 007.0 2 274.8 2951.1 3 254.4 3 406.4 3 458.8 3 472.0 3 768.4 4 176.9 4 146.2 4959.1 5637.5 6 957.6 7 188.4 6 868.2 6 064.8 5 057.4 4603.1 4 303.2 4970.5
34.0 35.2 35.4 35.5 35.5 35.6 36.2 36.3 37.4 38.2 38.5 39.3 40.0 39.8 39.3 38.9 39.0 37.5 35.5 35.2 35.2 35.2 35.1 35.4
162.6 58.3 -71.1 64.6 -44.5 -7.6 81.6 5.2 -154.3 4.4 -204.8 59.9 -24.9 -80.5 42.4 -10.1 -3.8 -14.0 16.6 3.0 -34.5 -30.7 8.7 11.0
116.2 177.9 14.5 3.0 15.6 -26.1 -11.7 42.9 8.0
2.0 3.7 35.9 37.0 -5.2 — 11.8 76.6 128.2 112.5 172.8 68.1 -62.0 -75.8 -47.0 -57.5
15.2
16.7 27.0 22.0 33.2 27.1 28.6 4.4 6.4 9.1 7 3
Table 11.2 Fiscal year
Contd. Corporate income (¥bn.)
Corporate tax revenue (¥bn.)
(2)/(l)
(1)
(2)
(3)
Case 2: ¥0.1-10 billion of paid-in capital 1973 4072.1 1.445.9 1974 4483.7 1 675.2 1975 3 055.2 1 182.8 1976 3 766.8 1 461.4 1977 4480.2 1 742.9 1978 5 092.8 1 984.5 1979 6 183.2 2416.1 1980 7 280.7 2 844.8 1981 7330.6 2 935.6 1982 7413.5 3 048.2 1983 7583.6 3 115.0 1984 8 250.5 3 440.5 1985 8 390.0 3 562.8 1986 8 309.5 3519.8 1987 9 191.7 3 854.2 1988 LO 726.4 4 424.6 1989 11 770.0 4 865.4 1990 11393.6 4527.1 1991 11 728.8 4 409.0 1992 10436.0 3912.9 1993 9 070.7 3401.5 1994 8 439.5 3 164.8
35.5 37.4 38.7 38.8 38.9 39.0 39.1 39.1 40.0 41.1 41.1 41.7 42.5 42.4 41.9 41.2 41.3 39.7 37.6 37.5 37.5 37.5
Taxfree reserve (¥bn.)
Accelerated depreciation (¥bn.)
(4)
(5)
133.4 117.3 -55.8 80.5 50.7 -46.7 -11.1 39.9 -76.3 -108.3 -107.3 -26.3 -29.5 -29.5 -23.7 -1.2 27.6 -4.5 29.7 11.2 10.5 4.3
60.3 91.9 -0.3 -26.2 27.1 -42.9 -40.0 -41.
-35.3 -32.4 -23.5 -30.1 -9.8 -33.6 -16.2 13.8 -7.7 -3.7 -5.4 2.5 -17.1 -1.4
Special deduction for overseas technical service' (¥bn.) (6)
3.5 7.6
6.0 9.5 8.8 10.9 13.0 17.8 16.8 14.6 17.7 17.3 13.3 8.1 6.0 6.5 7.7 8.3 3.8 4.9 8.3 4.3
Tax credit for experimental and research expenses (¥bn.)
(2)-(7) (l) + (4) + (5) + (6)
(7)
(8)
12.0 19.0 9.0 11.8 10.1 11.6 15.1 18.1 21.4 24.8 24.0 21.9 24.6 23.3 19.3 19.9 20.2 21.3 22.5 18.5 11.2 9.4
33.6 35.2 39.1 37.8 37.9 39.3 39.1 38.7 40.3 41.5 41.4 41.6 42.3 42.4 41.9 41.0 41.1 39.5 37.3 37.3 37.4 37.4
1995 1996
8 629.9 9010.5
3 236.9 3378.9
Case 3: ¥10 billion and over paid-in capital 1973 1 078.4 3 074.8 1974 1 151.3 3 108.9 1975 2 650.2 1 014.1 1976 3 434.5 1317.5 1977 4296.1 1 647.1 1978 1 683.7 4359.7 1979 1 958.1 5 080.9 1980 2359.0 6089.1 1981 2 740.4 6 934.9 1982 7395.2 3 007.0 1983 7901.8 3 207.3 1984 3 882.7 9412.0 1985 11459.2 4 829.6 1986 4 866.0 11 572.7 1987 5400.1 12929.9 1988 6409.8 15646.1 1989 17811.1 7 300.0 1990 19811.1 7819.1 1991 6 926.4 18445.3 1992 15274.8 5 727.9 1993 4 866.4 12977.2 1994 4079.1 10877.6 1995 4031.4 10750.4 1996 5 806.0 15482.8
37.5 37.5
18.2 -21.4
6.3 -13.6
4.0 2.9
9.4 14.0
37.3 37.5
35.1 114 38.3 38.4 38.3 38.6 38.5 38.7 39.5 40.7 40.6 41.3 42.1 42.0 41.8 41.0 41.0 39.5 37.6 35.7 37.5 37.5 37.5 37.5
165.7 151.9 — 6.7 117.3 47.9 -61.1 10.7 32.4 124.2 -32.8 -80.9 61.1 -2.8 16.1 -14.8 88.7 244.3 153.7 268.2 -24.1 111.2 189.6 7.7 105.1
106.0 103.5 -22.2 -41.4 39.6 -59.8 -60.7 -80.6 6.5 -59.0 -72.0 -69.4 -26.8 -34.5 -7.7 33.1 15.0 -3.3 52.3 4.2 -19.8 -18.5 91.7 3.5
19.0 24.6 12.9 19.9 31.1 25.7 25.8 30.6 30.2 38.5 35.8 37.7 48.8 37.2 33.1 39.5 47.7 53.4 37.5 38.0 29.6 34.1 29.9 13.1
14.9 20.4 4.7 10.5 16.6 20.6 30.8 38.3 54.9 63.6 53.2 57.6 89.0 68.0 49.3 74.8 111.6 133.8 162.2 92.1 28.9 23.1 24.4 63.4
31.6 33.4 38.3 37.0 36.9 39.0 38.3 38.2 37.8 40.1 40.5 40.5 41.3 41.4 41.3 40.1 39.7 38.4 36.0 36.9 36.9 36.6 36.8 36.8
Source: Unpublished data presented to the Diet by the Tax Bureau, MOF.
Fig. 11.2 Effective corporate tax rates before and after adjustment for special tax measures, by size of corporation, 1973-1996: (a) nominal basis; (b) actually paid basis Source: Table 11.2.
and accelerated depreciation than smaller corporations. For instance, tax-free reserves such as retirement allowances or bonus payments are used mostly by large corporations. As a result, the regular trends in the tax rates of the three types of corporations on a statutory basis are thrown into disorder when we turn to the actually-paid tax rates. Note should be taken of the lowest tax rate paid by the largest corporation in the early 1970s.
Corporate Tax Levels and Tax Incentives
197
ARE CORPORATIONS OVERTAXED?
The Corporate Tax Burden The level of corporate tax burden is important for a number of reasons.1 It may affect the savings rate and the cost of capital, and in turn the economic growth rate. Thus, different corporate tax burdens may lead to differential growth rates by different countries. In addition, in a world economy where international capital can move relatively freely, corporate tax burdens may contribute to determining the inflow and outflow of capital (see Gravelle 1983). Despite the immense importance of the subject, no attempts have been made to clarify the relationship between the corporate tax burden and other economic phenomena. I am not so ambitious as to propose any conclusive result, but simply approach the subject by estimating the average rate of corporate tax burdens.2 It is possible to gain a simple estimate of the corporate tax burden by using the statutory structure of the corporate income tax. The MOP has traditionally used this formula to clarify the tax burden in the corporate sector. The corporate tax had a somewhat complicated structure until 1989, consisting of multiple rates as shown in Chapter 10. For the corporate tax burden, the MOF usually computes what we shall call a nominal rate of tax burden (II) on a statutory basis. If we take the split tax system before 1989, the formula of the corporate tax burden is as follows:
where f r = tax rate on retained profits, td=tax rate on distributed profits (both of these are national taxes), t/ = tax rate of local corporate income tax, and f;, = tax rate of local enterprise tax. If we use the statutory tax rate data, II is 52.92 per cent in 1984-6 (i.e. tr= 0.433, fj = 0.333, fe = 0.173, ffc = 0.12). In this calculation, it is assumed that corporations pay out dividends of 30 per cent of their before-tax earnings. Simply calculated on the statutory rate basis, this is the nominal figure of corporate tax burden presented by the MOF. Thus, nominal tax rates are readily obtainable by using the MOF formula, but these are not representative of the actual tax burdens paid by corporations. The MOF estimate merely covers the sum of national and local taxes on corporate income on a statutory basis. Put another way, the effects of tax incentives offered
1 In discussing corporate tax burdens, I imply that corporations bear the ultimate burden of the corporate tax without any forward or backward shifting. 2 Most popular discussion has so far focused on the effect of income taxation on the cost of capital. There are several papers worth noting pertinent to the Japanese experience: see e.g. Shoven and Tachibanaki (1988), Tajika andYui (1987), and King (1986).
198
Corporate Taxation and Taxes on Capital
through the special tax measures are not taken into consideration. If such incentive effects are included in corporate tax burdens, the results obtained are very different. Keidanren (the Federation of Business Organizations) has criticized the MOF's concept of nominal tax rates, showing that Japan has the highest level of corporate tax burden of any major industrial country (Kubouchi 1984). This is due mainly to the past curtailment of special measures for tax incentives. The Keidanren estimate has induced a controversial debate among many economists, including the MOF staffs.3 Keidanren insisted that the corporate tax burden should be estimated to take account of the effects caused by tax incentive policies. Thus, it calculates the effective rate of tax burden in the corporate sector based upon the tax amounts actually paid by corporations. Specifically, the MOF estimate II is changed into the Keidanren's effective rate Ile, by applying certain coefficient to the former:
where Ts = revenue loss from special tax measures, and Tf= final tax payments. In Table 11.3, the results of the different estimates of corporate tax burden are indicated in an international comparison. The most conspicuous result is that the smallest gap between the nominal and effective rates of the corporate tax burden appears in Japan. The statutory-base figure presented by the MOF does not provide a realistic picture of the actual tax burden. If greater importance is placed on the Keidanren estimate, the Japanese tax burden in the corporate sector is higher than in any other industrialized country. This reflects the fact that the curtailment of special tax measures in the 1980s has been accelerated especially in the corporate sector. In contrast, governments in other major countries have taken the initiative to introduce bold measures for tax relief to
Table 11.3 An international comparison of corporate tax rates, 1984
MOF Keidanren a b c
Japan
USA
UK
W. Germany
France
52.92 51.57
51.18 32.28a
52.00 18.06b
56.52 49.84
50.00 45.70C
1985 figure. 1982 figure. 1980 figure.
Source: Kubouchi (1984). 3 On this point, Hollerman (1984) states: 'At the present time, a controversy is being waged between Keidanren (Japanese Federation of Economic Organizations) and the Ministry of Finance concerning the level of corporation taxes, in which Keidanren maintains that Japanese corporations are more heavily taxed than those in the USA. This controversy symbolizes the new state of events in Japan.'
Corporate Tax Levels and Tax Incentives
199
promote economic revitalization, such as the ACRS (accelerated cost recovery system) in the USA, or the initial allowance in the UK. It is evident that the relative level of effective tax rates in Japan and the other major countries (e.g. the USA) has been reversed in the past decade (see Gravelle 1983). Now we shall compare the corporate tax burden in Japan with that in the USA. Since Japan and the USA are closely linked in business transactions, the level of tax burden in the corporate sector must affect substantially the flow of capital and the movement of business activities. Therefore, it is important to explore whether Japan's corporate tax burden is heavier or lighter than that of the USA in a meaningful way. Figure 11.3 compares nominal rates of national and local corporate taxes in Japan with those in the USA for the years 1970-93. These are considered to be ex post rates based upon actual tax payments and reported corporate income for tax purposes. Clearly, Japan shows an upward movement of tax rates until 1989 while the USA shows a declining trend until recently. This result reflects the fact that Japan has continued to increase basic rates of corporate taxation since 1970, whereas the USA has lowered its corporate tax rates and expanded investments tax credits. However, nominal corporate tax rates have been declining since the late 1980s in both Japan and the USA, although Japan's rate has fallen more swiftly. However, such a comparison using nominal tax rates to contrast the relative tax burdens of two countries is of little significance economically. In the first place, the two countries have adopted a wide variety of special tax measures, such as tax-free reserves, accelerated depreciation, and tax credits. As a consequence, the types of
Fig. 11.3
Nominal tax rates of corporate taxes, Japan and USA, 1970-1993
Nominal tax rates arc defined as the ratio of actual tax payment (including both national and local corporate taxes) to reported taxable income. Source: Tajika and Yui (1999), calculated from the following statistical data: NTA, Actual Performance of Corporate Firms in View of '[ax Statistics; Ministry of Home Affairs, Actual Reports of Local Public Finance; US Internal Revenue Service, Statistics of Income: Corporation Income Tax Return; and US Economic Reports of the President, yearly.
200
Corporate Taxation and Taxes on Capital
taxable income that are reported differ to a considerable extent. Second, different increases in the price level in each country influence the real tax burden of the corporate sector. We have already examined the first of these from several standpoints, but have not yet touched on the relation between the corporate tax burden and inflation. Since Japan and the USA have experienced different levels of inflation in the past decade or so, we need to include inflationary effects as well as incentive provisions when comparing the corporate tax burdens of the two countries. For the computation of the tax base, corporate net income is calculated by subtracting the expenses incurred in a firm's receipts from the gross receipts or sales. If all expenses are paid when sales are made, the difference between receipts and expenditures poses no problem in providing a correct concept of net income. There are, however, two complications (see Pechman 1987, 173): (1) business expenses are often paid long before sales are realized, and (2) expenses and sales are generally assessed at different price levels. Thus, a variance in accounting gains and losses appears because of inflation when calculating the tax base of corporate taxation. There are mainly four components included in accounting methods that are affected by inflation: tax-free reserves, depreciation allowances, net financial assets and liabilities, and inventory valuation. In order to define a concept of income adequate for taxing corporate profits, we need to have economic income that is not embodied in current tax law but is derived from 'accretion' or 'economic power'. In the case of business firms,4 this is equivalent to the variation in the real market value of corporations before profits are distributed to shareholders for a specific period. When we try to compare the level of corporate tax burden between Japan and the USA, it is necessary to calculate the effective tax rate, that is, the ratio of actual tax paid to economic income, not reported taxable income. The concept of economic income provides a common base to measure the corporate tax burden under different tax systems. An interesting attempt is made by Tajika and Yui (1999) to estimate effective tax rates for both countries on the basis of the economic power concept, as a result of making all adjustments to correct for inflation. We see first the relation between economic income and reported taxable income, and then the deviation between these two incomes. The incomes are related as follows: Economic income = taxable income + tax-free reserves + depreciation + net gains of financial liabilities + inventory valuation + others.3 Table 11.4 indicates the ratio of each item relative to economic income, neglecting the last item. To begin with, gains resulting from tax-free reserves are added to 4 If this concept is applied to individuals, it is equivalent to the sum of personal consumption plus the net increase in the value of the individual's assets during the taxable period, according to the Haig-Simons formula (see Ch. 5). ' 'Others' include both effects of income adjustment and local taxes.
Corporate Tax Levels and Tax Incentives
201
Table 11.4 Taxable and economic incomes of non-financial corporations, Japan and USA, 1970-1993 (%)
Year
Taxable income -^ economic income (1)
Tax-free reserves (2)
Depreciation (3)
Net gains of financial liabilities (4)
Inventory valuation (5)
Japan 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
70.0 70.5 67.9 58.1 60.5 87.4 70.1 79.5 83.7 75.7 75.7 90.5 88.1 85.6 84.3 84.0 84.4 81.0 84.6 81.0 84.3 82.4 85.0 79.5
4.2 3.1 4.3 4.3 5.3 2.2 3.3 5.0 2.3 2.4 1.5 0.9 -1.8 -0.3 1.8 1.8 0.3 1.1 1.6 2.3 0.4 1.6 0.5 1.9
14.7 18.6 15.0 8.6 6.2 -0.9 6.9 6.4 7.3 5.5 4.1 5.5 6.2 8.6 7.5 8.9 10.6 10.5 9.8 6.9 7.6 8.3 11.7 5.1
5.7 -2.3 10.7 40.4 37.3 8.9 20.2 5.4 -3.0 24.0 27.2 0.6 0.5 -2.4 0.5 -3.8 -13.7 -3.9 -0.8 5.2 2.2 1.0 -8.4 0.3
-1.1 0.6 -4.7 -15.0 -14.7 -12.9 -6.6 -2.7 1.9 -12.6 -13.8 -5.4 -1.6 1.6 0.6 3.6 10.0 5.3 0.1 -0.9 -1.8 -0.4 1.5 2.8
USA 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982
100.4 97.1 95.4 101.3 107.3 101.2 100.5 98.7 104.6 107.7 104.4 104.1 93.1
0.8 0.6 0.4 0.6 1.6 -0.5 0.2 1.2 -0.1 0.5 1.6 0.2 1.2
-1.3 -0.5 1.4 1.1 -3.6 -7.6 -5.8 -3.7 -6.5 -7.6 -9.4 -8.2 -3.4
-0.5 -0.3 -0.3 0.4 6.3 4.5 1.7 1.1 1.3 3.7 4.4 -0.4 -1.1
-8.5 -4.4 -5.4 -14.9 -27.3 -6.1 -6.5 -6.9 -8.3 -14.1 -12.5 -7.2 -2.9
Corporate Taxation and Taxes on Capital
202
Table 11.4
contd. Taxable income -reconomic income
Year
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
(1)
87.2 82.1 76.3 73.4 86.2 88.8 87.3 90.0 95.5 88.6 87.4
Tax-free reserves (2)
Depreciation
-0.5 0.8 -0.1 0.4 -0.1 0.4 0.4 1.2 0.6 0.6 -0.2
financial liabilities
Inventory valuation
(3)
(4)
(5)
4.5 9.1 15.4 16.3 9.2 7.1 6.2 2.2 -2.8 3.4 5.3
-0.5 -1.2 0.3 0.7 -0.8 -0.3 -0.3 -0.0 -0.2 -1.0 -0.9
Net gains of
-2.3 -1.6 -0.3 1.1 -3.3 -3.4 -1.8 -1.4 -1.5 0.8 0.2
Note: Columns (2)-(5) are computed as ratios of economic income. The difference between (1) and (2) + (3) + (4) + (5) implies other effects, such as income adjustment and local taxes. Source: As Fig. 11.3.
taxable income in column (2). Next, since depreciation is based on bistorical costs, gains and losses resulting from depreciation allowances should be recognized in column (3) as inadequate depreciation in the period in which they accrue. Also, real gains and losses on net financial assets and liabilities in column (4) should be included in the calculations of income. Lastly, the effects of inventory valuation due to price changes are added to other items in column (5). Using the Tajika-Yui estimates, we find a number of interesting facts which explain some of the differences between economic and taxable incomes in Japan and the USA. First, economic income has always been larger than taxable income in Japan, while in the USA the former is often smaller than the latter, especially during periods of high inflation from the mid-1970s to the early 1980s. Second, tax-free reserves have constantly got taxable income narrower than economic income in Japan. This was true in the 1970s. By contrast, they have not had so strong an impact in the USA. Third, taxable depreciation has been allowed more often than economic depreciation allowances for tax purposes, resulting in gaps between the two incomes. Japan generally allowed generous depreciation for tax purpose, generating a narrower tax base. On the other hand, the USA had a shortage of depreciation allowances until 1982, resulting in larger taxable income. However, as a result of introducing the ACRS in 1982 under the Reagan Administration, the use of accelerated depreciation contributed greatly to reducing the tax base after 1983. Fourth, inflation plays an important role in causing net gains of financial liabilities. Overall, the main factors that cause deviations between economic income and taxable income are net gains of financial liabilities in Japan while they are of little importance in the USA. Fifth, in Japan inventory evaluation also played an
Corporate Tax Levels and Tax Incentives
203
Fig. 11.4 Average effective tax rates of corporate taxes, Japan and USA, 1970-1993 Average effective tax rates are equal to the ratio of actual tax payments (national and local) to economic income. Source: As Fig. 11.3.
important role in expanding taxable income due to gains from inventory caused by inflation in the 1970s, but thereafter its role turned to the reverse, reflecting inventory losses due to the slowdown of the wholesale price level after the mid-1980s. On the other hand, inflation in the USA has generally tended to make taxable income less by generating inventory gains. Turning now to consider the corporate tax burden calculated on the basis of economic income, Figure 11.4 shows the average effective tax rates in Japan and the USA during the period 1970-93. While in the USA tax rates continue to fall in the long run, those in Japan tend to move upward. Before 1980 Japan's tax rates were relatively lower than those in the USA (with the exception of 1975), but they began to exceed the US level after 1980. Particular attention should be paid to the widening gap between the effective tax rates of the two countries in the 1980s and the 1990s. In fact, the deviation in the tax rates expanded to as much as 20 per cent in 1984. Two major factors behind this phenomenon are the fact that gains on depreciable assets resulting from accelerated depreciations have sharply increased in the USA, and that net gains of financial liabilities have become minor in Japan. As is clear from the evidence, Japan's corporate tax burden is now much heavier than the US tax burden if one calculates the tax burden using the economic income concept as a common base. In this chapter, we have analysed the corporate tax burden in view of tax incentive policies. No attempt has been made to clarify the effectiveness of tax incentives on the promotion of private investment and savings. A further task is required to investigate the relationship between tax incentives and their policy goals, including the promotion of corporate investment, although such a study would be very difficult to achieve in quantitative terms (see, e.g. Takenaka 1986).
12
Inheritance and Gift Tax Estate transfers, bequests, and gifts are generally regarded as appropriate objects of taxation. There are two main objectives for taxing these items. First, such transfer taxes may check the accumulation of undue concentrations of wealth and thus may promote a more equitable distribution of economic resources. Second, taxes on property transfers raise revenue. They tend, however, to comprise only a small percentage of the total tax revenues in advanced countries, despite the imposition of high statutory marginal tax rates. Therefore, greater importance has been attached to the first objective in justifying the imposition of taxes on property transfers. In societies where property is privately owned, the government is responsible for protecting the property rights of individuals; but at the same time, it intervenes in the transfer of property from one generation to the next. Taxes on property transferred from individuals to their heirs have long been considered a desirable form of direct taxation, and constitute one of the oldest forms of taxes. In Japan, such taxes have been used exclusively at the national level from the inception of the tax system. We will consider two main topics in this chapter. In the first section, we explore the structural features of the Japanese inheritance and gift taxes in an historical perspective. In the second, we consider the role of property transfer taxes in affecting the tax burden and its redistributional effects. STRUCTURAL FEATURES
Accession Tax
The modern estate tax took effect in 1905 and was based on the value of the decedent's property. It was retained until 1950, when the government adopted the accession tax proposed by the Shoup Mission. The Mission recommended the accession tax as one of the best forms of transfer taxation for the two objectives noted above. The basic framework of the accession tax is as follows (see Shoup Mission 1949, 143-55). 1. It is a cumulative tax levied on the recipients of gifts and bequests. 2. It is graduated progressively according to the total amount of gifts and bequests received by a given individual. 3. The manner of its application is similar to that of the gift tax: when a gift or bequest is received, it is added to the total of taxable gifts and bequests previously received, and tax is computed on that total according to the current set of rates.
Inheritance and Gift Taxes
205
Tax is also computed at the current set of rates on the previous cumulative total, and the difference between the two taxes becomes the amount of tax currently due. In theory, there are two major forms of transfer taxation, depending upon how the tax base is determined and when the transfers are made (see Pechman 1987, ch. 8; Institute of Fiscal Affairs 1978, ch. 15). One is an estate tax on the privilege of transferring property at death, based upon the size of the entire estate. The other is an inheritance tax, in which tax is imposed on the privilege of receiving property from the decedent. This tax is based on the size of each individual share of the estate. In addition, the gift tax has been established to complement taxes on estates and inheritance. Without the gift tax, an individual would distribute his estate to his successors before he died in order to avoid or mitigate the burden of transfer taxes. When the accession tax was proposed, the Shoup Mission pointed out that it had several advantages over the separate estate and gift taxes. The main points of the Shoup Mission's argument are as follows. 1. The tax burden will be distributed more equitably among heirs. Under the accession tax, the total tax on the entire property is lower if the estate is divided among two or more heirs than if it is left all to one person. Under the estate tax, on the other hand, the tax is about the same, regardless of the number of beneficiaries. The former result is to be preferred, since a given sum of money spread thinly among a number of people does not give rise to as much taxpaying as the same sum concentrated in the hands of one person. Thus, the accession tax promotes a wider distribution of wealth than does the estate tax. The cumulative feature of the accession tax prevents circumvention of this intention through multiple gifts. It also encourages donors to make gifts and bequests to those who have not already received gifts and bequests from others. 2. Under the existing estate and gift tax law, in which each tax has its own separate rate scale, it pays the individual to split his giving carefully between gifts during life and bequests at death. If he achieves just the right combination, he minimizes his total gift and estate taxes. Under the accession tax, it makes no difference to the total tax whether the gift is made during life or at death. The accession tax is thus a neutral tax; the gift and death tax combination is not. The basic nature of the accession tax is evident from the following statement: The accession tax, so far as it applies to transfers at death, is simply an inheritance tax, and as such has a long record of experience in many countries. So far as it applies to gifts during life, it is similar to the well-known gift tax. The accession tax that we recommend for Japan is therefore a combination of two familiar taxes, and affords no hazard in the way of experimentation with new types of taxation. (Shoup Mission 1949, 145.)
Basically, the accession tax is closer to an inheritance tax than an estate tax and so is regarded as a transfer tax derived from a modification of the inheritance tax principle. If each receipt of the inheritance or gift is taxed separately in its unmodified
206
Corporate Taxation and Taxes on Capital
form, this will lead to serious defects. For instance, two individuals would pay different taxes on equal inheritances if one received them in a lump sum and the other received them from several decedents. Under the accession tax, this defect can be remedied. Total lifetime acquisitions of a given individual through inheritances and gifts are levied on the basis of a cumulative formula. One calculates the tax in any one year by subtracting the tax paid on earlier acquisitions from the tax on total acquisitions that one has received.
Present Methods of Taxing Property Transfers In spite of the Shoup Mission's strong support, the accession tax was enforced for only four years and was repealed in 1953. It was disintegrated into an inheritance tax and a gift tax, similar to the unmodified inheritance tax described above. One of the chief reasons for the change was the presence of serious deficiencies in administrative practices. The accessions tax was too sophisticated to be managed adequately with a poor level of tax administration. For instance, it was almost impossible for tax officials to determine total lifetime acquisitions, which was necessary for computing the tax base. Under a new system starting in 1953, heirs and donees paid their own taxes, based on the value of property left or transferred to them. As stated earlier, varying distributions among the same number of heirs could affect the tax payment to a substantial degree. This situation was considered undesirable, and stimulated the government to reform taxes on property transfers once more. In 1958 a hybrid system of an estate tax and an inheritance tax was instituted, and this system has remained basically intact up to the present time. This form of transfer taxation is unique, in comparison with the estate tax in the USA and the UK, or the inheritance tax in Germany and France. The main aim of combining the two taxes was to balance the tax burden on estates of the same size with the same number and types of heirs, even if the estate was distributed differently among the heirs. To begin with, let us explain briefly the present calculation method of the inheritance tax, separating the computation of the tax base from that of the tax amount in 1992 (see MOF Tax Bureau 1992; Gomi 1985). As indicated in Figure 12.1, the inheritance tax is levied on the gross estate acquired through inheritance or bequest minus non-taxable property,1 liability, and funeral expenses, that is, on the net estate. If an heir receives properties by gift from the decedent within three years of his death, the value of such property transfers is included in the tax base. Thereafter, a basic exemption of ¥48 million+ (¥9.5 million X number of statutory heirs) is deductible and the remaining property constitutes the amount of taxable inheritance. 1 In addition to the non-taxable treatment of life insurance, personal accident insurance, retirement, and similar allowances received by heirs, property transfers through inheritance or bequests to religious, charitable, educational, and scientific organizations are not subject to the inheritance tax.
Inheritance and Gift Taxes
Fig. 12.1
207
Computation of the Japanese tax base
Source: Data submitted to the Tax Advisory Commission.
Figure 12.2 summarizes the process of computing the tax amount. The amount of taxable inheritance (A in the figure) is divided among the statutory heirs (assuming wife, son, and daughter) in accordance with percentages prescribed in the Civil Code (i.e. -, -, and -) (B, C, D in the figure), irrespective of the actual shares of their inheritance.2 Then, tax rates are applied to each share computed in this way, result ing in the tax amounts (parts a, b, and c). Total inheritance tax liability is obtained by combining the shares of the individual heirs (part E). The total tax is then distributed among the heirs in proportion to the amounts they actually receive (parts d, e, and f). The inheritance tax is, however, increased additionally by 20 per cent for heirs other than spouses, children, and parents, although the total tax including this addition should not exceed 75 per cent of the heirs' actual share. Furthermore, there are several tax credits available to certain heirs. Of most importance is the credit for the spouse, computed as follows. If a surviving spouse receives properties, X can be credited against the inheritance tax (part d in the 2
Methods of dividing an estate among the statutory heirs are stipulated by the Civil Code as follows:
(1) spouse -, lineal descendants (children) —; (2) spouse -, lineal ascendants (parents) • ; (3) spouse4, brothers and sisters -; in the case of Figure 12.2, we assume two lineal descendants.
208
Corporate Taxation and Taxes on Capital
Fig. 12.2
Computation of taxes due
Source: see Figure 12.1.
figure) he/she actually pays, where
X=
total inheritance tax liability (part E)
(1) one-half of total value of taxable property (minimum ¥80m.) or (2) the amount of property actually received by the spouse x
Total value of taxable property
Inheritance and Gift Taxes
209
(With regard to (1) or (2), the smaller of these is applied.) The credit for the spouse enormously increases the amount of property that married couples may transfer free of tax. In fact, this method of credit usually provides complete exemption for transfers between husband and wife. (The appendix to this chapter contains a detailed description of how to calculate the relevant tax amount.) Likewise, three tax credits are available to other individuals under specific conditions. As mentioned above, properties received by gift from the decedent within three years of his death are included in the tax base, but the gift tax for such properties is credited against the inheritance tax to avoid double taxation. This is called the tax credit for gift tax paid. If an heir is under 20 years of age, his inheritance tax is reduced by ¥60000 for each of his years under 20. Similarly, if an heir is handicapped, his inheritance tax is decreased by ¥60000 (¥120000 in the case of the severely handicapped) for each year before he reaches 70. Next, the gift tax is levied on properties received in the calendar year with a basic annual exemption of ¥600 000. The value of the property is assessed on the basis of the current price or value at the time of acquisition, but several items are excluded from the tax base, e.g. properties acquired from relatives for living and educational expenses, and those used for religious, charitable, scientific, or other public purposes. In addition to the basic exemption, a lifetime allowance of ¥20 million is allowed for gifts of residential property to a spouse who has been married to the donor for over 20 years. Also, if a severely handicapped person becomes the beneficiary of money, securities, or other properties under the trust contract, a gift tax is not imposed on the beneficiary up to the value of ¥60 million. Rates, Exemption, and Credits Tax rates are applied to each share of the taxable inheritance (B, C, and D in Figure 12.2) or taxable gifts received by the donee. In Table 12.1, inheritance and gift tax rates are shown under the 1975-87 and 1998 tax law. Both rates begin at 10 per cent and rise to a maximum of 75 per cent in 1975-87 and 70 per cent in 1998, but the gift tax has much steeper progressivity than the inheritance tax. Indeed, the tax rate in 1998 reaches 70 per cent at ¥100 million under the gift tax, while it does so only at ¥2 000 million under the inheritance tax. The reason for the steeper progressivity of the tax on gifts is that it was substituted for the accession tax, which would have applied at a single set of rates to the cumulative gifts and bequests that a given individual receives during his life. These tax rates have not been adjusted frequently since 1958 when the new inheritance tax was adopted. At that time, the progressive rate scheduled started at 10 per cent on the first ¥0.3 million and rose to a maximum of 70 per cent above ¥100 million, with 13 brackets. Thereafter, there has been only a minor alteration in 1966. Similar adjustments have been made on the gift tax. In recent years the inheritance tax tends to be flatter in reducing the bracket number; that is to say, it was only 9 brackets in 1998 as against 13 in 1975-85.
210
Corporate Taxation and Taxes on Capital
Table 12.1 Tax rates (%)
10 15 20 25 30 35 40 45 50 55 60 65 70 75
Inheritance and gift tax rates, 1975-1987 and 1998 Taxable inheritance (¥m.)
Taxable gifts (¥m.)
1975-87
1998
1975-87
1998
under 2.0 2.0-5.0 5.0-9.0 9.0-15.0 15.0-23.0 23.0-33.0 33.0-48.0 48.0-70.0 70.0-100.0 100.0-140.0 140.0-180.0 180.0-250.0 250.0-500.0 500.0 and over
(under 8.0) (8.0-16.0) (16.0-30.0) (30.0-50.0) (50.0-100.0)
under 0.5 0.5-0.7
(under 1.5) (1.5-2.0) (2.0-2.5) (2.5-3.5) (3.5-4.5) (4.5-6.0) (6.0-8.0) (8.0-10.0) (10.0-15.0) (15.0-25.0) (25.0-40.0) (40.0-100.0) (100.0 and over)
(100.0-200.0) (200.0-400.0) (400.0-2 000.0) (2 000.0 and over)
0.7-1.0
1.0-1.4 1.4-2.0 2.0-2.8 2.8-4.0 4.0-5.5 5.5-8.0 8.0-13.0 13.0-20.0 20.0-35.0 35.0-70.0 70.0 and over
Source: National Tax Administration, Annual Report, 1987 and 1998.
In contrast, the basic exemption has been increased ten times since 19583 until now. A drastic change occurred in 1975 (the 1987 level), when the level of basic exemption was sharply increased from the modest level of 1973-74: i.e. from ¥6.0 million+ (¥1.2 million X number of statutory heirs) + a maximum of ¥6.0 million for the spouse to ¥20.0 million + (¥4.0 million X number of statutory heirs). Similarly, the basic exemption on the gift tax has been adjusted occasionally: ¥0.2 million in 1958, ¥0.4 million in 1964, and ¥0.6 million in 1975. Tax credits under the inheritance and gift taxes have also been increased on a few occasions. The basic exemption of inheritance tax was raised twice in 1992 and 1994, and at present it is ¥50.0 million + (¥10.0 million X statutory heirs). Until 1987, the number of decedents leaving estates with taxable value had steadily increased, mainly reflecting the absence of tax changes in more than a descade (see Table 12.2). The percentage of such decedents was only 3.4 per cent in 1970, but it had risen to 7.9 per cent in 1987, although there was a large drop between 1974 and 1975. In 1975, the basic exemption was raised drastically, causing a sharp decrease in the number of taxable decedents and in the amount of the inheritance tax relative to the value of all taxable property. Adjustments on the basic exemptions and tax rates under the inheritance and gift taxes are thus long overdue. The bubble boom clearly affected the value of taxable property and inheritance tax due. In fact, it is noteworthy that both substantially expanded around the period 1987—92. Since the collapse of the bubble, they have shrunk to a considerable extent. 3
In 1958, a basic exemption was allowed for ¥1.5m. + (¥0.3 X number of statutory heirs).
Table 12.2
Trends of tax payments in the inheritance tax, 1958-1996
Nos. deceased*
1958 1965 1970 1974 1975 1980 1982 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 a
The coverage of taxation
(B)/U) (%)
(A)
Objectives of taxation (= taxable decedents) (B)
684 189 700 438 712962 710510 702 275 722 801 711 883 740 247 752 283 750 620 751172 793 014 788 594 820 305 829 797 856 643 878 532 875 933 922 139 896211
5284 13407 24454 32898 14593 26797 35922 43012 48 111 51847 59008 36468 41655 48287 56554 54449 52877 45335 50729 48476
0.8 1.9 3.4 4.6 2.1 3.7 5.0 5.8 6.4 6.9 7.9 4.6 5.3 5.9 6.8 6.4 6.0 5.2 5.5 5.4
Total value of taxable property
Nos. of statutory heirs per decedent
Amount (¥b.)
Per decedent (¥000)
(C)
4.17 4.27 4.23 4.31 4.17 4.11 4.05 4.03 3.99 3.93 3.68 3.89 3.86 3.81 3.86 3.81 3.79 3.72 3.69
36.7 209.1 701.1 1 896.6 1 512.1 3021.5 4472.9 5428.7 6 246.3 6 763.7 8 250.9 9 638.0 11768.6 14105.8 17841.7 18820.1 16754.5 14545.4 15299.8 14077.4
Inheritance Tax amount (¥bn.)
Per decedent (¥000)
(D)/(Q (%)
884 3 111 5488 13303 13521 16417 17620 18620 19250 20 142 24307 42855 57448 61 149 70 112 62626 52514 46450 42835 39970
12.7 19.6 19.1 23.1 13.0 14.6 14.2 14.3 14.8 15.4 17.4 16.2 20.3 20.9 22.2 18.1 16.6 14.5 14.2 13.8
(D)
6345 15597 28670 57651 103618 112755 124518 126214 129831 130456 139826 264 286 282 526 292 124 315481 345 646 316858 320 843 301 599 290 399
4.7 41.0 134.2 437.7 197.3 439.9 633.0 776.9 926.1 1044.3 1434.3 1 562.9 2393.0 2952.7 3965.1 3409.9 2776.8 2105.8 2173.0 1937.6
Data based on Statistics of Population Census (Ministry of Health and Welfare). Other data are derived from Annual Report of the National Tax Administration. Source: Data submitted to the Tax Advisory Commission.
212
Corporate Taxation and Taxes on Capital
As is evident from the formula of calculating the inheritance tax, the greater the number of statutory heirs, the lighter the tax burden. In recent years, it has often been argued that some people try to avoid the inheritance tax by increasing the number of legally permissible statutory heirs before they die. Although the number of statutory heirs per decedent has declined in the long run (see Table 12.2), this type of tax avoidance is becoming more frequent. In fact, we find many cases in which individuals attempt to adopt grandchildren, nephews, and nieces as sons and daughters for tax purposes. The major part of the inheritance tax base consists of land, housing, business property, stocks and bonds, cash and deposits. Land, such as farming fields and residential sites, occupies the largest share—67.4 per cent of the total, followed by 14.4 per cent for bonds and stocks in 1990. Cash and deposits occupy 7.8 per cent of the tax base, housing 4.5 per cent, and business property 0.4 per cent.
THE ROLE OF I N H E R I T A N C E AND GIFT TAXES
Impact on the Tax System How significant have taxes on property transfers been in the tax system as a whole? The inheritance tax makes up only a small percentage of national tax revenues (see Table 12.3), but its relative share expanded from 0.5 per cent in 1950 to 3.9 per cent in 1998. Moreover, this share is the largest among major industrialized countries: in 1996-97 the corresponding figures were 2.0 per cent for the USA, 0.7 per cent for the UK, 0.6 per cent for Germany, and 2.4 per cent for France. The burden of the inheritance tax varies as the number of heirs differs. Two cases of the inheritance tax burden as measured by the value of taxable property are shown in Table 12.4. No tax is paid by the spouse because of the tax credit for spouses, and more children can make the tax burden lighter by increasing the amount of the basic exemption. For estates with taxable value of ¥3 billion, the inheritance tax amounts to 26.8 per cent of the taxable value when bequested to a spouse and one child, but to 23.0 per cent when turned over to a spouse and four children. It is difficult to judge whether the burden of the inheritance tax is actually heavier for a spouse with one child, as there are several other factors that affect the tax burden. The treatment of the gift tax seems to be generous, and it is hard to prevent tax avoidance on gifts from one living person to another. Likewise, land and closely held businesses are allowed to be greatly undervalued so as to lower the burden of the inheritance tax.4 4 Farm land was given special tax treatment under the inheritance tax until fiscal year 1991. An heir who receives agricultural land from a farmer through inheritance or bequest was allowed to exempt the difference between the market value of agricultural land and the present discounted value of agricultural income on land if he continues lo cultivate that land as a farmer for the next 20 years (see MOF Tax Bureau 1991, 132).
Inheritance and Gift Taxes
213
Table 12.3 Inheritance tax revenues, 1950-1998 Fiscal year
National tax revenues (¥bn.)
Inheritance tax" Amount (¥bn.)
1950 1955 1960 1965 1970 1975 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998
570.2 936.3 1801.0 3278.5 7773.2 14504.3 28368.8 30455.1 32003.1 34162.1 36774.8 39150.2 42851.0 42360.4 52193.8 57136.1 62779.8 62779.8 57396.4 57114.2 54000.7 54963.0 55226.1 57905.2 60345.1
2.7 5.6 12.3 44.0 139.1 310.4 440.5
552.1 664.5 786.1 877.3 1061.3 1369.6 1513.0 1830.9 2017.8 1918.0 2583.0 2746.2 2937.7 2669.9 2 690.3 2419.9 2395.0 2351.0
% of total 0.5 0.6 0.7 1.3 1.8 2.1 1.6 1.8 2.1 2.3 2.4 2.7 3.3 3.6 3.5 3.5 3.1 4.1 4.8 5.1 4.9 4.9 4.4 4.1 3.9
a
Including the gift tax. Source: National Tax Administration, Annual Reports.
As regards the gift tax, the tax burden is computed as the ratio of the tax paid to the amount of taxable gifts. For the past decade, the gift tax has averaged about 12 per cent of the taxable value. Redistributive Effects on Wealth In addition to the individual income tax, taxes on property transfers, such as inheritance and gift taxes, may influence the size distribution of wealth to some extent because of progressive tax rates. Despite their minor proportion in the total tax revenue (i.e. 3—4 per cent at the national government level every year), these taxes play an important role in redistributing economic power in connection with their goal of
Table 12.4 Value of
taxable property (¥m.)
(1)
The inheritance tax burden, 1998 Spouse's share (¥m.)
(2)
35 70 100 50 200 100 150 300 500 250 500 1000 1000 2000 1500 3000 2500 5000 Basic exemption
Spouse and 1 child
Spouse and 4 children
Tax paid by spouse (¥000) (3)
Tax paid by child (¥000) (4)
0 0 0 0 0 0 0 0 0
1850 14300 30800 72300 203 800 503 800 803 800 1 450 300
(3)+ (4) (1) (%) (5) 0 1.9 7.2 10.3 14.5 20.4 25.2 26.8 29.0
u
Notes: Columns (5), (8): ratio of total taxes to total taxable inheritance. Source: As Figure. 11.1.
¥70m.
(6)+ (7)
Tax paid by spouse (¥000) (6)
Taxes paid by children (¥000) (7)
(1) (%) (8)
0 0 0 0 0 0 0 0
0 0 7850 20000 52000 157000 414 500 689 500 1 304 500
0 0 3.9 6.7 10.4 15.7 20.7 23.0 26.1 ¥100m.
Inheritance and Gift Taxes
215
avoiding an excess concentration of wealth. Hence inheritance and gift taxes can be treated as a supplementary part of income tax in terms of redistributive taxation. Therefore, it is worth investigating the extent to which they affect wealth distribution. I shall use the same procedure on wealth distribution that I followed to determine the impact of the individual income tax, i.e. the Lorenz curve and Gini coefficients. For this estimation, it is essential that corresponding figures for income and tax yields be used. Necessary data are available from the Annual Report of the National Tax Administration. Let us consider the case of the inheritance tax. The available information on tax statistics poses some problems for the purpose of our study. For example, data on the wealth distribution of statutory heirs do not exist. Since taxes on property transfers take the form of an inheritance tax in Japan, it is necessary to investigate how the wealth of statutory heirs, obtained through inheritance and bequest, can be affected by taxation. Despite this requirement, no data are available on heirs, and only data on decedents are found in the tax statistics. Accordingly, we are compelled to use the figures concerning the wealth of decedents subject to the inheritance tax. The redistributive effect of the inheritance tax in this analysis is not related to how heirs inherit the estates in practice. If we ignore such an important limitation on the available statistics, we can obtain the number of taxpayers,3 the value of property, and the tax yields necessary for our estimations. The results are summarized in Table 12.5, in which the equalization coefficient
216
Corporate Taxation and Taxes on Capital Table 12.5 Redistributive effects of the inheritance tax, 1958-1996 Gini coefficient
1958 1960 1962 1964 1966 1968 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Before tax Rb
After tax R,
0.438 0.460 0.477 0.491 0.439 0.449 0.472 0.485 0.476 0.487 0.490 0.403 0.407 0.411 0.414 0.409 0.429 0.450 0.456 0.463 0.461 0.471 0.472 0.494 0.515 0.521 0.520 0.538 0.521 0.497 0.482 0.477 0.529
0.377 0.396 0.401 0.416 0.370 0.383 0.398 0.405 0.402 0.403 0.408 0.350 0.357 0.360 0.363 0.357 0.374 0.400 0.408 0.415 0.413 0.421 0.421 0.435 0.456 0.451 0.449 0.465 0.460 0.441 0.432 0.429 0.485
Source: Calculated from Administration.
Annual Reports of the
Equalization coefficient (%) 13.92 13.90 15.93 15.38 15.73 14.64 15.72 16.48 15.45 17.15 16.67 13.31 12.39 12.53 12.43 12.73 12.96 11.00 10.55 10.37 10.46 10.50 10.92 11.81 11.36 13.51 13.64 13.42 11.68 11.38 10.20 10.08 8.42 National
Tax
variation of
Inheritance and Gift Taxes
217
higher brackets, because part or all of the exemption increase will cut away the taxable value of property at the highest bracket the taxpayer faces. Thus, the sharp elevation of the basic exemption in 1975 clearly effected a weakening of the equalizing power of the progressive rate structure under the inheritance tax. Furthermore, the movement of the before-tax distribution of wealth increased in the 1980s. This implies that the wealth distribution has become more uneven.7
APPENDIX Since the Japanese inheritance tax is extremely complicated, it might be beneficial to show how to calculate the tax amount by using a numerical example under the present tax code (see MOF Tax Bureau 1997,131-2). Example of computation of inheritance tax (¥1 000) Decedent: M r X Heirs: Mrs X Messrs Yl, Y2 (decedent's children) Total value of the properties bequeathed: Value of the properties inherited by each heir: Mrs X MrYl MrY2
¥800 000 (A) 400000 (B) 200000 (C) 200000 (D)
(1) Amount of basic exemption for inherited properties ¥50 000+ ¥ 1 0 0 0 0 X 3 =¥80000
(£)
(2) Tax base (A-E) ¥800 000-¥80 000 = ¥720 000
(F)
(3) Total amount of tax Tax base
Statutory share of inheritance
¥720000 (H)X (Mrs X) X 1/2 =¥360 000 ¥720000 ( H ) X (Mr Yl) X 1/4 = ¥180 000 ¥ 7 2 0 0 0 0 ( f f ) X ( M r Y 2 ) X 1/4 = ¥180000 Total amount of tax (G + H+I) =¥258400
Corresponding tax amount ¥144800 (G) ¥56800 (H) ¥56800 (I) (/)
1 Similarly, the redistributive effects of the gift tax can be calculated on the basis of the same data sources and procedures. In the final results, the before-tax distribution of wealth (i.e. the value of transferred property) tends to grow more uneven in the long term. On the other hand, the value of tp fluctuated greatly, ranging from 22.07 per cent in 1965 to 4.24 per cent in 1985. This pattern of ip fluctuations is quite different from that of the individual income tax and the inheritance tax during the same period. It seems that comparisons over time may be dangerous because data on the gift tax are less accurate and reliable.
218
Corporate Taxation and Taxes on Capital
(4) Tax amount to be paid by each heir
8
See X on p. 208 to calculate tax credit for the spouse.
13
Land Tax Reform A sharp rise in land prices began in the late 1980s, causing a number of land-related problems within the Japanese economy and society. In 1990, heated arguments occurred in the government and the private sector in an attempt to seek an effective policy with respect to reducing land prices. From the expectation that land taxes could play a substantial role in slowing down the increased tempo of land prices, particularly in urban areas, land tax reform emerged as an important measure. On 30 October 1990, the Tax Advisory Commission submitted to Prime Minister Toshiki Kaifu a tax report entitled 'Basic Recommendations on the Ideal Framework of Land Taxation', based upon very intensive deliberations by the Subcommittee on Land Taxation, chaired by myself, which had continued for seven months. These recommendations had the ambitious target of restructuring the land tax system as a whole and displaying its basic direction in the coming decade or so. Allowing for political compromises that were made during deliberations in the LDP Tax Council, the proposal was presented to the 120th Diet session as the Land Tax Bill, and became effective from 1992. The main objective of this chapter is to clarify the fundamental nature of land tax reform that was developed by the Subcommittee on Land Taxation in 1990. First of all, the background of land tax reforms in Japan is traced. The second section reviews the existing land tax system and stresses the necessity of restructuring land taxes in the Japanese tax system. Third the proposed reform packages are outlined about taxes on land-holding, land acquisition, capital gains from the sale of land, and the special issues of agricultural land tax. The fourth section summarizes major points of the Land Tax Bill proposed by the government, and offers some appraisals with special reference to political considerations of the LDP Tax Committee. Lastly, the aftermath of land tax reform is explored in relation to the post-bubble recession.
B A C K G R O U N D OF LAND
TAX R E F O R M
Land Price Hikes: Effects and Causes
The increase in land prices in the late 1980s marked the third run of steep hikes after World War II. As depicted in Figure 13.1, the first run occurred from the second half of the 1950s, reflecting the sustained, high-level rate of economic growth. This chapter draws primarily on Ishi (1991).
220
Corporate Taxation and Taxes on Capital
Fig. 13.1 Long-term trends of land prices (nationwide, residential land), 1956-1998 Likewise, the second land price hike swept across the nation early in the 1970s, triggered by the then PM Kakuei Taraka's grandiose vision of 'Rebuilding Japan's Archipelago' which induced large-scale land speculation throughout the country. Over the past 40 years, land prices have increased continuously with one exception in 1975 when severe monetary restraint was intensively undertaken to reduce the sharp rise in land prices. Given the consistent rising tempo of land prices, a so-called 'land myth' has gradually proliferated with the belief that land prices will never drop. Land itself is considered one of the most valuable assets in the capital market. Behind the 'land myth', land prices have increased at an annual rate far faster than basic economic indicators, such as nominal national income, real income of the working household, and the price level. For instance, the urban land price index of the six largest cities shot up 128 times between 1955 and 1989, while real income of the working household merely rose seventeen times and the consumer price index five times during the same period. In particular, from 1985 to 1990 in urban areas land prices roughly tripled. This abnormal upswing of urban land prices was not necessarily induced by changes in real economic activity. Instead it is widely acknowledged that a substantial part of increased land prices reflected a 'bubble' phenomenon due to the speculative expectation of further price increases. Several problems emerged from such land price hikes, particularly in the metropolitan areas. First, the widening gap between the haves and have-nots (i.e. asset
Land Tax Reform
221
gap) became conspicuous and impaired the people's sense of social fairness. Second, it became almost impossible for the average worker to purchase his own residence by earned income. Generally speaking, housing prices in 1988 were seven times the annual salaried income of average workers, while counterpart figures were three to four times in the USA, the UK, and Germany. In large metropolitan areas where land prices are much higher than the national average, the possibility of becoming an owner-occupier is becoming an unattainable dream for salaried workers. Third, land prices in the Tokyo metropolitan area are so excessively high that it is impossible for foreign, as well as local, firms to enter into that area for business activities. This was the reason why the USA raised the land problem as a non-tariff barrier during the Structural Impediments Initiative (SII) talks in the 1980s. The key factor behind the land problem is that land is held for its asset value alone. This is unique to Japan, though it may hold true for Korea and Taiwan. Even in the early 1990s when land prices began to decline due to the end of the 'bubble economy', it was believed that land was still the most profitable form of asset. Thus, land provided the best opportunity of obtaining appreciation in asset value for individuals as well as corporations. Ideally, land was supposed to be used for housing, office space, factories, etc. Under such patterns of land use, the unit-land price theoretically represents a discounted value for its expected stream of annual returns over a stretch of a certain number of years. This may be called the rental value or net annual value. In Japan, however, market prices of urban land have far exceeded this theoretical value in the past, mainly because land is used as a means for speculation on its asset value apart from its use value (see Ishi 2000, ch. 4). Why did urban land take on such asset value? No doubt, the past steep rise in urban land prices was primarily caused by the advantages of holding land as an asset. There are three reasons behind this (see, for instance, Noguchi 1990). First, the government's land-use plan is ambiguous and too loosely regulated. In Western countries, land use is well-defined by category for commercial, industrial, housing, or agricultural land usage. In addition, it is impossible to divert the established land use without the approval of relevant authorities. Land regulation is thereby strictly enforced. By contrast, in Japan the ill-defined character and ambiguous regulation of land use is prevalent. For example, commercial buildings can easily be constructed in residential areas officially designated for residential purposes only. Agricultural land within the city limits can also be converted into housing or commercial land. Obviously, such lenient regulations have contributed to promoting the huge asset value of urban lands. Second, attention should be paid to the 'easy money' that has persisted since the second half of the 1980s. Under the banner of easy money, financial institutions were encouraged to increase land-related lendings which boosted land prices. After monetary policy moved in the direction of tighter credit in 1990, urban land prices began to fall, as shown in Figure 13.1. This implies that an 'easy money' policy has frequently promoted the sharp rise of land prices in the past. The third factor is the very light tax burden on land. Currently land taxes are levied during the three stages of land acquisition, holding and transfer, but it is widely
222
Corporate Taxation and Taxes on Capital
believed that these taxes have been borne lightly by landowners. No doubt, the generous treatment of land taxation has played a vital role in lifting the asset value of land. Remedies to be Taken Thus, various measures must be taken to resolve these land problems, including regulations on land use, land-related lending activities, and the land tax system. In other words, a comprehensive land policy is needed, which must be initiated by co-operation of the relevant ministries (e.g. MOF, MOC, NLA, etc.). It is important to stress that the land tax system alone cannot be an effective means to remedy land-related issues. The overall reform of land taxes should be accompanied by other reforms of the legal aspects of land use and regulation. Accordingly the Basic Land Law which is generally believed to establish the framework of solving the land problem was enacted in December 1989. Of most importance, this law asserts that all lands should be used to enhance public welfare, in spite of their being privately owned. This way of regarding land as a sort of'public good' would have a revolutionary impact on land ownership and might contribute to an increase in the land supply. In Japan, private ownership of land has so far been protected from expropriation, and this encouraged the idea that a landowner can do what he likes with his land. In contrast, land is thought of as public property in Western nations and cannot be sold freely in pieces by private landowners. Moreover, buildings in these countries cannot be constructed in ways that do not fit into the neighbouring surroundings. After establishing the Basic Land Law in Japan, the fundamental concept of landownership would essentially be changed. This has a close bearing on the enforcement of land taxation in any new land policy. Recognizing the need for curing these serious land issues, the government of Japan decided to establish the Subcommittee on Land Taxation in April 1990 under the Tax Advisory Commission. Consequently, a comprehensive study on the land tax system was conducted by this Subcommittee on the basis of such tax principles as equity, neutrality, and simplicity. This study was made in accordance with the ideas expressed by the Basic Land Law and with other land policies as noted before. The Subcommittee met at least once a week, including domestic and overseas research trips before finally proposing its tax report in October 1990. In this report accompanied by two earlier interim reports,1 the Subcommittee emphasized the following two points as central to the discussion of land tax reform. First, it was necessary to levy an appropriate tax burden on land assets from the standpoint of tax equity. Consequently, this contributed to the efficient utilization of land. Second, land taxation could play an important role, as a part of land policy, 1 See, Subcommittee on Land Taxation (\990a,b). The first interim report clarified the major issues to he remedied in the course of reviewing the current land tax system while offering basic reform targets. The second one sorted out the views expressed by the Subcommittee members concerning land tax reform.
Land Tax Reform
223
in promoting an increase in the land supply via more efficient land utilization. It was also believed that a heavier burden on land would perhaps prevent speculative land transactions. In what follows, the major proposals for land tax reform are presented with their related arguments, based on the final report of the Subcommittee.
REVIEW OF THE E X I S T I N G LAND TAXES
Defects of Land-Holding Taxes Before proceeding to the new tax reform plan, it is necessary to explain briefly the current land tax system and to draw attention to the shortcomings to be remedied. Table 13.1 outlines three types of land taxation on acquisition, holding and transfer. Eight taxes in three separate categories are listed as a system of 'land tax'. Of these there are four that have a connection with the current tax reform debate. The most important issue in the existing land tax system is the extremely light property tax burden levied by municipal governments on land-holdings (the property tax is sometimes called the fixed assets tax in Japan). Such a light tax burden essentially reduces the cost of holding a piece of land, and in turn tends to cause an increase in the value of land as an asset. In addition, it tends to lead to the insufficient use of land, reflecting the almost negligible cost of land-holding. Therefore, there is much support for augmenting the tax burden on land-holding by any means. The current property tax offers an effective policy instrument for this purpose. Currently, the Japanese local property tax system imposes a main land-holding tax, as well as a special land-holding tax. The property tax is levied annually on the assessment value of three taxable assets (i.e. land, buildings, and depreciable assets) at the rate of 1.4 per cent.2 There are two special reliefs for the tax base of residential land, i.e. (1) a reduction of the tax base for residential use to one-half, and (2) a reduction of the tax base of small-scale land for residence (up to 200m ) to one-quarter. Table 13.2 indicates the effective tax burden of the land-holding tax from 1970 to 1995, in which effective tax rates are calculated relative to land assessment values in terms of the official valuation price set by the NLA. Compared with the statutory rate of 1.4 per cent, the property tax burden shows a range of 0.09 to 0.24 per cent and began to decline from the start of the most recent land price hike in 1985 until 1990. Even when the two other taxes were added to the property tax, the low tax burden of holding land remained almost the same, but since 1991 such an effective tax burden has constantly risen, reflecting the recent reform of estimating land evaluation; that is, the ratio of property tax assessment to official valuation of land price has been decided as 70 per cent by the MOHA. 2
rate.
In addition to the standard rate of 1.4 per cent, an upper limit of 2.1 per cent was allowed as a surtax
224
Table 13.1 Stage
Acquisition
Holding
Transfer
Corporate Taxation and Taxes on Capital Outline of land taxation, 1990 Tax item
Outline of the tax Tax base
Tax rate
Others
Land-acquisition tax (Prefectural tax)
Assessment value for the property tax
4/100
Special land-holding tax on acquisition (Municipal tax)
Acquisition cost
3/100
The tax is imposed on buildings besides land The amount corresponding to the land acquisition tax is credited
Registration and licence tax (National tax)
Assessment value for the property tax
Inheritance tax (National tax)
Assessment value for the inheritance tax
50/1000 (rate applied to sales and purchase) 10-70% (progressive tax rate)
Property tax (Municipal tax)
Assessment value for the property tax
Standard rate 1.4/100 Ceiling rate 2.1/100
City planning tax (Municipal tax)
Assessment value for the property tax Acquisition cost
Ceiling rate 0.3/100
Special land-holding ta> on holding (Municipal tax) Income tax (National tax) Corporation tax (National tax) Inhabitant's tax (Prefectural and municipal tax)
1.4/100
The tax is imposed on all inherited assets. The gift tax supplements the inheritance tax The tax is imposed on buildings and depreciable assets in addition to land The tax is imposed on buildings in addition to land The amount corresponding to the property tax is credited
See the 'Outline of land transfer taxation', in Table 13.4
Land Tax Reform
225
Table 13.2 Effective tax burden of the land-holding tax, 1970-1995 (%) Year
1970 71 72 73 74 1975 76 77 78 79 1980 81 82 83 84 1985 86 87 88 89 1990 91 92 93 94 1995
The property tax (land)
Total land-holding taxes
Land assessment
Land assessment
0.09 0.10 0.09 0.11 0.11 0.13 0.15 0.16 0.16 0.15 0.18 0.16 0.17 0.18 0.18 0.19 0.17 0.13 0.12 0.11 0.10 0.13 0.16 0.17 0.21 0.24
0.14 0.15 0.13 0.17 0.23 0.21 0.23 0.24 0.25 0.23 0.26 0.23 0.25 0.26 0.26 0.26 0.24 0.18 0.17 0.16 0.15 0.18 0.23 0.24 0.30 0.33
Note: Total land-holding taxes include the city planning tax and the special land-holding tax in addition to the first column. Land assessment value is set in accordance with the official valuation price. Source: Economic Planning Agency, Annual Report on National Accounts, Ministry of Home Affairs, Data on Local Taxes, each year.
How do we explain the gap between statutory and effective tax rates that we observed before 1991? This is mainly due to the underestimated land valuation for the property tax, as well as the special relief for residential land mentioned above. So far the valuation of land has posed puzzling problems.3 The basic standard of land
3 In Japan, there is no counterpart of the UK Valuation Office which is a unified institution on a nation-wide basis for official land assessment.
226
Corporate Taxation and Taxes on Capital
Table 13.3 The ratios of property tax assessment to official valuation of land price, selected years Year
1979
1982
1985
1988
1991
National average (%) Coefficient of variation
61.4 18.3
67.4 16.9
52.1 23.2
47.2 23.7
36.3 34.2
Note: Ratios are calculated by using the highest land prices in major cities of 47 prefectures, and are simply averaged. Coefficient of variation is the ratio of variants to average value, based upon land prices in 47 prefectures. Source: Data prepared by the Ministry of Home Affairs for the Subcommittee on Land Taxation. Also, author's calculation.
valuation is given by an official valuation price (chika kbji kakaku) which is calculated on 1 January and published every year by the NLT. As seen in Table 13.3, the property tax assessment for land produced a huge deviation from its official valuation price. Every three years the local authority has to attempt to reassess land prices on a nation-wide basis (the number of properties are 160 million) in order to construct a new taxable base for the property tax. In 1979, the property tax on land was assessed on the basis of 61.4 per cent of its official valuation price, but the trend of undervaluation was as low as 36.3 per cent in 1991. Of course, as land prices rose, both the property tax assessment and the official valuation price were raised in conjunction with the rising tempo of land prices in the market. However, the former did not increase as rapidly as the latter.4 This is the reason why a gap has emerged between the two land prices, which in turn has lowered the property tax burden of land-holding. An inspection of land valuation at individual places in 47 prefectures revealed that the higher the land price hike by region, the lower the assessment for tax purposes. For instance, the ratios in 1991 were 21.9 per cent in Tokyo special wards5 and 14.6 per cent in Osaka-city, as compared with the 36.3 per cent national average ratio used in Table 13.3 and a more than 50 per cent ratio for other local cities. Given the fixed level of a 1.4 per cent tax rate, this implies that local governments have made adjustments in order to avoid raising the property tax burden.6 The coefficient of variation in Table 13.3 moves upward from 16.9 to 34.2. The rising trend of coefficients shows the expansion of regional disparity in the undervaluation of property tax assessments for land, partly because land prices themselves have a diverse rising pattern regionally and thereby increase the bias in assessing the 4
The official valuation price of land itself tends to be estimated at only 70 or 80% of its market price. Tokyo is composed of two areas: the 23 special wards in the central area and the other suburban cities and towns. 6 It is widely acknowledged that the effective rate of the property tax (i.e. actual tax payment relative to market price of land) is extremely low, say 0.05-0.06 per cent in the Tokyo metropolitan area. The comparable figures in New York and London are 0.8-1.0 per cent for houses or flats and 2.5-4.3 per cent for office and business space (see data prepared for the Subcommittee on Land Taxation). 5
Land Tax Reform
227
land value for tax purposes. Thus, the important targets in land tax reform were how to increase the land-holding tax and how to reform the land assessment process in order to reduce the asset value of land and stimulate more efficient land utilization. Inheritance and Capital Gains Taxes The second problem in the existing land tax system is the inheritance tax. The inheritance tax itself is not a proper category of land taxation. However, since land is the most important form of inherited assets (i.e. about 70 per cent of the estate total), the current structure of tax on inheritance has a close bearing on the land problem. As shown in Table 13.1, the inheritance tax is levied on land acquisition by taxing property transfers together with other inherited assets, such as stock, deposits, etc. Such a tax is paid by heirs, not decendents.7 The inheritance tax is based on the assessment value set by the NTA, and the tax rate is currently progressive from 10 to 70 per cent with standard basic exemptions. Similar to the property tax, special reliefs are granted for small-scale land (up to 200 m 2 ), where the specific residential-use and business-use assessment values are lowered to 80 per cent and the other-use case to 50 per cent. In relation to Japan's land problem, inheritance in the form of land benefits taxpayers to a great extent, because land is taxed for inheritance at 70 per cent of its official valuation price which itself is believed to be only approximately 70 per cent of its market price. Thus, the inheritance tax assessment for land would be about half of its market price, while inherited stock and bonds are subject to tax on the basis of their full market prices. Evidently, such tax benefit has in recent years played a role in accelerating the asset value of land. As a part of land tax reform, it is necessary to restore the 'balance of valuation' between land and non-land inherited assets. The third problem is the capital gains tax on the sale of land as a tax on land transfers. The current system of land transfer taxation is outlined in Table 13.4. There are three taxes levied on capital gains on land: the individual income tax, the corporate tax (both national taxes), and the inhabitants' tax (prefectural and municipal taxes for both individuals and corporations). The basic structure of capital gains tax on land is highly complicated, reflecting a number of stop-gap changes taken by the government in the past to achieve specific land policy targets. Essentially, taxation on capital gains from land sales depends on the length of the holding period of the transferred land. In the case of individuals, both short-term and long-term capital gains are taxed at specific flat rates, separate from other income. A heavier tax rate is levied on short-term capital gains from land held for 5 years or less than on long-term gains from land held for more than 5 years. Heavier taxation on short-term capital gains was implemented as a kind of penalty tax against speculative land transactions. When the gains fall on business 7 Japan's inheritance tax is a hybrid system of an estate tax and an inheritance tax, which is unique in comparison with other countries. See, for more detailed discussion, Ch. 12.
228
Corporate Taxation and Taxes on Capital
Table 13.4 Outline of land transfer taxation, 1990 Holding period (on 1 Jan. of the transfer year)
Capital gain
2 years or less
Over 2 years, 5 years or less
Separate taxation Tax amount is the larger of the following (a) gain X 40% (12%) (b) marginal tax amount as if treated as comprehensive taxation X 110%
Individual
Business income or miscellaneous income
Corporation
Separate taxation Same as above Tax amount is the larger of the following (a) gain X 50% (15%) (b) marginal tax amount as if treated as x 120% comprehensive taxation Gain is taxed at the rate of 30% in addition to ordinary corporation tax
Over 5 years Gain after special deduction is separately taxed by applying the following rate, 20% (6%) to the part of the gain below ¥40m. 25% (7.5%) to the part of the gain over ¥40m. Ordinary comprehensive taxation
Gain is taxed Ordinary at the rate of corporate 20% in addition taxation to ordinary corporation tax
Note: The rate in brackets shows the inhabitants' tax rate.
income, say in the case of a real estate agency, very short-term capital gains are subject to a larger penalty-type tax. As to corporations, additional taxation on both short-term and very short-term capital gains is applied at the rate of 30 per cent and 20 per cent respectively in addition to the ordinary corporate tax. On the other hand, long-term capital gains are aggregated with other income and the regular tax rate is applicable to the gains. The current system has developed through trial and error over the past two decades. This is particularly true in the case of individuals. Before 1968, capital gains on land were treated under the ordinary individual income tax code, although half the gains were taxed in favour of averaging irregular income. However, special
Land Tax Reform
229
provisions were introduced in the 1969 tax reform to promote the supply of land and to discourage speculation on land sales. Until 1975, the new tax rates were applied to capital gains on land, separate from other income; 10 per cent in 1969-71, 15 per cent in 1972-73, and 20 per cent in 1974-75 respectively on longterm capital gains, and 40 per cent on short-term gains. This provision of the 1969 tax reform had far-reaching effects on the transfer of landownership from individuals to corporations, but the total effect of promoting an increase in the land supply was not observed as markedly as had been expected. To make matters worse, the special treatment of capital gains on land substantially sacrified tax equity (see Chapter 6). To appease the critical view of inequitable taxation, the government in 1976 began to make modifications towards heavier tax burdens in the individual income tax for long-term capital gains on land. Since then, taxable long-term capital gains have been divided into two classes: above and below ¥40 million, applying different flat rates to each, as seen in Table 13.4. As a result, although some modifications have been made from time to time, flat but higher rates of tax have been applied to long-term capital gains. Nonetheless, long-term capital gains have been taxed more lightly than other ordinary income, such as earned income. By contrast, the 40 per cent penalty tax rate on short-term gains has remained unchanged since it was established in 1969. Similarly, for corporations, additional taxation on short-term capital gains on land has continued in the same way as when it was introduced in 1973. In 1987 a new scheme of very short-term gains held for less than 2 years with the tax rate of 30 per cent was installed. One of the most controversial issues in the corporate tax on capital gains is that deficit-operating corporations can be totally exempt from the tax payment, even if they made large capital gains from land sales. It has been noted that some corporations intentionally avoid their corporate tax burden by turning themselves into so-called deficit-operating businesses.8 In addition to the basic structure of capital gains tax on land for both individuals and corporations, tax relief proliferated over the past two decades. Particular attention should be paid to the following two points. First, there are special deductions and reduced rates concerning transfers of residential land and buildings by individuals or transfers of land for specific policy targets. For instance, when residential land or buildings owned by individuals for over 10 years are transferred, reduced tax rates are applicable to the gains. Also, special deductions are available to land transferred for expropriation. The second point is that the special rule for 'business assets roll-over', where certain land or buildings are sold in place of other land or buildings by the same person or firm, is allowed with some limitations. This roll-over 8
The ratio of deficit-operating corporations to total corporations in Japan exceeded more than 50 per cent for a long time in 1985 (see Ishi 1989fc, 160-1). From the conventional view of ongoing concerns, this seems suspicious.
230
Corporate Taxation and Taxes on Capital
rule was adopted to promote decentralization and urban renewal. Under the special rule, the capital gains tax on the old asset can be deferred for its payment allowing the new asset to carry over the basis of the old asset. The existing system of capital gains tax on land was obviously far from satisfactory from a standpoint of tax fairness, neutrality, and simplicity. Thus, the capital gains tax as well as the land-holding tax became important goals of land tax reform. Taxes on Agricultural Land Finally, the fourth problem of land taxation concerns the special treatment of taxes on agricultural land. Currently, agricultural land benefits greatly from preferential treatment of both its property tax and inheritance tax. Table 13.5 summarizes the past records of the property tax in Urbanization Promotion Areas, from which we can observe a de facto state of neglect concerning the tax burden on agricultural land. In principle, agricultural land within the Urbanization Promotion Area9 of the three Metropolitan areas is treated as residential land for property tax purposes. However, if farmland of more than 990m 2 was worked continuously over a 10-year period, the tax due on residential land as an excess tax amount over the tax calculated for agricultural land was deferred. Thus, during the relevant period, the tax was calculated on the basis of the assessment of agricultural land, whose burden is almost negligible.10 Furthermore, such a deferred tax amount was entirely exempted if the continuation of the agricultural operation was confirmed every 5 years. Likewise, the inheritance tax had a special tax deferment measure on agricultural land in general on the ground that the land should not be divided into small pieces in order to preserve the efficient operation of agriculture. Of course this measure was applicable only when the heir continued to be a farmer. The tax amount on the assessment value of the land above that of the invested capital for agriculture was deferred, and in addition the deferred amount was exempted if the heir continued to be a farmer for more than 20 years. Both the property tax and the inheritance tax deferment systems of payment were considered as key factors in preventing agricultural land, particularly in the urban y
In 1971 when the City Planning Law was established, Urbanization Promotion Areas were demarcated from Urbanization Control Areas, and were regarded as urban areas where even the existing agricultural land therein should be converted to residential use within a 10-year period. However, backed by vested farmer interest groups the government of Japan has continued to observe, until now, lenient treatment in favour of the land in question. 10 The MOF presented the following prototypical case of agricultural land taxation within the Tokyo Urbanization Promotion Area. Example, 3600 m2 of land site located in Setagaya-ward: (a) An official valuation price: ¥3.42bn.; (V) Property tax assessment: for residential land, ¥576m.; for agricultural land, ¥2.32m.; (c) Property tax: for residential land, ¥4.03m. (effective lax rate 0.1%); for agricultural land, ¥30000 (effective tax rate 0.0009%). The effective tax rate is calculated as the ratio of (c)l(a).
231
Land Tax Reform Table 13.5 1971-1991
The property tax on agricultural land in Urbanization Promotion Areas, 1971
Area
Nationwide
Type of farm
Type 'A farm (applicable after 1971) Type'B' farm (applicable after 1972)
Type 'C' farm (applicable after 1975)
1972
1973
1974
1975
1976-1981
1982-1991
Limited to the designated cities in the three metropolitan areas Not enforced
Many cities paid subsidies by their discretion Not en- Many forced cities paid subsidies by their discretion
The tax can be reduced by municipal ordinance to those farms that are actually cultivated and appropriate to be preserved as farms for more than 3 years Not enforced
© Taxed as residential land except for land with value less than 30 000 yen per 3.3 m2. @ Deferment of tax on farms continuing long-term agricultural operations was introduced, which applies to farms actually cultivated and appropriate to be preserved as farms for more than 10 years
Notes: 1. Type 'A' farm: farm with assessment value per 3.3 m2 in 1972 over average price or ¥50000. Type 'B' farm: farm with assessment value per 3.3 m2 in 1972 below and over average land price (except for farm less than ¥10000). Type 'C' farm: farm with assessment value per 3.3 m2 in 1972 below half of average land price or ¥10 000. 2. Designated cities in the three metropolitan areas (190 cities in 1988). a. Wards in Tokyo, and 105 cities in Ibaragi, Saitama, Tokyo, Chiba, and Kanagawa. b. 28 cities in Aichi and Mie. c. 56 cities in Kyoto, Osaka, Hyogo, and Nara.
areas, from facilitating the planned conversion to residential land. In fact, there are many cases in the suburbs of Tokyo where ordinary land is 'disguised' as agricultural land by the planting of trees, such as nut or persimmon. This attitude of farmers is motivated by speculation in anticipation of a future price hike on their own land.
232
Corporate Taxation and Taxes on Capital
Thus, in order to encourage the effective utilization of land as a means of solving the land problem, it is widely believed that the preferential treatment of taxes on agricultural land should be eliminated to a considerable extent.
A NEW
L A N D - H O L D I N G TAX AND
O T H E R TAX
REFORMS
Should Land-Holding be Burdened more? The members of the Subcommittee on Land Taxation all agreed from the start to increase the tax burden on land-holding. It was thought desirable to levy a heavier burden on land-holding based on its asset value mainly for two reasons: (1) to reduce the advantages of land as an asset and to secure equitable taxation on land-holding, and (2) to encourage the efficient use of land by increasing the cost of holding land. As regards the selection of policy measures for this purpose, there were basically two alternatives considered. They were: 1. The review and improvement of the existing tax system at the local government level. (a) The property tax (b) The special land-holding tax 2. The creation of a new land-holding tax at the national government level. (a) A tax on idle or underutilized land (b) A tax on unrealized capital gains on land (c) A general tax on land-holding The Subcommittee at an earlier stage of deliberation began with the examination of how well the existing tax system could be restructured in order to raise the tax burden on land-holdings. However, local authorities strongly maintained that the property tax should not be employed to levy a heavier burden on land-holdings as a means of reducing the asset value of land. This reflected their basic attitude of benefit taxation. They believe that the property tax should be collected by municipal governments to cover the cost of local public services to the inhabitants. Such a tax has no direct bearing upon the increased asset value of land caused by price hikes. As a consequence, a lower value of property tax assessment was rationalized on the ground that it should basically differ from either the market price or the official valuation price of land.11 Similarly, with regard to the special land-holding tax, it was pointed out that the government could not use this tax effectively in accordance with the initial objective 11 Another weakness of increasing the land assessment value for property tax is that it would produce a vast regional disparity of tax revenues between large cities like Tokyo, Osaka, or Nagoya, and depopulated areas like Hokkaido where land prices have actually dropped. Since the current intergovernmenta fiscal transfer system has no horizontal adjustment of equalization payments from rich to poor areas, there would be no means to restore such revenue disparity between the different areas.
Land Tax Reform
233
of the land-holding tax as mentioned above. In 1973, the government had introduced the special land-holding tax on idle and underutilized land for the particular policy goals of preventing speculative land transactions and promoting the efficient utilization of land. As time went by, however, the scope of tax-exempt exclusions substantially expanded, and the revenues continued to fall, despite the increasing trend of land prices. The most difficult aspect of the special land-holding tax was obviously the strict definition of 'idle or underutilized land'. Arbitrary judgement essentially intervened into the actual enforcement of this tax. Thus, the Subcommittee came to the conclusion that it would be impracticable to rely on such a specific policy-related tax to increase the cost of holding land in general. In the end, the first alternative for improving the existing tax system had to be abandoned. Turning to the creation of a new tax on land-holding, the Subcommittee had three choices, as listed above. As for the first category of tax on idle or underutilized land which was greatly supported by business groups, the same problem existed as with the special land-holding tax. No doubt, a technical difficulty emerges from setting objective standards for judging the degree of land utilization, chiefly because effective and clear-cut land-use planning is lacking at present. Therefore, the Subcommittee had to drop the option of the first type of land-holding tax at a relatively early stage in the discussion. Concerning the choice of the second alternative, particular attention was paid to the economic role of unrealized capital gains on land mainly owned by corporations. Such gains have expanded more rapidly than nominal GNP for the past 30 years, and the relative ratio of gains to GNP continued to rise from 45 per cent in 1960 to 115 per cent in 1988. It is often pointed out that corporations using the increased price of land greatly benefited from the advantages of higher equity-financing. The Subcommittee considered that a tax on unrealized capital gains on land would be one possible desirable land-holding tax, but it was not finally accepted because of one drawback, i.e. only land held for a longer time-period would be subject to tax while newly acquired land would be exempted from such a tax. Accordingly, the third category of a general tax on land-holding was selected by the Subcommittee on the grounds that it would be most suitable to levy the tax on land based on the asset value regardless of utilization, location, and so on. The tax was regarded as the best way of achieving the basic objective of ensuring equitable taxation on land-holdings and a reduction of the advantages of land as an asset. In the process of the Subcommittee discussions concerning the introduction of a new tax, a detailed scheme in quantative terms was not concretely prepared. The final decision on constructing a fundamental framework for the new tax was moved into the political arena of the LDP Tax Committee. The Subcommittee on Land Taxation merely presented a basic design for the new general land-holding tax as follows: 1. The tax should be a national tax levied on land, based on a uniform assessment of the asset value itself in accordance with the landowner's ability to pay.
234
Corporate Taxation and Taxes on Capital
2. For this purpose, the tax base should use the inheritance tax land valuation which is reassessed on a nation-wide basis every year. 3. The tax rate should be set at a level adequate to diminish the asset value of land and, at the same time, to enable people to carry on business continuously without any serious damage. 4. Considering the nature of levying a tax on the asset value, as a basic exemption a tax threshold should be introduced. 5. Some exclusions from taxation should be set to include residential land in principle, small-scale stores, government-owned land, and public-interest land used for hospitals, social welfare facilities, etc. 6. The number of taxpayers due to the tax threshold and exclusions mentioned above will be restricted mainly to corporations. Evidently, in view of its basic goals and nature the new tax is different from the property tax which, depending upon the principle of benefit taxation, collects revenues to cover the cost of local public services. Moreover, the scope of both taxes is quite different: the new tax is targeted on a specific type of valuable land while the property tax is uniformly imposed on the land of the entire country. Thus, one tax should basically be considered to be independent of the other.12 However, local authorities began to protest strongly at the introduction of a new national tax on land-holding, as the deliberation in the Subcommittee was approaching an end. Obviously, they were afraid that the existing property tax at the municipal governments' level might be replaced in the future by such a new tax as the land-holding tax. Thus, apart from their previous attitude, they agreed to increase the local property tax burden by raising the assessment value of land for property tax purposes. Also, they agreed that the special land-holding tax should be increased to promote the efficient use of idle and underutilized land by lowering the tax threshold or extending the scope of taxable areas. In any event, the resolution of this conflict between the national and local governments was transferred to the political arena of the LDP Tax Committee. OTHER LAND TAX REFORMS
Compared with the heated debate on land-holding tax, other tax reforms were concluded in the Subcommittee without serious difficulty or conflict. With regard to inheritance tax reform, it was decided that the assessment value of land should be raised to correct the imbalance between land and non-land inherited assets. As noted earlier, the lower valuation of land for the inheritance tax has clearly 12 A two-tier taxation on land-holding exists in the Australian tax system; one is the land value tax set by the state government while the other is rates, similar to the Japanese property tax, set by the municipal government. The justification for levying the same land-holding tax at both upper and lower levels of government is the same for both countries. See New South Wales Tax Task Force (1988), Walsh (1990), Land Tax Review Group (1990).
Land Tax Reform
235
precipitated the rising trend of land prices. It is argued that the inheritance tax assessment on land should be raised to 80 per cent of the official valuation price from the current 60-70 per cent. On the other hand, since this produces a substantial increase in inheritance tax revenue, it is necessary to reduce the tax burden by widening the tax brackets of the steep progressive rates and lifting the tax threshold. Taxation of capital gains on land was proposed in order to effect drastic changes in the current tax system. As indicated by the Basic Land Law, land whose value tends to be augmented by the effects of the external economy, such as road construction, transportation, etc., has the basic feature of serving the public welfare. Thus, it is equitable to levy a heavier burden on gains from the sales of land than on other income. The increase of such a tax is expected to restrain speculative and transactions and in turn put downward pressure on land prices. Traditionally, lessening the tax burden on capital gains has predominantly found support among many economists, as a means to promote an increase in the land supply. This had generally been believed since the 1969 tax reform in which capital gains on land were taxed at a low flat rate, separate from other income. As the Subcommittee properly pointed out, however, such treatment would impair tax fairness among taxpayers and furthermore its policy effect would be very questionable. On the contrary, the lenient capital gains tax seems to be counter-productive, reflecting the fact that the demand for land rose and the advantages of land as an investment asset advanced due to generous treatment of capital gains in the past. Previous reforms of the capital gains tax made it appear likely that there would be some easing of the tax burden in the coming years. Thus, landowners tended to hold off sales of land, given the changeable state of the tax system. After reconsidering this former policy stance, the Subcommittee proposed to increase the tax burden on capital gains for individuals and corporations. As regards gains from land transfers by individuals, from a tax equity point of view the rate applied to long-term capital gains on land should be raised to be comparable with the rate for earned income. On the other hand, to check speculative transactions the tax rate for short-term gains should remain the same. Following the same reasoning, it is appropriate to levy a heavier tax burden on corporate long-term gains while the present penalty-like taxation is maintained for both very short-term and short-term gains from land transfers. As mentioned above, there were arguments that corporate capital gains on land sales should be taxed perfectly separately from other income, in order to prevent deficit-operating businesses from avoiding their tax payments. But the Subcommittee's view was inconclusive, mainly because the deficit-operating corporations in question could not reasonably be distinguished from ordinary corporations that are forced to sell land. In addition, the Subcommittee proposed to curtail the level of special deductions in favour of capital gains on land, since a large part of gains tax is eroded by such special tax measures. This is important in making the tax burden more equitable and acceptable. Also, it was proposed to substantially eliminate the business asset roll-over scheme. The roll-over from long-term land-holdings to depreciable assets
236
Corporate Taxation and Taxes on Capital
has been criticized as stimulating speculative purchases of flats in the central part of the big cities. This, in turn, is detrimental to solving the land problem. Before any of the aforementioned discussion in the Subcommittee, the thorough amendment of agricultural land tax was largely concluded in accordance with the 'Comprehensive Land Policy Plan' proposed by the Japanese government during the US—Japan SII talks. The Subcommittee decided to take measures to solve the very political issue that had been left pending for a considerable time. Of these, it was recommended that the line between agricultural land within Urbanization Promotion Areas should basically be redrawn distinguishing (1) 'agricultural land to be preserved' from (2) 'agricultural land to be converted into residential use'. The former should be taxed as agricultural, not residential land to strengthen the measures to ensure its preservation. Of course, such demarcation should clearly be defined by city planning and the conversion into other land-uses be legally restricted. The inheritance tax deferment system should also be applicable to 'agricultural land to be preserved', given both strict demarcation and conversion restrictions. It was necessary as well to tighten the eligibility requirements of 'continuous agriculture activity' which is a prerequisite for this preferential treatment. On the other hand, land to be converted should be taxed as if it were residential land, implying that the property tax on such land would be increased sharply, forcing 'farmers' to discontinue agricultural activities within Urbanization Promotion Areas. Moreover, from the standpoint of tax equity, it was more important to eliminate the deferred inheritance tax payment exemption that is allowed on land used for agriculture for over 20 years. Thus, if these reform proposals were carried out, they would be of great help in increasing the supply of land and dispelling the sense of unfairness among taxpayers. Farmers who had no intention of cultivating their land for agriculture would be persuaded to part with their land-holdings. By contrast, those who chose to work their agricultural land on a permanent basis would find it very difficult to convert it to residential land. When they chose to quit their agricultural activity, they would be requested to sell their land to the local authority at a 'reasonable' price.
SUMMARY OF THE LAND TAX REFORM PROPOSED BY THE
GOVERNMENT
Seven Major Points In accordance with the conventional tax process in Japan,13 the final detailed terms of the land tax reform were decided in December 1990 by the LDP Tax Committee, as a result of political compromises due to the pressure of vested interest groups. The Tax Advisory Commission merely proposed, in October 1990, the basic direction for 13 Concerning the relationship between the Tax Advisory Commission and the LDP Tax Committee within the tax process, see Ch. 1.
Land Tax Reform
237
land tax reform without referring to either the quantitative magnitude of tax rates, the exemption levels, or other specific components. As usual, the Commission's proposals were forwarded to the LDP Tax Committee, and in turn the Japanese government submitted the necessary land tax reform legislation to the Diet in February 1991 after receiving the approval of the ruling political party. The seven major points of land tax reform, which were formulated clearly in terms of the Tax Bill and have now come into effect, are summarized below. 1. Introduction of a National Tax on Land-Holding called the Land Value Tax A new land-holding tax as the land value tax to be levied on the ownership or the leasehold of lands throughout the entire country was introduced by the national government. Both individuals and corporations would have to pay the tax based on a uniform level of assessment corresponding to the inheritance tax land value. Several exemptions were allowed as follows: land held by the government and public corporations, land used directly for purposes closely tied with public welfare (e.g. hospital, railways, social welfare facilities, etc.), agricultural and forest land, and land with an assessment value below ¥30 000 per m2. Moreover, residential sites up to 1 000 m2 (only primary dwelling) and buildings rented for residential use (excluding company houses for board directors) were also exempted. In addition to these exemptions, the larger of the following amount was allowed as a basic deduction: (1) ¥1 billion (¥1.5 billion for individuals and small and medium-sized firms), or (2) ¥30 000 multiplied by the area in square metres of land held, excluding the exemptions listed above. A tax rate of 0.3 per cent (only 0.2 per cent in 1992) was applied to the total assessment value of all land held (except exemptions) minus the basic deductions by a taxpayer on 1 January each year. The tax amount could be deducted as an expense when computing the taxable income for the individual income tax (in the case of business income) and the corporate tax. The date of enforcement was 1 January 1992, and the new land value tax would be reviewed at least every 5 years in connection with the possible increased level of land assessment as noted below. 2. Promotion of Increasing the Property Tax Land Valuation To remedy the extremely low land valuation of the current property tax, land valuation increase would begin the periodic revaluation from fiscal year 1994. However, the planned revaluation for fiscal year 1991 was implemented as usual. The basic aim of the new revaluation was to narrow the gap to within a certain margin of the official valuation price.14 However, a specific relief programme would be required to ease the effect of the sudden tax increase on individual residential land. Representative assessment points for land valuation (i.e. the Road Rating Assessments) were publicized frequently in fiscal year 1991 in order to make clear to 14 Implicitly, it is expected that the revaluation of land for the property tax assessment should reach a target of approximately 70 per cent of the official valuation price.
238
Corporate Taxation and Taxes on Capital
taxpayers the assessment of the land values. In addition, the number of publicized assessment points were to be increased systematically from fiscal year 1994. To mitigate tax increases arising from increased land assessment, a reduction in the inhabitants' tax was to be put into effect. 3. Improvement of the Special Land-Holding Tax The special land-holding tax and the property tax were generally increased to promote the efficient utilization of land and to prevent speculative land transactions in two ways. First, the tax was levied on idle land in excess of 1 000 m 2 within areas designated in city planning as a 'District for Identifying and Promoting the Utilization of Idle Land'. The tax rate was 1.4 per cent, applied to the land value in terms of the current price or acquisition price, whichever is higher. Second, the scope of the special land-holding tax was extended, lowering the minimum threshold, and excluding open-space parking-lots, outdoor athletic facilities, etc., from exemption. Furthermore, the exemption would be eliminated for lands within Urbanization Promotion Areas acquired on or after April 1982 and held for over 10 years. 4. Increased Land Valuation for the Inheritance Tax The land valuation for the inheritance tax was raised in 1992 in order to weaken the preference for land as an asset. The target rale was approximately 80 per cent of the official valuation price, compared with the rate of 60-70 per cent in the past. At the same time, the annual valuation date was changed from the present date of 1 July each year to 1 January for the official valuation. To compensate for the increased tax burden that accompanied the higher valuation of land, an inheritance tax reduction would be implemented by either reducing progressive tax rates or raising the tax threshold in fiscal year 1992. 5. Tax Increases on Capital Gains for Individuals and Corporations The individual income tax rates for long-term capital gains in cases of the transfer of land or buildings owned by individuals for longer than 5 years were raised from 20 per cent (6 per cent) below ¥40 million and 25 per cent (7.5 per cent) above ¥40 million to flat rates of 30 per cent (9 per cent) on a uniform basis (percentage figures in parentheses are the inhabitants' tax rates). Basically, this change aimed at restoring a tax-equity balance with earned income, while diminishing the preference for land. On the other hand, tax rates for short-term capital gains remained unchanged. Likewise, the corporate tax on capital gains from land sales was substantially changed from the former system. Very short-term capital gains on land held no longer than 2 years were taxed separately from other income at the rate of 30 per cent in addition to the basic rate, and even deficit-operating firms had to pay taxes under this method of separate taxation. This aimed to reduce the tax avoidance by firms 'deemed' as deficit-operations.
Land Tax Reform
239
As for long-term capital gains on land held longer than 5 years in addition to the application of ordinary corporate taxes, a rate of 10 per cent was levied. Such a new tax was also applicable to deficit-operating firms in an effort to curtail tax avoidance. These changes in the capital gains tax on land aimed at reducing the land asset value and at stopping speculative transactions. 6. Other Changes Related to the Capital Gains Tax on Land Contrary to the proposals of the Tax Advisory Commission, the Land Reform Bill contained certain preferential treatments towards capital gains tax on land, as a means to promote an increase in the land supply. One reason behind this was the 'lock-in effects' feared by the government which have resulted from a heavier tax burden on capital gains. The scope of roll-over relief was also expanded in cases of industrial relocation and decentralization. When transferred land was used for this purpose, the roll-over of business assets was allowed at a deferment ratio of 90 per cent instead of the ordinary 80 per cent ratio. Moreover, the special deduction (¥50 million) for capital gains on expropriated land was perpetuated, apart from the present temporary measures. The same treatment was allowed for transfers of agricultural land to residential and commercial uses. Special reduced tax rate measures were applied to further encourage an increase in the land supply for residential use. To begin with, when individuals transferred land owned for longer than 5 years to the national or local governments for development projects, the flat tax rate on capital gains was reduced to 15 per cent (5 per cent) from the original level of 20 per cent (6 per cent). When residential land or buildings owned by individuals for longer than 10 years were transferred, a more preferential reduced tax rate on capital gains was applied as follows: where the present tax rates were 10 per cent (4 per cent) below ¥40 million and 15 per cent (5 per cent) above ¥40 million, the new rate structure was set for gains below or above ¥60 million. 7. Improvement of Taxation on Agricultural Land within Urbanization Promotion Areas The special treatment of agricultural land in both the inheritance tax and the property tax were abolished from the standpoint of tax equity and of promoting an increase in the land supply. For this purpose, not later than the end of 1992, agricultural land to be preserved within Urbanization Promotion Areas in the three Metropolitan areas had to be designated in city planning under the Productive Green Tract Area System. At the same time, the conversion of this agricultural land into other uses was restricted by law. In such cases, the deferment system of the inheritance tax was not applied to agricultural land that was not designated as a Productive Green Tract Area, although some transitional measures were available. By contrast, agricultural land to be preserved was eligible for this deferment system, but the special treatment of exempting
240
Corporate Taxation and Taxes on Capital
the deferred tax amount on agricultural activity which had continued for 20 years was eliminated. The system for checking the continuity of agricultural activity became more stringent. Second, agricultural land that was not designated as a Productive Green Tract Area was treated as residential land, and taxed as residential land for property tax purposes. Thus, the special property tax reduction measure covering agricultural land suitable for long-term continuous agricultural operation was abolished.
AP p R A I S A L Since the LDP proposed its reform package on land taxation in December 1990 which automatically became the government's tax bill, there have been a number of criticisms particularly from economists. Most of the criticism focuses on the new land-holding tax (i.e. the land value tax), because the new tax compared with the original proposal of the Tax Advisory Commission was quite watered down. On the other hand, the other land tax reform packages, such as the inheritance tax, capital gains tax on land, etc., were kept more or less similar to the Commission's proposals. Thus, attention should be paid to comments concerning the LDP's land value tax. Stated simply, the new land-holding tax was emasculated by many modifications that were made for political reasons by the LDP Tax Committee. The original plan, embraced by the members of the Subcommittee on Land Taxation, had sought a tax rate of 0.5-1.0 per cent with a tax threshold of ¥100 million. According to the initial framework, this new land tax could collect about ¥1 000 billion from 300 000 individuals and corporations. The LDP's new tax with an initial tax rate of 0.2 per cent (0.3 per cent after the first year) and much broader exemptions applied mainly to corporations. Residential land 1 000 m2 or less, owned by individuals, was excluded from the new landholding tax. The new tax does not apply to land valued below ¥30000 per m2. Corporations capitalized at ¥100 million or less with land-holdings valued at a minimum of ¥1.5 billion are subject to tax, as are larger corporations with land valued at ¥1 billion. Under the LDP plan, the tax rate was much lower and the exemption level much higher than had originally been planned. While the LDP plan was under deliberation, chairman M. Shiokawa suggested levying the new tax at an annual rate of 0.5 per cent on land valued at ¥500 million or higher, regardless of the size of the business. This idea might have been acceptable, given the initial difficulty of introducing such a new tax. However, even that plan drew immediate criticism from many LDP members who said the tax burden would be too heavy for small and medium-sized businesses. Finally, the LDP members, backed by several strong business lobbies, pressed for the abandonment of the chairman's proposal and for a reduction in the minimum level of the new tax.
Land Tax Reform
241
The new land-holding tax is so light and includes such broad exemptions that economists say it is not likely have much effect on resolving the land price hike problem in Japan. As a result of political modifications, the new tax will apply to only 50 000 corporate and other larger landowners, producing only ¥300 billion to ¥400 billion in tax revenue annually at the 0.3 per cent rate. In fact, it is said that the new tax comes as a relief to some in the financial and real-estate markets. Many doubted the LDP's land-holding tax would be effective in raising land ownership costs and discouraging speculation. In particular, the opposition parties called for stronger measures to cope with land problems. Needless to say, the new land-holding tax is therefore far from satisfactory. Although it is acceptable as a first step, further reform will certainly be necessary in the future. RECENT REFORMS TOWARDS MITIGATING LAND TAX BURDEN
Enforcement and Repeal of Land Value Tax The land value tax was introduced into the national tax system in April 1992, based on the movement of land-tax reform starting from 1989. The main aim of such a new tax was directed to bringing down the overheated level of land price triggered by bubble phenomenon. The coverage of land-owners subject to land value tax was much smaller than it was originally planned, because of both a wider basic deduction and several exemptions. In fact, the total number of taxpayers on the basis of tax returns amounted to 40532 in 1992, consisting of 28455 corporations and 12 077 individuals. Originally it was planned to be approximately 300 000. After the introduction of the land value tax in 1992, its revenue continued to drop until its year of repeal in 1998. As shown in Figure 13.2, the land value tax started at the tax rate of 0.2 per cent, not 0.3 per cent, producing ¥520.1 billion. Special relief was applied for the first-year tax rate to initiate a new tax smoothly. No doubt, this was taken as a result of political compromises among politicians and business groups, and the MOP. Tax revenue has constantly decreased from the ¥780.2 billion at the tax rate of 0.3 per cent in 1992 to ¥160.1 billion in 1997. Meanwhile the tax rate was lowered to 0.15 per cent in 1996. Such a tempo of tax decrease at the fixed tax rate of 0.3 per cent was obviously caused by the continuous fall of land prices, as indicated in terms of the land price index in Figure 13.2. Land prices were more than halved during the period of 1992-97. It is not clear whether the adoption of land value tax could lower the land price level or not during the period in question. Although a causal relation between two factors cannot be determined in practice, it would be reasonable to conjecture that the land value tax has not been related to the downward movement of land prices in recent years, merely because the absolute amount of tax revenue was too small to influence the land transaction in the market. In parallel with the decrease of tax revenue, the number of taxpayers also continued to decline to 21716 in 1996 (i.e. 18 138 corporations and 3 578 individuals).
242
Fig. 13.2
Corporate Taxation and Taxes on Capital
Movement of land tax revenue and land price
From the beginning, the land value tax was substantially criticized by businessmen, particularly those in land-intensive industries, such as department stores, real estate, etc. They had a very big complaint about the newly increased land tax burden, and stressed the specific distorted effects of a new land-holding tax on land transaction and business management. However, the quantity of land transactions in the market has not fallen since 1992 when the land value tax was established. As a matter of fact, land transactions continued to rise from 310539 in 1992 to 417 593 in 1996, in terms of the annual registry necessary to sell and purchase land. It is important to note to what extent companies have been burdened by the land value tax. With this aim, the tax burden ratio is calculated as that of land value tax revenue to average current profits15 for the top 100 companies of tax returns. In 1996, the relevant ratios were as follows. As far as these data are concerned, the majority of leading companies were within the range of 0-5 per cent. Thus, it does not appear that the land value tax has greatly overburdened and damaged the business management of many corporations. (See Table 13.6.) As the Japanese economy plunged into prolonged recession after the collapse of the bubble, the repeal of land value tax began to be maintained politically in the context of fiscal stimulus measures (see, for more detailed discussion, Ishi 2000,
13 Current profits include land value tax revenue, and are averaged for the past ten years. Data are available from the Tax Advisory Commission.
Land Tax Reform Table 13.6
243
The burden ratio of land value tax
The tax burden ratio (%)
The number of companies
0-5 5-10 10-15 20-25 Others
86 5 ] 1 7
ch. 6; Bayoumi and Collyns 2000). In order to restore the depressed state of the economy, the continuous drop in land prices needed to be stopped. If not, non-performing assets in the financial sector would increase, reflecting the fall in land prices. Although it was quite uncertain to judge how effective the suspension of land value tax would be, the government finally decided to freeze the collection of such a tax from April 1998 for a while. This treatment is considered to be de facto abolished politically. Changing Capital Gains Tax for Land Sales In parallel with the adoption of land value tax in 1992, great stress was placed on the reinforcement of capital gains tax for land sales. In a situation of rising land prices, tax reduction on capital gains would normally be undertaken to encourage land transactions in the market and in turn to bring down the overheated price level. In theory, this would be correct. However, in the process of land tax reform at the Tax Advisory Commission when the bubble phenomenon was prevailing, it was asserted that capital gains taxes on land sales ought to be raised to some extent in order to curtail land value in the transaction, because a lower tax burden on capital gains was considered as being of great help to increase land value. Capital gains taxes are applied for the land transactions of both individuals and corporations. Figure 13.3 depicts the case of individual income tax including local inhabitants' tax in which the realized capital gains from land continuously owned for over five years are levied at flat rates, separate from other incomes. Before 1989, the tax rates applicable to taxable capital gains (i.e. gross gains — special deduction) were generally 26 per cent for ¥40 billion or less, and 32.5 per cent when exceeding ¥40 million. For 1992-94, they were uniformly applied at 39 per cent after eliminating the demarcation of two income portions. This was obviously undertaken to strengthen the tax burden on capital gains. Thereafter, as is shown in Figure 13.3, capital gains taxes have begun to be lowered, reflecting the sharp drop in land prices and the slump in land transactions. In 1995, the ¥40 million demarcation was revived with a new zone of 32.5 per cent, and after 1996 the movement towards mitigating the tax burden was furthered.
244
Fig. 13.3
Corporate Taxation and Taxes on Capital
Long-term gains taxes of individuals for land sales after 1989
Note: Special type case indicates the so-called 'desirable' transfer of land to the government for public use from the landlords. In this case, generous tax treatment is given to them. Taxes included are both the national income tax and local inhabitants' tax. Source: Data from the Tax Advisory Commission.
By contrast, in the special case in which governments, national or local, purchase land for public use from the landlords, lower tax rates were applied to realize capital gains. On the other hand, short-term capital gains from land owned for less than five years have constantly been taxed more heavily to curb speculative transactions in land. In addition, super-short-term capital gains from the sale of land owned by real estate brokers for not more than two years have occasionally been taxed more heavily. When corporations earn capital gains from the sale of land, the corporate tax is levied under the special separation scheme. In Figure 13.4, the changing patterns of corporate capital gains taxes are summarized after 1982. Depending upon the holding period of land, the tax rates applicable to capital gains are varied. The special tax rate to be added over and above the ordinary corporate tax rate is shown, except one separate taxation case of ultra-short-term capital gains between lanuary 1992 and January 1996. After January 1987 when the land price hike became conspicuous due to the 'bubble', capital gains were heavily taxed under the special additional or separate tax rates in view of the ultra-short-term holding period. In particular, the separate tax rate of 30 per cent described above is a very heavy tax burden on ultrashort-term gains aimed at preventing transactions from speculative motives. Even deficit-operating corporations should pay such a tax.
Land Tax Reform
Fig. 13.4
245
Corporate capital gains taxes for the sales of land after 1982
Note'. Each percentage indicates special tax rates to be added over and above the ordinary corporate tax rate, but * is the separate tax rate applicable to taxable capital gains of all corporations including deficitoperating ones. Source; The same as Figure. 13.3.
After January 1996 the heavy taxation on corporate capital gains began to be mitigated, reflecting the declining trend in land prices. Business groups positively insisted of a reduction of the heavier tax burden on their capital gains, pointing out that such taxation should be removed quickly to restore vigorous transactions in the land market. It is quite uncertain whether or not the frequent tax changes on capital gains have brought about a desirable direction of shifting land prices during relevant periods since the late 1980s. The Property Tax as a Land Tax In line with the adoption of land value tax, property land assessment began to be raised along with increasing the land-holding tax. As noted before, the new scheme of land assessment for the property tax was introduced by the MOHA from fiscal 1994, fixing 70 per cent of official land price as the taxable land value in order to narrow the gap between two assessments. However, a sudden hike of land assessment for the property tax needed adjustment to mitigate the jump in the tax burden. In Figure 13.5, a schematic chart is drawn to show the increased property tax burden in relation to (1) the official valuation of land, (2) property tax assessment, and (3) taxable land assessment. A sharp rise in official evaluation of land can be observed from 1982 to 1991 until the bubble phenomena were stifled, while the taxable land assessment was moving steadily keeping at a much lower level. The lower
246
Fig. 13.5
Corporate Taxation and Taxes on Capital
A schematic chart of property tax assessment
and steady movement of taxable land assessment along the upward baseline was intentionally brought about by the underestimation method applicable to modified ratios in order to avert sudden tax increases. However, since 1997 the property tax assessment has automatically been linked with 70 per cent of official land assessment to strengthen the function of the land-holding tax. As a result, the 1994 land assessment for the property tax should have been fixed at the B point where the official assessment of land—the A point— on 1 January 1993 (i.e. the basic date for the 1994 assessment) was multiplied by 0.7 (i.e. A X 0.7). If so, a very big gap between the B point and the upward baseline of taxable land assessment would suddenly come out. Theoretically, the B point ought to be taken as the taxable land assessment in accordance with the MOHA commitment for 1994—97, but in practice the baseline position during the period in question was chosen to avoid a sudden tax increase due to the new assessment method. Given the declining trend of land prices as shown in Figure 13.5, the property tax assessment will be approaching the taxable land assessment before long, but the D point that was adopted for 1997-2000 did not coincide with the baseline for tax purposes.
Land Tax Reform
247
The property tax has practically increased along with the uptrend of taxable land assessment from 1994 to today while land prices have continued to fall considerably. Taxpayers who have witnessed the drop in land prices in the market have expected lowered tax payments, too, but they have been constantly disappointed and have legitimate grounds for complaint.
This page intentionally left blank
Part IV
Indirect Tax System
This page intentionally left blank
14
The Traditional Framework of Indirect Taxes Nowhere in the world were indirect taxes as unimportant as in Japan in the past. Throughout the postwar period, the importance of indirect taxes as a source of revenue declined markedly in comparison with that of direct taxes. In recent years, however, there has been great interest in improving the whole system of indirect taxation, especially by means of the introduction of a broad-based consumption tax. This chapter deals with matters that were very relevant to tax reform movements in the late 1980s, and is devoted to a detailed analysis of the structure of the prevalent indirect tax system mainly during the period before the introduction of VAT in April 1989. Needless to say, the adoption of VAT had a close bearing upon the changes in excise taxes themselves. The discussion is intended to prepare the way for a consideration of the tax reform debate in the next chapter. There are two main points to be covered. One is an outline of the indirect tax system and a consideration of some of its characteristics in international comparison. The other is a clarification of the fundamental framework of major excise taxes, such as alcoholic beverages, tobacco, petrol, and commodity taxes.
J A P A N E S E INDIRECT TAXES
Types
It is first necessary to investigate briefly what is meant by 'indirect taxation'. There are several criteria for classifying taxes as direct or indirect, but the conventional distinction is based upon what is called the 'incidence' of taxation. Despite a controversial debate over this classification, we shall follow the conventional definition for its simplicity. Under this criterion, indirect taxes are initially imposed on taxpayers who are expected to shift the burden on to others. Direct taxes, on the other hand, are paid by those who are intended to bear the ultimate tax burden. Thus, indirect taxes can be defined as a form of consumption taxes.1 Broadly speaking, there are two types of consumption taxes: general and specific. An example of a general consumption tax is the general sales tax, which in turn can be divided into single-stage or multi-stage taxes. The single-stage tax comprises three major types: manufacturers', wholesalers', and retailers' sales taxes. Each is 1 An expenditure tax is a form of consumption tax, but it cannot be regarded as indirect taxation. Thus, I omit it in the following discussion.
252
Indirect Tax System
levied on the sales of goods and services at each stage of production and distribution. Sales taxes are collected from the seller and shifted forward on to the consumers. They are still used extensively in several countries; for instance, at the retailers' level in the USA and at the wholesalers' level in Australia. Principal types of multi-stage sales taxes are the value added tax and the turnover tax. The value added tax is a relative newcomer, but is now widely used all over the world. It is levied on the difference between what a firm sells and what it purchases; i.e., on the net value added by each firm. In contrast, the turnover tax is levied on the value of the transaction when one firm turns over a commodity to another firm at each stage in the economy. Unlike the value added tax, the turnover tax is not used in any country nowadays, mainly because it is an old-fashioned type of consumption tax. The turnover tax is generally criticized for accumulating the tax amount from one stage to another through pyramiding. Another form of consumption tax is the excise tax, which is selectively levied on the sale of a particular commodity or group of goods and services. Selective excise taxes commonly used throughout the world include those placed on alcoholic beverages, tobacco, petrol, and consumer durable goods. Customs duties are also included in this category, but they are not discussed here because the role of customs duties is not regarded as a revenue-raising one. In Japan, there was no general consumption or sales tax until March 1989. This is the country's most prominent feature in comparison with the indirect tax systems used in other advanced countries. Thus, indirect taxes were collected solely as selective excise taxes, and this spawned a variety of issues in the context of necessary tax reform in the late 1980s. Excise Taxation Let us focus on excise taxes exclusively here, because the consumption tax (Japan's VAT) will be discussed later (see Chapter 15). During the postwar period, the Japanese indirect tax system developed exclusively on the use of excise taxes on selective goods and services. Both national and local governments levy excise taxes on particular commodities, but here we shall consider only the structure of the national excise system. (See Chapter 18 for a discussion of local indirect taxes.) In Japan, excise taxes are conventionally classified into four major groups by the Ministry of Finance. Although this classification is not of great use from an analytical point of view,2 we shall follow the conventional grouping (see Table 14.1). There are a couple of points worth noting to clarify some features unique to excise taxation in Japan before introducing VAT in 1989. To begin with, taxes on alcoholic beverages and tobacco have traditionally not been conceived in relation to
1
As a criterion for classifying excise taxes for analytical purposes, greater importance should be attached to aspects of negative externalities, sumptuary target, pollution, luxury products, etc.
Traditional Framework of Indirect Taxes
253
Table 14.1 Percentage distribution of national indirect taxes, selected fiscal years (%)
1. Excises on alcohol, tobacco and
sugar (a) Alcoholic beverages (b) Tobacco (c) Sugar 2. Selective taxes on goods and services (a) Commodity tax (b) Playing-cards tax (c) Admission tax (d) Travel tax 3. Taxes on stamp, security transactions and bourse (a) Stamp tax (b) Securities transaction tax (c) Bourse tax 4. Earmarked taxes (a) Petrol tax (b) Local road tax (c) Liquefied petroleum gas tax (d) Motor vehicle tonnage tax (e) Aviation fuel tax (f) Promotion of power- resources development tax (g) Petroleum tax 5. VAT 6. Others'3
1934-6
1960
1975
1985
1990
1999C
62.6
51.4
29.0
26.9
17.8
13.7
27.0 25.3 10.3
30.2
20.5 7.5 1.0
18.0 8.5 0.4
5.4 5.4a — — — 12.0
12.5 10.1 — 2.0 0.4
16.3 15.4 — — 0.8
15.1 14.3 — — 0.7
7.6
12.7
19.8
11.7 6.0 — — — — — — 16.3
9.4 4.3 — — — — — —7.2
10.4 —
6.2 1.3 — 14.8 12.5
10.7
13.2 6.2 —
11.5
1.6 0.3
7.2 — —
2.2 — — — —
30.3 18.6 3.3 0.8 6.6 0.3 0.7
30.9 15.7 2.9 0.3 5.5 0.6 2.2
20.6 9.1 2.2 0.1 4.0 1.8
19.8 12.9 1.4 0.1 1.3 0.1 1.7
—
—
3.7
3.0
2.3
1.7 — — — — — — — —
17.7
3.5
4.5
0.3
0.4
—
—
—
—
28.0
49.3
20.0
13.6
11.7
7.3
17.3
9.9
* Textile excise tax plus taxes on soft drinks. Include customs, crude oil tax, etc. c Preliminary. Source: Data submitted to the Tax Advisory Commission.
b
negative externalities. In the USA and Western Europe, it is generally accepted that alcohol and tobacco taxes may be justified to control the consumption of items that are considered unhealthy (see Cnossen 1986; Shoup 1983). In Japan, however, the idea of external diseconomies associated with the consumption of alcohol and tobacco has not received wide attention (although current trends are pointing in the opposite direction). Thus, taxes on alcoholic beverages and tobacco, combined with that on sugar, are defined as taxes on non-essential or luxury items considered proxies for taxpaying capacity. They have long been recognized as being chiefly for revenue purposes.
254
Indirect Tax System
Second, the commodity tax is typically given as an example of an excise on luxury consumption, but it is grouped into selective taxes on goods and services with other unrelated taxes. It is noted that admission and travel taxes are two of very few examples of taxes on the sale of services, although their revenues are minor. The third category consists of three taxes levied on stamp, security transactions, and bourse. Lastly, excises in motoring and related fields are earmarked for specific appropriations of government-provided services. These taxes were designed as service charges for the use of road, airport, 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. Obviously, the relative weight placed on excise taxes has changed substantially in the past several decades. In the prewar period, the dominant share of excise tax revenue was occupied by taxes on alcoholic beverages, tobacco, and sugar. Other taxes were almost negligible in significance except for the stamp tax. In particular, there were no earmarked taxes such as road user charges. Turning to the postwar excise system, we find several interesting facts which clarify the unique structure of excise taxation in Japan. 1. The two 'traditional excises', alcohol and tobacco taxes, have shown a long-run declining trend. Moreover, the sugar excise yields very little revenue. Since its role as a sumptuary tax has disappeared today, the sugar tax should be eliminated from excise taxation. 2. Excises in the motoring field have greatly expanded and reached the top rank in 1975 and 1985. In 1988, however, revenues from these excises fell, mainly reflecting a lower import price of crude oil caused by the sharp rise in the value of the yen. 3. Stamp and security transaction taxes have steadily increased their relative importance in the excise system. In particular, enormous increases of their revenues were incurred in 1988 because of abnormal price rises in stock and land markets in the state of'bubble economy', although this phenomenon may shortly end. To sum up, the Japanese indirect tax system before fiscal 1988 is explained by two factors: (1) there is no general consumption tax, and (2) excise taxes are imposed on a narrowly restricted range of commodities and nearly exempt services from the revenue base. Since the Japanese indirect tax was made up of selective excise taxes, it is necessary to study in detail the major component of excise taxation. We shall discuss the four major excises in the remaining part of this chapter. In Table 14.1, the data for the two years 1990 and 1999 are added to those before the adoption of VAT. After VAT was created, the relative reliance on excise taxes was obviously diminished to a considerable extent, although total revenues from indirect taxes were virtually increased. It is noticeable to find that selective taxes on goods and services in the light of the commodity tax disappeared as a result of being integrated into the VAT. In addition, both the security transaction tax and the bourse tax were repealed from fiscal 1999.
Traditional Framework of Indirect Taxes
255
MAJOR EXCISE TAXES ON GOODS AND SERVICES
Alcoholic Beverages
The tax on alcoholic beverages has developed as one of the most important in the national tax system since the Meiji Era. In fact, around the turn of this century, it produced more than one-third of total national tax revenues. Initially, it must have been designed for revenue purposes, because alchoholic beverages can be regarded as one of the best fields for the imposition of excises. The basic requirements for levying excises on a particular commodity for revenue purposes are large sales volume, inelastic demand, ready definability, and no close substitutes. Alcoholic beverages satisfy all these requirements. However, as economic development progresses, the case for the alcohol tax levied primarily for revenue purposes gradually becomes less clear. Thus a new condition, such as its role as a sumptuary tax, is emphasized in order to justify its use. For historical reasons, the alcohol tax established very efficient administrative procedures and operations. Since the early years of the Meiji government, the licence system was firmly adopted for the production or sale of alcoholic beverages, for two reasons. One was to secure the alcohol tax revenues, and the other was to guarantee their quality. Producers or sellers were required to obtain a licence for each kind of alcoholic beverage and for each production or sale premises from tax offices. At the end of fiscal year 1997, the number of licensed manufacturers amounted to 5245, and of licensed liquor sellers 180030 (see MOF Tax Bureau 1997,181). The alcohol tax is levied on domestic alcoholic beverages shipped from manufacturing premises and on imported ones which are drawn from bonded areas. It is a typical case of a manufacturers' 'excise', which is assumed to be shifted forward on to consumers. Alcoholic beverages were classified into ten categories,3 for tax purposes. In 1988, a specific rate was applied to each type, but if the price of some (imported whisky, wine, etc.) exceeded a certain amount (i.e. a maximum non-taxable price for an ad valorem tax), the ad valorem rate was imposed on the full price in place of the specific rate. The alcohol tax still maintained the largest share of indirect taxes, generating 16.0 per cent of national indirect taxes and 4.5 per cent of total national taxes in 1988. However, as seen in Figure 14.1, the relative weight of the tax has been declining over the long run. As a percentage of alcohol consumption, it fell from 61.3 per cent in 1950 to 30.9 per cent in 1990. Similarly, its relative share of total tax revenues decreased from 18.5 to 3.1 per cent. The burden of the alcohol tax is generally calculated as the effective tax rate relative to retail prices. Some examples in 1987 are given in Table 14.2 in comparison 3 The ten categories of alcoholic beverages are as follows: sake, sake compound, shochu, (distilled grain alcohol), mirin (cooking alcohol), beer, wine, whisky, brandy, spirits, and miscellaneous liquors.
256
Indirect Tax System
Fig. 14.1 Alcohol tax as a percentage of alcohol consumption and national tax revenues, selected fiscal years Source: Usui (1987, 34-5, 60) and data from the MOF.
with those in 1992 after substantial changes in the alcohol tax in 1989. Clearly, there was considerable difference in the level of tax burden among beer, whisky, wine, and other beverages. The alcohol tax system faced several problems until 1989. The basic framework of the present system was established by the 1962 tax reform, and remained almost unchanged. It produced structural problems which needed to be remedied in the light of current economic and social changes. In particular, there had been a long-standing dispute concerning the alcohol tax between Japan and the European Community (EC in the 1980s). The EC pointed out several problems which it would like the Japanese government to remedy promptly in order to reduce trade friction and promote international tax co-ordination. There are two points related to the Japan-EC disputes. The first concerns the grading system, according to which each of the ten categories of taxable alcoholic beverages was further classified into subcategories. Eor instance, whisky was divided into special, first, and second grades, each of which was taxed at a different rate (see Table 14.2). The point is that there was a big gap in the effective tax rates among the three grades of whisky. In addition, the grading system seemed to be arbitrary in categorizing each type of alcoholic beverage. One of the major complaints from the EC was that imported whisky was arbitrarily grouped in the 'special grade' category for no convincing reason and was therefore at a disadvantage vis-a-vis Japanese competitors.
Traditional Framework of Indirect Taxes
257
Table 14.2 Alcohol tax rates as a percentage of retail prices, 1987, 1992-2000 Alcoholic beverage
Sake Special grade 1st grade 2nd grade Shochu Group A Group B Beer Wine Whisky Special grade 1st grade 2nd grade
Tax rate (%) 1987
1992
1997
2000
40.1 26.9 14.1
— 20.7 16.7
— 17.9 19.8
— 17.9 19.8
14.4 8.7 48.8 5.4
21.3 13.5 44.1 7.0
26.9 18.1 46.5 7.6
35.8 31.8
50.3 45.0 28.3
36.3 — —
38.9 — —
22.8
Source: Data from the MOP.
The second issue concerns the adoption of an ad valorem tax rate for certain items. The ad valorem rates were very high at the time: 150 per cent on a special grade of sake or 220 per cent on a special grade of whisky or brandy. Since imported whisky was generally categorized as 'special grade', it was automatically taxed at the very high rates of ad valorem tax. This resulted in complaints from the EC that imported spirits were disadvantaged in Japan's market.4 Clearly, the alcohol tax system then included many defects which needed to be remedied. In particular, the total number of categories needed to be reduced considerably, and a single category introduced for all imported and domestic whisky of the same quality. Also, the ad valorem rate needed to be abolished to make the rate structure less complicated. Furthermore, the grading system should be abandoned in order to allay suspicions of unfair treatment by overseas competitors. In response to these complaints the alcohol tax was considerably restructured when VAT was introduced in 1989. The grading system for sake and whisky was repealed, and at the same time rate structures were simplified, as seen in Table 14.2. In addition, the ad valorem rate was eliminated to adjust for the adoption of the VAT rate on alcoholic beverages. After the mid-1990s, the conflict between the EU (European Union) and Japan increased because of the tax rate on whisky vis-a-vis shochu. The EU strongly 4 The EC pointed out that 'The current system has been so designed to give Japanese whiskies in the "First" and "Second" grade categories a guaranteed market, free from high-quality import competition, and is so perceived by all countries endeavoring to export whisky to Japan" ('Dampened Spirits?', Look ]apan,Nov. 1987).
258
Indirect Tax System
insisted that whisky should be taxed at the same rate, because it is the same kind of alcoholic beverage (i.e. a spirituous liquor) as shochu. On the contrary, the Japanese government resisted, saying that shochu was quite different from whisky, being a low-class distilled drink produced in local regions of Japan with no competition against world-class imported whisky (i.e. Scotch). However, as a result of WTO (World Trade Organization) arbitration, the tax rate on shochu has been raised while that on whisky has been lowered, to take effect from 1 October 2000 (see Table 14.2). Tobacco
The harmful effects of tobacco consumption are as well known as those of alcohol consumption. However, not until recently have taxes on tobacco been approached with the primary aim of reducing consumption; they have been collected mainly for revenue purposes. In the last several years, the prohibitive nature of smoking has prevailed from a standpoint of negative externalities, but it still has no bearing upon the setting of tax rates. Similar to the tax on alcoholic beverages, taxes on tobacco have a very long history, having formed a part of the indirect tax system since 1877. Until April 1985, tobacco had been produced and sold exclusively by Japan's Monopolized Public Corporation. Instead of excises on tobacco, the special charge was levied on monopoly profits of the public corporation. This was equivalent to taxes on tobacco in a quasi-excise form. When the Monopolized Public Corporation was transformed into Nippon Tobacco Product Industry in the process of privatization in 1985,3 a tobacco excise tax was substituted for the special charge on monopoly profits. Now, tobacco products are allowed to be imported from overseas. The tobacco excise tax is levied on manufacturers of Lobacco products, or on imported products withdrawn from bonded areas, in a way similar to the alcohol tax. Tax rates are applied in a combination of ad valorem and specific rates, usually in an 80-20 ratio. While the UK and Germany favour a specific tax, Japan relies more heavily on the ad valorem tax similar to France and Italy. The tax base of the ad valorem tax is the retail price of tobacco and is subject to the authorization of the MOP. This implies that the MOP can cause revenue under ad valorem taxation to rise with the price of tobacco. This is why the ad valorem rate is favoured by the MOP for revenue purposes. In addition to being a tax on tobacco, an ad valorem rate may be justified as being a tax on quality, flavour, brand image, packaging, etc. The remainder of the revenue is raised at specific rates. As in the USA, excises on tobacco are used at both national and local (prefectural and municipal) governments. Table 14.3 shows the total burden of the tobacco 3 According to the Nippon Tobacco Product Industry Acl, the government should own more than half the new shares of the corporation. However, for the time being, until the business comes into full operation, the government's holding of new shares is required to exceed Iwo-thirds of the total.
Traditional Framework of Indirect Taxes
259
Table 14.3 The tobacco tax burden on retail prices, 1987 and 1998 (¥ %)
Retail price (A) Tax burden National Ad valorem Specific Sub-total Local Ad valorem Specific Sub-total National and local Ad valorem2 Specific5 Total (B) Ratio B/A
1987
1998
220
230C
46.00 20.64 66.64
62.52
44.80 20.00 64.80
62.52
90.80 40.64 131.44
125.04
59.7%
57.1%d
Note: The most popular brand of cigarette in Japan, Mild Seven, was used for calculations. "''' The relationship between these two rates is 70:30, not 80:20, because a temporary increase in the specific rates was enforced (i.e. ¥0.45 per one cigarette) between 1 May 1986 and 31 Mar. 1988. c Retail price including VAT. d This is the ratio of (B) to retail price with no VAT (¥219.05). Source: As Figure 14.1.
excise tax relative to the retail price in 1987 before the introduction of VAT. The tax-inclusive price of one package of 20 cigarettes is ¥220, including 131.44 of tax, evenly split between national and local government shares. As a percentage of the retail price, the tax burden is 59.7 per cent. The rates of cigarette taxes, which account for over 95 per cent of the revenue derived from tobacco products, are compared in 12 countries in Table 14.4. The ratio of tax to the retail price of the most popular brand ranges from 40 per cent in the USA to 86 per cent in Denmark. Japan was ranked second-lowest next to the USA in 1985. The relative position in 1991 and 1997, although data are not available for all relevant countries, remained unchanged. Japan's ratio is lower than those in European countries. The tobacco excise tax has consistently declined as a relative share of national taxes. In 1950 it occupied 20 per cent of the total, but it fell to as low as 2.2 per cent in 1988. This declining trend has been caused by both the reallocation of tobacco taxing power to local governments in 1954 and the long-run decrease in tobacco consumption.
260
Indirect Tax System Table 14.4 Proportion of cigarette taxes in retail prices, selected countries, 1985,1991, and 1997 Country
Belgium Denmark France Germany Greece Ireland Italy Japan Luxemburg Netherlands UK USA
Tax rate (%) 1985
1991
1997
70 86 75 73
n/a n/a 71.9 73.1 n/a n/a n/a 56.8 n/a n/a 64.7 26.8
n/a n/a 70.8 66.3 n/a n/a n/a 54.9 n/a n/a 74.5 35.2
61 74
72 56.7" 67 72 74 40b
a Based on Mild Seven, the most popular brand. The consumption tax burden is excluded. b Rounded figure, including the NY city tax. Source: Cnossen (1986), Usui (1987), andMOF estimates.
Unlike the alcohol tax, the tobacco excise tax was established only in 1985 and faces basically no problems at present. However, when VAT was introduced into the indirect tax system, some adjustment was made in the tobacco excise tax. As a result, the rate was limited to the specific one at the manufacturing stage, since VAT jointly levies an ad valorem tax on tobacco consumption at the wholesale and retail stages. On December 1998, the special tobacco excise tax was created to redeem partially the accumulated long-term debt of the Japanese National Railroad as a result of political compromise. The tax rate is ¥820 per thousand cigarettes, generating ¥260 billion of tax revenue. Petrol and Other Related Items The petrol tax is the second largest consumption tax, accounting for 12.9 per cent of indirect taxes and 5.5 per cent of total national taxes in 1999. These percentages have steadily increased in recent years, e.g. 9.3 and 2.4 per cent each in 1992. As compared with the traditional excises on alcoholic beverages and tobacco, it is a relatively new tax, having become effective in May 1949 as a national tax for raising general revenue. At an initial stage local governments requested that this new levy on petrol be made a part of their tax structure, but it was decided that the tax could
Traditional Framework of Indirect Taxes
261
be most conveniently collected at the prime importation and domestic refinery points,6 and hence should be administered at the national level. From the beginning, some people proposed to have the revenue from the petrol tax allocated exclusively to road repair and construction. At that time, however, the method of earmarking the tax as specific revenue for such designated expenditures was impractical, mainly because roads were used mostly by carts and bicycles, not motor vehicles. As the postwar economy revived, road construction became an important goal. In 1953 one-third of the petrol tax revenues was transferred from national to local governments to provide financial resources for improving the local road service. In addition, in 1955 the local road tax was formally established as an earmarked tax for road repair and construction in regional areas. Since then, the basic structure of taxes on petrol has remained unchanged. At present, the petrol tax is collected together with the local road tax when petrol is shipped from refineries or withdrawn from a bonded area. Specific rates are applied to both taxes: ¥48 600 and ¥5 200 per kilolitre of petrol for the petrol and local road taxes, respectively, in 1998. These revenues are allocated to the Special Account for Road Construction and Improvement. Needless to say, taxes on petrol may be a reasonable proxy for the use of roads. Since they reflect the varying consumption of road services per vehicle-kilometre, they may be regarded as a good measure for determining road-user charges. It is impossible, however, for these taxes to distinguish adequately between types of road services. In view of the variable maintenance charge associated with usage, other car-related taxes have been introduced: a liquefied petroleum gas (LPG) tax in 1965, and a motor vehicle tonnage tax in 1971. Table 14.5 shows general features of four car-related taxes at the national level.7 The liquefied petroleum gas tax is also a fuel tax on LPG. Since LPG is a close substitute for petrol, it was created to keep the tax burden on fuels fair. Moreover, it is also argued that the motor vehicle tonnage tax may play some role as a fair approximation of measuring social cost, such as noise, air pollution, and congestion. This acts as a sort of registration or car inspection fee on the weight per vehicle to cover economic user charges. Total revenues of these four taxes produced the substantial amount of ¥4096.2 billion in 1998, 18.5 per cent of national indirect taxes. Almost 70 per cent of this revenue was generated from taxes on petrol (i.e. the petrol tax plus the local road tax). To what extent do the petrol and other related taxes impose a burden on petrol consumers and passenger car users? In the past, the MOP made an attempt to estimate the burden of these taxes in international comparison in 1985-86. Table 14.6 6
At that time, two import and four domestic refinery points were considered for collecting revenue. Local governments also have four car-related taxes, such as the motor vehicle tax, the car purchase tax, and the light oil delivery tax at the prefectural level, and the light vehicle tax at the municipal level. 7
262 Table 14.5
Indirect Tax System Car-related taxes, 1998 Tax base
Taxable stage
Tax rates
(1)
(2)
(3)
Petrol tax
Petrol
Shipment from manufacturer
¥48 600 per kl
Local road tax
Petrol
Shipment from manufacturer
¥5 200 per kl
284.8 (1.3%)
LPG tax
LPG
Shipment from a filling station
¥17.5 per kg
30.0 (0.1%)
Motor vehicle tonnage tax
Passenger cars, lorries, buses, etc.
Registration or car inspection
Specific rate per tonnage
1 120.0 (5.1%)
Total
Tax revenues, 1998 (¥bn.) (4) 2661.4 (12.0%)
4096.2 (18.5%)
Note: Car-related taxes are limited to national taxes. Figures in parentheses are percentages of total indirect taxes. Source: Data from the MOE
Table 14.6
Proportion of taxes on petrol in the retail price, selected countries, 1985-1986
Retail price (A) (Date) Tax amount (B) petrol excises (C) VAT (D) Effective tax rates (%) B/A CIA DIA
Japan (¥«)
USA ($/gal.)
UK (P«)
W. Germany (M/ 100-0
France (F/100£)
115.30 (Nov. 1986)
86.00 (Sep. 1986)
36.73 (Apr. 1986)
93.40 (Oct. 1986)
542.00 (Jan. 1985)
22.94 17.00
24.17 19.38
57.47 46.00
357.10 272.10
5.94a
4.79
11.47
85.00
61.5 49.2 12.3
65.9 50.2 15.7
53.80 53.80
46.7 46.7 —
26.7 19.8 6.9
65.8 52.8 13.0
Note: (B) = (C) + (D). a
The retail sales tax in New York State. Tax rates are those on 1 Jan. 1987.
Source: As Figure 14.1.
shows VAT and excises on petrol in five countries. There are large variations in the level of taxation between the three European countries and the USA. Indeed, taxpayers in the UK and France pay 2.5 times more in effective tax rates than those in the USA. Japan's tax burden lies in the middle in both cases. When compared with
Traditional Framework of Indirect Taxes
Fig. 14.2
263
Total tax burdens of car-related taxes on the standard passenger car: international
comparisons, 1997 Notes: 1. The standard passenger car is assumed to have a 2000cc engine, to weigh 1.5 tons, to have 6 years of depreciation, and to consume 1 200 litres of petrol annually. 2. The retail price of such a car is ¥2 609m. (US$16 540, £18 700, DM49 105 and F144 000). 3. Exchange rates are used: US$ = ¥110, £ = ¥174, DM = ¥73, F = ¥21. 4. Tax rates are those on 1 April 1997, including both national and local taxes. Source: Unpublised data from the MOP, Tax Bureau.
European countries, petrol taxes are almost equal to each other, but there is no component of VAT on petrol in Japan. Thus, the level of the petrol tax burden is lowered against the European cases. Figure 14.2 depicts the total burden of car-related taxes on the standard type of passenger car among the same countries in 1997, on the assumption that the taxpayer purchases the car and uses it for one year. The tax system in the USA favours the purchase and the use of motor cars. In contrast, France bears the highest combined burden of car-related taxes. Japan has a lighter burden than European countries, although it has to bear a heavier burden on registration fees and user charges (i.e. motor vehicle tonnage and motor vehicle taxes). The present system of taxes on vehicles and vehicle-related sources is too diversified and sophisticated. Some of the taxes should be abolished or merged with others to make the whole tax system on motor cars and petrol less complicated. In particular, a simple alternative form of excise taxation is needed to improve the current tax system.
264
Indirect Tax System Commodities
The commodity tax was introduced in 1937 as a form of excise tax in wartime, for two purposes: (1) to raise wartime revenues, and (2) to discourage consumption of luxury items. Thus, from the outset it was characterized by excises on luxury commodities. During the wartime period the revenue objective took first priority, and taxable items were greatly expanded (i.e. from 10 in 1937 to 107 at the peak in 1944). When the Shoup Mission proposed tax reform, they recommended that these excises be continued as an indirect way of taxing personal ability to pay as indicated by luxury or semi-luxury expenditures. At the same time, however, the Mission recommended that commodities used wholly or chiefly in business such as adding machines, should be completely exempt from the commodity tax (see Shoup Mission 1949,168-70). Since then, the commodity tax had grown up to become one of the most important of indirect taxes until it was merged into VAT in 1989. Having deviated a little from a simple luxury excise, it was justified as a consumption tax that imposes on taxing capacities expressed in consumption, expenditure for particular commodities. Accordingly, the commodity tax experienced periodic changes in tax rates and taxable items, which were required to respond to the pattern of economic and social change. For instance, the nature of the luxury or commodity to be taxed varied frequently during the high-growth period; but the tax structure itself basically remained unchanged. Generally speaking, the commodity tax was mostly a series of manufacturers' excises levied on a number of specified articles. Taxable commodities were divided into two categories—i.e. class 1 and class 2—according to their nature and stage of tax imposition. The tax for commodities of class 1 was levied at the retail stage, that of class 2 at the manufacturing stage. Different tax rates were applied to each of the retailers' and the manufacturers' prices. In the latter case, tax rates were equivalent to much lighter retail rates. In 1988, ten items such as jewels and fur products were being taxed at the retail stage at 10-15 per cent tax rates. On the other hand, at the manufacturing level 75 items, including cars, cosmetics, cameras, and electrical appliances, were being taxed at rates ranging from 5 to 30 per cent. In 1988 the commodity tax raised the second-largest proportion of indirect taxes next to the alcohol tax: 14 per cent of the total. This revenue was collected on 85 taxable items, but three-quarters of it came from taxes on motor cars and electrical home appliances. Table 14.7 illustrates the top ten items by tax amount, accounting for 73.0 per cent of the commodity tax. If the list was extended to cover the top 20 items, these taxes account for nearly 90 per cent of total revenues. This implies that the major revenue-producing items for commodity excises were relatively few, while the remainder was composed of many miscellaneous commodities, each generating minor revenues.
Traditional Framework of Indirect Taxes
265
Table 14.7 Commodity tax revenues collected from the top ten major items, 1986 Tax revenues
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
Passenger cars Air-conditioners Precious stones TV sets Light combined passengercargo motor vehicles Video tape recorders Refrigerators Cosmetics Vacuum cleaners Car air-conditioners
Total Commodity tax
(¥bn.)
(%)
675.0 108.4 77.6 65.6 51.3
40.9 6.6 4.7 4.0 3.1
48.1 47.4 45.5 43.7 41.2
2.9 2.9 2.8 2.6 2.5
1203.8 1650.1
73.0 100.0
Source: Data submitted to the Tax Advisory Commission.
The rate structure consisted of two rates in class 1 and five in class 2, although motor cars and related items were treated separately. From among varied rates, a standard rate system was adopted in 1962. Currently, standard rates are 15 per cent in both classes. Commodities levied at the standard rate produced larger tax revenues than commodities taxed at other rates (see Table 14.8). Both 30 and 20 per cent in class 2 were higher than the standard rate and were regarded as luxury tax rates. In contrast, reduced tax rates were applicable to necessities through lower-than-standard excise rates. Apart from the statutory rate structure, it is interesting to see how much of a burden the commodity tax represents within the retail price. Major cases are illustrated in Table 14.9, ranging from the highest rate of 11.9 per cent on a passenger car down to 3.0 per cent on soft drinks. Even if tax rates are high on a statutory basis, these commodities reveal much smaller tax rates relative to the retail price because they are taxed on the manufacturer's price since they belong to class 2. Comparing two different rates in Tables 12.8 and 12.9, we can observe a gap between the formal and actual tax burden on specific commodities. For example, manufacturers must pay a 20 per cent tax on air-conditioners, but in actual practice this burdens the consumer by only 10 per cent at the retail level. The commodity tax was an indirect tax unique to Japan, and, in fact, it would be difficult to find an equivalent tax in other advanced countries. In the Japanese indirect tax system, the commodity tax was the closest substitute for a general consumption tax.
266
Indirect Tax System
Table 14.8 Rate structures of the commodity tax, 1986 Tax rates
Major items only
Tax amount in 1986 (¥bn.)
Retail stage: Class 1 1 5% Precious stones, pearls, fur products 10% Carpets Manufacturing stage: Class 2 30% Motorboats, golf clubs, billiard tables 20% Air-conditioners, refrigerators, slot machines 15% Electric fans, washing machines, dishwashers 10% Radio sets, magnetic tape recorders 5% Microphones, coffee, cosmetics Motor-vehicle-related items Passenger cars 30%a Small passenger cars 18.5% Light passenger cars 15.5% Motorcycles 10% Small motorcycles 5%
3
Relative share (%)
158.6 3.8
9.6 0.2
34.0 193.5
2.1 11.8
353.3
21.4
43.1 85.8
2.6 5.2
777.9
47.1
1650.1
100.0
Temporarily 23% owing to the Special Tax Measures Law.
Source: As Figure 14.1.
Table 14.9 1987
The commodity tax as a percentage of retail price, major items,
Items
Retail price3 (¥)
Tax (¥)
Effective tax rates (%)
Passenger cars (2000cc) Air-conditioners TV sets (20-inch) Video tape recorders Pianos Cameras Cosmetics Soda drinks
2000000 200 000 150000 200000 600 000 100000 5000 100
237299 20000 11934 16956 53217 8739 231 3
11.9 10.0 8.0 8.5 8.9 8.7 4.6 3.0
a
Based on most standard price category of commodities.
Source: Data from the MOF.
Traditional Framework of Indirect Taxes
267
When the value added tax was introduced, the commodity tax was intended to provide a good base for the new tax and to be merged with it. The commodity tax was levied on a narrower basis of selective goods; it did not cover services, and produced distorted effects in resource allocation. Major issues in narrowly based commodity excises were solved by introducing a broad-based indirect tax.
15 The Value Added Tax
In Japan, the value added tax (VAT) was finally introduced in the tax structure in April 1989, after long-standing trial and error by the government. The government had flirted with VAT since the late 1970s to stress the important role that VAT coul play in tax system. Whenever attempts were frustrated to adopt VAT, different names for it were applied so as to defuse political tension, i.e. the general consumption tax in 1979, the sales toxin 1987, and the consumption toxin 1989. Towards the latter half of the 1980s, the first and the second sweeping tax reforms were proposed in succession and carried out by the initiatives of the Nakasone and Takeshita cabinets, respectively. Needless to say, the main reason for pursuing these reforms was to solve unsettled issues with the introduction of VAT. Japan's VAT, named the consumption tax, came into existence in the second round of these tax reforms along with the worldwide movement to adopt VAT. The major aim of this chapter is to clarify the picture of Japan's VAT before and after its introduction. It is divided into five sections. The first is devoted to preliminary discussion on the key trend of VAT in the world and the historical background of VAT. The second deals with the main features of a new VAT in detail with relevant problems, followed by the third section in which economic aspects and administrative consequences of VAT are discussed. Finally, recent amendments and further reform to improve VAT are summarized, and the basic features of local VAT that wa adopted in 1997 are explored in detail.
THE
KEY TREND OF I N T R O D U C I N G VAT
Recent Developments The first emergence of VAT in the modern history of taxation was in the French tax system in 1954. VAT in France was introduced to take the place of turnover tax and levied at all stages of transactions to impede tax evasion and avoidance. It formed the basis of the EC-type VAT, and VAT is now used not only in Europe but in man other countries in the world as more or less a common form of broad-based consumption tax (Webber and Widavsky 1986, 547-9).
This chapter is primarily based upon Ishi (1992a).
The Value Added Tax
269
A. T. Tait (1988, 3) emphasizes the key trend of introducing VAT as follows: The rise of value-added tax (VAT) is an unparalleled tax phenomenon. The history of taxation reveals no other tax that has swept the world in some thirty years, from theory to practice, and carried along with it academics who were once dismissive and countries that once rejected it.
VAT has so far developed rapidly in many countries with a wide variety ranging from more complete VAT to less-than-complete VAT. Basically, fundamental forms of VAT are levied by first calculating value added (sales minus purchases) at all stages from manufacturers to retailers and then applying a tax rate to that value. This is, however, not necessarily always the case. For instance, some countries, particularly in Latin America and Africa, do not apply VAT at the retail stage, while other countries are using a VAT through the wholesale stage or are applying it only at the manufacturing level. If we broaden the scope of VAT to cover such imperfect types, there are more than fifty countries in the world that have already introduced VAT in their tax structure.1 Of most importance is the fact that there has been no country except south Vietnam where VAT was abolished after being adopted (see Shoup 1990, 1). It is stressed that VAT is not a sort of temporary or provisional tax but a well-established tax with a permanent place in the existing tax system of major countries. Trends ofOECD Member Countries From a practical point of view, we are more interested in analysing VAT in major industrialized countries. Table 15.1 summarizes the past trends after adopting VAT in OECD countries. Depending upon the years of introducing VAT, twenty-four Table 15.1
Past trends of VAT among OECD countries
1. Advanced countries for VAT— 12 countries: Austria (1973), Belgium (1971), Denmark (1967), France (1968), W. Germany (1968), Ireland (1972), Italy (1973), Luxemburg (1970), the Netherlands (1969), Norway (1970), Sweden (1969), UK (1973) 2. Less advanced countries for VAT—7 countries: Canada (1991), Greece (1987), lapan (1989), New Zealand (1986), Portugal (1986), Spain (1986), Turkey (1985) 3. Non-VAT countries—5 countries: Australia (from luly 2000), Finland, Iceland, Switzerland, USA Note: Figures in parentheses stand for year of introducing VAT. Source: OECD (1988, 30) and others.
1 This would perhaps pose a conceptual problem for VAT. Of course, it is possible to define VAT t perfect form in a strict sense (see Tait 1988, 6 and 7).
270
Indirect Tax System
member countries are classified into three groups. Japan is placed in the second category, as well as Canada. Most countries in the first group adopted VAT around the year 1970 since they were committed to implement VAT on entering the European Community (EC) as a member. VAT is considered to be essential for border-tax adjustments. In the 1980s, many of the countries with no VAT attempted to introduce such a form of tax. Greece, Portugal, and Spain introduced it in 1986-87 as a prerequisite for entry into the European Community. Turkey did the same in 1985. In the Pacific countries, the pros and cons of a VAT have long been a major issue; only New Zealand has success fully established a form of VAT, the 'goods and service tax' in 1986. During the US tax reform of 1984-86 the possibility of introducing VAT was discussed, but this wa rejected. It seems that the US government would have difficulty in instituting such a tax at the federal level because the state governments have employed the retail sales tax as a dominant tax.2 Similarly, in Australia, Canada, and Japan, the governments proposed introducing such a tax at various times, but such proposals encountered strong opposition and rejection. However, Australia and Canada, as well as the USA, had already adopted a type of broad-based sales tax on manufacturers or wholesalers, and reform has meant trying to transform an incomplete sales tax into a more satisfactory alternative, i.e. the value added tax (see Canada 1987; Australia 1985). In a strict sense, Japan remained the only country that had no form of broad-based indirect tax until 1989. In this respect, Japan's tax reform lagged behind the mainstream of the global reform. In succession to Canada, Australia has finally decided to adopt VAT (i.e., goods and services tax) from July 2000, based on the GST law on 8 July 1999. Regardless of the fact that the timing in each country differs, why have countries adopted VAT? Broadly speaking, there are four reasons: (1) The existing turnover tax was not satisfactory, (2) Discriminatory border taxes had to be abolished because of a customs union, (3) Other taxes, such as income taxes, had to be cut mainly because relevant countries were dissatisfied with their existing tax structure, (4) Accumulated fiscal deficits required additional sources of revenue. (1) and (2) were good reasons for VAT-advanced countries to introduce such a tax earlier than the second group of countries. On the other hand, the reasons for adoption by VAT-less-developed countries seem to be more closely connected with (3) and (4). Japan originally attempted the introduction of VAT to make up for fiscal deficits, but later more stress tended to be placed on the fact that the evolution of the tax system did not keep pace with the development of the economy, say ageing.
2 The tax levied on the value added at each stage of a transaction is proportional to the value of the final sale to the consumer. Thus, a retail sales tax would theoretically be equivalent to the value added tax.
The Value Added Tax
271
Worldwide Indirect Tax Reform Generally speaking, in the 1980s there was a trend among major industrial countries towards switching from a relative reliance on direct taxes to a reliance on indirect taxes: a broad-based indirect tax was introduced in countries with no VAT, and in a number of European countries where VAT had already been implemented the governments raised the rate of VAT to offset part of the revenue loss from income tax changes.3 Such a reshaping of the indirect tax system on a world scale had a close bearing upon Japan's sweeping tax reforms starting from the mid-1980s. Of central importance to both the first and second stages of the tax reforms was, as stressed before, an overall reform of the indirect tax structure in the form of a broad-based indirect tax. Japan's method of reforming the indirect tax system corresponds to tax reform movements abroad (see Messere and Owens 1986; OECD 1987; Messere ed. 1998). Japan's indirect tax system was characterized by reliance upon selective excise taxes until 1989 (see Chapter 14). Why was indirect tax reform necessary in Japan, too? The existing selective excise taxes posed several problems which needed to be remedied. First, the inherent arbitrariness of a tax on selective goods and services tended to discriminate between taxed items and non-taxed items and to produce distortions in consumer choices.4 In addition, the tax base of such a tax was extremely narrow. This is the most important disadvantage of a selective excise tax system. The more broadly based the tax, the less distorting it becomes. Moreover, higher tax rates were required to raise needed revenue from a narrower tax base. Second, the existing indirect tax system was outdated. Although the economy was becoming more service-orientated, the government could not fully tax service items. In fact, only a small portion of services were being taxed, i.e. the travel tax, the admission tax, entertainment tax, etc. Since consumption patterns were diversifying rapidly, it was almost impossible for selective taxes accurately to reflect tax-paying capacity. It was difficult to correlate the income level of consumers and their purchase of costly goods with higher tax rates. Third, taxes levied on selected items were liable to be criticized from abroad, triggering international conflict. Typical examples were the complaints by various countries regarding the tax on imported whisky and wine and the commodity tax on watches and automobiles. Fourth, selective excise taxes had failed to secure adequate sources of revenue. The relative share of indirect taxes showed a long-run decline in total tax revenues. Thus, the tax system as a whole was placing too heavy a reliance on direct taxes.
3 A typical example was observed in the UK tax reform of 1979 in which the rate of VAT increased from 8 to 15 per cent in place of reducing income taxes with equal revenues. 4 The following are typical examples of an arbitrary selection of taxed or untaxed items: furs, gold artefacts, surfboards, water skis, ordinary furniture, coffee, and cocoa are all taxed, while deluxe textiles, tennis equipment, ordinary skis, high-quality apparel, and tea are not.
272
Indirect Tax System
Given the basic framework of the indirect tax system with no VAT, it was not advisable to increase tax rates and tax items so as to secure more needed revenue in the future. The proposal would have yielded no substantial gains because it would have tended to worsen the distortions and arbitrariness of the existing indirect taxes. In order to eliminate these problems, an overhaul of indirect taxes proved desirable, along with the introduction of a new broad-based consumption tax. H I S T O R I C A L B A C K G R O U N D OF J A P A N S VAT
Prehistory Although there was no VAT in the Japanese tax system, we experienced, or at least expected, the adoption of broad-based indirect taxes twice during the earlier postwar period. The first instance was the turnover tax of 1948-49, which was enacted to raise needed revenues at that time. The turnover tax became effective in September 1948, and was repealed in December 1949. Although it was short-lived, it was really Japan's only experience of a broad-based indirect tax. It was applied to the value of transactions (i.e. gross sales) at all levels—producing, wholesaling, and retailing—at a rate of 1 per cent. Some transactions were exempted from the tax: export transactions, sales of securities, transport, transactions involving staple foods, sales of agricultural forestry, livestock, or marine products by the original producer, rents, and certain other less important items. Furthermore, small traders with gross sales of less than ¥30 000 per month were tax-exempt. Until early 1949, the turnover tax was collected principally from the advance sale of stamps to the vendors. Since this advance payment raised a great deal of complaint among taxpayers, the government was obliged to switch to cash payment on the basis of a tax return submitted by the taxpayer. On theoretical and practical grounds, the turnover tax is the least refined category of the sales tax. If a business firm starts with its own raw materials and then produces and sells the finished goods, this vertically integrated firm greatly benefits from the turnover tax. On the other hand, an individual group of independent firms, which may include raw material producers, manufacturers, wholesalers, and retailers, is at a disadvantage. While the vertically combined firm pays the tax only once, independent firms have to pay the tax at each turnover. This is a great defect, and it tends to encourage the vertical integration of independent firms. In addition to such theoretical shortcomings, there was strong criticism by taxpayers of the Japanese turnover tax initiated at that time. In the first place, it was claimed that the turnover tax, which required taxpayers to submit a statement of gross sales, assisted the government in imposing the income tax on self-assessed businessmen, since this feature enabled the tax authorities to make intelligent guesses as to taxpayers' net income. Second, knowledge of gross sales enabled the tax office to investigate the use of illegal pricing via the black market. Not surprisingly,
The Value Added Tax
273
taxpayers organized protest movements to complain against the continuation of the turnover tax. As a result, the tax was repealed within 15 months of its enactment.5 The second instance was a VAT recommended by the Shoup Mission in 1949 (see for an expanded discussion, Shoup 1989). It was proposed as a prefectural tax, but its application was postponed several times, and in 1953 it was abolished without having ever been put into operation. Since it was an unusually rare and innovative idea at that time, it is interesting to study the outline of the VAT contained within the Shoup recommendations (see Shoup Mission 1949,197-204). The main purpose of introducing VAT was to replace the enterprise tax (prefectural and municipal), which it was feared placed too heavy a burden on the business firm. The enterprise tax was an independent prefectural tax, but its revenue was shared with municipalities, usually on a 50-50 basis using a municipal surtax.6 The tax was levied on the profits (i.e. net income) of the preceding year, and in principle the taxable profits were those computed for purposes of the national income tax. Thus, the enterprise tax was simply a surtax on the national tax. It appears that the enterprise tax was not intended to be shifted to consumers because net income taxes were assumed to be non-shiftable. This implies that the business firm had to absorb the tax burden. The cumulative burden of the national and local tax system on the net income base was considered to be too heavy. The business proprietor of the unincorporated enterprise was assessed on the following three taxes: 1. a national income tax on his net income after basic exemptions and credit for dependants; 2. a local income tax (i.e. a prefectural-municipal inhabitants' tax); 3. the enterprise tax. The combined marginal rate of these three taxes reached nearly 70 per cent for proprietors even with a modest income. The Shoup Mission proposed a new enterprise tax, i.e. the value added tax, mainly in order to lower to some extent the cumulative burden on net income as a tax base. Accordingly, it recommended that the tax base be expanded to include the sum of profits, interest, rent, and the payroll. Obviously, the scope of the proposed tax base was equivalent to that of the income type of VAT. The income type of VAT shows a direct correspondence of the taxable value added to the economic activity of the year. The tax base can be computed by adding the factor rewards accruing to labour and capital. Instead of the addition method in computing the income type of VAT, we can also subtract all purchases from other 5 The Shoup Mission also proposed the elimination of the turnover tax (the 'transactions tax', as they called it): 'Altogether, the transactions tax is one of the least promising members of the present Japanes tax system. It cannot be raised substantially without grave economic disadvantages. )t is not likely to be taken very seriously by either tax administrator or taxpayer at its present rate, and thus may decay gradually into a tax that is largely unenforced' (Shoup Mission 1949, 167—8). 6 The standard tax rate was 7.5 per cent for the prefectures and 7.5 per cent for the municipalities, a total rate of 15 per cent.
274
Indirect Tax System
firms from total gross sales and reach the same magnitude of tax base. Depreciation is subtracted, not the amount spent purchasing investment goods that year. The Shoup recommendations pointed out two advantages of VAT. First, it is free of the distorting effects of the turnover tax, which tends to stimulate the growth of vertical integration. Second, in comparison with the enterprise tax on a strict net profit base, VAT does not discriminate against the use of capital, especially in the form of labour-saving equipment. According to the tax proposals made by the Shoup Mission, the rate of VAT should be set somewhere around 4-6 per cent to assure the prefectures a certain amount of revenue, i.e. ¥44 billion. The Mission also recommended that farmers and small business concerns be exempted from VAT. Why was VAT rejected by the Japanese government and taxpayers? The reason was quite simple. The concept of 'value added' was very difficult to understand, and many could not imagine what type of tax it would be. Also, many taxpayers were wary of the Mission's suggestion of a VAT because 'the new tax is a bad tax'. Withou understanding its basic nature and significance, people were sceptical of its rapid enforcement. The atmosphere in those days did not permit the acceptance of an innovative, new tax. The General Consumption Tax Either the turnover tax or Shoup's income type of VAT was noteworthy but premature. Controversial issues were raised by the undertaking to adopt a VAT-style 'general consumption tax' proposed by the Ohira cabinet in 1979. Since then, there has been a great deal of argument on whether or not a broad-based indirect tax ought to be introduced. The history of Japan's VAT is full of turmoil. Before VAT emerged in the tax structure, the Japanese government had been tempted, shaken on the brink, and succumbed several times. Although the attempts to introduce VAT failed, the attraction of VAT still remained irresistible.7 Table 15.2 summarizes the historical process from the starting-point of the first trial in 1978 until the implementation of a new VAT in 1989. The whole period can be divided into three sub-periods, based upon the different cabinets. First of all, let us look at the general consumption tax. The need to introduce such a tax arose from the necessity to find sources to reduce the rising fiscal deficits caused by the oil crisis of 1973 (see, for a detailed discussion, Ishi 1986a, 2000). Strategies to curtail fiscal deficits are based on tax increases, expenditure cuts, or a combination of both. Initially, the government and the MOP adopted the bold strategy of creating a new VAT. The new tax was named 'the general consumption tax', which is a tax-credited type of VAT that has no invoice. 7
The behaviour of the Japanese government compared with the Korean experience is summed up by Tail as follows: 'The Korean VAT, introduced in July 1977, was the first VAT in Asia and as such was a major innovation; after all, the Japanese had toyed with the idea of a VAT since the 1953 Shoup Mission, but had nervously postponed such a dramatic change in taxation. The Korean experiment was a brave initiative' (Tait 1988,204).
The Value Added Tax Table 15.2
275
History of tax reforms in Japan before 1989 Event
Date I. The Ohira Cabinet 1978 8 September 26
December
1979 7 21
October December
The Tax Advisory Commission proposed the introduction of a broad-based indirect tax. The general consumption tax named in the LDP's Tax Reform Plan of 1979. General election for the Lower House. Both Upper and Lower Houses passes a resolution abolishing the general consumption tax.
II. The Nakasone Cabinet 1984 19 December The necessity of a sweeping tax reform in the near future was accepted by both the Tax Advisory Commission and the LDP Tax Council in their proposals for the annual tax reform for fiscal 1 985. 1985 20 September Enquiry from Prime Minister Y. Nakasone to the Tax Advisory Commission on the first sweeping reform. 1986 25 April Interim report on the individual and corporate tax reform from the Tax Advisory Commission. 6 May Basic reform on the individual and corporate tax endorsed by the LDP Tax Council. 6 July General election for both the Lower and Upper Houses. 29 July Three types of new indirect taxes proposed by a Subcommittee of the Tax Advisory Commission. 28 October Final report on a sweeping tax reform issued by the Tax Advisory Commission with the combined package of income tax reduction and a new VAT. 5 December LDP Tax Council decides on the basic principle of tax reform. A new VAT was named the sales tax. 1987 16 January Cabinet approve bill on the fiscal 1987 tax reform, including the sales tax. 10 February Tax bill sent to the 108th Diet session. 27 May The Diet closed. Tax bill withdrawn and the sales tax abolished. 31 July Revised bill on the individual tax reform presented to the Diet. 25 September Diet passed revised bill incorporating income tax reduction and the elimination of the tax-free savings system. III. The Takeshita Cabinet 1987 12 November Enquiry from Prime Minister Takeshita to the Tax Advisory Commission on a second stage of the tax reform process.
Indirect Tax System
276
Table 15.2
contd.
Date
Event
1988 Mid-February to early March 25 March 28 April
14
lime
15 28 29
lune June July
16
November
24
December
1989 1
April
Public hearings held by the Tax Advisory Commission in 20 places throughout the country. Preliminary report from the Tax Advisory Commission. Interim report from the Tax Advisory Commission to clarify basic forms of a new broad-based indirect tax. Tax report from the LDP Tax Council named such a tax as the consumption tax. Final report from the Tax Advisory Commission. Cabinet approves 'The Outline for Tax Reform'. Tax bill, including the consumption tax, sent to the 113th provisional Diet session. Tax bill passed by the Lower House with minor amendments (Japan Socialist and Communist Parties did not appear in the Diet). Tax bill passed by the Upper House (a majority of opposition parties were absent for voting) . A new VAT, named the 'consumption tax' introduced.
Source: Data submitted to the Tax Advisory Commission.
The Tax Advisory Commission proposed the introduction of a VAT in order to close the fundamental structural gap between government expenditures and revenues in December 1978 (see Tax Advisory Commission 1978). The new tax was designed in 1979 as follows. A uniform tax rate of 5 per cent would be levied on the amount of value added (gross sales minus purchases) for all business firms, and the exemption level would be fixed at ¥20 million in sales volume, with estimated revenues accounting for ¥3 000 billion in fiscal year 1980. Whether or not a VAT should be adopted became a crucial point of dispute in the general election to the Lower House in October 1979. Prime Minister Masayoshi Ohira risked proposing the new indirect tax during the election campaign, but he was finally forced to back down because of strong opposition. As a result of the tax increase proposed by the LDP, the party lost a number of seats and received a major setback. Why did this effort meet with such vigorous resistance? There seem to be two main reasons. First, although general taxpayers were not so hard set against the new tax, a great deal of opposition came from small businesses in the wholesale and retail trade, most of which are the main supporters of the LDP. 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 or evade income tax as they had been doing. Second, there were major complaints regarding inequitable tax burdens and the wasteful use of government expenditures. Taxpayers asserted that these problems should be solved before the new tax was introduced.
The Value Added Tax
277
After the political failure of Prime Minister Ohira (he died soon after, in 1980), the issue of adopting a broad-based indirect tax became a political taboo. Before proposing a new indirect tax again, steps had to be taken to promote further reduction of the large deficits. Since major increases in taxes proved to be unacceptable, the government was obliged to alter its policy options. It was compelled to reduce the current fiscal deficits by cutting expenditure. This was a direct response to the complaints of general taxpayers, who demanded spending cuts as a precondition for the general consumption tax. The Emergence of the Sales Tax After 1985, the question of sweeping tax reforms had become the most important political issue for the general public. The reform process began in September 1985, when Prime Minister Nakasone proposed an enquiry to initiate the most sweeping tax reform since the Shoup recommendations in 1950. This was the start of the first stage of the tax reform process (see Ishi 1986c). A year and a half later, Nakasone's tax proposals were frustrated by his political mistakes, and, as was argued above, he achieved only half of his original plan. His main ideas, however, were retained by Prime Minister Takeshita, and the basic framework of tax reform proposals remained almost untouched. Thereafter, nationwide interest focused on what could be called the second sweeping tax reform process. No doubt, Nakasone's enthusiasm for tax reform was spurred by the successful example of US President Reagan's tax reform. He must have expected that the success of American tax reformers would accelerate the Japanese tax reform. He behaved very prudently in proceeding with the tax reform in the form of an overhaul of indirect taxation, because he decided to challenge the political taboo brought about by former Prime Minister Ohira. In fact, Nakasone did not permit the Tax Advisory Commission to consider a broad-based indirect tax until its interim report was issued in April 1986. Moreover, during the campaign before the general election in July 1986, he pledged not to institute a 'large-scale indirect tax that cannot obtain the approval of the public'. After a landslide victory for the LDP, he urged the Commission to construct a package of tax reforms including a new, broad-based indirect tax. It is thus necessary to trace the general features of the various alternatives. A sale tax proposed in the first stage of tax reform emerged as a combination of three types of new indirect taxes proposed by the sub-committee of the Tax Advisory Commission in July 1986. It is important to examine the nature of these taxes, since they were regarded as possible alternatives to the broad-based indirect tax. When the sub-committee members investigated alternative types of indirect taxes, several factors were taken into account: (1) conformity with Japanese business practices; (2) simplication of tax payment and administration; (3) elimination of accumulating tax burdens; and (4) the importance of fairness and neutrality. Based upon these requirements, three types of new indirect taxes were proposed.
278
Indirect Tax System
Type A: Manufacturers' sales tax. All manufactured goods would be subject to the tax, but raw materials, machinery, and other goods used in the process of production would be tax-exempt. Two sub-categories were added, depending upon whether the taxpayer prepared a list of tax-exempt items. Type B: Retail sales tax (officially called the 'sales tax with exemption for inter-firm transactions'). This tax would be levied once, when goods and services were sold at the retail stage to consumers because inter-firm transactions would be exempt in principle. Type C: The Japanese-style value added tax. This tax would be imposed on the value added to a good or service at every stage of processing from manufacture to retail i.e. on the difference between the value of sales and the value of purchases. This would be a multi-stage sales tax similar to the EC-type VAT. These three alternatives were considered to provide an outline for a broad-based indirect tax. Types A and B were both single-stage sales taxes, while type C was a form of multi-stage sales tax. Although the alternatives were designed to satisfy the four requirements mentioned above, they differed in several respects and have various merits and demerits. After heated debates among the members of the Tax Advisory Commission, type C was finally selected, in the final report issued in October 1986, as the best form of broad-based indirect tax that could be adopted in Japan. This tax had two principal advantages over the two alternatives, in that it had a broader tax base and led to neutrality in resource allocation. On the other hand, its main disadvantage was that it would involve more taxpayers than the alternatives at every stage of transactions. The LDP Tax Council also agreed, in December 1986, on the choice of type C, and named it the sales tax instead of 'Japanese-style VAT'. The details of sales tax were decided upon through political compromise. Its main features were: 1. 2. 3. 4.
a consumption type of VAT with the tax credits determined according to invoice; a single 5 per cent rate with a 'zero-rating' on exports; 51 tax-exempted items; exemption of firms whose annual sales are less than ¥100 million.
In the process of shaping the detailed framework of the sales tax, tax-exempted items were greatly expanded from the original plan of seven items—as was the general consumption tax in 1979. Furthermore, the exemption level of certain firms was raised to ¥100 million in terms of annual sales because of strong political pressure from small traders.8 These compromises greatly impaired the advantage of universality inherent in the broad-based indirect tax. Such special considerations incurred difficulties and complications in both administering the tax and obtaining compliance from 8 This exemption level was clearly much larger than other comparable cases, i.e. approximately 12-20 times in comparison with the counterpart figures in the EC countries and Korea. Even the 'general consumption tax' in 1979 included merely ¥20m. in exemptions for certain firms.
The Value Added Tax
279
taxpayers. In the end, the contentious implementation issue prevented the government from obtaining general support for introducing the new tax. From the beginning, there was a great deal of misgiving about the future success of this reform plan because of a key political mistake made by Nakasone. His LDP colleagues and groups of key supporters of the party felt betrayed, and they, as well as the opposition parties, accused Nakasone of going back on his promise. Once the legislation reached the Diet in February 1987, dramatic opposition arose from all sides, including even strong supporters of the LDP, and Nakasone was forced to withdraw the bill in May 1987. Once again, VAT proved as unacceptable as it had been in 1979. Tax increases remained as politically taboo as ever. M A I N F E A T U R E S OF THE
CONSUMPTION
TAX AS A NEW VAT
Strong Anti-Movement The second step of tax reform began with Prime Minister Takeshita's enquiry to the Tax Advisory Commission in November 1987. Tax reform could have been the crowning achievement of Nakasone's term, but he failed. Thus, when he appointed Takeshita as his successor, he evidently wanted Takeshita to devote all his efforts towards achieving tax reform. After accepting the Prime Minister's enquiry, the Tax Advisory Commission opened its deliberations for the second stage of the tax reform process and began to consider how to realize the postponed reform package. Since Nakasone had been criticized for not offering the issue up for nationwide debate, the Commission endeavoured to attain public support by holding public hearings in twenty places throughout the country in February and March 1988. Further public hearings took place in five places in mid-April. In April 1988, following the sales tax, another form of broad-based indirect tax was proposed, once again under the name 'consumption tax'. Of the more controversial issues in the reform package, the most important concerned this tax. Ultimately, support for the Takeshita tax reform by the general public depended primarily upon the pros and cons of introducing the consumption tax, despite the fact that direct tax reductions substantially exceeded the increased tax burdens. Past experience suggested that political considerations rather than economic ones were the deciding factors in obtaining public support for VAT. However, as far as results of public polls were concerned, strong anti-sentiment against VAT was persistently prevalent up to the final stage (see Table 15.3). Although it is difficult to compare the results of the various polls because of the different ways of making the enquiries, the majority of respondents in the range of 50—70 per cent responded 'No' to VAT. Reflecting such a mood, the opposition parties strongly opposed the consumption tax bill in the Diet and as a consequence the bill was passed with difficulty by a single-handed voting by the ruling LDP. Even after implementing the consumption tax from 1 April 1989, the opposition parties still promised to repeal it.
280
Indirect Tax System Table 15.3 Results of public poll for or against the introduction of the consumption tax by mass-media Responses (%) Dates 3 8 11 26 5 19 23 12
July 1988 July 1988 Sep. 1988 Sep. 1988 Oct. 1988 Dec. 1988 Feb. 1989 Mar. 1989
Source
Yes
Nikkei NHK Mainichi Nikkei Asahi Nikkei
34 50 20.8 52.6 26.6 15 58 32 56 16 65 27 62 17 71 Should be introduced as scheduled 12% Not introduced 57% Introduction should be postponed 24%
Yomiuri
Nikkei
No
Others
Note'. Responses are given to such a common enquiry as 'whether or not the consumption tax should be introduced'. The sum of the figures is not necessarily 100% because multiple choices are admitted in some polls.
Engineering Details Learning from their previous experience of the sales tax controversy, government officials widely believed that the key issue in selecting the best indirect form of tax among various alternatives was the breadth and uniformity of its coverage. The most important advantage of making the new indirect tax broad-based was that such coverage would make the tax simpler and less onerous to business. Moreover, a broad-based tax would permit lower tax rates. In contrast, a narrower base would produce arbitrary distinctions in the tax system, which in turn tend to increase compliance costs, cause distortions, and make the tax burdens less transparent. This was true in the case of the sales tax. Therefore, the primary goal in shaping the new indirect tax was to make the tax base as broad as possible but with a single tax rate. From this viewpoint, the consumption tax was successfully designed in spite of political pressures. 1. Tax-exempted goods and services were limited to only a few items, i.e. some of the education, medical care, and welfare programmes; 2. The exemption level for firms was lowered to as little as ¥30 million in terms of annual sales. It is clear that the first point enabled the new indirect tax to apply to almost all consumer goods and services. In general, it is difficult, for political reasons, to implement a new indirect tax without some exemptions. Even the broadest of VAT
The Value Added Tax
281
usually exempts some items, such as medical and educational services, residential rents, etc. No doubt, each exemption introduces distortions and makes a new indirect tax more complicated. The coverage of the consumption tax seems to be among the broadest in the world, bearing a closer resemblance to the goods and services tax in New Zealand. In this regard, the consumption tax is worthy of praise in view of its base-broadening effect and its neutrality, although the exemption level of firms is still higher than other cases. Many VAT countries use multiple tax rates, but Japan's consumption tax has one rate plus a zero rate on exports. Generally speaking, multiple-rate structures offer a greater opportunity to fit VAT to various social and political targets, but then VA tends to lose its simplicity. Under the consumption tax introduced by the Takeshita tax reform, the rate is single one, and the world's lowest, at 3 per cent. Obviously, such a low tax rate does not require any multiple-rate aid to achieve special ends. In some senses, the rate is too low to diversify the rate structure for commodity differentiation and improving tax compliance. In contrast to the successful reformulation of the broad-based indirect tax, three disadvantages have emerged from the consumption tax: 1. no invoices (essential to the tax credit method) are provided for; 2. a special simplified scheme (similar to forfeit) for the computation of tax is built in; and 3. a unique vanishing exemption with 'marginal deduction' is adopted. Invoices admit the use of the tax-credit method, universally preferred in all VAT countries. If invoices are compulsory, the sum of the taxes already paid by other firms on purchases by the firm in question can be traced. Each invoice for a purchase from another firm indicates the total amount of an input tax. Firms collect all such invoices during each period (three months or one year) and aggregate the input tax shown on them. This is the amount credited against the firm's own gross tax in order to calculate VAT payable by the firm. The consumption tax with no invoices must rely on the accounts method of VAT. To compute a firm's VAT, total purchases are subtracted from total sales by using it book-keeping records. The balance by subtraction is then subject to the rate of VAT. Although the sales tax was proposed with the invoice-credit method, the consumption tax was designed with the accounts method, without use of invoices. Since there is no requirement to issue an invoice in Japan, how can the tax official be sure that the firm has not overstated its input tax? It is officially stated that, instead of investigating invoices on a transactional basis, this can be accompanied via the audit of accounts which are usually kept for individual and corporate income taxes. There seem to be two reasons behind this choice. In the first place, the accounts method was officially praised for enabling simpler computation of tax. Second, this simpler method is sometimes preferred for not having a built-in incentive for cheating. It is also politically advantageous, in that it would obtain the support of small
282
Indirect Tax System
distributors who are threatened by the prospect of their income being exposed by invoices with no tax credit breaks. The tax credit method in the proposed sales tax was clearly rejected because it was believed to be equipped with this built-in safeguard which tended to be disadvantageous to taxpayers. Almost the same can be said for adopting the special simplified scheme for computing the tax, a measure that favours small firms. Firms whose annual sales are less than ¥500 million (now ¥200 million after amendment twice, as seen later) were allowed to employ this method to enhance tax compliance. Instead of directly calculating the total value of purchases from other firms, certain fixed percentages (i.e. 10 per cent for wholesalers and 20 per cent for other traders, which are changed now) were multiplied by total sales values and the results deemed to be subject to a 3 per cent rate. For example, tax amounts of non-wholesalers can be computed for the total value of sales X 0.2 X 0.03, which is equivalent to the turnover tax with a rate of 0.6 per cent. The number of firms liable to such special rules statistically amounted to 96.7 per cent of the total in 1986. The special simplified scheme for computation may unduly render the figure smaller than the true value added at each stage. The turnover tax is not a true VAT. Thus, this may prove to be a flaw in a new VAT system because the turnover tax liability can be diminished by vertical integration, as stressed above. This possibility, however, does not seem to be practically crucial at the moment because the special simplified scheme was not available to bigger firms with annual sales of ¥500 million or more. Therefore, more importance might be placed on the reduction of administrative burden of true VAT on small and medium-sized firms. The option of turnover tax would certainly help them to calculate easily the tax due (see MOP 1990,161). The vanishing exemption method was also introduced to give the relief provision to smaller traders. Those whose annual sales do not exceed the maximum limit of ¥60 million (later ¥50 million) above the exemption level of ¥30 million can benefit from this method in terms of tax credit. In this case, the tax due gradually reduced to zero by a 'marginal deduction' that takes the form of tax credit. The calculation is as follows:
Let us assume a small trader at the retail stage with only ¥40 million of annual sales. Since he is eligible to use the special simplified scheme, the tax payable without 'marginal deduction' would be 0.6 per cent x ¥40 million, i.e. ¥240000. Under the calculation described above, the tax credit is (¥60 million— ¥40 million) •+• ¥30 million x ¥240 000, i.e. ¥160000. Thus, the tax due is only ¥80000 (i.e. ¥240000¥160000), which is one-third of the tax amount without this relief. The tax due becomes equal to 0.2 per cent of sales, not 0.6 per cent. In the case of a much smaller trader, say, with ¥30 million of sales, the tax vanishes at that level because the tax credit is essentially equal to the tax otherwise due.
The Value Added Tax
283
Table 15.4 Various tax rates applicable to different sales—the case of non-wholesale firms, 1989. Annual sales (¥m.)
Tax rate % 3
Methods Regular rate
500
0,6
Special simplified scheme
60
0-0.6
Vanishing exemption due to 'marginal deduction'
30
0
Exemption
0 Note: Drawn from the explanation of Shoup (1990).
The tax becomes negative at sales below ¥30 million, but there is no refund to the taxpayers.9 Taking account of these relief provisions for small and medium-sized firms, various rates of the consumption tax can be applied to different levels of annual sales. Table 15.4 shows four rates applicable to each stage of sales, depending upon special tax provisions in favour of smaller firms and traders. Relation to Existing Indirect Taxes In conjunction with the creation of the consumption tax, the whole system of indirect taxes needed to be restructured to harmonize with the existing excise taxes and the allocation of revenue sources between the national and local governments. As seen in Table 15.5, five national excise taxes were repealed as well as three local taxes. All these taxes were replaced by the new consumption tax on a nationwide basis. Of most importance was the absorption of the commodity tax into a new tax. The multiple-rate structure of the commodity tax was thus changed into a single rate of 3 per cent on all taxable commodities; as a temporary measure, only 9 On this point, Shoup states that 'So far as I am aware, this type of vanishing exemption is not found in other VATs, coupled with the tax-on-sales option, except, to a degree, in the new Canadian goods and services tax' (Shoup 1990, 438).
284
Indirect Tax System
Table 15.5 Tax variations in the national and local tax system under the Takeshita tax reform Items
National
Local
Creation Repeal
Consumption tax Commodity tax, playing-cards tax, sugar excise tax, admission tax, travel tax Individual income tax, corporate tax, inheritance tax, liquor tax, tobacco excise tax, petroleum tax tax, bourse tax, securities transaction tax, stamp tax
Consumption transfer tax Electricity tax, gas tax, timber delivery tax
Amendment
Inhabitant's tax, enterprise tax, real property acquisition tax, entertainment tax, tax on consumption at hotels and restaurants
Source: Data from the MOF.
passenger cars were taxed at the higher rate of 6 per cent during the transitional period.10 In addition, several other excise taxes were adjusted to coexist with the consumption tax. In principle, major commodities such as alcoholic beverages, tobacco, and petroleum are subject to both selective excise and general consumption taxes in VAT countries. The same procedures were applied in restructuring the indirect tax structure in Japan. Alcohol and tobacco taxes were initially applied at specific rates only at the manufacturer's level, and then an ad valorem rate of 3 per cent was levied as the consumption tax for distributors and retailers. Despite the combined taxes with different rates, total tax burdens were adjusted to remain unchanged. Taxes on petroleum, however, consisted of both the existing and new taxes, causing a heavier tax burden. Also, the securities transaction tax was reduced because of the increased capital gains tax on the sale of securities. Attention should also be paid to intergovernmental revenue transfers. Two measures were proposed when the consumption tax was introduced. In local taxes, several excise taxes were repealed or amended, producing a substantial loss of revenue. To compensate for revenue sources lost due to local tax cuts, one-fiftieth of the consumption tax was handed over to local governments as a consumption transfer tax, which in turn was allocated regionally among prefectures and municipalities according to population. Second, 24 per cent of the consumption tax at the national level (exclusive of the consumption transfer tax) was appropriated for tax-sharing grants—the local allocation tax (see Chapter 18). The overhaul of Japan's tax structure was nothing less than drastic. A reform of such magnitude had not been undertaken since the beginning of the Meiji period. 10 This measure was taken to adjust for too sharp a tax cut on certain commodities, i.e. the commodity tax rate on passenger cars ran as high as 23 per cent. Cutting such a high rate suddenly to 3 per cent was regarded as too drastic a change. At first, this temporary measure was intended to expire by the end of fiscal year 1991. However, it was decided to postpone it for two years at the reduced tax rate of 4.5 per cent when the budget for fiscal year 1992 showed a revenue deficit.
The Value Added Tax
285
When the Takeshita Tax Bill was approved by the Diet in 1988, some amendments were made to the consumption tax as well as the individual income tax (see Chapter 17). The consumption tax was slightly altered as follows. Although it became effective from April 1989, the first half-year up to 30 September was planned to be a period of preliminary enforcement. Reflecting a lack of familiarity with the new tax in Japan, the National Tax Administration was requested to enforce it less strictly, putting emphasis on public relations, guidance, and consultation.
ECONOMIC ASPECTS AND CONSEQUENCES
Effects of VAT on Prices
In all countries the public and the government have anticipated the increase in retail prices caused by the introduction of VAT with anxiety. Japan was no exception. On 1 April 1989, VAT known as the consumption tax was introduced in Japan to replace eight excise taxes including the commodity tax. Since the new VAT was applied to areas that were formerly exempt, in particular to foodstuffs, the change-over to VAT was in itself expected to raise prices to some extent. In theory, the likely impact of VAT on prices would just induce the once and for all expected increase in retail prices due to the 3 per cent of VAT rate. In practice, however, it was anticipated that traders might take advantage of the confusion to raise their prices unduly. Given such anticipation of price increases, the government took measures to introduce the consumption tax smoothly several months before the date of enforcement. First, a Special Council on the Transition to a new tax was established in the Cabinet on 6 January 1989 to consider any transitional problems, such as price increases, arising from the implementation of the tax. Through this council, each ministry and agency was requested to cooperate and to promote the enforcement of VAT. Second, the government carried out an extensive advertising campaign in newspapers, posters, TV, VTR, and radio. The main purpose was to reduce the public's fear of price hikes and to restrain possible cases of overcharging by traders. In addition, the government opened a telephone enquiries service where taxpayers could get help with the calculation of the tax due, and consumers could report complaints about prices.11 Third, the Economic Planning Agency strengthened the operation of the pricemonitoring system. Monitors' staffs were doubled, and prices of a wide range of commodities were monitored during the period before and after the introduction of VAT. In a word, the government did all in its power to effect the smooth adoption of the new tax. 1
' For the first four months from April 1989, the number of enquiries by telephone services on prices were 4 4 1 2 in April, 630 in May, 241 in June, and 140 in July. A substantial number of complaints were related to overcharging and charges by tax-exempt traders (sec Economic Planning Agency 1990a).
286 Table 15.6
Indirect Tax System Trends of the consumer price index (CPI) in a nationwide area (%)
Period
Total
Major items Foodstuffs
Fiscal year 1988 Fiscal year 1989 1989 Jan.-Mar. Apr.—June Apr.
May June July-Sep. Oct.-Dec. 1990 Jan.—Mar. Apr.-June
0.8
2.9
Public utilit)/
General goods and service
3.8 3.1
0.0 1.5
0.7 3.2
1.1 (-0.1)
2.8 (-5.6)
0.4 (0.0)
1.0 (-0.3)
2.8 (1.8)
4.1 (3.2)
1.3(1.0)
2.4(1.2) 2.9(0.5) 3.0(0.2)
1.5(1.0)
5.1 (2.6)
1.2 (0.9) 1.4 (0.2) 1.4 (0.2) 1.4 (0.3) 1.5 (0.2)
2.9 (2.6) 2.8 (2.0) 3.0 (0.6)
2.7 (0.5) 2.6 (0.5)
5.7 (-2.1) 0.4(0.1) -1.8(0.7)
3.3(1.0)
9.9 (5.7)
2.4 (0.6)
8.8(2.1)
1.6(0.1) 1.5(0.9)
3.2(0.1)
3.2 (0.2) 3.3 (0.8) 3.5 (-0.1) 2.3(1.4)
Note: Figures are percentage changes relative to the same period in the previous year. On the other hand percentages in parentheses are those relative to the previous quarter or month after seasonal adjustments. Source: Economic Planning Agency, 'On Price Trends after the Introduction of the Consumption Tax', 7 Oct. 1990.
In order to see the effect of the consumption tax on prices, percentage changes in the consumer price index (CPI) are presented in Table 15.6. The CPI in total rose by 2.9 per cent in fiscal year 1989 which was substantially higher than 0.8 per cent in the previous year. It appears that the new VAT was responsible for a rise in the CPI during fiscal year 1989. Special attention should be paid to the April-June quarter or the month of April 1989 when the new VAT came into effect. The CPI increased by 2.8 and 2.4 per cent relative to the same periods in 1988 respectively, but this was in part accounted for by foodstuffs. Basic price rises continued at a reasonable level until the end of 1989. In 1990, prices of foodstuffs suddenly rose considerably, but this was not related to the implementation of VAT. Increases in the general price level of public utilities, such as gas, electricity, transportation, education, etc., were much lower than other items. This was due to the fact that related authorities including local government had been reluctant to pass VAT on to public utility prices, despite strong insistence by the national government for forward-shifting. Further information may be derived from Table 15.7 in which we present the results of the sixth monitoring. Twenty-nine items targeted for monitoring are grouped in two categories: (1) items that VAT taxed for the first time, and
The Value Added Tax Table 15.7 monitor
287
Effects of VAT on prices of major commodities: results of the sixth price
1990
1989 Apr. 21 items chargeable under the new VAT Food (10) 2.7 General goods (6) 2.9 Services (5) 2.8 8 items with no commodity tax -6.0 Durable consumer goods (5) -1.4 Others (3)
May
June
Aug.
Nov. Feb.
0.3 0.2 0.5
0.1 0.2 0.2
0.1 0.0 0.1
0.1 0.1 0.1
0.1 0.0 0.1
-0.3 -0.3
-0.1 0.0
-0.2 0.0
0.0 0.0
-0.1 0.0
Note: The increasing rate of prices is the percentage change relative to the previous month. Source: As Table 15.6.
(2) items on which the commodity tax was levied no more. It was expected that there would be an increase in prices in the first category, while in the second there would be price reductions at least in the margin of the gap between the commodity tax and VAT. Based upon evidence of selected months following the change over, food, general goods including clothing, and services rose by 2.7-2.9 per cent for the first month, but at less than the VAT rate of 3 per cent. By contrast, durable consumer goods prices decreased by 6.0 per cent. Within one year, the rate of price increases varied as anticipated. All in all, the Japanese experience may be considered as an example of a smooth change-over to VAT with little net effect on prices. A combination of government guidance on prices, official monitoring, and public awareness may be behind this fact. Forward-Shifting Another important issue relates to forward-shifting. Put another way, to what extent should or could VAT be passed on to prices? Of course, VAT does not need to be shifted forward. Depending upon macroeconomic conditions (i.e. restraint of money supply or sluggish aggregate demand), it is quite possible to take the alternative way by which traders might be forced to absorb the burden of VAT. In Japan, however, it was proposed that the full amount of VAT liability should be ultimately passed to consumers by every trader, considering in principle the basic nature of Japan's VAT (see Tax Advisory Commission 1988). This implies that the enactment of VAT ought to be accompanied by repeal of the supply and demand rule in the markets. There was a general awareness that VAT, as a broad-based indirect tax, should be borne by everybody in a similar fashion, because it was thought of as a necessary revenue source to be appropriated towards an ageing society in the future.
Indirect Tax System
288
Apart from theoretical considerations, both traders and consumers expressed fears about passing on VAT. On the one side, traders, in particular smaller traders, constantly expressed their misgivings that they could not pass on the full amount of VAT, given competitive market conditions. If so, they must absorb some portion of VAT, reducing their profits, or passing it backward in the form of lower factor rewards. On the other hand, consumers were likely to be aware of a major tax change and to expect the worst. They anticipated that traders might tend to raise prices exorbitantly resulting in price increases exceeding the rate of VAT. Moreover, they considered that even tax-exempt traders might pass VAT on to prices. Since VAT was implemented in the midst of these mutual anxieties, it is of the great interest to examine how perfectly or to what extent the VAT could be shifted forward. There are several kinds of research data available to us, which were mostly collected through a questionnaire method by several ministries and agencies. The MITI, mainly in charge of industrial policy, was very keen to investigate the effects of VAT on prices after the imposition of the consumption tax. The MITI attempted to obtain from about 4 000 traders monthly reports about price changes in 147 major goods and services items affected by VAT. Table 15.8 shows to what extent traders at each stage could pass forward the VAT. Enquiries are classified into three categories: 'Almost shifting', 'To some extent', and 'Almost not shifting'. As far as this study was concerned, 84.9 per cent of the total affirmed the feasibility of forward-shifting. In particular, nearly 100 per cent of manufacturers successfully passed the VAT burden on to prices. Conversely, retailers had some difficulty in shifting it forward to consumers. As compared with the results of the month in parentheses, it is clear that it was becoming easier for traders to pass on VAT. From its investigation, the MITI came to the conclusion that 'forward-shifting progressed almost smoothly'. It can easily be seen that shifting would be different for larger traders than for smaller traders. In particular, retailers must be affected by business size: the bigger Table 15.8
Extent of forward-shifting, May 1989 (%)
Stages
Extent of shifting the new VAT on to prices
Unknown
No. of respondent traders
Almost shifting
To some extent
Almost not shifting
Manufacturer Wholesaler Retailer
99.8 (98.6) 98.8 (96.6) 75.9 (73.0)
0.2 (0.4) 0.6(1.1) 9.8 (10.0)
0.0 (0.0) 0.5 (1.3) 14.1 (16.6)
0.0 (1.0) 0.1 (1.0) 0.2 (0.4)
406 1 119 2402
Total
84.9 (83.6)
6.2 (6.0)
8.8 (9.8)
0.2 (0.6)
3927
Note: Percentages are the relative share of respondents to the question 'To what extent can you shift forward the consumption tax on prices?' Figures in parentheses indicate relevant results, April 1989. Source: MJTI, 'On the Shifting of the Consumption Tax" (in Japanese), 28 July 1989.
The Value Added Tax
289
retailers are, the easier it would be to shift the VAT to consumers. Table 15.9 has been prepared to show this. The enquiries were the same as in Table 15.8, but retail traders are divided into three classes, depending upon the level of their taxable sales. Four selected months are taken to indicate possible changes in responses from questionnaires. There are three interesting points worth noting. First, large traders above ¥500 million taxable sales were able to pass on the tax relatively easily, the tempo accelerating during the first half of the year. Second, medium-sized traders felt the feasibility of shifting to a lesser extent, but more than 80 per cent were successful in passing on their VAT liability. Third, smaller traders found it more difficult to shift their VAT burden on to prices; in their case shifting was lowered from 60.0 per cent at the beginning to 43.6 per cent in July. The Fair Trade Commission also published the results of monitoring the relationship between the rate of VAT and price changes.12 Particular attention was paid to whether the 3 per cent of VAT rate could be passed on to prices in full or not. In Table 15.10 three cases are illustrated as (a) full-shifting, (b) over-shifting, and (c) under-shifting at the retail stage. It is interesting to find that 83.1 per cent of all
Table 15.9
Extent of forward-shifting by firm size at the retail stage, fiscal 1989 (%)
Taxable sales (¥m.)
Traders with sales of more than 500 Traders with sales of 30-500 Traders with sales of less than 30
Month
Apr. July Oct. Jan. Apr. July Oct. Jan. Apr. July Oct. Jan.
Unknown
Extent of shifting the new VAT on to prices Almost shifting
To some extent
Almost not shifting
84.8 95.8 98.8 96.3 73.0 83.7 84.5 82.9 60.0 43.6 43.7 49.8
5.2 2.2 0.6 0.6 10.8 9.3 9.1 9.6 14.0 18.4 18.9 16.6
9.4 2.0 0.6 0.6 15.8 6.9 6.3 7.4 25.9 37.6 37.4 33.6
0.6 — — — 0.4 0.1 0.1 0.1 0.1 0.4 — —
No. of respondent traders
672 753 652 679 1102 1025 1090 1 102 609 560 636 590
Note: As Table 15.8. Source: MITI, 'Actual Performance of Implementing the Consumption Tax' (in Japanese), 7 Aug. 1990.
12 Through 1000 monitors, two researches were carried out (i.e. (1) 19-25 March, and (2) 19-25 April) about 20000 commodities in order to investigate price changes and pricing methods (i.e. VAT-quoted pricing or not). Whether or not 3 per cent VAT rate was passed on was derived from the comparison of the price level at two different dates (see Fair Trade Commission 1989).
290 Table 15.10
Indirect Tax System Full-, over-, and under-shifting at the retail stage by business type, April 1989
Type
No. of goods atid services More than 3% on prices
Less than 3% on prices
663 (90.0) 5819(87.9)
40 (5.4) 420 (6.3)
34 (4.6) 378 (5.7)
737 (100.0) 6617(100.0)
4303 (90.5)
251 (5.3)
202 (4.2)
4756(100.0)
4028(68.9)
597(10.2)
1 224 (20.9)
5849(100.0)
1571 (89.6)
74 (4.2)
108(6.2)
1753 (100.0)
3% on prices
Department stores Large-scale supermarkets Small and medium-size supermarkets Small and medium-size retail shops Co-op and agricultural co-operatives Others Total
166(83.0)
14(7.0)
16550(83.1)
1 396 (7.0)
20(10.0) 1966(9.9)
Total
200(100.0) 19912 (100.0)
Note: Figures in parentheses are per cent distribution in each column. Source: Fair Trade Commission, 'On the Results of Monitoring Price Changes and Pricing Methods after the Consumption Tax was Implemented" (in lapanese), 23 May 1989.
retailers could pass on the full amount of VAT. Over-shifting accounts for only 7 per cent. Department stores shifted almost perfectly their VAT liability on to price while retail shops experienced under-shifting in a substantial way (i.e. 20.9 per cent). Other types of businesses showed more or less similar phenomena in the patterns of shifting. The shifting of the new VAT might, therefore, be considered to have been satisfactory as a whole with the exception of smaller traders, such as small-sized firms. The ease or difficulty of shifting VAT on to prices may have a close bearing upon alternative pricing methods. If prices are without VAT (i.e. Sotozei: VAT-exclusive pricing), consumers are then exposed to the shock of VAT (i.e. ¥100 plus ¥3 tax) when they pay, but traders can pass on VAT by agreement with consumers. By con trast, if prices include VAT (i.e. Uchizei: VAT-inclusive pricing), the tax burden is concealed from the taxpayer (i.e. ¥103 including ¥3 tax). Over-shifting may occur in this case. Some people like the VAT content of a price to be shown separately from the tax-free price in order to understand fully their tax liability. They prefer the Sotozei. On the other hand, others may choose to be aware of the full cost of a good including tax and may not want to be faced with an additional payment at the cashier's. This is the Uchizei case. From the beginning of implementing the consumption tax, the government left the choice of the two different pricing methods to the traders themselves, and it is of interest to know which of the alternative pricing methods was adopted by the majority of traders. The Fair Trade Commission showed an interesting result from
The Value Added Tax
291
its investigation held in April 1989 (Fair Trade Commission 1990) as follows: Total items VAT-quoted price Without VAT (Sotozei) With VAT ( Uchizei) VAT-concealed price (no mention of VAT)
19912 (100.0) 17412 (87.4) 13930 3482 2500 (12.6)
At the time of investigation, the ratio of pricing including VAT as against pricing without VAT was 20:80. Most traders preferred the clearcut pricing of the Sotozei method to convince consumers of forward-shifting. Judging from my own experience, it seems, however, that the ratio has been reversed in recent years, consumers preferring to pay the tax without noticing the tax burden, i.e. the Uchizei case. Administration and Compliance Evidently, the introduction of VAT was a major upheaval in the tax system, causing both administrative and compliance considerations. The government was greatly concerned with the practical difficulties of administering the tax, while traders believed that it would impose very heavy costs on those obliged to account for it. In Japan, more importance has been placed on traders' compliance than tax administration by the VAT authorities. As noted above, a new VAT was introduced, eliminating the existing excise taxes (see Table 15.5). Thus, there was no need to establish a completely new authority for collecting VAT, and the existing tax staffs were empowered to administer the new tax. Since they are highly trained staffs, moving to VAT did not involve extensive retraining and recruitment. In general, th consumption tax organization was neither understaffed nor overburdened, and in turn tax administration as a whole remained almost unchanged13 after the implementation of VAT. On the other hand, the traders' reaction posed more important problems that depended upon the businessman's feelings about the compliance costs of VAT. Traders anticipated that VAT was the most serious burden that had been officially created on business. Thus before the new tax started, there were a number of fears about the likely burden of compliance costs among traders in general, including computer facilities. Despite prevalent fears in advance, the implementation of VAT seems to have been much smoother than had been anticipated. Generally speaking, both traders and consumers behaved well and co-operated with the tax authorities, although anti-VAT sentiment still remained strong among the general public. Several facts 13 In 1991, however, the organization of the NTA was reshuffled to a great extent, associating the administration of VAT with individual and corporate income taxes. In particular, this was promoted to make it easier for taxpayers to file returns for both the consumption tax. and income taxes.
292
Indirect Tax System
became evident from reviews attempted by the NTA to investigate actual performance of the consumption tax for the previous year.14 First, as to how taxpayers calculated the tax payable: 19.6 per cent of the respondents said that they paid the tax on a self-assessed basis without any help from tax professionals, because they could understand almost fully the amount of VAT to be paid; by contrast, the remaining 80.4 per cent relied upon professional knowledge to compute the tax due. Obviously, the larger the firms are in terms of annual sales, the more likely they are to rely upon the skills of professionals. Thus, the costs of compliance with government requirements were higher than previously. Second, 72.0 per cent of total respondents complained about additional compliance costs for paying VAT, while 28.0 per cent responded that there was no cost required to comply with tax offices. Among the more detailed items of new costs for the majority group, the top three were (a) purchase of tax-related stationery goods, (b) the provision of information for customers, and (c) the change of bookkeeping records. Computer hardware and software were less costly than had been expected. Third, it is interesting to focus on to what extent the special simplified scheme was used. Table 15.11 shows the relative proportion of VAT taxpayers using this special provision during fiscal year 1989. A total of 67.7 per cent of all traders applied to use the special simplified scheme, a proportion smaller than expected. No doubt, smaller traders tended to employ this scheme more intensively. Even tax-exempt traders with less than ¥30 million in taxable sales made an entry as taxable firms so as to obtain tax-credit on purchases.15 Fourth, note should be taken of the realized ratio of value added in gross sales by type of industry. This is important because rough approximate percentages (i.e. 10 per cent for wholesalers and 20 per cent for the rest) are available to traders to compute deemed value added instead of the tax base if they apply for such a special provision. As seen in Table 15.12, value added ratios spread over a wide range from 14.8 per cent of wholesale to 45.6 per cent of real estate. From this, it is evident that the special simplified scheme impaired the fairness of taxation by type of industry. In particular, service-related industries can greatly benefit from special relief in terms of a deliberate lower ratio of taxable value added. Lastly, it is of great interest to examine the consequences of using the vanishing exemption system. As indicated in Table 15.11, the number of eligible traders in the range of ¥30-¥60 million in taxable sales accounted for 546900 in 1989. Among them, 510277 traders employed vanishing exemption whose ratio is 93.3 per cent. The average amount of tax credit with 'marginal deduction' is ¥119 000 compared with average tax otherwise due in this range—¥135000 (=¥45 million X 0.03). Obviously, the tax payable has been reduced to ¥16 000 (=¥135 000-¥119 000). 14 The NTA attempted a substantial amount of reviews to clarify actual consequences of VAT one year after its enforcement (see NTA 1990«, b). 13 Of course, another reason should be related to the gap of annual sales between the base and tax years (see note b in Table 15.11).
The Value Added Tax
293
Table 15.11 Small and medium traders applying to the special simplified rule as a percentage of total taxpayers, 1989 Taxable sales3 (¥m.)
Total taxpayers
(2)/(l)
(1)
Traders with special simplified rule (2)
Less than 30 30-60 60-100 100-200 200-300 300-400 400-500 Sub-total More than 500
212881 546900 363063 342 924 134238 73510 46871 1507506 212632
132697 442882 279435 256337 95848 49147 27523 1 151172 24449
62.3 81.0 77.0 74.8 71.4 66.9 58.7 76.4 11. 5b
Total
1 933019
1308318
67.7
(3)
a
Figures in 1988 (previous year). Since traders using the special simplified scheme are registered during the base period (i.e. two years before the relevant tax period—in 1987), their taxable sales in 1988 exceeded the threshold of less than YSOOm., reflecting a growing business activity in the past. Source: Ministry of Finance, 'On Data Collected to Investigate the Actual Performance of the Consumption Tax', 7 Aug. 1990. b
Table 15.12 Value added ratio by type of industry under the application of the special simplified scheme Type
Ratio
Type
Ratio
Wholesale Retail Construction Mining Agricultural, Forestry, and Fisheries
14.8 21.4 23.1 29.0 31.0
Manufacturing Services Transformation and Telecommunications Real Estate
33.4 39.3 45.2 45.6
Source: As Table 15.11.
AMENDMENTS AND
FURTHER
REFORM
Towards Potential Amendments Contrary to earlier fears and anti-sentiment against the consumption tax, the new tax settled down smoothly in the Japanese tax system only a year after it was enacted. Politically, however, controversial arguments still continued between the opposition parties and the ruling LDP. The former strongly persisted in demanding
294
Indirect Tax System
the repeal of VAT from the beginning of its implementation, while the latter gradually had to respond to such political attacks by rectifying some parts of the tax. What were the controversial issues between the LDP and the opposition parties on the original VAT? The three points in dispute were as follows: 1. Regressive tax burden among lower income-earners. 2. Cash-flow benefits for a period before tax revenues are handed over to the tax authorities. 3. Windfall revenue gains under the special simplified scheme with deemed value added. All these are related to the major complaints of consumers when such a tax was implemented. As a consequence, these controversial issues provided the broad basis for the amendments to the current consumption tax effected from 1 April 1989. The first point concerning regressivity remained predominant for a long time among the opposition parties supported politically by the anti-VAT movement. Whenever Japan's VAT was proposed in past tax reforms, the most criticism was con centrated on its basic nature as a regressive tax and the distributional consequences. This was the reason why the number of exempted items in the sales tax had to be increased from the original seven to fifty-one to mitigate political pressure under the Nakasone cabinet. Generally speaking, it is widely acknowledged that regressivity could be lessened to some extent if some essential goods and services (e.g. food, education, health, housing, public transport, etc.) were exempted from taxation. Needless to say, the enlarged scope of exemptions distorts the fundamental structure of VAT, impairing uniformity and scope which is of great advantage to tax neutrality. The government persistently maintained that distributional issues could be better served by progressive income taxation and by carefully targeted transfer payments to the poorer households. Politically, however, the increased use of exemptions from VAT has been getting more popular support from consumers. Second, greater attention has been put to cash-flow advantages where traders have the use of tax revenues before passing them on to the tax authorities. Originally, traders were allowed to file a return and pay the tax due twice a year (including once for interim payment) if their tax payable in the previous year exceeded ¥600 000. Otherwise, tax payment was only once a year in the case of small traders. This rule implies that a specified 'grace period' is substantially long enough to cover more than several months. During such a period, traders who collect the tax on behalf of the tax authorities and are allowed to retain it can obtain, in effect, interest-free loans. Perhaps it would enable them to earn additional interest by investing retained cash flow in the portfolio market. The benefits that traders derive during the 'grace period' have been criticized greatly by consumers in view of tax equity. As a third point, great emphasis has been placed on inequitable aspects of windfall revenue gains caused by the application of the special simplified scheme and vanishing exemption. In particular, deemed value added (i.e. 10 per cent for wholesalers and 20 per cent for the rest) poses a problematic underestimation of tax base.
The Value Added Tax
295
As seen in Table 15.12, big differences can be observed between the approximate ratio and the actual ratio of value added by type of industry. Evidently, service-orientated industries really enjoyed revenue gains as a result of the wide margin between the two relevant ratios. Similar to the second issue, such gains accruing to the traders were considered to be unfair by the general public. Process of Amending the Original VAT
In the face of strong anti-VAT protests in general, the government was greatly concerned to see how smoothly the new tax could be implemented. Particularly, the anticipated general elections to Upper and Lower Houses were very important to judge publicly the response to the introduction of VAT. Table 15.13 tabulates events after the consumption tax was enforced in April 1989 until amendments were finally made in October 1991. Successive events during the period in question occurred as a result of political struggles and compromises. When the election to the Upper House was held in July 1989, the LDP's political powers were lessened to a great Table 15.13 1989 1 28
April June
23 29
July Sept.
24 1 16 1990 18 6
Nov. Dec. Dec.
19
April
26
June
1991 25
April
8 1
Chronology of events after implementing the consumption tax
Feb. March
May Oct.
The introduction of the consumption tax The Tax Advisory Commission established subcommittee on a follow-up of the consumption tax. The election to the Upper House. Opposition parties presented to the 116th Extraordinary Diet the first 'Tax bill for repealing the consumption tax'. Interim report from subcommittee of the Tax Advisory Commission. LDP decided basic principle of amending the consumption tax. The Diet was closed. Tax bill was withdrawn. The general election to Lower House. Cabinet approved the 'Tax bill for the amendment of the consumption tax which was presented to the 1 1 8th Extraordinary Diet. Opposition parties presented to the Diet the second 'Tax bill for repealing the consumption tax. The Diet closed, and two different types of tax bills were withdrawn. Llowever, a loint Committee on Tax Issues was established at the Diet to discuss the future treatment of the consumption tax in co-operation with both the LDP and opposition parties. Joint Committee presented to the Diet the 'Amended tax bill of the consumption tax. Amended tax bill passed the Diet. A new consumption tax started.
Source: Data submitted to the Tax Advisory Commission.
296
Indirect Tax System
extent after it lost a substantial number of seats. Throughout the campaign, continuance or repeal of the consumption tax was the main issue of dispute between the LDP and opposition parties.16 If the LDP had failed once again in the general election to the Lower House in February, 1990, the consumption tax would have disappeared from the Japanese tax system simply for political reasons. Fortunately, Japan's VAT was maintained as a result of a political victory of the LDP. The process of rectifying the structure of the consumption tax was divided into two stages. The first was derived from a 'Tax bill for the amendment of the consumption tax approved by the cabinet in March, 1990, but it was not successful. The other was based upon an 'Amended tax bill of the consumption tax' presented by the Joint Committee in April 1991, which led to the revised form of the current consumption tax. During this period, the opposition parties submitted twice a 'Tax bill for repealing the consumption tax' which was strongly against the continuation of VAT. The first amendment plan was mainly composed of three features as follows: 1. Food is fully exempted only at the retail level, while half of VAT rates (i.e. 1.5 per cent) is applied to inter-firm transactions on the same product. 2. The scope of exemptions is expanded to cover (1) birth expenses, (2) cremation and burial costs, (3) certain goods and services for disabled persons, (4) certain welfare services, (5) education, and (6) housing rents. 3. The number of tax returns and payments increased from once a year to four times for larger traders to shorten the 'grace period'. As is evident from Table 15.13, the amended scheme of VAT did not come into effect when the tax bill was withdrawn at the 118th Extraordinary Diet in June 1990. We must conclude that this was really a good decision. If the revised form had passed the Diet, a very strange style of VAT would have been created. In fact, special treatment of food (i.e. the combination of zero rate and reduced rate on different stages of transactions) is not feasible compared with other VAT countries. This strange idea resulted in a political conflict with the opposition parties that had constantly promised to repeal VAT or at least exempt food. The next amendment, which was a successful trial, was derived from a joint agreement between the ruling and opposition parties. As a consequence, the consumption tax was altered to a considerable extent on the basis of the fundamental direction mentioned above. Main features of the new VAT are summarized in the following three points. 1. A specified 'grace period' is shortened by increasing the number of tax returns and payment from twice a year to four times. This applies only to large traders whose tax due exceeds more than ¥5 million. 16 Of course, there were other issues in dispute, besides the consumption tax, such as the scandal involving cabinet members and the elimination of protection for rice products.
The Value Added Tax
297
2. Windfall revenue gains are lessened by establishing new requirements; (a) Deemed ratio of value added is diversified into four categories; 10%—wholesale 20%—retail 30%—agriculture, forestry, fisheries, mining, construction and manufacturing 40%—others, such as transport, telecommunications, real estate, restaurants, etc. (b) The maximum level applicable to the special simplified scheme is reduced from the current ¥500 million to ¥400 million in terms of taxable sales. (c) The same holds for the vanishing exemption whose maximum level is reducing to ¥50 million from the current ¥60 million. However, the exemption level remained unchanged at ¥30 million, although many asserted the necessity of reducing it to as low as ¥10-20 million. 3. Exempted goods and services are expanded in a way similar to the first amendment plan listed above. As a result, the new consumption tax had two side-effects. No doubt the first and second points contributed towards the improvement of VAT, but VAT was worsened by the third point, the eroding tax base. The new tax began in October 1991, although it should remain unchanged for a time. At that time it was expected that the next big issue would perhaps be whether tax rates should be raised to the range of 5-10 per cent.
The Debut of Local Consumption Tax Since the consumption tat was introduced in 1989 by the national government, local governments have begun to have big complaints about their share of revenue sources. As is seen in Table 15.5, local governments repealed three excises and amended two excises at the local level to be integrated into the (national) consumption tax. Instead of them, these lost revenues were automatically compensated for by intergovernmental transfers from the national government in terms of the consumption transfer tax. Although revenue sources are guaranteed by a new device, the consumption transfer tax cannot provide any independent taxes which local governments collect by their own system of tax offices. It is strongly protested by them that local autonomy cannot be maintained as long as major taxes are not collected under an independent institution of tax collection, separate from that of national government. As a consequence, local governments insisted that the share of local independent tax sources should be increased when consumption tax rate was lifted in 1997 up to 5 per cent from the current 3 per cent. There was in principle no problem to shift financial sources corresponding to the consumption transfer tax from the national to local governments. The problem was what type of tax instrument should be employed for this purpose. To construct such
298
Indirect Tax System
a new tax formula, the Subcommittee on Local Consumption Tax was established in 1994 within the Tax Advisory Commission which was composed of about ten academic tax specialists. As a result of controversial debates among subcommittee members, the local consumption tax was designed and was passed in the Diet in autumn 1994. The most difficult issue concerning such a new tax was how well it was possible to formulate the basic nature of local consumption tax in the discussions of the Subcommittee. The current consumption tax at the national level is levied in the nationwide territory of Japan, whose revenues are not necessary to be divided into each region. Theoretically speaking, it is impossible that such a multistage sales tax as VAT could be introduced into the local tax system, because any taxing point is often different from the ultimate place where consumers will bear the tax. Thus, any adjustment must be made inter-regionally from one taxing place to another. At the outset, the MOHA maintained that consumption taxes paid by traders in A region should be attributed to the local government in A region. This implied that the percentage consumption tax would be similar to an income-type of VAT, which is assumed to be non-shiftable to ultimate consumers. No doubt, this was off the mark because the current consumption tax was introduced as a consumption-type of VAT. Finally, this view was abandoned. As a result of heated arguments in the Subcommittee, some compromise was made between the MOP and MOHA in order to create a new tax. The basic framework of the local consumption tax under consideration is summarized in Figure 15.1.
Fig. 15.1
Mechanism of the local consumption tax
The Value Added Tax
299
1. 2. 3. 4. 5. 6. 7.
Taxing authority; prefectural government Taxpayer; the same as the (national) consumption tax Tax base; the same as above. Tax return; the same as above. Tax rate; 25 per cent of the consumption tax. Tax collection; national tax offices. Liquidation of taxes among interprefectures; local consumption taxes paid to each prefecture are redistributed to relevant prefectures by employing consumption-related data on a prefectural basis. 8. Transfers to municipal governments; after liquidation, half of prefectural revenues is transferred to subordinate municipalities in response to the number of the population and of workers.
Obviously, a new local consumption tax is far from satisfactory. Despite the original idea proposed by local governments, this is not an independent local tax, separate from the national tax. Tax collection is still under the control of national tax offices, although tax inspection can be admitted to a local tax authority to some extent. It must be stressed that the present form of local consumption tax should be thought of as an imperfect and provisional one. It seems that there are three directions towards improvement. First, independently of the (national) local consumption tax, the local goods and services (GS) tax should be designed after expanding the tax base more broadly with an inclusion of expenses for hotels, restaurants, entertainment, etc. Second, the present tax should be turned into an income-type of VAT, integrating the amended local business tax. For this purpose, the current form of local business tax must also be changed into a new tax whose base should be broadened to cover not only profits but wages, interest, rent, and depreciation. Third, apart from a multistage consumption tax similar to the present VAT, the retail sales tax might be taken into consideration in the local tax system, like that imposed by US state governments.
16
Design of Environmental Taxes Worldwide attention may, in the near future, be paid to whether or not environmental taxes should be practically introduced into the Japanese tax system. This argument is, however, still immature, considering the recent sluggish growth of the Japanese economy. Nevertheless, as the economy enters the phase of recovery in the future, the pros and cons of environmental taxes, like the value added tax (i.e. the consumption tax in Japan) will no doubt become one of the most controversial policy issues in the twenty-first century. Today, Japan is expected to take initiatives to solve global pollution problems via bilateral or multilateral negotiations in the international community through its financial resources. The main aims of this article are twofold. One is to clarify the background of environmental issues and protection in the past. The other is to explore possible ways of designing a desirable form of environmental taxes from the standpoint of economic and environmental policies. Based upon these considerations, the introduction of a new carbon/energy tax is recommended in Japan to solve global environmental problems in the future. GENERAL BACKGROUND OF ENVIRONMENTAL POLICIES
Pollution and Control During the earlier period of postwar reconstruction in the 1950s and 1960s, great efforts were made to achieve rapid economic growth through business investment. Unfortunately, this growth-oriented policy performance was adopted by both private and public sectors without proper attention to the environment. As a consequence, it led not only heavy pollution and irreversible damage to the natural environment, but also resulted in serious health problems, such as Minamata or Itai-Itai diseases, Yokkaichi asthma, etc. Public concern prompted the adoption of pollution protection policies to avoid great damage to the natural environment as much as possible. In concrete terms, the Basic Law for Environmental Pollution Control was enacted in 1967, and furthermore the Environmental Agency was established in 1971 as a primary organizatio in charge of the environment, under the Prime Minister's office of the government. Both contributed a lot to providing the main basis and impetus for major achievements in relation to pollution control efforts in environmental conservation. This chapter is based on Ishi (1995a).
Environmental Taxes in Japan
301
Structural changes in the Japanese economy, caused by two oil shocks in the 1970s, greatly affected the process of environmental conservation and control. As a result of the increased price of crude oil, energy-saving behaviours were adopted in industrial activities with great emphasis on the environment, reducing the pollution burden to a great extent. Heavy-polluting industries were socially criticized and obliged to be equipped with anti-pollution measures. Oil crises were evidence that the Japanese economy was founded upon a very vulnerable base of import dependency for basic raw materials. Thus, with the constraint of energy and raw materials, the rate of real economic growth essentially decreased from 10 to 5 per cent. In turn, a slowdown of Japan's growth coupled with the promotion of pollution control measures and increased energy efficiency in the 1980s led to a reduction in the environmental problems. On this point, the OECD commended environmental policies in Japan as follows: Over the past two decades Japan has had the largest economic growth of G7 countries, while substantially reducing emissions of a number of pollutants in the atmosphere and toxic substances in water, and further containing the growth of other pollutants and of waste production. For instance, while economic growth increased over the period by 122 per cent, SOX emissions decreased by 82 per cent and NOX emissions by 21 per cent, the best performance among OECD countries. This decoupling was achieved through economic structural changes, increased energy efficiency and effective environmental policies. These successes have proved that environmental policies and economic development can be mutually supportive; the competitiveness of Japanese industry has not suffered overall and has even benefited in some sectors (e.g. the automobile industry and the pollution control equipment sector). (OECD, 1994,182.)
As noted above, pollution control policies in Japan have been quite efficient with a number of success stories. As far as pollution control is concerned, Japan's performance has recently been highly evaluated by other industrialized countries. However, these remarkable results of pollution control are limited to specific regions and sectors. In fact, great efforts have been made to abate the damages of air, water or waste pollution in the natural resources. The Emergence of Global Environmental Problems Apart from the regional scope of environmental problems, new forms of pollution and environmental deterioration have emerged since the 1980s. Long-standing pollution problems in the specific regions are still acute, but at the same time global environmental issues now emerge, from acid rain to global-warming, ozone-layer depletion, etc. As environmental problems are rapidly widening their scope on a global scale, the existing pollution control measures cannot cover the entire spectrum of environmental policies. Particular attention is now paid to the environmental damage of global warming, caused by emissions of greenhouse gases mainly due to carbon dioxide (CO2) emissions. It is generally predicted that, given the projected level of CO2 emissions,
302
Indirect Tax System Table 16.1
CO2 emissions by major countries, 1991 CO2 emission (million M/T)
USA USSR China Japan Germany UK Canada Others
Total
Per cent distribution (%)
4932 3581 2543 1091 970 577
21.7 15.8 11.2 4.8 4.3 2.5
410 8569
1.8 37.9
22673
100.0
Note: Data are (jO2 emissions from i:ndustrial processes in terms of million tonnes. Source : World Re'sources 1 994-95, Table 23.1.
the average temperature might rise by around 4 degrees over the next 100 years. This trend is expected to continue. Although these estimates are no doubt uncertain, one of the most important objectives of environmental policies is obviously on the reduction of CO2 emissions to some target level. As a matter of fact, an international agreement that CO2 emissions should be reduced as compared with the 1990 level by the target period of 2008-12 was made by the COP3 (the third Conference of Parties) to the UN Framework Convention on Climate Change at the Kyoto meeting in 1997. According to the Kyoto protocol, Japan has to cut its CO2 emissions by 6 per cent, USA by 7 per cent, and the EU by 8 per cent. As regards the reduction of CO2 emission in the world, Japan is also primarily responsible for achieving this target by using some policy instruments. Table 16.1 shows that Japan's relative share of total emissions of CO2 in the world was 4.8 per cent in 1991. Japan was therefore ranked as the top fourth nation, following the USA, the USSR, and China. In 1995, the ranking of Japan was the same with 4.8 per cent of its relative share. Today, it is widely acknowledged that Japan should play a leading role in reducing CO2 emissions, not only for itself, but for all other countries in order to solve the problem of global warming. Since the Rio Environment Summit was held by the United Nations in June, 1992, the necessity to develop a new framework of the basic law for environmental policies as a whole has widely been recognized among the general public in Japan. Although the passage at the Diet was delayed by the dissolution of the Lower House, the Basic Environment Law finally became effective on 19 November 1993. The contents of the Law are far from clear-cut, and it contains vague interpretations in many parts, reflecting policy struggles among several related ministries and agencies. In particular, there have been repeatedly controversial arguments as to the use of economic instruments which implies the introduction of a new environmental
Environmental Taxes in Japan
303
tax (say, a carbon tax). The MITI has strongly resisted the adoption of economic instruments via market mechanisms in environmental policy, mainly reflecting the MITI's concern over its negative impact on the development of international trade and economic growth. Thus, the Environmental Agency in charge of enacting the Law was obliged to recede from this initial position to a great extent. In Article 22 of the Basic Environment Law, the term 'economic measures not 'instruments', is referred to as a policy 'to prevent interference with environmental conservation.' Main points are stated as follows. The government shall appropriately conduct surveys and researches on the effectiveness of implementing such measures with regard to prevention of interference with environmental conservation and on the effects of such measures on the Japanese economy; and should it be deemed necessary to implement such measures, the government shall make efforts to seek the understanding and cooperation of the people with regard to utilization of such measures to prevent interference with environmental conservation. In this case, should such measures be implemented for global environmental conservation, the government shall consider international collaboration so as to appropriately ensure the effectiveness of such measures.1
In spite of the critical views noted above, the Law plays a very important role in defining the basic nature of environment as a public good rather than a free good. It is generally believed that it has provided the fundamental base for discussions on the introduction of economic instruments in light of taxes and charges. Towards the Use of Economic Instruments The environmental tax, regardless of the way in which it is formulated, is one of the typical economic instruments. Therefore, discussions in favour of environmental taxes have been developed, closely tied with the recent trend of supporting the adoption of market forces and economic instruments. Anticipating the future expansion of global environment issues, an important consequence is that economic and environmental policies cannot be separated. Effective integration between the two is needed to solve environmental problems on a global scale from the standpoint of greater economic efficiency. No doubt, it is important to evolve a wider use of market mechanisms through economic instruments in order to improve this efficiency. Since the early 1980s, the move towards the use of economic instruments in environmental policy has increased sharply, in particular among OECD countries. Japan follows this trend with a substantial time lag, and is now trying to catch up with the advanced level of adopting economic instruments performing in some OECD countries, say the Nordic countries. According to the general classification of economic instruments, there are four types: (1) taxes and charges, (2) subsidies, (3) tradable permits, and (4) deposit-refund 1 See an English excerpt from the Basic Environment Law (Environmental Agency, 1993). The original text in Japanese is much more difficult to understand. 2 This is strongly promoted by the OECD. Sec (e.g. OECD 1991a,fc).
304
Indirect Tax System
systems.3 Our primary concerns are with taxes and charges among these four types for several reasons mentioned below. Economic instruments should be contrasted with direct regulation within the broader scope of environmental policies. Traditionally, in carrying out environmental policies top priority has so far been placed on direct regulation of environmental damage and pollution in Japan. The MITI still attaches greater importance to this policy and is therefore likely to continue. The 'command-and-control' type of regulation has often generated prompt and remarkable results of pollution abatement, coupled with strict enforcement of emission-restraint standards by government intervention. A number of successes in overcoming the issues of industrial pollution were induced by direct regulation.4 Thus, the first priority tended to be put upon the regulation-type of environmental policy in Japan. As the second environmental policy, subsidy policies have frequently been employed in the form of 'tax expenditures' (i.e. disguised subsidies), rather than direct subsidies from the expenditure side of budget. There are no direct subsidies for environmental protection given to private enterprises with a few exceptions (e.g. R&D in a specific area) in Japan. Typical cases of tax expenditures are tax exemption, special or accelerated depreciation, tax credits and special deduction to encourage the development or use of certain techniques or technologies for pollution abatement and energy saving.5 'Environmental policy' in Japan simply means such indirect subsidies as tax concessions, which are listed in special tax measures of national and local tax systems.6 These two instruments that we have traditionally employed could not be justified exclusively at present, given the current state of environmental problems. First of all, the emergence of global warming makes direct regulation through government intervention ineffective or almost impossible. The effect of regulation is regionally restricted to a narrow district, not enlarged to a global region. Second, there is one important practical consideration that counts against the use of environmental subsidies. These subsidies tend to provide a form of protection for the industries concerned, and it may be relatively easy for protectionist pressures to increase subsidies for the reason of environment conservation. Obviously, such indirect or concealed 3 See e.g. OECD (1989, 1993). In addition to these four types, sometimes other categories such as market creation, financial assistance or enforcement incentives are employed in OECD documents. As regards the Japanese environment policy, see Kazu Kato (1993). 4 One example is shown in the case of NO., reduction for passenger cars. By strong regulation of emission gas control at the manufacturing level, emissions of NO^ have drastically decreased by a big margin of 92 per cent for five years from the enforcement of this regulation in 1972 (Source: data presented to Central Environmental Commission by the Ministry of Transport). 5 Typical examples are (1) special initial depreciation for solar and energy-saving equipment, pollution-preventing equipment, and recycling equipment, and (2) tax reduction for air, water, and noise abatement equipment, asbestos emission reduction facilities, oil desulphurization facilities. 6 Special tax measures in relation to environmental damage abatement are widely applied not only to personal and corporate income taxes, the inheritance tax, the stamp duties and register-licence tax in the national taxes, but also to the fixed asset tax, special land-holding tax and business-site tax in the local taxes.
Environmental Taxes in Japan
305
protection in the guise of environmental policy is against the Polluter-Pays Principle (PPP) adopted by OECD countries.7 In order to rectify these drawbacks of traditional types of environmental policy, greater stress is now placed upon the role of tax instruments. Within the broad category of economic instruments, tax instruments are only one of many options. However, in view of environmental market mechanisms which use price incentives to encourage individual decisions benefiting the environment, taxes are regarded as one of the most efficient instruments in environment policy. This is likely to be supported by many when environmental problems are widely extended to a global scale, as is stressed repeatedly. Within this new phase of global environmental issues, environmental taxes need to be explored and incorporated into tax structures in the context of integrating economic and environmental policies. It is important, however, to note that tax instruments and other types of environment policies (i.e. regulation and subsidies) should reinforce and support each other. This implies that the present situation is characterized by the prevalence of a 'mixed system', in which environmental taxes complement regulation according to the type of environmental pollution.
TAXATION AND ENVIRONMENT
Restructuring the Existing Tax System When we consider the possibility of designing the environmental tax in the Japanese tax system, there are a number of distinct types. In order to clarify alternative types of the environmental tax, two circles have been drawn in Figure 16.1: one shows the position of the existing tax system (dotted line), and the other that of new forms of environmental tax (bold line). Depending upon the demarcation of energy and non-energy taxes in the existing tax system, three zones are derived from Figure 16.1. Zone I and II constitute the use of existing taxes, while zone III implies the creation of a new environmental tax, such as a carbon tax. Let us first focus upon energy taxes in zone I. Energy taxes are only one environment-related tax in the existing tax system, but their revenues are mainly earmarked for road construction and others without reference to environmental policies in Japan. Energy resources in Japan are almost all derived from petroleum 7 The PPP is explained as follows: The principle to be used for allocating costs of pollution prevention and control measures to encourage rational use of scarce environmental resources and to avoid distortions in international trade and investment is the so-called 'Polluter-Pays Principle.' This principle means that the polluter should bear the expenses of carrying out the above-mentioned measures decided by public authorities to ensure that the environment is in an acceptable state. In other words, the cost of these measures should be reflected in the cost of goods and services which cause pollution in production and/or consumption. Such measures should not be accompanied by subsidies that would create significant distortions in international trade and investment (see OECD (1972)).
306
Indirect Tax System
Fig. 16.1 Alternatives of the environmental tax
and its derivatives which are totally dependent on overseas markets. Various types of taxes are imposed on different forms of energy under different considerations, including not only economic but political reasons. In Figure 16.2, current structures of energy taxes are summarized in terms of tax rates, revenues, and earmarking in fiscal 1998. Taxes are levied at the three stages from imports to petroleum derivatives for end-users. At the first stage of imports, the customs duty is imposed on crude oil. Then, the petroleum tax is levied on crude oil, and furthermore on imported LPG and LNG, and domestic natural gas. Finally, other related taxes are imposed on petroleum gas, petrol, light oil, and jet fuel, but heavy oil, kerosene, and naphtha are completely non-taxable apart from the tax burden at the earlier stages (i.e. custom duty and petroleum tax). Tax rates applied to each tax base are all specific rates. Although not indicated in Figure 16.2, any form of coal, imported or domestic, is not subject to taxation. Tax revenues collected at each stage are earmarked for four public expenditures: (1) road construction, (2) airport construction, (3) petroleum subsidies, and (4) coal subsidies. The largest revenues among all energy taxes are produced by the petrol tax. Consequently, petrol is most heavily burdened in total petroleum-related products in Japan. As argued previously, present energy taxes have no bearing upon environmental policies, but are exclusively collected for the purpose of securing financial sources of road construction. This is evident from Table 16.2 where the tax burden per ton of CO2 emissions is calculated. Different sorts of fossil fuels contain different carbon contents, and discharge different emissions of CO2 into the air. Crude oil and derivatives, which mainly constitute present energy taxes, are among the important fossil fuels, and are thought to generate a great deal of carbon and in turn CO2.8
Environmental Taxes in Japan
307
Fig. 16.2 Tax rates, revenues and earmarking of energy taxes, fiscal 1998 Note: Figures in parentheses are revenues of each tax and earmarked expenditures in billion yen. Source: Ministry of Finance (1998).
If we consider energy taxes as a tax instrument in environmental policy, tax structures must be designed carefully in view of greater efficiency to reduce both CO2 emissions and potential global warming. 8 According to the estimates by the Environmental Agency concerning CO2 emissions by fuels, petroleum occupies the largest share 57.1 per cent of the total, followed by coal 24.4 per cent, and natural gas 9.1 per cent in fiscal 1989.
308
Indirect Tax System
Table 16.2 Tax burden per ton of CO2 emissions, 1992 By energy
Category of energy taxes
Crude oil Custom duty Petroleum tax Petroleum tax Light oil transaction tax Petroleum tax Kerosene Aviation fuel tax Jet fuel Petrol tax Petrol Local road tax Petroleum tax Naphtha LPG Petroleum-gas tax Petroleum tax Natural gas Imported LNG Petroleum tax
Tax rates
One thermal unit of energy
(1)
(2)
CO2 emissions per thermal unit of energy (g/lOOOkcal) (3)
Taxes per CO2 emissions (¥/t) (4)
2 390¥/kl 350¥/kl 2 040¥/kl
9 400 kcal/1 9400kcal/l 9 400 kcal/1
80.23
3169
0¥/kl
8 000 kcal/1
80.46
0
24 300¥/kl 0¥/kl 26 000¥/kl 53 800¥/kl
9 200 kcal/1 8 000 kcal/1 8 700 kcal/1 8 400 kcal/1
78.39 77.47 76.65 76.58
33694 0 38989 83635
0¥/kl 18170¥/t
8 000 kcal/1 76.05 12 000 kcal/kg 68.33
0 22160
720¥/t 720¥/t
9 800 kcal/kg 56.39 13 000 kcal/kg 56.39
1303 982
0¥/t
6 350 kcal/kg 99.60
0
Heavy oil Light oil
Coal Note: Source: Data from the Environmental Agency.
As is evident from Table 16.2, however, the tax burden of each type of energy is not fully related to the level of CO2 emissions. The largest burden is imposed on petrol, followed by jet fuel and light oil, in spite of relatively smaller CO2 emissions per thermal unit of energy. By contrast, coal, heavy oil, and naphtha are fully exempted from energy taxes while they generate larger amounts of CO2 emissions. These results reflect the fact that energy taxes cannot play any significant role in achieving environmental protection on a global scale. Thus, if we were to try to convert the present energy taxes into a new form of environmental taxes, it would be necessary to restructure the existing tax system to a considerable extent. Next, let us move to the use of the existing tax system concerning non-energy taxes (zone II in Figure 16.1). At present, there is no generalized system of non-energy taxes to be regarded as the environment tax in Japan. According to the survey of existing tax instruments among OECD member countries, there are a number of cases of using existing taxes, which have generally been introduced for non-environmental reasons in the past but now become increasingly important from environmental considerations (see OECD 1993). For example,
Environmental Taxes in Japan
309
special attention should be paid to taxation of road transport and motor fuels. Existing taxes on road transport include vehicle-related taxes, such as (1) sales taxes on new motor vehicles with higher tax rates, (2) special taxes on the registration or use of motor vehicles, and (3) tax deductibility for less environmentally damaging cars. Similarly, motor fuels are generally subject to a number of different non-energy taxes: e.g. (1) higher rates of VAT or general sales tax on petroleum and motor diesel, (2) excise taxes on motor fuels, and (3) environmental damage taxes (say, CO2 tax), and fuel storage taxes. Another attention is paid to taxes on goods and services in the area of nonenergy taxes for environmental protection. Typical examples include (1) taxes on agricultural fertilizers and pesticides, and (2) product taxes on batteries, plastic carrier bags, and disposable drinks containers. Many countries in the OECD have already begun to use these types of environment-related taxes widely at the practical level. Generally speaking, however, there is no idea yet to convert or remodel these existing taxes into a new environmental tax in Japan. Apart from the typical types of environmental taxes listed above, we can point out a couple of pollution charges about the use of the existing system in Japan. Pollution charges, which are used to provide direct control over environmentally sensitive activities, are usually imposed on sources of pollutant emissions (or effluents), or on the users of pollution control equipment. Specific charges are levied upon air or water pollution, waste and aircraft noise mostly by local governments. In particular, it is noted that the air pollution charge is connected with the air pollution-related health damage compensation system. Based upon the PPP, the compensation system is devised to settle the conflicts between the polluter and the victims on the basis of civil liability. This arrangement is separate from the social security system (see, for detailed discussion, Kazu Kato, 1993). The air pollution charge is calculated in terms of sulphur oxides (SO,.), and is collected from industrial sites and business firms who release any amount of SOX into the air. Pollution victims are compensated by a fund whose revenue is raised by this charge (80 per cent) and the automobile tonnage tax (20 per cent). To sum up, the current tax structure is not effective from the standpoint of minimizing environmental problems in Japan. There is, however, considerable scope for environmental targets to be reflected through the restructuring of existing taxes rather than the introduction of wholly new taxes.9 Designing New Forms of Environmental Taxes Next, we shall shift our attention to new forms of environmental taxes (zone III in Figure 16.1), apart from the use of the existing tax system. There are two reasons to promote the introduction of such a new tax in Japan. 9
Tradable permits and a deposit-refund system, which are often referred to as other types of economic instrument, have not yet been put into operation in Japan.
310
Indirect Tax System
For one thing, it is politically almost impossible to convert existing energy taxes, most of which are earmarked for financing road construction, into an environmental tax. There would be no support among politicians and bureaucrats to use even a part of, say, the petrol tax for the purpose of environmental protection, reflecting strong pressure of vested interest groups. Thus, an idea of designing a new tax would be more appropriate and feasible. The other reason is that external pressure in the international community will increase to facilitate the incorporation of a new tax (e.g. a carbon tax) into the national tax system. Either bilateral or multilateral negotiation may push Japan to execute its responsibility to abate environmental damage by tax instruments in the light of international policy coordination. When we consider any desirable type of environmental tax, it is important to tax both fossil energy sources in view of its objective of limiting CO2 (i.e. carbon) emissions and all forms of energy in view of its promoting the efficient use of energy. For this purpose, the EC proposed to adopt a tax on carbon emissions and energy; i.e. carbon/energy tax (see EC 1992). Let us explore the possible forms of carbon/energy tax in Japan, based on the EC proposal. The sources of energy concerned are fossil fuels (coal, oil, natural gas, and their derivatives) and electricity generated by hydroelectric installations and nuclear power stations. It is assumed that the new tax is imposed on relevant energy sources at two stages of importation and consumption with a tax rate of $10 per barrel of crude oil equivalent.10 Primary energy sources are taxed at both stages, while taxes on electricity and city gas (i.e. a second form of energy) are only levied at the consumption level. The results of estimates in 1992 are summarized in Table 16.3, which are divided into the carbon tax and the carbon/energy tax. The tax base is a composite one, with 50 per cent being accounted for by the carbon content of the energy source and 50 per cent by its energy content in terms of thermal units. The carbon tax is estimated as an extreme case of carbon/energy tax, assuming 100 per cent carbon component. There are four points to be noted. First of all, total revenues of a new tax would account for ¥3.6-3.8 trillion in any case, which was about 6 per cent as a percentage of national taxes in 1992. It is expected to generate substantial amount of tax revenues, ranked as the fourth in total revenues, following the individual income tax, the corporate tax, and the consumption tax.11 Second, at the stage of importation, more than half of total revenues are collected from crude oil, and furthermore nearly 80 per cent from both crude oil and coal. If the new tax is levied at this stage, tax practices would be administered very efficiently with the least cost of tax collection. 10 The EC proposed that the tax should be phased in gradually, starting at $3 a barrel of oil equivalent on 1 January 1993 and rising by $ 1 a barrel a year to $ 10 a barrel of oil equivalent in the year 2000. 11 In fiscal 1992, the individual income tax generated 26.0 trillion (41.4 per cent), the corporate tax ¥18.4 trillion (29.3 per cent), and the consumption tax¥4.6 trillion (7.4 per cent). Percentages in parentheses are relative shares of each tax in total national taxes.
311
Environmental Taxes in Japan Table 16.3 Hypothetical types of carbon tax and carbon/energy tax, 1992 Type 2: At the Consumption Stage
Type 1 : At the Stage of Importation
Carbon/energy
Carbon/energy Energy Sources
Carbon tax %
¥b
Coal Material General Hard Crude oil NGL Petroleum Petrol Naphtha Kerosene Light oil A heavy oil B heavy oil Oil coke LPG Natural gas LNG
Energy Sources
tax
¥b
%
¥b
Carbon tax %
¥b
%
567.0 370.4 19.8
15.9 10.1 0.6
507.2 325.0 17.4
14.3 9.2 0.5
Coal Material General Hard
69.9 365.5 11.5
2.0 10.2 0.3
62.6 320.4 10.0
1.6 8.4 0.3
1 872.3
52.4
1872.5
52.9
Coke
245.0
6.8
200.3
5.2
54.6
1.5
55.3
1.6
39.0
1.1
53.6
1.4
321.0
9.0
209.9
5.5
5.7 142.2 16.6 3.7 0.2 0.1 45.0 127.6
0.2 4.0 0.5 0.1 0.0 0.0 1.3 3.6
5.7 144.1 16.6 3.7 0.2 0.1 39.0 136.8
0.2 4.1 0.5 0.1 0.0 0.0 1.1 3.9
174.0
4.9
174.0
4.6
2.5
0.1
2.6
0.1
356.8
10.0
360.4
9.4
14.3
0.4
17.0
0.4
60.4
1.7
61.2
1.6
336.5
9.4
401.4
11.2
31.4 224.6 346.3 238.9 1.4 460.2 57.8 39.9 151.6
0.9 6.3 9.7 6.7 0.0 12.8 1.6 1.1 4.2
31.7 225.5 345.7 237.3 1.4 449.8 67.0 34.6 162.4
0.8 5.9 9.0 6.2 0.0 11.8 1.8 0.9 4.3
10.4
0.3
12.5
0.3
LNG
251.1
7.0
299.5
7.8
City gas
123.6
3.4
144.5
3.8
New energies
0.0
0.0
33.7
0.9
Hydro power
0.0
0.0
89.1
2.3
Nuclear power
0.0
0.0
230.5
6.0
3582.8 100.0
3820.2
100.0
Coke furnace gas Blast & revolving furnace gas Crude oil NGL Petroleum Petrol Naphtha Jet fuel Kerosene Light oil A heavy oil B heavy oil C heavy oil Oil refinery gas Oil coke LPG Natural gas
Total
tax
3 576.0 100.0
3 542.0 100.0
Total
Note: The tax at $10 per barrel of crude oil equivalent is levied on all sources of energy by the carbon component and/or the energy (thermal) component; 10% : 0% in the carbon tax, and 50% : 50% in the carbon/energy (ax. Source: Author's estimates from Energy Statistics {Sogo Energy Tokei) (MITI: 1992).
312
Indirect Tax System
Third, by contrast, at the consumption stage, taxes are distributed more widely in various forms of energy sources, including city gas and electricity. The relatively larger shares are occupied by coal (general), petrol, and C heavy oil. Fourth, obviously the carbon tax excludes the taxes on electricity by nuclear power. This implies that the carbon tax tends to promote the construction of nuclear power stations, which would be criticized by anti-nuclear groups. To avoid such a criticism, the adoption of carbon/energy tax at the consumption stage might be more recommendable as a new environmental tax than a simple form of carbon tax. Key Issues Raised by Environmental Taxes Before concluding the arguments in this chapter, it is necessary to consider some aspects of introducing a new environmental tax mentioned above. Although there are a number of key issues to be worth investigating, some reference is made to the following four points concerning the distribution of the environmental tax burden and the extent of its negative impact on economic growth. First, the introduction of carbon/energy tax is likely to have a significant distributional effect, because of the importance of energy expenditures in the budgets of poorer households. Figure 16.3 illustrates this fact in terms of environment-related expenditures as a percentage of annual income at different levels of incomes. Two expenditures of electricity and fuels, and motor-vehicles and other transportation
Fig. 16.3 Patterns of household environment-related expenditures as a percentage of annual income, 1992 Source: Author's calculation from Office of the Prime Minister (1993).
Environmental Taxes in Japan
313
evidently indicate declining trends of relative shares in household budgets as income levels rise. If the tax base of a new environmental tax at the consumption level is equivalent to these expenditures, a flat rate of tax would generate a heavier tax burden on lower income earners. No doubt, the basic characteristics of the environmental tax are regressive. It is thus important to consider whether it might be possible to seek a policy package so that distributional issues that might otherwise produce a significant obstacle are offset or at least mitigated. Second, and related to the first point, focus is upon the use of the revenues from environmental taxes. This issue is very important and will become more controversial in Japan. Given the regressiveness of the carbon/energy tax, tax revenues, partially or wholly, should be first appropriated to reduce the individual income tax, including local inhabitant taxes, and not to increase overall tax pressure on households. A portion of the revenues might also be used to decrease the corporate tax burden from an economic standpoint of mitigating the detrimental burden caused by environmental taxes. When such a new tax is introduced as part of a revenue-neutral package, the political acceptability will be enhanced to a greater extent. Third, another alternative use of the revenues should be for financing of environmental expenditures in the public sector. Currently there exist a number of environmental taxes and charges earmarked by specific programmes and funds in many countries. Obviously, this earmarking problem will be decisively important in Japan when a new environmental tax is adopted, because relevant ministries and agencies are very keen to secure financial sources from this tax. It is clear that there are always conflicting interests among related bureaucratic groups in the use of the revenues. Generally speaking, earmarking of such a new tax to some environmental objective, including ODA (official development assistance) for environmental protection, will be of great help to increase the transparency of the measure and to generate political support. Although earmarking is, in the long run, likely to induce undesirable rigidities in the budgeting process, it should be stressed as a transitory solution to make the introduction of the tax more acceptable. Lastly, attention should be paid to the macroeconomic effects caused by the introduction of environmental taxes, although it is not the central focus of this chapter. In addition to the direct impact of additional tax payments on the distribution of household incomes, environmental taxes are expected to have wider repercussions on the economy: i.e. effects on employment, price, production, industrial structure, international competition, and so on. Many firms as well as the MITI12 tend to express their misgivings about the detrimental effects of environmental taxes on the working of the Japanese economy. Macroeconomic effects have so far been investigated using a variety of econometric models. Indeed, now in Japan, simulation results of approximately ten models 12 The MITI has established a couple of task-forces or working groups under the Industrial Structure Council to study the economic effects of environmental taxes. Broadly speaking, the basic attitude of the MITI is to attain the target of solving global warming by encouraging corporations and households to adopt energy-saving measures without any restriction and tax on CO2 emissions.
314 Table 16.4
Indirect Tax System Implicit carbon taxes, 1988 ($ per ton of carbon) US
Japan
Germany France Italy UK
Canada
65 0 0 28
130 2 0 79
212 23 0 95
351 38 0 229
317 80 0 223
297 0 0 106
108 0 0 52
Implicit subsidy and price support for the coal industry Subsidies for coal industry
—
2
28
25
—
10
—
Price support
—
15
49
—
—
36
—
Implicit carbon tax Oil and oil product Gas Coal Total
Source: Peter Hoeller and Markku Wallin (1991), p. 23.
analysing the possible effects of carbon tax on the future trend of the Japanese economy (see, for more detailed discussion, Amano 1992) are available. It is very difficult to reach any decisive conclusion about analytical results, but the consensus of research to date seems to be that the macroeconomic effects of carbon tax are likely to be quite limited. For example, a certain study group at the Environmental Agency has recently pointed out13 that a carbon tax rate might be set around ¥9 000-35 000 in 1995 per ton of carbon in order to keep CO2 level of the year 2000 at the same level as that of 1990, assuming there are no other instruments than a carbon tax. The impact of such a new tax on the Japanese economy would perhaps slow down real growth rate by about 0.01-0.5% annually. The other estimates show more or less similar results of slighlly reduced growth rate by the introduction of the carbon tax (see, for instance, Kuroda 1993). Given these empirical results in quantitative terms, we can safely conclude that the macroeconomic effects of the environmental tax would possibly be smaller than generally expected. Taking account of the issues noted above, any form of a new environmental tax (say a carbon/energy or carbon tax) should be adopted in the Japanese tax system in conjunction with a future tax reform. As seen in Table 16.4, the level of implicit carbon tax is still lower than international standards, following the US and Canadian cases. The tax burden on carbon emissions could be raised to a substantial degree if we consider prevailing rates in the European countries. In view of tax incidence, the ultimate burden of such a tax should be theoretically shifted to consumers through the price mechanism. Thus, the tax might be desirable to be imposed on the consumption level of energy sources. However, considerations of international competitiveness might be important to mitigate or exempt the 13 See the recent interim report of the study group on the economic system under global warming, April 1994. Also, see Gregory et al. (1991).
Environmental Taxes in Japan
315
burden of environmental tax for industries which are highly energy-intensive (e.g. steel industry). In conclusion, unlike the use of existing taxes, the introduction of a new environmental tax would be strongly recommended in the following form: Carbon/energy tax, whose contents are evenly divided, Taxing point at the consumption level of energy sources, The reduction of the existing taxes to offset the undesirable effects and/or the increase of environmental expenditures to reinforce the environmental effects of the tax.
This page intentionally left blank
PartV
Recent Tax Developments
This page intentionally left blank
17
Rebuilding the Tax System The rebuilding of the Japanese tax system began with two main tax reforms, starting in the mid-1980s. One was tax reforms for 1987-93, consisting of the first and second sweeping tax reforms. The other was that for 1994-99 in relation to fiscal expansionary policies after the collapse of the bubble in the early 1990s. In each of the periods in question, tax reform was the subject of a vigorous policy debate in Japan. An understanding of recent tax developments must start with a complete picture of the tax reform movement, because the results obtained from those reforms, as well as the issues remaining, have considerably affected the workings of the contemporary tax system. This chapter intends to explore the reasons leading up to a need for major reforms over the past two decades and the contents of reform packages. There are three main sections. The first aims to clarify key trends which formed the impetus for initiating three stages of the tax reform movement and the process of resolving tax disputes. The second section is devoted to clarifying the whole package of the Takeshita tax reform, followed by reform plans of three direct taxes—the individual income tax, the corporate tax and the inheritance tax. Detailed analysis of the consumption tax is undertaken in Chapter 15 to bring the movement of tax changes as up to date as possible. Third, we discuss in detail the recent tax reform movement developed in the 1990s, which was carried out to stimulate the depressed state of Japan's economy, rather than as a genuine style of tax reform.
BASIC BACKGROUND
Fiscal Deficits and fiscal Reconstruction
To understand the reasons behind the movement for tax reforms, we should begin by considering the structural changes that have occurred in the Japanese economy since the first oil crisis of 1973. From 1973 to the mid-1980s, Japan experienced fundamental economic changes and encountered significant problems in adjusting to them. Among the most important of these changes were the sharp reduction in the country's economic growth rate, large imbalances between domestic saving and investment, a sustained rise in fiscal deficits, and large current account surpluses in the balance of payments (see, for general discussion, Ito 1992; Takenaka 1992; Ishi 2000). Why is the year 1973 worth noting as a starting point? The outbreak of the oil crisis clearly revealed the vulnerability of a country poorly endowed in natural resources.
320
Recent Tax Developments
This triggered significant structural and institutional changes in the Japanese economy. From our standpoint, the most important change was the emergence of huge fiscal deficits (see Figure 3.2) after 1973. Economists disagree as to the main causes of the deficits in the government sector (see for example Noguchi 1987a, Lincoln 1988). Nevertheless, we must seek any causes, direct or indirect, for the large fiscal deficits resulting from the major burst of new spending on social welfare programmes and the lack of tax revenues reflecting the slowdown of economic growth (see Ishi 1982, 1986a). With the emergence of expanding fiscal deficits after the mid-1970s, the MOF became concerned about the rising deficits and began to stress the importance of achieving a balanced budget.1 Since then, reducing the fiscal deficit has become one of the most crucial objectives of budgetary policy, and even of overall economic policy.2 Eliminating fiscal deficits is officially called 'fiscal reconstruction'3; that is 'fiscal consolidation' in more general terms. The original impetus for tax reforms came from the need for fiscal reconstruction. Generally speaking, the MOF's desire to reduce fiscal deficits has had considerable support from big business, labour union leaders, and the general public. However, strong opposition has been generated, particularly from macroeconomists, the MITI, and the EPA.
Tax Increases or Expenditure Cuts? At first, the government and the MOF attempted to raise tax revenues in the form of a new VAT, i.e. the general consumption tax, in order to curtail accumulated debt. Such a bold strategy, however, was completely frustrated when a tax of this kind was politically rejected in the general election of 1979 (see Chapter 15). Then, from the summer of 1979 onwards, the MOF constantly pressured the government to hold down non-entitled expenditures (except debt services and revenue-sharing grants) when drawing up the initial budget. Accordingly, initial budget guidelines, in which limits were set to restrain increased spending, were strictly applied to the 1 Local governments also suffer from increased deficits, but their deficits do not pose as serious problems as the national government has experienced. Thus, it is appropriate to focus on the behaviour of the MOF and the national government in explaining the trends of deficit problems. 2 The MOF gave four arguments in response to those who favoured expansionary fiscal policies, following the Keynesian principle: (1) the sharp rise of debt service costs (redemption and interest payments), (2) the crowding-out effect in capital markets, (3) the debt burden on future generations, and (4) the increase of inefficient and wasteful expenditures (see e.g. the MOF's documents collected in the pamphlet, Considering Fiscal Reform, Feb. 1983). 3 The exact definition of'fiscal reconstruction' refers to the reduction of the amount of 'deficit-covering bonds', not 'construction bonds', in the general account of national government, by the targeted fiscal year. The Suzuki administration set a specific deadline of fiscal 1984 at the outset, but the date had to be postponed because of big revenue shortages after the second oil shock. Following this failure, the Nakasone administration set the target to curtail deficits by fiscal year 1990, which was finally attained in 1991 due to expanded revenues caused by the 'bubble economy' (see Ishi and Ihori 1992).
Rebuilding the Tax System
321
preliminary budget requests from each ministry. These guidelines had been set loosely at around 20-30 per cent in the 1960s, but the ceiling was sharply dropped to 10 per cent in 1980 and was lowered further to zero or negative increases after the mid-1980s.4 These expenditure cuts were carried out effectively in conjunction with an administrative reform movement. This was a lesson that the government had learned from the opposition to the 'general consumption tax'. The administrative reform movement was initiated primarily by big businesses, in particular Keidanren (the Federation of Economic Organizations), and was supported by the general public. In October 1980 the Ad Hoc Commission for Administrative Reform (Rinji Gyosei Chosakai—the name 'Rincho' is most commonly used as an acronym) was initiated, chaired by the late Toshio Doko, a senior business leader who had just retired as chairman of Keidanren. The main target of Rincho was to accomplish 'fiscal reconstruction without tax increases' on the basis of the assumption that the burden of reducing fiscal deficits should depend wholly on cutting expenditures. As a basic strategy, Rincho called for a 'starving-out' of government expenditures through a drastic review of the expanded administrative mechanism (see Kumon 1984). Rincho submitted five reports from July 1981 to March 19835 and recommended a number of important reforms to trim overly expanded portions of the government bureaucracy (see the chronological table compiled by Lincoln 1988, 119). Its major proposals were: 1. privatization of three major public corporations (the Japanese National Railroad, the Nippon Telegraph and Telephone, the Japan Tobacco and Salt); 2. cuts in spending on public works; 3. reduction of the number of government employees; 4. restraint of social security benefits; 5. simplification of certification and inspection procedures for manufactured products; and 6. merger of some government agencies. Furthermore, Rincho officially presented the aim of reducing the spending level and promoting fiscal reform, with no tax increases during the period of budget compilation every year. The Rincho proposals were considerably implemented, and were successful in curbing the growth of government, although they remain far from perfect. Needless to say, the general mood for administrative reform has played a vital role in the process of fiscal reconstruction, producing an environment of fiscal austerity. 4 During the fiscal years 1984-88, a —10 per cent ceiling was applied to current expenditures and —5 per cent to capital expenditures. In fiscal 1988 the latter was changed from a negative to a zero ceiling. 5 During this period, Rincho published five reports to promote administrative reform. The most comprehensive records relevant to Rincho's activities can be found in Administrative and Management Centre (1982).
322
Recent Tax Developments
Fig. 17.1 Percentage changes of non-entitlement expenditures over the preceding year, 1970-2000 Note: The figure in fiscal 2000 is based on its initial budget. The scope of the government is limited to the general account to the national government. Preliminary figures are used for fiscal 1999 and 2000. Source: Data from the MOF.
There was an overwhelming concern about the interrelations between cuts of wasteful expenditures, reductions in the large deficits, and drastic administrative reform during the first half of the 1980s. As a result of such policies, the growth of government expenditure has indeed been restrained. In particular, non-entitled government expenditure did not grow at all from fiscal 1983 to fiscal 1986 (see Figure 17.1).6 Under these circumstances, special attention should be paid to the reform of health insurance in 1984 and of the social security system in 1985, which contributed a lot to eliminating social welfare spending, although it evoked a great deal of criticism from the general beneficiaries. However, in the 1990s non-entitlement expenditures began to expand to buoy up the depressed state of the economy after the collapse of the bubble. Likewise, the rising trend of fiscal deficits was reversed from around 1980 in terms of the bond dependency ratio (the percentage of bond issues to total expenditures: see Figure 17.2). Overall, the whole process of fiscal austerity under the administrative reform movement is worth assessing if we focus only on the specific aspect of fiscal reconstruction (see, for an expanded discussion, Ishi and Ihori 1992).7 Fiscal 6 Behind this successful story, attention should be paid to the phenomenon of fiscal 'window-dressing' to restrain the growth rate of non-entitlement expenditures (see Ishi 1986fo, 146—7). 7 Of course, a variety of criticism has been raised against fiscal austerity. Of greatest importance is the MOF's usual failure to generate a tnacroeconomic policy response. The MOF's continued concern for
Rebuilding the Tax System
Fig. 17.2
323
Trends in the bond dependency ratio, 1970-2000
Note: The ratio is that of national bonds to total government expenditures in the general account of the national government. The figure in fiscal 2000 is provisional in the initial budget. Source: See Figure 17.1.
reconstruction was successfully achieved by the end of 1980s (i.e. deficit covering bond ratio was zero), but once again the bond dependency ratio has rebounded in relation to successive fiscal stimuli programmes (see Ishi 2000, ch. 6). In contrast to the control of government expenditures, how has the revenue side of the budget performed? Was it possible to implement fiscal reconstruction without tax increases in the true sense of the term? The answers are, No. Along with severe spending constraints imposed by Rincho to promote the goal of reducing deficits, the MOF began to fall back on various small measures to increase tax revenues. Let us return to Table 3.3. As noted before, we can observe a switch from tax reductions to tax increases from 1979 to the mid-1980s. The first measure was to terminate the annual reductions of the individual income tax until a substantial reduction was allowed in 1984. Automatic increases in income tax could be achieved by bracket creep and inflation (not seen in Table 3.3), in addition to minor increases brought about by structural change. Second, and more important, the corporate income tax was increased through a variety of minor adjustments, in particular the elimination of special tax measures (see Table 17.1). The basic rate was also lifted to the margin of 1.3 per cent as a temporal measure to raise revenue for the years 1984—86. The third measure was to make use of existing indirect taxes as much as
austerity has induced both irritating debate over stimulating domestic demand and foreign criticism of Japan's macroeconomic policy, when the economy was weakening and current account surpluses were accumulating sharply (see Ishi 1986fe; Yashiro 1987).
324
Recent Tax Developments Table 17.1 Tax increases of national corporate income taxes, 1980-1986 Fiscal year 1980
Tax increases" (¥bn.) Reduction of the maximum amount of reserves for retirement allowances, etc. Basic rate increased from 40 to 42%, the deductible ratio of the reserve for bad debts (banking and insurance) lowered, etc. The same treatment, applied to the reserves for bad debts (retail, wholesale, and manufacturing), the restriction on obtaining an extension of the tax due, and reduction of the maximum percentage that could be deducted as a reserve for price fluctuations Lowering the deductible ratio of reserve for bad debts again (banking and insurance) Provisional increase of basic rates for two years; repeal of measure extending the tax due Increase of special rates applied to co-operatives and non-profit organizations, and a decrease in the deductible ratio of reserve for bad debts Partial elimination of special rule for carry-over of loss; the postponement of provisional rate increases, repeal of reserve for price fluctuations, etc.
1981
1982
1983 1984 1985
1986
371.0
640.0
324.0
36.0 414.0 194.0
51.0
a
The amount of tax increases are on a full year basis. Source: Data submitted to the Tax Advisory Commission.
possible for revenue purposes. As illustrated in Table 17.2, major domestic indirect taxes were frequently increased.8 These changes were typical of the MOF's efforts to increase taxes in the first half of the 1980s, assuming that the fundamental framework of promoting fiscal reconstruction without tax increases could be maintained. To be sure, the MOP did not pursue major reforms which would have greatly altered the basic tax structure, but it did engage in a great deal of minor tinkering with the existing system. The Emergence of Sweeping Tax Reforms The history of the first and second tax reform is tabulated in Table 15.2 from beginning to end with special reference to the process of introducing the consumption tax (see Chapter 15). The basic characteristics of the reform are evident from Nakasone's 8
The tax reductions for 1982-83 in Table 3.3 all come from a custom tax cut.
Rebuilding the Tax System Table 17.2
325
Tax increases of national domestic indirect taxes, 1980-1984
Tax
Year
Price change or rate increase
Amount (¥bn.)
Alcoholic beverages Tobacco
1981 1984 1980 1986 1981 1984 1981 1981 1980 1983 1984
Rate increase Rate increase Price change Rate increase Rate increase Rate increase Rate increase Rate increase Rate increase Rate increase Rate increase
314.U 351.0
Commodity Stamp Security transactions Promotion of power resources developments Petroleum
135.0 123.0 56.0 421.0 59.0 99.3 68.4 134.0
Note: See Table 14.1. Source: As Table 14.1.
enquiry to the Tax Advisory Commission on 20 September 1985, in which he stressed the following points (see Tax Advisory Commission 1986, appendix p. 1). 1. The current tax system, founded on the Shoup recommendations in 1950, has brought about a lot of distortions and imbalance in all aspects of direct and indirect taxes. 2. Since these irregularities were caused by substantial social and economic changes, the tax system should be thoroughly restructured to respond to those changes that have already appeared and those that may develop in the future. 3. The basic principles are 'equity', 'fairness', 'simplicity', 'choice', and 'vitality'. In October 1986, the final plan for the first sweeping tax reform was presented to him by the Commission, and the next step of the legislative process was embarked on. However, this reform plan again failed, like Ohira's proposal for the general consumption tax, chiefly because Nakasone made a big mistake politically in the 1986 general election campaign, as was discussed previously. As a result of political turmoil, Nakasone was finally obliged to withdraw the bill in May 1987. Here it is worth noting the main elements of the first stage of the tax reform package:9 1. cuts in the individual income tax totalling ¥2 700 billion ($17.25 billion, with 156 yen to the dollar); 9 In addition to the strong accusation of Nakasone's dishonesty, many observers had misgivings with the reform package. Homma's estimates on the tax burden by income class had much impact, in that they revealed that higher income-earners (above ¥6m. annual income) would benefit more from the tax reform, even if the overall tax burden were revenue-neutral on a macro basis (see Homma 1987).
326
Recent Tax Developments
2. the eventual reduction of the nominal corporate tax rate (combined national and local rates) to less than 50 per cent from the current 52.9 per cent, but stiffer taxation of reserve funds or distributed profits—¥1 800 billion ($11.53 billion); 3. termination of tax-free interest on savings—¥1 000 billion ($6.44 billion); 4. the introduction of a broad-based indirect tax (later named the 'sales tax'), possibly along the lines of the EC value added tax—¥3 500 billion ($22.44 billion). The package was designed to be 'revenue-neutral' (i.e. (1) + (2) = (3) + (4) in the above four items), so as to neither raise nor lower tax revenues at least in its first year. It was conjectured that the effects on the overall economy would be minimal, reflecting the revenue neutrality at the government level.10 After Nakasone withdrew the legislation, the idea of tax reform did not die, but the potential for raising tax revenues by means of a new indirect tax went completely awry. Over the summer of 1987, a revised bill was deliberated among the LDP and opposition parties, and submitted again to the 109th Diet. It was a modest package of income tax reductions and the elimination of the tax-free saving system, and it did not include the institution of a broad-based indirect tax and the corporate income tax reform. This bill was finally passed in September 1987, and took effect in fiscal 1988. Overall, it was more of a tax cut than the tax rise that the MOP had been hoping for. Reductions of marginal tax rates were much more modest than the original proposals, having retreated to the range of 10.5-60 per cent in 12 income brackets from the original plan, which proposed 10-50 per cent in only six brackets. The only important structural change was to realize the MOF's long time dream of imposing tax on interest income from tax-favoured savings accounts. However, a substantial part of the first stage of the tax reform process was postponed. The second stage of tax reform started again, initiated by Prime Minister Takeshita's enquiry to the Tax Advisory Commission in November 1987. In his enquiry, Takeshita placed emphasis on the following two points: 1. the current tax system should be throughly restructured to secure stable revenues for an ageing society; 2. for this purpose, a well-balanced tax system should be achieved through an appropriate mix of taxation on income, consumption, and property. Obviously, the Takeshita tax reform was extremely important in the history of the Japanese tax system. In fact, Japan's VAT was finally created by this trial. Also, there was a widespread conviction that, once a value added tax was established, the MOF would begin increasing the tax rate and use it as the main vehicle for raising total taxes. These are two important reasons for the development of a strong resistance to Nakasone's tax reform. For comments on the vociferous opposition to the proposed tax reform, see Miyajima (1987) and Nishibe (1987). 10 According to the MO1; estimate, this reform package with 'revenue-neutral' approach would increase by 0.1 and 1.8 per cent the real GNP and consumption deflator, respectively, and would decrease by ¥0.5bn. the current account surpluses when compared with the case of no tax change (see Tax Advisory Commission 1986, appendix p. 144; also Ballentine 1986, 24-8; Yoshida 1987).
Rebuilding the Tax System
327
Catching up with Global Tax Reform Movements No one would deny that tax reforms in Japan were related to a global movement of tax reform that had occurred since the late 1970s. In this respect, of greatest significance was the psychological boost from the US tax reform in the 1980s. It is interesting to note the similarity of past tax reform issues addressed by governments in different countries, although the economic, social, and tax structures they face differ. There are a number of criticisms more or less common to most tax systems at that time (see Messere and Owens 1986; US Treasury 1984; Canada 1987; Ito and Krueger 1992): 1. The distribution of the individual income tax is unfair, and this conflicts with the principle of horizontal equity. 2. High marginal tax rates distort private economic decisions. People may be discouraged from working harder and saving more. 3. Capital income receives preferential treatment, owing to tax preferences and loopholes. 4. The present tax system is too complicated for the taxpayers and the tax authorities to manage in an efficient way. 5. The corporate tax structure stimulates business firms to undertake investments for tax rather than economic reasons. 6. There is too heavy a reliance on direct taxes, which tend to induce the distortions described above. Many of these problems are not new, but in a number of countries they became more urgent, reflecting the almost universal recognition of the inequities and distortions created by their tax systems. As a consequence, tax reform proposals were designed to meet several broad objectives: fairness, neutrality, and simplicity. A number of industrialized countries were undertaking major tax reforms (see Pechman 1988), and the tax policy authorities in Japan followed this global movement. In particular, the success of the US tax reform in 1986 greatly affected Japan's reform plans, except in regard to its thinking about the value added tax. The Japanese government was surely intent on catching up with, or at least not missing out on, the ongoing tax reform movement. The fundamental strategy of making tax plans was based upon the following trends, which are descernible among many reforms: 1. Individual income tax rates are reduced, particularly in the top bracket, thereby widening the tax base. 2. Corporate tax rates are also cut, eliminating tax preferences while at the same time enacting some base-broadening measures. 3. An increased reliance is placed on a broad-based indirect tax. These three strategies played a vital role in restructuring the tax system in the 1980s, and the most important among them was the shift from reliance on direct to indirect taxes.
328
Recent Tax Developments THE
S E C O N D TAX R E F O R M P L A N AS A W H O L E
Fundamentals of the Second Tax Reform The second stage of the Takeshita administration's tax reform was intended to revive the postponed part of the first tax package. After the original Nakasone tax reform was shelved in May 1987, the government decided to enforce only a limited package consisting of individual income tax reductions and the repeal of the 'Maruyu' tax-free saving system. As a result of a political compromise between the LDP and the opposition parties in September 1987,H tax reform was partially effected beginning in fiscal 1988 (see Aoki 1988; Nagano 1988). Two proposed reforms, however, were postponed: (1) the introduction of a broad-based indirect tax, and (2) a cut in the corporate tax rate. These were taken up by the Takeshita administration as its political responsibility. Along with the transition from the first stage to the second, during the years 1987-88 the fundamental conditions for achieving tax reform had changed. Land and stock prices soared rapidly during this short time of a 'bubble economy', widening the gulf between rich and poor. This trend was accelerated by the fact that investment income derived from selling land and stock was taxed very leniently under the tax preferences system. These phenomena aroused complaints among taxpayers regarding the inequitable burden of the existing tax system (see, for more general discussion, Ishi 2000). Continued economic expansion yielded large revenue increases in 1988. Even the target of'fiscal reconstruction' (i.e. no issuance of deficit-covering bonds in the general account of the national government) no longer looked unattainable. If tax revenues had been falling and no other means were available for raising the needed revenues, the people might have agreed that higher tax burdens were indispensable. However, this was not the case. Abundant revenue sources prevented the government from proposing the prompt adoption of a new broad-based indirect tax for collecting revenues. As a consequence, in order to make the tax reform debate more palatable, emphasis had to be shifted away from obtaining the revenue lost owing to income tax cuts towards the removal of unfair tax burdens. Accordingly, the perceived lack of fairness among taxpayers became a decisive factor in promoting the necessity of tax reform. In particular, wage and salary workers had long criticized the income tax in view of its unfavourable treatment of their employment income as opposed to other incomes (e.g. business or capital income), and had always complained that the income tax burden was too heavy. This view was constantly expressed by labour unions on behalf of their members. Thus, relieving such a perception of unfairness among working groups became an important goal of tax reform. 11 A limited tax package plan was constructed by the ad hoc Special Committee of the Lower House. Its members were composed of both the LDP and three opposition parties, excluding the Communist Party.
Rebuilding the Tax System
329
Table 17.3 Reform package of tax increases and reductions:3 tax bill proposed in July 1988 (¥bn.) Tax reductions
Tax increases
Direct tax reduction
5600
Creation of a new
Individual income tax
3 100
consumption tax
Corporate income tax Inheritance tax
1800 700
Others (repeal of the reduced tax rate on dividends, amendment of foreign tax credit, etc.)
1200
Repeal of certain selective excise taxes
3 400
Total
9000
Total
6600
5400
3
Includes both national and local taxes. Source: Data submitted to the Tax Advisory Commission.
The government had two aims for the second tax reform. One was to correct deficiencies in the existing tax system from an equitable point of view; the other was to create a new broad-based consumption tax in order to rebuild the indirect tax system and to secure long-term, stable revenue sources. To mitigate the strong resistance to the latter aim, the government reduced income and corporate tax burdens substantially. Great efforts were made to design the least objectionable package of tax reform (see Ishi 1988k). An Outline of Tax Mix Reform Package In July 1988, a new tax reform plan was sent to the Diet incorporating a ¥24000 billion net tax reduction (see Table 17.3). This formed a striking contrast to the Nakasone tax reform, in which total tax amounts were to remain 'revenue-neutral'. This choice of a net reduction in tax amounts was unexpected, considering the MOF's traditional goal of raising revenue while maintaining fiscal austerity. There are two special reasons, however, that explain why the decrease of tax revenues became feasible at that time. First, the failure of the Nakasone tax reform forced the government to make such reform more attractive by offering a large-scale cut in direct taxes. Second, the government could afford to cut taxes to some extent without creating new tax sources, given the ample revenues derived from recent economic expansion. The key points of the reform package are summarized as follows: 1. major reductions of both the national individual income tax and the local inhabitants' tax by lowering progressive tax rates and increasing the amount of exemptions and deductions;
330
Recent Tax Developments
2. cuts in corporate tax rates; 3. introduction of a new broad-based indirect tax, the 'consumption tax'; in conjunction with this reform, many selective excise taxes were repealed or amended; 4. reduction of the inheritance tax; and 5. new tax imposition on capital gains on the sale of stocks, and concomitantly rate cuts in the securities transaction tax. The main feature was the mixture of individual income tax reductions and indirect tax increases via the introduction of a new consumption tax. Great emphasis should be placed on this mixed reform package, which should result in a change in the tax mix towards indirect taxes, based upon a revenue-non-neutral approach. Thus, the ¥5 600 billion reduction of direct taxes came from cuts in the individual income tax, the corporate tax, and the inheritance tax combined, while tax increases were derived mostly from the introduction of a consumption tax. As a result, the net amount of the tax cut was ¥2 400 billion. An Outline of the Individual Income Tax Reform The main objective of individual income tax reform under the second sweeping tax reform was to reduce the tax burden at both the national and local levels by lowering progressive tax rates drastically and enlarging fundamental exemptions and deductions. For some people, income tax reform may result in a slightly increased tax burden as a result of minor adjustments made for the elimination of unequal tax burdens. In comparative terms, however, tax reductions were much larger than tax increases for most income-earners. Table 17.4 summarizes total revenue changes on a full-year basis. Income taxes were reduced to a great extent by lowering tax rates. As noted below, a natter rate structure was designed to alleviate income tax burdens in favour of middle-income wage-earners. The basic exemption, the exemption for spouses, and the exemption for dependants were simultaneously raised from ¥300000 (¥280000) to ¥330000 (¥300000). (Figures in parentheses refer to the local inhabitants' tax. The same will be true for the following figures.) The main aim of the increase was to raise the tax threshold and to give more of the benefit of income tax cuts to the lower-income classes. Similarly, special exemptions for the handicapped,12 widows (or widowers), and working students were also expanded, from ¥250 000 (¥240 000) to ¥270 000 (¥260 000), respectively. Moreover, two additional measures favoured specific taxpayers. One was a largescale increase (more than double) of the special exemption for spouses, which applied only to one-earner couples, from ¥165 000 (¥140 000) to ¥350000 (¥300 000). The other was an additional exemption of ¥100000 (¥50000) for dependants aged 12 In the case of a severely handicapped person, the special deduction was raised from ¥330000 (¥260000) to ¥350000 (¥280000).
Rebuilding the Tax System
331
Table 17.4 Revenue changes of income taxes (¥bn.), full-year basis Items
Nat. individual income tax
Local inhabitants' tax
Total
Tax reductions Lowering tax rates Increasing personal exemptions Expanding the special deductions for spouses Tax increases New imposition of capital gains tax on stocks Partial removal of tax concessions given to medical doctors
-2255.0 -1319.0 -529.0
-881.9 -404.3 -275.2
-3136.9 -1723.3 -804.2
-407.0
-202.4
-609.4
+ 757.0 + 695.0
+ 14.6 + 2.7
+ 771.6 + 697.7
+ 62.0
+ 11.9
73.9
Source: MOF, Outline of Tax Reform (Zeisei Kaikaku Taiko), 28 June 1988.
16-22. In particular, middle-income wage- and salary-earners were expected to benefit greatly from these special measures, if the family met specific requirements for these tax preferences. The special exemption of spouses took effect in fiscal 1988, with the partial enforcement of income tax cuts in the Nakasone tax reform. This measure was introduced to alleviate the widespread perception of unfairness among wage and salary workers. Self-employed taxpayers were allowed to allocate income to family employees as part of the privileges of filing a 'blue return', while wage and salary workers were not permitted to split their incomes with their spouses. It was generally emphasized that housewives support the earnings of their husbands, and that unequal tax treatment between the self-employed and salaried workers should be eliminated. Thus, a special exemption for spouses, in addition to the general exemption for them, was adopted for salaried workers of a one-earner couple, rather than permitting the 'income-splitting' between husband and wife.
Tax Base Broadened? It was widely acknowledged that reform of the individual income tax should be based upon the reduction of marginal tax rates through a broadening of the income tax base. This goal was very important, given the necessity to reform the current income tax system in terms of fairness, neutrality, and simplicity (see Pechman 1984, 1986; US Treasury 1984).
332
Recent Tax Developments
Needless to say, base-broadening measures lead to tax increases, and in turn induce strong resistance from special groups of taxpayers. There are two broad approaches to widening the tax base. The first is to bring into the tax base income sources that were previously excluded; typical examples include capital gains, interest, and dividend income. The second approach is to eliminate favourable tax treatment of particular sources of income, i.e. the use of special exemptions and deductions, the deductibility of expenses, fringe benefits, etc. Obviously, broadening the income tax base has a close bearing upon the correction of inequitable tax burdens. The existing narrowly defined tax base often resulted in an unequal treatment of two taxpayers with equal incomes, simply because one received income in ways that were tax-favoured. The primary goal of the Takeshita tax reform was to correct the unfairness of the existing income tax system. Accordingly, the income tax base should have been widened by removing exclusions, special exemptions, deductions, etc. Would the income tax base be broadened to correct such inequities under the reform plan? Unfortunately, the answer is No. In its interim report of April 1988, the Tax Advisory Commission officially pointed out the following six cases as typical examples of inequitable taxation: 1. capital gains tax on the sale of stocks; 2. special tax treatment of the income that medical doctors receive from social insurance programmes; 3. taxation of'deemed' corporations; 4. the deductibility of corporations' interest-payments-associated land purchases; 5. taxation of non-profit organizations, in particular religious corporations; 6. taxation of deficit-operating corporations. The first three examples are related to the individual income tax, while the last three are relevant to the area of corporate taxes. Capital gains tax on the sale of stocks alone was virtually restructured under the Takeshita tax reform, as mentioned previously (see Chapter 8). Non-taxable treatment was eliminated, and such capital gains were taxed in two alternative ways: (1) self-assessed declaration method at a rate of 26 per cent (including 6 per cent for the local inhabitants' tax), and (2) the withholding tax at a flat rate of 20 per cent, separate from other incomes. It is, however, clear that the new system was still too loosely worded to resolve burdens via the capital gains tax, although it could be seen to some extent as the first step towards the final goal of global taxation. In particular, the repeal of the device designed to tax those who were continuous and voluminous dealers in stocks would increase the unfairness in the system, because it unreasonably favours those who have earned substantial income from capital gains and who should pay at least 1 per cent of the withholding tax rate on the sale price of stock. Minor adjustments were made to rectify the tax benefits allowed to medical doctors. These consist of special tax measures which deduct an excessive amount of business expenses from the income paid to doctors from the social insurance
Rebuilding the Tax System
333
programme. Unfortunately, the benefits will suffer almost no damage from such minor adjustments.13 No attempt was made to eliminate the taxation of 'deemed' corporations at that time, which benefited self-employed persons since they could subtract expenses twice in the computation of their income tax base (see Chapter 5). It was necessary also to broaden the tax base via the removal of unnecessary exemptions and deductions. In general, special exemptions and deductions for specific policy goals, such as social benefits, tend to favour certain taxpayers over others with equal incomes. In the tax reform, no special exemptions and deductions were removed, even though some were certainly unnecessary (e.g. deductions for life insurance premiums, and fire and other casualty insurance premiums). Instead, some of the unnecessary exemptions (e.g. exemption for working students) were even increased to benefit special groups of taxpayers. To make matters worse, the special exemption for spouses was almost doubled and additional exemptions for dependants were admitted for those between the ages of 16 and 22. These new exemptions eroded the income tax base for specific taxpayers, and created new candidates for unfairness. In fact, two-earner couples protested against the increased exemption for spouses, which applied only to one-earner couples, calling for two-earner exemptions as well. The second stage of the tax reform process made no major progress in broadening the tax base so as to achieve greater equity. It seems that the base-broadening features of the income tax reform in lapan are not as far-reaching as in other countries such as the USA. Flattening Tax Rates Leaving the tax base intact, planners of recent tax reforms placed greater emphasis on achieving substantial rate reductions for individual taxpayers. This choice was made primarily because, whereas vested interests and specific beneficiaries strongly resisted the elimination of eroding provisions which would widen the tax base, no one opposed a reduction in rates. Broadening the tax base entails a heavy political cost, and so far no Japanese politician has been willing to tackle the problem. As a basic strategy for reducing the income tax burden, the progressive rate structure has been flattened since the first stage of the tax reform process. This reform was welcomed by many, for two reasons. First, the flat tax scheme, pure or modified, has become popular around the world during the past several years (see Messere and Owens 1986; Pechman 1988; Messere ed., 1998). In particular, the US income tax reform, in which only three rates (i.e. 15, 28, and 33 per cent) were maintained 13 Two tax preferences were allowed for medical doclors to compute the income tax base. One was a graduated lump-sum deduction of 52, 57, 62, 70, and 72 per cent from their remuneration for medical care provided under the social insurance programme. Such a deduction was used in lieu of actual necessary expenses, but it was considered to be excessive. The other was the exemption of their remuneration from the local enterprise tax. Under the Takeshita tax reform only 52 per cent of the special deduction subject to the highest-income groups was removed, leaving the other tax concessions intact.
334
Recent Tax Developments
Table 17.5 A modified flat rate structure of the income tax under the proposed tax reform plan, 1988 National
Prefectural
Taxable income class (¥m.) Less than
Over
3.0 6.0 10.0 20.0 20.0
Tax rates (%)
Taxable income class (¥m.)
10 20 30 40 50
Less than Over
Municipal
5.0 5.0
Tax rates (%)
Taxable income class (¥m.)
2 4
Less than Over
Tax rates (%) 1.2 5.0 5.0
3 8 11
Source: As Table 17.4.
around the late 1980s, greatly affected the thinking of Japanese tax reformers. In Japan this had a very attractive appeal politically, also. Second, a main motivation of reforming the progressive income tax was to remove the negative effects of bracket creep under the steep rate schedule. For this purpose, it is of great value to lower the marginal tax rates across the board and to decrease the number of income brackets. The statutory tax rate schedules had so far been altered to benefit middleincome-earners. The 1984-86 rate schedule, which ran from 10.5 to 70 per cent, was replaced by a flatter tax rate schedule in 1987-88 (see Table 5.5). The number of income brackets was reduced from 15 to 12, with the top rate lowered from 70 to 60 per cent. In September 1987, a part of the Nakasone tax plan was put into effect in which individual income tax cuts were achieved by widening income brackets. From the outset, however, this reform was considered a temporal measure before the final tax rates were arrived at; thus, it was expected that an even flatter rate structure would be proposed during the second stage of the tax reform process. In Table 17.5, the tax rates proposed by the Takeshita tax reform are shown, ranging from 10 to 50 per cent in only five income brackets. The level of taxable income at which the top rate of 50 per cent is applied has been reduced to ¥20 million.14 We may call this schedule of statutory tax rates a Japanese 'modified flat tax'. Similarly, the progressiveness of the local inhabitants' tax was decreased. The rates of the municipal inhabitants' tax were reduced from the range of 3-12 per cent with seven income brackets to 3-11 per cent in three bands. Also, rates of the prefectural tax were decreased, from 2-4 per cent over three brackets to only 2 and 4 per cent. 14 As a result of an agreement between the LDP and opposition parties on the means of instituting income tax cuts in July 1988, an additional top rate of 60 per cent over ¥50.Om. was revived as a temporary measure during 1988. This procedure was desired by the opposition parties to maintain vertical equity. It was really a political compromise.
Rebuilding the Tax System
335
Given the actual state of the Japanese economy and society, the flattening of progressive rates has several advantages over the steep progressive income tax. If base-broadening is accompanied by a flatter rate structure, such a tax would reduce the inequality of tax treatment of taxpayers with equal incomes, would abate possible distortions of economic decisions such as those determining the labour supply, saving and risk-taking, and would alleviate some of the complexities that resulted from the steep progressiveness of the income tax system. Furthermore, the reduction of income brackets and the widening of income bands would mitigate problems inherent to a progressive rates structure, e.g. bracket creep, bunching of income, and incentives to avoid or evade taxes. These advantages should be weighed against the distributional implications of a flatter rate structure.13 Obviously, a flatter structure is of greater benefit to wealthy taxpayers at higher income brackets than to low-middle-income taxpayers. Vertical equity is impaired, depending upon the extent to which progressiveness is modified. Any flat tax proposal must result in a conflict between the goal of distributional equity (vertical) and that of equal tax treatment of equals (horizontal). Most Japanese agree that people with higher incomes should pay a greater portion of their income in tax than those with lower incomes. They agree with this idea in principle; but the proper pattern of progressive tax burden in terms of the effective tax rate is a matter on which opinions differ. Even under the modified flat rates, the Japanese tax structure still has five rates as compared with two or three in the UK, and the USA; moreover, the top rate in Japan is the highest in the world.16 Thus, the extent of vertical equity would not be affected in Japan as much as in other countries by the adoption of a flatter tax structure. The Takeshita Tax Bill, which had been proposed to the 113th Extraordinary Diet on 29 July 1988, was finally approved by the Lower and Upper Houses on 16 November and 24 December 1988, respectively. In order to pass the Tax Bill some amendments were made to the original plan as political compromises between the LDP and opposition parties, although they were minor and did not change the amount of tax reductions greatly. In particular, the following three amendments are noted in relation to the individual income tax: 1. The special deduction for retirement income was increased (effective from January 1989). In order to increase the special deduction to ¥15 million (currently ¥10
15 Even if the rate structure becomes flat, pure or modified, it would theoretically be possible to maintain progressivity in terms of average effective rates. A flat tax could be applied to the increased taxable income as income rises, given a fixed amount of tax threshold. Low-income-earners would pay less, and high-income-earners more, even under a linear tax schedule (see Kay and King 1986, 215-20). 16 In 1988 Japan's top rate was 76 per cent, higher than that in any other country: it was 75 per cent in Sweden, 70 per cent in the Netherlands, 68 per cent in Denmark, etc. These rates are combined national and local tax rates (see Pechman 1988,4).
336
Recent Tax Developments
million) based on 30 years' employment, the level of such a deduction per year was raised in the national income tax as follows: Years of employment 20 years or less over 20 years
Current ¥250 000 ¥500 000
Proposed ¥400 000 ¥700 000
Similarly, local inhabitants' tax was allowed to raise the special deduction for retirement income. 2. Exemption for elderly bedridden dependants became effective from January 1989. The exemption for elderly bedridden dependants aged over 70 years for a taxpayer living with them was raised from ¥800 000 to ¥1 200 000. 3. The income tax reduction resulting from the above amendments increased by an additional ¥200 billion. Thus, total tax reduction under the Takeshita reform package will be revised to ¥9 200 billion, rather than ¥9 000 billion (see Table 17.3).
Reducing the Basic Rate of the Corporate Tax Tax reformers in various advanced countries have tended to lower the corporate tax rate while at the same time widening the tax base. Their thinking about corporate tax reform is basically the same as about income tax reform. In the Takeshita tax reform package, the major goal of corporate tax reform was to reduce the tax burden by decreasing the basic rate on retentions. As seen in Table 17.3, total revenue sources of corporate tax reductions were ¥1521 billion in the national corporate tax; similarly, the tax cuts in the local inhabitants' tax (corporations) and the enterprise tax were ¥282 billion. These sources were mainly used for the basic rate cuts.17 Tax rates on retentions in the national corporate tax would be reduced gradually over the next few years, as shown in Table 17.6. A lowering of these tax rates aimed to establish a corporate tax system in harmony with the international environment. Among advanced industrialized nations, Japan's corporate tax rates were the second highest, at 52 per cent in terms of effective rates (combined national and local tax rates) in 1988, following West Germany at 56 per cent. By contrast, the UK taxes corporations at only 35 per cent and the USA at 39 per cent (see Pechman 1988, 5). To attain parity with other countries, both the first and second stages of tax reform proposed that the level of effective rates be lowered to less than 50 per cent in the near future. The basic rate levied on ordinary corporations was gradually to be reduced, from 42 per cent in fiscal 1988 to 37.5 per cent after fiscal 1990. On the other hand, lower tax rates on small and medium-sized corporations were decreased only from 30 to 28 per cent. The tax rate of co-operatives remained unchanged, even though all other basic rates were 17 Another corporate tax reduction is to raise Ihe maximum limit for depreciable assets allowed as an initial deduction in full amount as expenses in the accounting period (from ¥100000 to ¥200000). Tax sources are used for ¥69bn. and ¥34bn. in the national and local corporate taxes, respectively.
Rebuilding the Tax System
337
Table 17.6 Basic rate on retentions (%) Fiscal year
Ordinary corporations Small- and medium-sized corporations Co-operatives, etc.
1988
1989
After 1990
42 30
40 29
37.5 28
27
27
27
Table 17.7 Removal of reduced tax rates on dividends (%) Fiscal year
Ordinary corporations Small- and medium-sized corporations Co-operatives, etc.
1988
1989
After 1990
32 24
35 26
37.5 28
22
25
27
reduced to some extent. Consequently, the difference between ordinary corporations and other firms was substantially narrowed in terms of the tax rates on retentions. In contrast to the general trend of lowering corporate tax rates, rates on dividends in the national corporate tax were increased step by step and applied as uniform rates after fiscal year 1990. The time schedule is shown in Table 17.7. Two measures contained within the reform plans rendered the existing corporate tax simpler in rate structure and moved it back towards the original Shoup recommendations; at least, they contributed to the simplification of the current tax system. In addition, tax concessions for dividends at the company level were terminated. Such special measures probably have little impact on equity financing and capital accumulation (see Chapter 10). In addition, several adjustments were planned to rationalize the corporate tax burden. They involved tax increases, which in turn enlarged the tax base. First, the tax on inter-corporate dividends was increased, and dividends by one corporation to another were subject to an additional corporate tax. Corporations were then allowed to deduct 100 per cent of the dividends they received from other corporations, but this percentage was reduced to 90 per cent in fiscal 1989 and to 80 per cent after fiscal 1990. The main reason for this alteration was to tax portfolio gains arising from the increasing cross-holding of stocks by corporations. However, if two corporations were closely affiliated (with a cross-shareholding of more than 25 per cent), the tax on inter-corporate dividends was waived.
338
Recent Tax Developments
Second, corporations were restricted in the amount of deductions they could take on interest payments for land purchases when computing the tax base. This measure made it impossible for them to deduct interest payments on loans for land purchases from taxable income as business expenses for four years. It aimed to discourage corporations from avoiding taxes by buying land with borrowed money. Third, the generous treatment of foreign tax credits was corrected to prevent an excessive amount of credits from being taken, a practice that was much criticized. It was relatively easy to lower corporate tax rates, despite the repeal of measures that had enabled corporations to reduce the rate they pay on dividends. The basebroadening measures, however, were insufficient to allow a substantial reduction of corporate tax rates. This posed practically the same problem as the individual income tax. In addition, no attempts were made to eliminate other tax-free reserves, such as the bonus reserve, so as to widen the tax base.
Inheritance Tax Reform Inheritance tax reform became a major issue during the second stage of the tax reform process in contrast to the first phase, in which it had hardly been mentioned. Given no tax changes since 1975, the enormous rise in land prices had imposed an unbearable burden on heirs who incurred the responsibility for paying inheritance taxes. This was especially the case in the Tokyo metropolitan area during the years 1987-88 (see Chapter 13). Therefore there were increasing demands to alleviate the burden of the inheritance tax. Four measures were suggested in the Takeshita tax reform plan to reduce the inheritance tax burden (including the gift tax). The total amount of the planned tax cut was ¥697 billion, as follows: Increasing the taxable threshold Reducing tax rates Lightening the burden on spouses Special deduction for small properties used for residential and business purposes
¥294bn. ¥201bn. ¥96bn. ¥106bn.
Total
¥697bn.
Tax reformers placed the greatest importance on an increase in the minimum taxable threshold for inheritance, since the value of taxable property had risen sharply along with land and other prices since 1975. As a result, the percentages of decedents exceeding the threshold had increased greatly (see Table 12.2). To cope with this problem, the level of tax threshold (¥20 million+ ¥4 million X the number of statutory heirs) was doubled to ¥40 million + ¥8 million X the number of statutory heirs in 1988. The tax rate structure was also changed. The top rate was reduced from 75 to 70 per cent with the tax brackets decreased. Special measures to mitigate the tax
339
Rebuilding the Tax System
burden of spouses was also enormously expanded to allow for increased exemptions. Moreover, tax credits for the handicapped and for minors under the age of 20 were raised to allow for inflationary adjustments. The level of exemptions for life insurance payments and the retirement income received by heirs by reason of the death of a decedent was more than doubled. The special deduction allowed for small properties used for residential and business purposes was also raised. In connection with the proposed reform of the inheritance tax system, the gift tax was also changed slightly. The top rate decreased from 75 to 70 per cent, similar to the inheritance tax. A person who had been married for 20 years or more could use the spouse deduction for transferring residential property, and this deduction was increased from ¥10 million to ¥20 million. The same treatment was given to the exemption for gifts to a severely handicapped person. (The deduction was doubled to ¥30 million.) Overall, there was not much argument for lightening the burdens of the inheritance and gift taxes in the tax reform scheme as a whole, although this issue was particularly important in relation to flattening income tax rates. Only inflationary adjustments were seen as an urgent enough measure to which it would be necessary to respond.
A N O T H E R TAX REFORM
MOVEMENT
IN THE
1990S
Successive Tax Reductions
After the tax reforms that ended with the adoption of the consumption tax (Japan's VAT) in April 1989, greater attention was paid to two additional issues regarding tax policy: (1) amendments to the consumption tax, and (2) the land tax reform at the beginning of the 1990s (see Chapters 15 and 13). In succession to these tax reforms, further tax reforms began to stimulate the Japanese economy after a prolonged recession; a recession triggered by the collapse of the bubble boom, generating nearly zero rates of real economic growth; i.e. 0.4 in fiscal 1992, 0.5 in fiscal 1993, and 0.6 in fiscal 1994. Thus, since the early 1990s, both individual and corporate income taxes have been involved in successive tax reductions as an important policy instrument of fiscal stimuli (see Ishi 2000, chaps. 4 and 6). The most marked feature was to introduce large-scale tax-cut policies to spur the depressed state of the economy in conjunction with a number of comprehensive economic recovery programmes with increased public investment. In November 1994, the government proposed a big tax package combining a cut in income tax at both national and local levels and a 2 per cent increase of the consumption tax rate (i.e. from 3 to 5 per cent, including 1 per cent of local consumption tax) over a stretch of four years. The main aims of this policy package were twofold; one was to buoy up the depressed economy and the other to secure revenue losses caused by income tax reduction during three years.
340
Fig. 17.3
Recent Tax Developments
Package scheme of tax reductions and increases, 1994-1998
Note: Individual income taxes include the local inhabitants' tax, too. Also, the 5 per cent of consumption tax rate contains 1 per cent of local tax portion.
As is seen in Figure 17.3, a temporary reduction of ¥5 SOObn. in individual income taxes was actually enforced in 1994, which it had been planned to terminate in the following year. Thus, the 1994 tax reduction was originally thought of as a singleyear cut on an ad hoc basis. However, as time passed, the original idea had to be amended in view of the necessity of continued expansionary measures, because the Japanese economy did not show any sign of serious recovery at that time. Against earlier expectation, it was decided that a part of the tax reduction (i.e. ¥3 SOObn.) should continue as a permanent measure in 1995, because the termination of the tax cut within only one year was considered to damage business recovery. The remaining portion of the original tax cut (¥2000bn.) was used as a temporary instrument and was postponed until 1996 for the same reason. To recoup such revenue losses, the consumption tax rate announced in the original tax package was to be raised from 3 to 5 per cent from fiscal 1997. The increase of 2 per cent was estimated as equivalent to the permanent reduction in income taxes (¥3500 bn.) in the tax system after 1997. However, before such a tax increase was adopted, total revenue of ¥165000bn. had already been curtailed in this reform package. Despite successive tax reductions on a non-neutral revenue basis, the Japanese economy still remained stagnant after 1997. What was worse, the stagnant condition of the economy increased, triggered by multiple factors, such as the 2 per cent increase in the consumption tax rate, the occurrence of the Asian economic crises, the big bankruptcies of Yamaichi Security Co., the Hokkaido-Takushoku Bank, and so on. In fact, the economy entered into a minus 0.4 per cent real growth rate in 1997 after recovery signs of 3.0 per cent real growth in 1995 and 4.4 per cent in 1996. Consequently, to combat the renewed slump, the Hashimoto cabinet had to initiate tax reductions once again, in which ¥2 OOObn. of individual income taxes were
Rebuilding the Tax System
341
put in force twice in 1998. At the same time a corporate income tax cut was also introduced with base-broadening measures, as will be argued later. The worsening state of business conditions greatly damaged the political life of PM Hashimoto who was obliged to resign as a result of the Upper House election in July 1998. A new cabinet of PM Obuti has completely switched the policy stance into anti-recession policies by using public investment and tax reductions, placing the first priority exclusively on economic recovery. Once again large amounts of individual and corporate income tax cuts were put into practice—in 1999 ¥1 lOObn. and ¥4 lOObn. each. At the same time, financial measures with a massive amount of capital injection (i.e. ¥60000bn.) were adopted in 1998 to stabilize the financial crisis in the private sector (see, for more expanded discussion, Ishi 2000, ch. 4). Despite the stabilization programme via fiscal and monetary measures, the Japanese economy had not revitalized by 1998. Indeed, real growth rate ran down to minus 1.9 per cent in 1998 in succession to a minus rate of economic growth in the previous year. A positive growth rate was realized from 1999; i.e., 0.5 per cent in 1999, and it is forecasted that it will be over 1.0 per cent in 2000. Tax Level and Structure Obviously, such big tax changes in the 1990s, as described above, must have greatly affected the tax level and structure of Japan's tax system. Similarly, the impact of post-bubble recession on tax revenues must have had some bearing upon them. To begin with, let us focus on the changing patterns of major tax revenues in the 1990s as compared with those in the 1980s. In Figure 17.4, the past trend of major national taxes is shown for 1983-99. Until around 1990 when the bubble boom collapsed, both individual and corporate income taxes were steadily increased. In fact, tax revenues almost doubled in between 1983 and 1989-90. In the 1990s, however, they began to decline due to two factors: tax reductions and economic stagnation. The individual income tax dropped from ¥26700bn. in 1991 to ¥15700bn. over time. Likewise, the corporate income tax showed a long-term decline from ¥19 OOObn. in 1989 to ¥10400bn. in 1999, although a slight upward movement could be observed in the mid-1990s. In particular, for the past four years, the declining trend has become conspicuous. By contrast, the consumption tax has constantly increased after being adopted in April 1989, although 3 per cent of tax rate remains at the same level. This implies that consumption-based tax can generate stable revenues irrespective of business conditions. In fact, the consumption tax increased revenue from ¥4 lOObn. in 1989 to ¥10400bn. in 1999, catching up with that of the corporate income tax. It has become one of the largest taxes in securing stable revenue. Next, moving our attention to the tax structure, let us investigate the main features of Japan's tax system. Figure 17.5 shows the tax structure of major countries for purposes of international comparison. Roughly speaking, about half of the total revenues of national and local taxes in 2000 were collected from direct taxation,
Fig. 17.4
Trends of major national taxes, 1983-1999.
Source: Data from the MOF.
Rebuilding the Tax System
Fig. 17.5
343
Tax structure of five major countries
Note: Both national and local taxes are included. Capital income taxes are included in individual income taxes. Source: Data from the Tax Advisory Commission.
that is, individual and corporate income taxes, 30 per cent from taxes on consumption, and the remaining 20 per cent from taxes on capital, say the inheritance tax or the property tax. This pattern has been created in recent years, switching from direct taxes to non-direct taxes, because both individual and corporate income taxes have substantially been reduced in succession for several years in view of fiscal stimuli. During the early 1990s, direct taxes still amounted to more than 70 per cent in total while taxes on consumption and capital shared less than one-third. Ironically speaking, successive direct tax reductions have unintentionally generated a heavier reliance on indirect taxation in the light of the consumption tax, which has been targeted for traditional tax reform in Japan for the past decades. Japan's tax system has a couple of features worth noting when making international comparisons, as seen in Figure 17.5. First, the tax burden in terms of tax-national income ratio is at the lowest level of 22.5 per cent among five major countries, followed by 27.5 per cent in the USA. This is in sharp contrast with that of European countries. Second, let us observe the composition of national and local taxes. Until the mid-1990s, the tax structure in Japan was quite similar to that in the USA (see, for instance, Table 1.1), relying upon a heavier direct taxation, but recently, as noted before, the relative importance of taxes on consumption has greatly increased, resulting in a total of 30.6 per cent which is higher than the USA. Third, the relative weight of corporate tax in Japan is 19.1 per cent, the largest share in the five countries, and this ought to be stressed as the most marked feature
344
Recent Tax Developments
of Japan's tax structure. Lastly, although the relative share of taxes on consumption has expanded, its share in Japan is still much lower than that in European countries. It is an important tax issue in the future whether or not it should be increased further. Individual Income Tax Reform Individual income tax reform began with a tax package in 1994 in relation to tax cuts, as stated earlier, and was part of a comprehensive economic expansionary programme. Reduction measures were divided in two distinctive ways. One was the application of a proportional ratio (e.g. 20 per cent) across the board to each income bracket to reduce the income tax burden without any institutional change of tax structure. This called for temporary tax reductions in a single year; i.e. ¥3 800b. in 1994, ¥1400bn. in 1995, ¥1 400bn. in 1996, and ¥2800bn. in 1998 (excluding local inhabitants' tax). The other was a tax cut on a permanent basis with institutional changes of a progressive rate structure and deductions; i.e. ¥2400bn. in 1995 and ¥2 600bn. in 1999. As seen in Figure 17.3, after enforcing this type of tax reduction, its effect will be continued in the future to reduce tax burdens. As a result of institutional tax cuts, the tax-rate structure became less steep during 1995-99 by raising each taxable income bracket at the same tax rate. As seen in Table 17.8, for instance, the income bracket applicable to 10 per cent rate is increased from ¥3.Om. in 1993-94 to ¥3.3m. in 1995-98. Likewise, when the individual income tax cut was instituted in 1999 the top rate was lowered to the level of 37 per cent from 50 per cent, the number of income brackets being curtailed from 5 to 4. In the process of tax-cut policy, exemptions and deductions have slightly changed, as summarized in Table 17.9. Basic exemption has remained untouched at the same level of ¥350000 since 1993 while exemption for spouses, special exemption for spouses,18 and exemption for dependants were raised from ¥350000 to ¥380000. On the other hand, special exemption for dependants aged 16-22 has steadily increased period by period, reaching ¥630000 in 1999. Furthermore, another type of special exemption for dependants was created for children younger than 15 years of age. Any increase in these exemptions and deductions led to raising the minimum taxable level, but more importance should be attached to the effects of temporary tax reductions during the period in question. The across-the-board tax cuts at a particular percentage obviously tended to raise the minimum taxable level by deleting taxable income from the lowest income classes. Thus, each temporary tax reduction, as well as the increased level of exemptions and deductions, raised the minimum taxable level from ¥3 277 000 in 1994 to ¥3 821 000 in 1999. 18 In addition to an ordinary type of exemption for spouses, certain taxpayers are allowed to use special exemption for a spouse on condition that the taxpayer's annual income does not exceed ¥10m., and he is living with a spouse whose income is less than ¥760 000. The taxpayer can subtract such an additional exemption in the form of a deduction corresponding to the spouse's income from his own income up to the upper limit.
Rebuilding the Tax System Table 17.8
345
Changes of tax rates, national income tax, 1993-1999
1999
1995-98
1993-94 Tax rates (%)
Taxable income classes (¥000)
Tax rates (%)
10 20 30 40 50
3000 6000 10000 20000 20 000-
10 20 30 40 50
Taxable income classes (¥000) 3300 9000 18000 30000 30000
Tax rates (%)
Taxable income classes (¥000)
10 20 30 37
3300 9000 18000 18 000-
Source: Data presented to the Tax Advisory Commission.
Table 17.9 (¥000)
Changes of major exemptions and deductions, national income tax, 1993-1999
Basic exemption Exemption for spouse Special exemption for spousea Exemption for dependants Special exemption for dependants aged 0-15 Special exemption for dependants aged 16-22 Minimum taxable level for wage-earners with wife and two children
1993-94
1995-97
350 350 350 350
350 380 380 380
350 380 380 380
500 3277
530 3 539
580 3616
1998
1999 350 380 380 380 480 630 3821
' Upper limit. Source: As Table 17.8.
How has Japan's individual income tax changed structurally in recent years? Summing up, in 1999, the top rate was reduced from 50 to 37 per cent with only 4 income brackets. Thus, the rate structure was much less progressive and flatter. It has been changed in a fashion similar to the US style with 5 brackets, the UK style with 3 brackets, and the French style with 6 brackets. On the other hand, the minimum taxable level of ¥3821000 in Japan is much higher than that of the USA (¥2 541 000), but it is more or less the same as that in Germany (¥4059000) and France (¥3254000). (Exchange rates are those on 1 August 1999). Overall, it may be pointed out that Japan's individual income tax has reached a sort of international standard, as compared with that before the two sweeping tax reforms of the 1980s. Corporate Tax Reform As compared with individual income tax reform in the 1990s, corporate tax reform at the national level has not been enforced so frequently. In other words, the corporate
346
Fig. 17.6
Recent Tax Developments
Effective corporate tax rates among major countries, 1999
Notes: lapan, the USA and Germany include local corporate taxes. Source: As Table 17.8
tax was changed only twice in the 1990s (in 1998 and 1999) in view of rate reductions and a broadening tax base. The higher level of corporate tax rate has persistently been criticized, particularly by business groups, for a long time. Figure 17.6 shows effective corporate tax rates, combining both national and local corporate taxes, in major countries. During the period 1990-97, Japan's tax rate showed the highest level of 49.98 per cent, followed by 48.55 per cent in Germany. Conversely, the USA, the UK and France had much lower tax rates. These tax rates are simply calculated on the basis of a statutory rate in terms of national and local corporate taxes,19 neglecting a lot of tax concessions, say tax-free reserves or special depreciation allowance. It is, however, widely acknowledged that they show a high corporate tax burden leading to complaints of the weakening international competitiveness of Japan's industries in the globalization process. As a consequence, the target of corporate tax reform has long been aimed at lowering such a high corporate tax rate to catch up with international standards. In fiscal 1998, the combination of the rate cut and base-broadening in the national corporate tax was adopted mainly for two reasons. One was to respond to persistent criticism from businesses by reducing the basic rate of national corporate tax from 37.5 to 34.5 per cent. The reduced tax rate for medium-small enterprises 19 In lapan, the local business tax is deductible as a business loss from the national corporate tax revenue. Thus, after adjusting for the deductible part of the local business tax against the national corporate tax, the effective tax rate should be recalculated (see the case of national tax rates in Figure 17.6). The same procedures are, more or less, employed in the case of the USA and Germany.
Rebuilding the Tax System
347
Table 17.10 Changes of the national corporate tax rates, 1990-1999 Accounting period
General tax rate (%)
Income demarcation
Reduced tax rate
April 1990 April 1998 April 1999
37.5 34.5 30.0
Below ¥8m. same same
28 25 22
Source: MOF, Tax Bureau, Primary Statistics of Taxation, February 1999.
was also lowered to 25 from 28 per cent as seen in Table 17.10. The other reason is that the MOP was reluctant to lose revenue sources due to tax-rate cuts and tried to compensate for this revenue loss by broadening the tax base with a view to improving tax equity and neutrality. The reform package was devised to cover the six years from 1998 to 2003, because some part of the base-broadening measures had to come into force year by year generating revenue gains. Major base-broadening measures in the light of tax-free revenues for 1998-2003 are as follows. Tax-free reserves, such as those for bonus, repairs and guaranteeing certain products, bad debts, and retirement allowance will have to be phased out step by step to mitigate sudden shocks on tax payment each year. 1. Reserves for bonus, repairs, and guaranteeing certain products will be phased out from fiscal 1998 to fiscal 2003. 2. The reserve for bad debts will be shrunk from fiscal 1998 to fiscal 2003 by applying only an empirical ratio of using realized bad debts, instead of a more generous percentage method in expectation. 3. The reserve for special repairs will be shrunk. 4. The reserve for retirement allowance will have to decrease by reducing the maximum amount to be accumulated against total retirement payment from 40 to 20 per cent gradually up to 2003. 5. The scope of depreciation allowance has been narrowed in certain areas, such as buildings or small-scale depreciable assets. Overall, according to the MOP estimation, tax rate reduction will diminish tax revenues of ¥1 380-¥1 442bn. while base-broadening will merely increase those of ¥1 122-¥1 228bn. each year on a full-year basis. Thus, such a package of corporate tax reform results in non-neutral revenue (i.e. net reduction) for a long stretch of time, despite the MOF's desire to make up for revenue loss due to rate cuts. In fiscal 1999, contrary to the 1999 reform, corporate tax reform was carried out by reducing tax rates to stimulate the stagnant state of Japan's economy. As indicated in Table 17.10, in succession to the previous year, corporate tax rates were lowered once again. The basic rate was curtailed substantially from 34.5 to 30.0 per cent, and similarly the reduced tax rate for medium-small enterprises from 25 to 22 per cent.
348
Recent Tax Developments
Consequently, the effective corporate tax rate reached a comparable level to other major advanced countries, as seen in Figure 17.6. In fact, the tax rate in 1999 was 40.87 per cent in Japan while it was 40.75 per cent in the USA, 30.00 per cent in the UK, and 40.00 per cent in France. Germany alone remained relatively at a higher level. Corporate tax-cut policy in 1999 was completely against the idea of MOF because it was fully financed by deficit-covering national bonds, but the long-standing target cherished by business groups has finally been accomplished. As the next step necessary to reform corporate tax structure, attention is being paid to the examination of both consolidated tax returns in parent-subsidiary companies and taxation in reorganizing a company into multiple forms. Both are now being deliberated in the subcommittee on Corporate Tax in the Tax Advisory Commission, and its final report will be published in a year or so.
18
Local Taxation and Intergovernmental Fiscal Relations
Local governments levy a variety of taxes to finance their needed revenues. However, owing to Japan's centralized fiscal system, they are considerably restricted by the national government in determining the level and nature of their fiscal activities. For example, fewer tax sources are allowed to local governments relative to the scope of services they are responsible for providing. This imbalance in the allocation of revenue sources necessitates the intergovernmental transfers of financial sources from national to local governments, a dependence that has been enhanced by various fiscal instruments. Thus, it is often argued that the extent of local autonomy (i.e. the ability of local governments to act independently of central control) is greatly impaired in Japan. The main issues regarding local taxation must be considered in the broader context of the interdependence between national and local governments. Therefore, this chapter may be different from others in that we have to consider a number of other topics not directly related to tax issues. Six topics in all are dealt with. First, the outline of the local government system in conjunction with the national government is described. To clarify the interdependence between the two, the shares of total tax sources and intergovernmental fiscal transfers are sketched roughly for preliminary discussion. Second, the whole system of local taxation is examined in detail. After the main features of local taxes are summarized, the function of the four major taxes is discussed. Third, other tax revenues, such as the local transfer tax and local borrowing, are briefly explained, and the nature of local dependence on central control via the permit system of local debt financing is clarified. The fourth and fifth topics concern two types of intergovernmental fiscal transfer systems: unconditional and conditional grants. The local allocation tax is equivalent to a tax-sharing or general-purpose grant for promoting fiscal equalization; in contrast, specific-purpose grants are provided for particular government services on a matching basis. Such grants stimulate increased local action in certain service areas, but they often succeed in restricting the independence of the local governments. To conclude the chapter, local fiscal behaviour and policy questions are investigated as a result of past empirical analysis.
This chapter draws substantially on Ishi (1986fc; 1995c).
350
Recent Tax Developments I N T E R D E P E N D E N C E BETWEEN THE NATIONAL AND LOCAL GOVERNMENTS
Evolution of the Local Government System Between the Meiji Restoration in 1868 and the end of the Second World War, local governments were made subordinate to the national government in order to unify the nation and to enable the national government to penetrate peripheral areas. It is generally argued that the influences of French and Prussian ideas were dominant in devising the system of local government. The governor of each prefecture was chosen from among officials of the Home Ministry (Naimusho: the predecessor of the present Ministry of Home Affairs, Jichicho). Local assemblies were left almost functionless, and authority was concentrated in the hands of local bureaucrats. However, a big change in the basic structure of local governments was brought about by the US occupation after the Second World War. Great stress was placed on the importance of local autonomy in a democratic nation, and the prewar system was completely restructured in order to encourage decentralization. The Ministry of the Interior was replaced by a much less powerful Local Autonomy Board at the outset, which in turn evolved into the Ministry of Home Affairs (MOHA). Prefectural governors were directly elected by local residents, and direct elections were also introduced at the municipal level. While the authority of national bureaucrats at the local government level was reduced, the functions of local assemblies were enhanced. In short, local government was given new and enlarged responsibilities and authority. As time progressed, however, the process of centralization was revived, and the national government began to adopt a policy of reversing many of these reforms. In the 1950s and the 1960s, a policy of rapid growth and industrialization became an important goal at both the national and the local level. This broad consensus clearly figured in the relative lack of resistance to the reverse process towards centralization. Innovative local programmes gradually withered while central powers expanded. This trend has continued to the present, although it has been broken occasionally by changes in the economy and society. Local passive attitudes were temporarily dropped in response to the appearance of urban congestion, the depopulation of less developed regions, and environmental pollution (see e.g. OECD 1986). These problems were all caused by rapid industrialization and were first faced by local governments. For several years beginning in the late 1960s, prompt responses to these problems produced a spurt of local innovations. However, this movement was relatively short-lived and ended with the oil crises in the 1970s. However, the movement towards local decentralization has persisted since the 1980s, particularly in academic groups and the local bureaucrats. Finally, these long-standing efforts resulted in the establishment of a local decentralization law in luly 1999, which must become effective in strengthening local autonomy to some extent.
Local Taxation
351
At present, the relationship between the national and local governments is still weighted predominantly in favour of the former, in many respects. In fact, the extent of authority, the revenue share, and the degree of responsibility of the national government are all greater than those of local governments. Local bureaucrats must heed functional superiors at the national level: since functional lines of authority are dominant, local officials must be responsive to the national officials in each area (e.g. public work, health, agriculture, etc.). It is widely acknowledged that local administration has become vertically fragmented.1 Structure and Size
In Japan the government sector is stratified into several levels, each having responsibility for a particular set of public functions. The main levels of the Japanese government are the national, the prefectural, and the municipal governments. The last two are called 'local governments', while the first is referred to as the 'national government'. The following statement may give a sense of the geography to Western readers. Japan is about the si/e of California and is divided into forty-seven prefectures, making prefectures much smaller than most American states and somewhat smaller than most European intermediate levels of government. Hokkaido, by far the largest prefecture, is somewhat larger than Maine but smaller than Indiana. The smallest prefectures are only half the size of Delaware. Compared to the Federal Republic of Germany, Hokkaido is bigger than Bavaria but the next largest German states are three or four times larger than the next largest Japanese prefectures. This is primarily a function of the smaller number of German states. Similarly, Italian regions are somewhat larger than Japanese prefectures because Italy has fewer regions. Britain and France have no comparable units. Japanese prefectures make up in population what they lack in land area. Tokyo Prefecture has a population similar to that of Pennsylvania or Illinois. The smallest prefecture, Tottori, is about the size of Vermont or Delaware in population. Prefectural populations are similar to those of German states or Italian regions, both in absolute terms and in the range of sizes represented (Reed 1986, 24).
On the sub-national level, Japan has a so-called two-tier system of local government. The total number of prefectures, 47, has not changed since the prewar period except for the addition of Okinawa (i.e. its forfeiture and return after the war). On the other hand, there are 3 229 municipalities, consisting of 671 cities, 2 558 towns and villages (in January 2000). These numbers have been achieved by constantly reducing the 10520 municipalities existing in 1945 through the mergers of cities, towns, and villages. We often refer to prefectures and municipalities jointly as 'local public bodies'. 1 In Japanese, the term tatewari gyosei (vertical consolidation) is generally used. In order to get more grants from each ministry of the national government, local governments are constantly forced to accede to the priorities of national bureaucrats.
352
Recent Tax Developments
In addition to the typical cases of local public bodies, there are several special arrangements. For example, twelve of the largest cities—Osaka, Nagoya, etc.—are treated as 'designated cities', separate from the others. All of them except for Hiroshima and Chiba have a population exceeding 1 million. These cities are responsible for carrying out some prefectural functions, although they are simply cities and are not excluded from the jurisdictional area of their respective prefectures. Another special case is the 23 'special wards' of the Tokyo Metropolitan area. These were established as a special type of local public body. However, some of the functions performed by municipalities elsewhere are also carried out by the Tokyo Metropolitan Government. Given this organization of the national-local government system, Japan's intergovernmental structure is typical of a medium-sized industrial democracy, more similar to unitary countries such as Britain and France than to federal nations such as the USA, Canada, and Australia. Each local government has budgeting accounts which compile the revenues and expenditures necessary for its activities. There are two types of accounts. One is the Ordinary Account for general administrative services such as education, fire, police, and so on. The other is the Public Enterprise Account for such services as housing, sewerage, and public transport. This budgeting system provides a common framework for comparing the financial situation among various local governments. These accounts must be reported to the MOHA, where they are statistically compiled each year and are utilized for the analysis of local public finance. Before the beginning of each fiscal year, the national government has to be apprised of the general local government situation in order to form basic fiscal policies. For this purpose, the national government is compelled by law to make official estimates of the total expected expenditures and revenues of all the local governments. Using this estimate as a basis, the national government is then expected to guarantee sufficient financial resources for each local government. The estimate and the revenue-sharing methods are consolidated into the Local Public Finance Programme (LPFP). This is an 'aggregated local public finance budget compiled of the annual budgets of more than 3 000 local public bodies'. It is considered to be comparable to the country's national budget. Figure 18.1 shows the current situation of tax shares and fiscal transfers in fiscal 2000 between the national and local governments on the basis of the LPFP and initial national budget data. Total tax revenues are ¥83.8 trillion, which are divided into national and local taxes. Before fiscal transfers, local taxes account for 41.9 per cent of total revenues. However, a substantial portion of national taxes is transferred to the local governments. Major fiscal transfers are of two broad types: unconditional and conditional. Unconditional transfers from the national to local governments are tax-sharing grants on a lump-sum basis financed by the local allocation tax (chiho-kofuzei). By contrast, conditional transfers are based on the condition that the recipient government must match a certain proportion of the transfer with its own expenditure. Furthermore, these grants are tied to specific types of expenditures
Local Taxation
353
Fig. 18.1 Tax shares between central and local governments, fiscal 2000 Note: Figures are based on the initial budget. Source: Data from the MOHA.
(education, social welfare, road construction, etc.), rather than being available for general purposes. Generally, these matching-type categorical grants are called specific-purpose grants (kokko-shishutsukin). In addition to these two types of fiscal transfers between the national and local governments, there are two other types of transfer. One is the local transfer tax, and the other is the 'transfer from the local to the national government1. The latter represents the local government's share in the financing of projects initiated by the national government. As can be seen in Figure 18.1, these two transfers amount to a much smaller sum than the items mentioned earlier. After reallocating the tax sources among different levels of the government, the final share of total tax revenues accruing to local governments increases to 73.9 per cent, as given by the ratio D/C in Figure 18.1. The ratio has remained almost unchanged for nearly a decade. This means that one-third of the national tax revenue is used at the local level.
354
Recent Tax Developments
Table 18.1 Total annual revenues3 of local government, fiscal 1998 Sources of revenue
Prefectures ¥bn.
%
Municipalities
Total
¥bn.
¥bn.
%
%
Local taxes Local allocation taxes Local transfer taxes Specific-purpose grantsb Local debt Othersc
17237.4 9272.8 128.6 10111.1 8665.0 10088.4
31.1 16.7 0.2 18.2 15.6 18.2
18684.8 8 776.2 466.6 5546.5 6562.0 14139.7
34.5 16.2 0.9 10.2 12.1 26.1
35922.2 18048.9 599.2 15657.5 15135.6 17505.5
34.9 17.5 0.6 15.2 14.7 17.1
Total
55503.3
100.0
54175.8
100.0
102868.9
100.0
a
Amounts are based on settled, not estimated figures. Specific-purpose grants to prefectures come from the national government, while those for municipalities come from both the national and prefectural governments. Total figures are net of the overlap between prefectures and municipalities. c Includes charges and fees, property revenues, contributions and miscellaneous items. Source: Ministry of Home Affairs (2000b). b
In order to clarify the financial position of the two-tier local government system, Table 18.1 summarizes the total annual revenues of each type of local government in fiscal 1998. Major fiscal sources are divided into six items. Local taxes account for the largest share, but are not sufficient to enable local governments to perform their fiscal activities under full 'local autonomy'. The proportion of final tax revenues collected by local government is only 31.1 per cent for prefectures, and 34.5 per cent for municipalities. The second-largest source of local revenues for prefectures are the specificpurpose grants, followed by the local allocation tax (i.e. unconditional tax-sharing grants). Local debts also occupy a substantial portion of total revenues at the prefectural government. On the other hand, municipalities raise the largest revenue from local taxes, followed by local allocation tax and local debt. Each category will be explained in greater detail below.2 LOCAL TAXES
Main Features A major problem of the present system is that local taxes are too small to enable local governments to perform their proper functions. In Figure 18.2 we can observe the past trends of the proportion of local taxes to total annual revenues. Although 2 Financed by these revenues, what functions do local governments perform? In fiscal 1997 the largest share of total local expenditures went to public works, 21.8 per cent of the total. Education was ranked next at 19.2 per cenl, followed by welfare programmes at 13.0 per cent. The largest three items occupy 54.0 per cent of the total.
Local Taxation
355
Fig. 18.2 Trends in the ratio of local taxes to total annual revenues at local level, 1955-1998 Note: Data are at the final settlement. Source: MOHA (2000).
there are tips and downs in the movement of the prefectural and municipal shares, the range of 30-40 per cent seems to have been maintained, in particular in the period 1960-80. In view of such a phenomenon, it was often mentioned that the degree of 'local autonomy' was simply '30 per cent', although the relative share of local taxes increased to over 40 per cent from the late 1980s to the early 1990s. Over a long-run period, municipal taxes have moved to a larger share than prefectural taxes. A major part of local government revenues are local taxes, which are classified in two categories: (1) prefectural taxes, levied by the 47 prefectural governments; and (2) municipal taxes, imposed by the more than 3 000 municipal governments. Table 18.2 shows the present system of local taxes in fiscal year 1999. Two primary local taxes at the municipal level come from the inhabitants' taxes and the property tax, while prefectural taxes are raised from the inhabitants' tax, enterprise tax, and local consumption tax. The inhabitants' tax is another form of local income tax, and the enterprise tax is equivalent to a tax on the net income tax of business firms. The structural features of these income taxes are used by both prefectures and municipalities. The remaining taxes produce minor revenues. The principal items, such as the inhabitants' tax, enterprise tax, and property tax, will be explained briefly later, the local consumption tax having already been studied in Chapter 15. On the working of the present tax system, two points must be made. First, in principle, each level of local government levies its own taxes, separate from the collection of national taxes. In the prewar period, the local surtax method, in which a 'piggyback' surtax was applied to the national tax on each item, played an important
356
Recent Tax Developments Table 18.2
Local tax collection by source, 1999
Sources
Tax revenues3
¥bn.
%
Prefectural taxes Prefectural inhabitant's tax Individuals Corporations Interests Enterprise tax Individuals Corporations Local consumption tax Property acquisition tax Prefectural tobacco consumption tax Entertainment tax Tax on consumption at hotels and restaurants Motor vehicle tax Mine-lot tax Hunter licence tax Prefectural property tax Motor vehicle acquisition tax Light-oil delivery tax Hunter tax
3598.9 (2499.0) (706.8) (393.1) 4 156.2 (271.1) (6051.4) 2462.6 605.3 273.9 89.7 109.5 1 742.3 0.4 1.8 13.1 467.3 1 297.2 1.3
(24.3) (16.9) (4.8) (2.6) (28.0)
Subtotal (A)
14819.5
100.0
Municipal taxes Municipal inhabitants' tax Individuals per capita Individuals on income Corporations per capita Corporations on income Municipal property tax Land Buildings Plant and machinery Small motor vehicle tax Municipal tobacco consumption tax Mineral product tax Special land-holding tax Spa tax Business office tax City-planning tax Others
8366.0 (117.9) (6261.0) (382.5) (1604.6) 9286.5 (3828.9) (3709.8) (1747.8) 115.3 858.9 1.7 60.8 23.0 320.5 1 364.6 0.2
(40.9) (0.6) (30.6) (1.9) (7.8) 45.3 (18.7) (18.1) (8.5) 0.6 4.2 0.0 0.3 0.1 1.6 6.7 0.0
Subtotal (B)
20476.2
100.0
Total (A + B)
35295.7
a
Kstimated figures. Source: Ministry of Home Affairs (2000a).
(1.7)
(39.0) 16.6 4.1 1.8 0.6 0.7 11.8 0.0 0.0 0.1 3.2 8.8 0.0
Local Taxation
357
role, but this was abolished in the postwar era in favour of independent taxation, to support local autonomy. In practice, however, all revenue sources are subject to control by the national government under the Local Tax Law (see MOP Tax Bureau 1992, 185-6; Jiji Sogo Centre 1988). The tax base and rates of major items are legislated by the Diet and can be altered by the proposals of both the MOHA and the MOP. This implies that a uniform rate is basically levied on the same tax base in all prefectures and municipalities. Despite strict uniformity, there are two options available to local governments. One is that they can collect a standard tax up to a maximum rate within limits set by the MOHA, subject only to the requirement that they have to inform the Ministry. In 1997 all except one prefecture raised corporate taxes to the maximum rate, but they did not increase personal taxes, for fear of the electoral consequences. The other option is concerned with the imposition of new taxes not listed in the law. Of course, local governments must seek the approval of the MOHA to levy such taxes and must show a special need.3 In 1997, 20 prefectures were given permission to use a non-listed tax such as the nuclear fuel tax on nuclear power plants. However, from fiscal 2000, 'the approval' of the MOHA has been replaced by more generous treatment of'the consultation' due to the enactment of Local Decentralization Law. Second, in levying the three levels of taxes, mutual co-operation is established among the municipal, prefectural, and national governments. For instance, when the municipal governments levy their inhabitants' tax on individuals, they collect the prefectural inhabitants' tax, too, using the same tax base. Information on taxable income necessary for computing the local inhabitants' tax is given by the national government. Although the individual and corporate income taxes accept the use of the same tax base by two (or even three) levels of government, local taxes cannot be deducted from national taxes except for the enterprise tax. This contrasts with the US local tax system (see Pechman 1987, 267-9). With no deductability of local taxes in calculating the national tax base, the national government makes adjustments to maintain the proper level of tax burden for each level of government through the allocation of tax sources. In fact, the individual and corporate income taxes at the local level are revised concurrently with revisions in national taxes. Prefectural Inhabitants' Tax The inhabitants' tax at the local level is collected by both prefectures and municipalities. The tax is levied on income in a manner similar to the collection of the national individual income tax. The prefectural inhabitants' tax was initiated in 1954 in order to counteract the repeal of value added tax proposed by the Shoup 3
The reason for this is explained as follows: 'The local public entity may levy special taxes on items for which the Local Tax Law has no particular provisions, provided that the Minister of Home Affairs approves such action. The Minister must give his approval insofar as such taxation does not impede trade among local public entities and as long as it does not duplicate other taxation by the local public entity concerned or by the National (State) Government' (MOF Tax Bureau 1987,185).
358
Recent Tax Developments
Fig. 18.3 Relative shares of four major taxes, 1955-1998 (a) Relative shares of two major items as a percentage of total prefectural taxes; (b) Relative shares of two major items as a percentage of municipal total taxes Note: Data are at the final settlement. Source: MOHA (2000).
Mission. Since there were no other sources that could replace the value added tax, the municipal inhabitants' tax was partially turned over to the prefectures. Figure 18.3 illustrates the long-run trend of both inhabitants' taxes as percentages of total tax revenue for each government. Starting at a low level of 16 per cent in 1955, the prefectural inhabitants' tax increased steadily to 30 per cent in the 1980s. By contrast, the municipal inhabitants' tax constantly maintained a higher level with faster growth, although it moved downwards due to the partial transfer of its revenues to prefectures in 1954. It has been noted that the inhabitants' taxes became one of the most promising methods of raising needed revenues for local governments,4 but they began to decline after the collapse of the bubble in the 1990s. 4 This phenomenon supports the Kay-King proposal to reform domestic rates by introducing a local income tax in Britain (see Kay and King 1986,146-51).
Local Taxation
359
Both taxes are alike, although there are some differences between the two. The prefectural inhabitants' tax is levied both on individuals and on corporations that have a domicile or business offices located within the prefecture. The individual inhabitants' tax occupies a greater percentage than the corporate inhabitants' tax in total prefectural taxes: in 1998 the former occupied 74 per cent of the total, while the latter only occupied 26 per cent, excluding inhabitants' tax on interest. The prefectural inhabitants' tax has three forms, each based on a different tax source: (1) per capita, (2) income, and (3) interests. In 2000 the prefectural governments imposed a per capita tax of ¥700 on individuals, and a ¥50 000-800 000 tax on corporations, according to the size of their paid-in capital. There are only two tax rates on income for individuals: 2 per cent on annual incomes of ¥7.7 million or less, and 3 per cent on those above that. For corporations, a standard rate of 5 per cent is levied on the national corporate tax. Among these principal types of taxes, the inhabitants' tax imposed on individual income is the most important. It is similar to the individual income tax at the national level, as the tax base is basically the same. However, the inhabitants' income tax is assessed on the income of a year previous to the income assessed in the national tax. Generally speaking, the individual inhabitants' tax is the best candidate for raising a large amount of local tax revenues, because it places the responsibility on as many inhabitants as possible to finance local public services. From April 1988, a new form, based on the third tax source, i.e. interest income, was added to the prefectural inhabitants' tax. This was due to the adoption of a flat rate of 5 per cent imposed on interest (similarly, 15 per cent of the national individual income tax) in the 1988 tax reform (see Chapter 8). Prefectural inhabitants' tax on interest has expanded rapidly since it was introduced in the local tax system, having a much larger share than the corporate inhabitants' tax. In 1999, it occupied 28.6 per cent of total prefectural taxes while the counterpart share was only 2.6 per cent in the corporate inhabitants' case (see Table 18.2). In spite of the uniformity enforced by the national government, taxpayers must pay their prefectural inhabitants' tax together with their municipal inhabitants' tax directly to the municipal tax offices, rather than combining them with payments made to the national government. This payment procedure is believed to be important for the maintenance of local autonomy, especially among local bureaucrats.3 Enterprise Tax The most important tax at the prefectural level is the enterprise tax (also called the 'business tax'), which accounted for 28.0 per cent of total prefectural tax revenues in 1999. Similar to the inhabitants' tax, the enterprise tax is collected from D In contrast, sceptics have criticized the separate payment of similar taxes. In particular, many supporters of administrative reforms call for the merger of national and local tax offices to save administrative costs (see Figure 1.2).
360
Recent Tax Developments
both individuals (i.e. unincorporated businesses) and corporations. The latter is much more important in terms of total 2000 tax revenues. The enterprise tax on corporations is generally imposed on net income (i.e. corporate profit), not on sales or turnover. In calculating the tax base of the corporate tax at the national level, the prefectural enterprise tax is allowed as a deduction. The standard tax rates on corporations in 2000 were as follows:6 Annual net income Below ¥4.0m. ¥4.0-8.0m. Above ¥8.0m.
Tax rate 5% 7.3% 9.6%
The maximum rate is permitted to be 1.1 times the standard rate. At present, it is being debated whether to broaden the enterprise tax base in order to secure stabler revenues. The reason behind this is the fact that the current profit-type of enterprise tax generates very changeable revenues. Thus, it is proposed by the Tax Advisory Commission that the relevant tax base should be broadened in one of the following ways: (1) income-type of value-added (i.e. wages plus profit plus interest plus rentals); (2) wages; (3) wages plus capital asset value, and (4) capital asset value. It will be politically decided in a year or so which of the four is chosen. For individual taxpayers, the enterprise tax is levied on income earned during the previous year. In computing taxable income, necessary expenses are deducted from gross receipts, but, unlike the individual income tax, deductions and exemptions are not applicable except for a proprietor deduction of ¥2.9 million. The standard rates on individuals, which range from 3 to 5 per cent, are applied to the four types of unincorporated enterprise. Like the enterprise tax on corporations, the prefectural government is permitted to raise the tax rate on individuals up to 1.1 times of the standard case.
Municipal Inhabitants' Tax As has been already noted, the inhabitants' tax has historically been limited to municipalities. For more than ten years after 1954, the relative share of the municipal inhabitants' tax in total municipal tax revenues was lower than it had ever been, since part of the tax revenue was being turned over to prefectural governments. Since the mid-1960s, however, the municipal inhabitants' tax has grown rapidly and has generated the largest revenues among municipal taxes. The municipal inhabitants' tax is more complex in structure than its counterpart at the prefectural level. Like the prefectural version, it is levied on both individuals 6 These tax rates are applicable to the ordinary income of ordinary corporations. Co-operative associ ations and liquidation income are usually taxed at lower rales than those listed.
Local Taxation
361
and corporations, but it is further divided into four categories: (1) a per capita tax on individuals, (2) a tax on individual income, (3) a per capita tax on corporations, and (4) a tax on corporate income. In 2000 the standard per capita taxes on individuals levied by municipal governments were ¥3 000 for cities with a population of 500 000 or more, ¥2 500 for populations of 50000-500000, and ¥2000 for other municipalities. The maximum amount can be increased to ¥3 800, ¥3 200, and ¥2 600, respectively, by the municipal government. The municipal tax on individual income is much more similar to the national individual income tax. Indeed, almost the same procedure is used for computing taxable income, which is then applied to a progressive tax rate structure. First, taxable income is calculated on the basis of the previous year's income as computed in the national individual income tax, and then a variety of deductions and exemptions are subtracted.7 For the most part, the amounts of deductions and exemptions are required to be lower than their counterparts in the national individual income tax. The reason for this is that greater benefit of public services can be provided to local taxpayers by local government, and therefore a greater tax burden is justified. The rate structure is rather progressive: in 1985 tax rates started at 2.5 per cent on the first ¥200 000 of taxable income and rose to 14 per cent above ¥49 million, with 13 income brackets, and in 1987 they ranged from 3 per cent on the first ¥600 000 to 12 per cent above ¥19 million, with seven brackets (see Table 5.4). In 1992, the rate structure was flattened, starting at 3 per cent on the first ¥1.6 million taxable income up to 11 per cent above ¥5.5 million with only three brackets. The same tax structure has been basically maintained up to now. In 2000, the three tax rates were 3 per cent less than ¥2.Om., 8 per cent for the income range of ¥2.Om. to ¥8.Om., and 10 per cent above ¥8.0m. The standard per capita tax on corporations imposed by municipal governments in 2000 ranged from ¥3 million for corporations with more than ¥5 billion paidin capital and 50 employees to ¥50000 for smaller corporations with less than ¥10 million. The maximum rates are allowed to be as much as 1.2 times the standard rate as mentioned above. On the other hand, the municipal corporate income tax is added to the national corporate tax at the standard rate of 12.3 per cent up to a maximum limit of 14.7 per cent. Property Tax
The property tax, which is called the fixed asset tax by the MOHA, is reserved for the use of municipal governments and raised 45.3 per cent of all municipal tax revenues in 1999 (see Table 18.2). The tax is imposed on owners of land, buildings, and tangible assets which are depreciable in individual and corporate income taxes (i.e. plant 7 Major items are the basic exemption, exemptions for dependants and spouse, the deduction for casualty losses and for medical expenses, and the deduction for social insurance premiums.
362
Recent Tax Developments
and machinery).8 Total taxes came 41.2 per cent from land, 39.9 per cent from buildings, and 18.8 per cent from depreciable assets in 1999. The property tax satisfies the requirements of a local tax, partly because the tax base is evenly distributed over the country and partly because it produces fairly stable revenues every year. However, it seems to be as badly administered as it is in most other countries. What is most crucial is that there are no reasonable assessments for property. In general, the value of the tax base for any particular property cannot be determined directly by market forces. Thus, property assessments are frequently arbitrary, causing an uneven distribution of the tax burden. The assessment of land and buildings is made every three years, and that of tangible business assets every year, in accordance with the assessment method determined by the MOHA. Substantial underassessment is more the general rule than the exception. It is generally supposed that valuations of land are based upon market value and those of buildings and equipment upon replacement cost minus depreciation. In practice, however, the value registered on the tax cadastre is used for assessments of land and buildings. Assessments are only a fraction of these values, and great disparity is seen among property assessments of equal value. Such poor assessments clearly create sentiments of unfairness among taxpayers and among different communities (see, for more detailed discussion, Chapter 13). Certain provisions have been instituted to prevent abrupt increases in the property tax burden. Since the market value of land was rising sharply in the late 1980s, it has been necessary to adopt a special method for determining gradually increasing assessments to avoid abrupt rises in assessed valuations.9 Under these special provisions, assessments lag behind the growth in values, leading in practice to a reduction in the tax burden on taxpayers. It is widely claimed that agricultural land has been assessed at a negligible level as compared with its market value. Real property below a certain assessed value of each type of asset is tax-exempt. The standard tax rate is 1.4 per cent, and the maximum rate is 2.1 per cent. In principle, the property tax is a municipal tax, but part of it is subject to a prefectural property tax, chiefly because the regional distribution of business assets is highly uneven. In addition to the tax on particular properties, there are two other related taxes on real estate. One is the city planning tax, which is levied by cities, towns, and villages for financing urban projects undertaken under the City Planning Law. The maximum rate is 0.3 per cent imposed on the assessed value of land and buildings calculated for the property tax. This tax is collected together with the property tax. 8 In addition to these assets, the property tax includes transfers as the fourth item. Property owned by the national government, prefectural governments, and public corporations are not taxed at the nationa and prefectural levels. However, municipal governments require these governing bodies to pay a fee instead of the property tax. These payments are a kind of transfer (in lapanese, nofukin), not a tax from the upper governments and public corporations to the municipalities. 9 Assessment ratios of land are raised gradually during three years from one assessment to the next, depending upon the extent of price rises of particular properties. As a consequence, the tax base for land has become much smaller than its market price.
Local Taxation
363
Table 18.3 International comparison of taxes on real estate, selected years (%) As % of national income
Japana 1975 1983 1997 USAb 1975 1983 1997 UKC 1975 1983 1997 Franced 1975 1983 1997 W. Germany6 1975 1983 1997
As % of total taxes
1.5 1.9 2.2
7.9 7.8 9.4
4.3 3.4 3.4
15.6 12.8 12.5
4.7 5.6 4.3
12.6 13.3 11.0
1.5 1.8 2.6
5.2 5.5 7.0
0.5 0.5 0.6
1.7 1.7 1.9
3
Property tax, city planning tax, and special land-holding tax. Property tax. Rates. d Foncier bati, Fonder non bati, and Taxe d'habitation etc. c Grundsteuer. Source: OECD (1998a,fo).
b c
The other tax is the special landholding tax levied by municipal governments. It was established in 1973 to check the sharp rise in land prices arising from speculation. Tax is applied at two separate rates: 1.4 per cent on the acquisition costs of land held on 1 January every year, and 3 per cent on that of land acquired within the year. Table 18.3 shows an international comparison of taxes on real estate in five major countries. The tax system seems stable in each country between 1975 and 1983 (or 1990 in Japan). The tax burden is relatively higher in the USA and the UK than in other countries. Japan is ranked in the middle, followed by France. These calculations reveal that taxes on real estate are moderate in Japan.10 10 Several other local taxes might be worth explaining, such as the motor vehicle tax and the light oil delivery tax among prefectural taxes, and taxes on electricity and gas and the municipal tobacco consumption tax among municipal taxes. However, all of them represent an insignificant portion of total tax revenues, so I have neglected them.
364
Recent Tax Developments OTHER REVENUE SOURCES
Local Transfer Taxes
In addition to the local taxes listed in Table 18.2, transfers from the national government have become a major tax source for local governments. A part or even the whole of specific national tax items are now transferred to local governments according to a given formula. Table 18.4 shows the six transfer taxes currently utilized by the government. In 1989 when the consumption tax (Japan's VAT) was introduced at the national level, some portion of its revenue was allotted to local governments in the form of the local transfer tax (see Chapter 15 for more detailed discussion). Thus, from 1989 a new tax, i.e. the consumption transfer tax, was added to the previous five items, but it was repealed in 1998, included in the local consumption tax. In the consumption transfer tax, 20 per cent of the national consumption tax was handed over to the prefectures and municipalities with the relative ratio of 6/11 : 5/11. Since this transfer tax was created to make up for revenue sources lost by elimination of local excise taxes, it is used as general revenue, not earmarked for specific use. As seen in Table 18.4, the largest share, 47.5 per cent of the total, comes from the local road transfer tax, followed by the motor vehicle transfer tax. The remaining three taxes are not large. All the taxes except the special tonnage transfer tax are 'earmarked' for relevant expenditures. These tax revenues can be regarded as de facto revenue sources for local governments. Theoretically, they should be collected as local taxes. From a practical point of view, however, it is rather difficult for local governments to levy them as ordinary Table 18.4
Local transfer taxes, fiscal 1998 (¥m.; %)
Consumption transfer tax Local road transfer tax Petroleum gas transfer tax Special tonnage transfer tax Motor vehicle transfer tax Aviation fuel transfer tax Total
Prefectures
Municipalities
Total
repealed
repealed
repealed
115591
171390
12729
1762
333
10416
282981 (47.5) 14491 (2.4) 10749 (1.8) 270709 (45.5) 16281 (2.7) 595210 (100.0)
270709 3959
12322
128611
466 599
Local Taxation
365
taxes, mainly because the tax bases are unevenly distributed among the various areas and tax collection by local authorities is inefficient. Therefore, the national government collects them as national taxes, and then transfers them, partially or wholly, to the local governments. The tax bases and method of transfer for each tax in 2000 were as follows. Local road transfer tax. The local road tax is levied on benzine and naphtha in petrol. Approximately 60 per cent of the local road tax is distributed to prefectures, and the remainder is transferred to municipalities in proportion to the length and area of roads in their jurisdiction. All the tax revenues are required to be used on expenditure relating to roads. Petroleum gas transfer tax. Half of the national petroleum gas tax levied on LPG is transferred to the prefectures and designated cities. None is transferred to the municipalities. Transfers are made proportionally on the basis of road length and area. They are earmarked for road construction and maintenance. Special tonnage transfer tax. The total of special tonnage tax revenues is transferred to the special wards in Tokyo and municipalities where international ports are located. This tax is levied on the net tonnage of ships using these ports which are engaged in foreign trade. In a way, the special tonnage transfer tax is a sort of a tax refund which is not earmarked for specific use, unlike the other four transfer taxes. Motor vehicle transfer tax. The motor vehicle tax is levied on owners of motor vehicles, based on licence certificates and the weight of the vehicle. The tax is transferred to municipalities. The transfer is made in proportion to the length and area of municipal roads and is used in the same way as the local road transfer tax. Aviation fuel transfer tax. Two-thirteenths of the revenues from the aviation fuel tax is allotted to municipalities and prefectures where airports are located. The aviation fuel tax is imposed on all aviation fuel and is used for improving airports and related facilities.
Local Borrowings
Although not directly related to local taxation, local borrowing is an area that should be discussed here in some depth. Under the Local Finance Law, local government expenditures must be financed by revenues other than local debt in order to maintain sound finance. Thus, the debt financing of local government is cautioned against, because the redemption and interest payments accompanying borrowing can become burdensome in future years. Local debt is defined as long-term borrowing by local governments for periods of more than one fiscal year. When local governments created debts, they were subject to strict regulations by the Local Finance Law and the MOHA. For example, the Local Finance Law restricted the use of local debt to the financing of certain types of expenditures, mostly limited to funding public utility enterprises and capital outlays, such as for gas and water supply, transport, road and harbour construction, etc.
366
Recent Tax Developments
When proposing to issue debt, the local government must include in its annual budget the following items: the purpose of the issue, the extent of the debt, the method of issue, and the conditions of interest charges and repayment. Positive support must therefore be obtained from the local assembly. The local government had then to seek permission from the MOHA, in the case of prefectures and designated cities, or from the relevant prefectural governor in the case of municipalities. There are official reasons for maintaining this 'debt permit' system. First, it is necessary for the national government to adjust the gap between the need for local debt and the financial funds available. Second, the national government can prevent limited funds from concentrating on the larger and powerful local public bodies at the expense of the smaller ones. Third, it is essential to maintain appropriate limits on local borrowing so that its financial burden will not prove too heavy in the future. In addition to the statutory permit system, the MOHA specifies certain conditions under which local governments cannot apply for permission to debt-finance. The MOHA does not give permission to any local government that has reached a critical level of fiscal deficits. When the debt charge ratio exceeds 30 per cent, a local government is designated as a 'financial reconstruction body', indicating that it has reached de facto bankruptcy in its financial management. In this case, local debt issues are no longer allowed.11 The local government must promptly devise a plan for fiscal reconstruction and have the plan approved by the MOHA. It will then be strictly supervised by the MOHA, but in return, special funds will be provided for paying the interest charges on the necessary loans. Both supervision and responsibility at the national level are interwoven to achieve a well designed balance in intergovernmental fiscal relations, although local autonomy is particularly restricted by this arrangement.12 As in most unitary nations, local borrowing is strictly regulated in Japan. In particular, the debt permit system of the MOHA reinforces central control. The MOHA firmly reviews requests for loans to prevent irresponsible borrowing, and often cuts the amount requested. The basic principle followed for the exercise of central control is to promote sound financial management of local government through the screening of local debt issues. For instance, when a local government sets its tax rates below the standard rate, it is impossible to get permission to obtain loans. 11 The debt charge ratio is the percentage of borrowing to general revenues (i.e. the sum of local taxes, local transfer taxes, and local allocation taxes), and it is an important factor in obtaining the 'debt permit'. When the ratio exceeds 30 per cent local debts are denied issuance except in the case of natural disaster restoration and local public enterprise. 12 Reed (1986, 32) points out very clearly the general features of Japan's local governments: 'In comparative terms, I would guess that Japanese local governments have less long-run financial flexibility than most. If one posits a solid local majority in favour of increased services, a local government in Japan would have more trouble raising the funds to implement the local mandate than would an American, British, German, or even a French local government. On the other hand, the Japanese system is well designed to enforce fiscal responsibility. The probability of a local government going bankrupt or getting itself in severe financial difficulties is probably less than in North American or Western Europe. France is a possible exception because there the prefect enforces fiscal responsibility. Japan is like France in the sense that the central government takes responsibility for enforcing proper financial practices on local governments. In other countries this responsibility lies more with the local electorate and the banking system.'
Local Taxation
367
Although the nature of local debt is basically unchanged, the 'debt permit' system was replaced by an advanced consultation system between the MOHA and local governments from April 2000 when the Local Decentralization Law was enforced. The change of procedure for issuing local debts seems to be less restrictive in favour of local governments, but it is still difficult to envisage the effects it will have in practice.
FISCAL EQUALIZATION AND LOCAL ALLOCATION TAX
The Importance of Fiscal Equalization A major issue pertinent to intergovernmental fiscal relations is that of equalization. The need for equalization arises out of the fact that, although all local governments have the same responsibilities for providing public services, they do not have the same financial capacities to assume those responsibilities. This is due largely to the economic disparities between various regions in the country. If the prevailing tax revenues were unadjusted, Tokyo and Osaka, for example, would have the potential to provide a much larger number and higher quality of public services than other areas. Table 18.5 shows the per capita amounts of the local tax (including prefectural and municipal taxes), the prefectural inhabitants' tax, and the enterprise tax in four rich and four poor prefectures. Obviously, the four rich prefectures, which are located in large metropolitan areas, have much larger tax revenues on a per capita basis than do the four poor ones. Poor prefectures derive revenues of less than 60 per cent of the national average. In addition, it is noted that the per capita prefectural tax in Tokyo is more than three and half times greater than that of Okinawa. The same difference is observed for the two principal taxes (i.e. prefectural inhabitants' and enterprise taxes). Although this table is relevant only at the prefectural level, the same phenomenon exists at the municipal level. Given these regional gaps in tax revenues, some means of fiscal equalization is necessary to provide local public services in poor areas. The most important means devised to handle this problem is the unconditional tax-sharing grant, that is, the local allocation tax system. The local allocation tax plays a central role in local public finance. The ideal financial resource for local governments is local tax revenues, because revenues can be raised by the local government itself without limitation. It is impossible, however, for local governments to rely entirely upon their own tax revenues, given the regional economic disparities and the necessity of levelling public services throughout the country. The local allocation tax was therefore established to equalize financial resources among different regional levels. The main aim of the present system is to secure a more even distribution of financial resources among local governments and to maintain local revenues at a level high enough to provide public services.
Table 18.5 Fiscal equalization at the prefectural level per capita, fiscal 1998 Local tax (1)
Prefectural inhabitants' tax
Enterprise tax (3)
(2)
Local allocation tax (4)
Per capita tax revenues after equalization (5)
Rich prefecture Tokyo Aichi Osaka Kanagawa
532935(1.87) 351425(1.23) 340452(1.19) 313104(1.10)
55533 36503 33606 34482
84302 51863 44642 31 214
1 16 11
238339(1.21) 159809(0.81) 149 825 (0.76) 120 849 (0.62)
Poor prefecture Okinawa Miyazahi Kagoshima Nagasaki Average
145341 (0.51) 174 644 (0.61) 175 562 (0.62) 177644(0.62) 285414 (1.00)
13808 16302 16756 17533 29031
15728 20 112 20336 20080 35615
162 440 178 684 162458 158 586 73676
230289(1.17) 259746(1.32) 243694(1.24) 233734(1.19) 196416(1.00)
Note: The dash in column (4) means that this prefecture is not given any amount of local allocation tax because of its wealthy financial capacity. The enterprise tax includes those for individuals and corporations. Column (5) indicates the results of prefectural tax, local allocation tax, plus local transfer tax on a per capita basis. Figures in parentheses are the proportion of the national average. Source: As Table 18.1.
Local Taxation
369
In Table 18.5, column (5), we can observe the adjusted tax amount (i.e. the prefectural tax plus the local allocation tax plus the local transfer tax) after equalization. The gap between rich and poor prefectures narrows; in fact, adjusted tax levels are even reversed between the two. The Evolution of the Equalization Tax System The original equalization tax system can be traced back to the local distribution tax in 1948. After the war, local governments were entrusted with the task of increasing their expenditures to restore the devastated country, to cope with new responsibilities transferred from the national government, and to improve the people's standard of living. However, owing to serious damages to the Japanese economy, revenues at the local level were extremely limited. Therefore several measures were taken by the national government to secure adequate revenues for local governments. One of these measures, the local distribution tax system, was established in 1948. In the 1950s, in place of this system, the government imposed a new programme of fiscal equalization, based upon the recommendations of the Shoup Mission. The local equalization grants system was introduced so as not to reduce local autonomy. This new system provided for open-ended equalization grants which matched the basic financial needs of local governments. In spite of this reform, local governments were still unable to raise sufficient revenues to meet the fiscal demand for their activities, the demand having remarkably increased during economic recovery. Local governments were thus obliged to make up for the revenue shortage by issuing debt, which in turn resulted in huge fiscal deficits and debt payments. The accumulated debts induced too heavy a burden on local governments, resulting in a disastrous situation. To cope with this fiscal difficulty, the national government provided a special temporary grant to local governments in 1954, and at the same time established a new equalization payment system. The new system was intended to guarantee stable revenues, and it forms the basis of the present local allocation tax system. The system has basically remained unchanged since 1954. The financial resources of the local allocation tax are derived from a portion of specific national taxes. In 1987, 32 per cent of the total revenue from the individual income tax, corporate tax, and the alcohol tax collected by the national government was allocated to the fund for fiscal equalization purposes. In this sense, the local allocation tax system is similar to a tax-sharing grant. If the revenues for equalization are not sufficient to maintain a reasonable level of public services, the allocation fund can be increased upon negotiation. In general, the tax-sharing ratio applicable to the three national taxes has been raised. The ratio has usually been raised when the relevant national taxes were reduced by the national government, causing revenue shortages in the local allocation tax. As is seen in Table 18.6, the tax-sharing ratio of the local allocation tax system has been increased ten times, starting from 22 per cent. In the past two decades, however, the
370
Recent Tax Developments
Table 18.6 Trends of the tax-sharing ratio, 1954-2000 Fiscal year
Individual income tax
Corporate income tax
Alcohol tax
1954
19.874
19.874
20.0
32.0 32.0
22.0 25.0 26.0 27.8 28.5 28.8 28.9 29.5 32.0 32.0 32.0 32.5 35.8a
1955 1956 1957 1958 1959 1960 1962-64 1965 1966-88 1989-96 1997-98 1999 2000-
32.0 32.0
Consumption tax
Tobacco tax
24.0 29.5 29.5 29.5
25.0 25.0 25.0 25.0
a
Determined in the policy-decision process. Source: Data from the MOHA.
ratio remained unchanged until 1999, because of the huge fiscal deficits at the national level. In addition to 32 per cent of three major national taxes, 25 per cent of tobacco tax, and 24 per cent of consumption tax (excluding consumption transfer tax) were included in 1989 in the tax-sharing sources under the local allocation tax system. Thereafter, the tax-sharing ratio of consumption tax was raised to 29.5 per cent. The local allocation tax is classified into two categories: the ordinary allocation tax and the special allocation tax. The former comprises 94 per cent of the total, while the latter occupies only 6 per cent. Obviously, the ordinary allocation tax plays a major role in fiscal equalization, but the special allocation tax is also important in compensating for the shortcomings of the uniformly calculated ordinary allocation tax. The computation formula of the ordinary allocation tax is so complicated that there is no counterpart in other countries. It is annually paid to local governments whose basic financial needs (N) exceed basic financial revenues (R). Thus, the ordinary allocation tax entitlement (T) is equal to the deficiency, N— R: T=N-R.
(18.1)
However, the total amount of the ordinary allocation tax, which is calculated in advance, does not necessarily cover the aggregate amount of the deficiencies of local
Local Taxation Table 18.7
371
Local allocation tax: receiving and non-receiving local governments, fiscal 1998 Receiving
Non-receiving
Total No.
%
100.0 100.0 100.0 100.0 100.0 100.0
No.
%
No.
%
Prefectures Municipalities Designated cities Cities Towns and villages
46 3114 12 571 2511
97.6 96.3 100.0 89.6 98.0
1 118
2.1 3.7
66 51
10.4 2.0
47 3232 12 637 2562
Total
3160
96.4
119
3.6
3279
Note: Tokyo, Aichi, Kanagawa, and Osaka are the prefectural non-receiving bodies. The special wards of Tokyo are included in the 'designated cities' category as a non-receiving body (the 23 wards are considered as one entity). Source: Ministry of Home Affairs (2000«).
governments whose basic financial needs exceed their basic revenues. This being the case, some modification is necessary in the calculation of the total allotted amount by using an adjustment coefficient a.13 The actual amount of ordinary allocation tax (T") granted to a local government is T" = (N-R)-aN.
(18.2)
The second term is added to adjust for the gap between Tand N~~R in (18.1). Local governments are split into two types, receiving or non-receiving, depending upon whether their basic financial needs exceed their basic financial revenues. As is shown in Table 18.7, 96.4 per cent of local governments were 'receiving local governments' in fiscal 1998. As typified by the case of Tokyo, the status of non-receiving implies that the government does not need any grants, reflecting its strong financial power. Important factors in calculating the ordinary allocation tax are basic financial needs and revenues. In what follows, we shall explain both of these factors in greater detail. 13
The adjustment coefficient a is calculated as follows:
In practice, a is nearly zero (say, 0.000235) although it is used every year.
372
Recent Tax Developments Basic Financial Needs
Basic financial needs indicate the amount of public services that it is necessary to provide from the standpoint of the national average. Thus, 'needs' do not have to correspond to actual expenditures by specific local governments; rather, reasonable and standard financial needs are computed, given the average condition of a model local government. At present, the model local government is conceived of as a prefecture with a population of 1.7 million and an area of 6900 square kilometres. Similarly, the hypothetical municipality is assumed to have 100000 people and an area of 160 square kilometres. In each case, a standard level and range of basic financial needs and revenues are assumed. In calculating basic financial needs, the fiscal activities of local governments are divided into six categories. In the case of prefectures, the categories are: police, public works, education, welfare and labour, industry and economy, and other administrative functions. Municipal expenditures are divided into similar categories. For each service item, basic financial needs are calculated according to the following formula: Basic financial needs = (indicators) X (unit cost) X (modification coefficients). Indicators are the units of measurement for each service item. For instance, the number of policemen is used to measure police expenses, and similarly, length is selected to calculate road and bridge expenses. It is not easy to estimate precisely the amount necessary to provide a specific public service by a single indicator. Despite this difficulty, basic financial needs are computed with a single unit of measurement, irrespective of economies or diseconomies of scale relating to the provision of a specific service. In order to adjust for these differential variable factors, modification coefficients are applied to each indicator by multiplying by the cost per unit. The basic financial needs of a local government are thus the total needs for all the services that it provides. It is necessary to explain further the nature of the modification coefficients. Without modification, the results of the basic financial needs calculations would not always be precise, because they reflect only the aggregate figures of all indicators multiplied by unit costs, which is considerably different from the real picture. For example, when population is used as an indicator, the greater the population, the less the cost per unit. Likewise, unit costs must differ from one area to another for a specific service, say, reflecting the climate. Therefore, unit costs must often be modified. Currently, modification coefficients are classified according to the following eight categories. (Detailed explanations are provided only where necessary.) 1. Class modification coefficients. A typical example is in calculating the financial needs of a high school. The indicator for a high school is the number of its
Local Taxation
373
students. However, educational expenses are likely to differ depending on the type of school (i.e. academic, engineering, agricultural). In such cases, class modification coefficients are applied to adjust for differences in unit costs. 2. Size modification coefficient. When economies of scale occur in the provision of public services, lower unit costs should be applied. For this adjustment, size modification coefficients are used. 3. Density modification coefficients. 4. Modification coefficients for special factors. Public services for local governments vary in accordance with differences in economic, social, and institutional factors in different regions. These factors must be taken into account when calculating financial needs. 5. Modification coefficients for cold areas. 6. Modification coefficients to allow for rapid growth of population. 7. Modification coefficients related to rapid decreases in the unit of measurement. This coefficient is applied, for example, to minimize any sharp reductions of the local allocation tax that would occur in the case of a rapid decrease in the population of a municipality. 8. Modification coefficients related to financial capacity. Basic Financial Revenues We shall next shift our attention to basic financial revenues, whose calculation is much simpler. Basic revenues consist of revenues that are used for the purpose of meeting the above basic financial needs. Their calculation is different for prefectures and municipalities. The formula is as follows: Basic financial revenues = evenues from ordinary taxes by standard rate
x^
+ revenues from the
local transfer tax
where (3 is 80 per cent for prefectures and 75 per cent for municipalities. The total amount of local tax revenues is not included in the above formula. Instead, basic financial revenues include 80 per cent (75 per cent for municipalities) of ordinary taxes and the full amount of the local transfer tax at the prefectural and municipal level. There are two reasons for adopting such prescribed percentages. First, it is impossible to measure completely the basic financial needs of all local governments by a uniform formula. Second, it is necessary to retain incentives for local governments to collect their own taxes. On the other hand, all revenues allotted from the local transfer tax are included, mainly because it is collected by the national government and has no relation to the tax collection efforts at the local level.
374
Recent Tax Developments
Additionally, some consideration should be given to the special allocation tax. This tax is granted when one of the following circumstances is fulfilled: 1. when special financial needs are incurred which are not covered by the ordinary allocation tax: e.g. campaign costs for local assemblymen elections, which take place every four years; 2. when local tax revenues are overestimated: since the ordinary allocation tax is based on the excess of basic needs over revenues, an overestimated local tax may cause a reduction in the ordinary allocation tax, which must be supplemented by the special allocation tax; 3. when increases in fiscal needs or decreases in tax revenues unintentionally take place during the period of the fiscal year: these phenomena tend to reduce the ordinary allocation tax, which should be made up for by the special allocation tax.
SPECIFIC-PURPOSE
GRANTS
Role
We should also consider the so-called specific-purpose grants (kokko shishutsukin)1* as an intergovernmental fiscal transfer from the national government to local governments. As was seen in Table 18.1, these grants play an important role in local government revenues. Their shares in total revenues were 18.2 per cent for prefectures, and 10.2 per cent for municipalities in fiscal year 1998. For municipalities, specific-purpose grants come from the two higher levels of governments, the prefectural and the national. Thus, both types of grant should be studied, but it will suffice here to investigate just the national grants. Prefectural grants are more or less similar. Most specific-purpose grants are supplied on a conditional, matching, basis. This is in sharp contrast to the local allocation tax, which is unconditional. Specificpurpose grants are used to secure a certain level of public services. The national government is responsible for equalizing the level of services for local governments. Therefore, the specific-purpose grants play a vital role in achieving uniformity for the provision of public services, given the compliance of local governments. Specific-purpose grants are considered to be financial resources paid by the national government to local governments as a part or the whole of the specific public expenditures. In some sense, they are cost-sharing grants. There are two types of disbursements from the national government: 1. disbursements to cover the national functions that are delegated to local governments and administered by them; and 14 The term is usually translated as 'national government disbursements' (see Jichi Sogo Centre 1983), but here we follow the translation of Yonehara (1981).
Local Taxation
375
2. disbursements to share, partly or wholly, the costs of providing certain services at the local level to promote national objectives. Let us examine each category in detail. The first concerns payments for agential tasks. A number of national government tasks are administered by local governments, including the election of national Diet members, the registration of aliens, the national census, the national pension, and other social welfare programmes. All these activities are entrusted to local governments for two main reasons: (1) the convenience of local residents, and (2) the effective use of expenses. In this case, the local governments perform, so to speak, agential tasks in place of the national government. Accordingly, the national government has to pay the full cost of public services incurred initially by the local governments. Thus, the full amount of payment is covered by the national government. Greater importance should be placed on the second category of specific-purpose grants, which is further divided into two sub-categories. One covers disbursements paid by the national government as an obligation for a part or the whole of specific public services. The other is to promote the execution of specific services for national goals, or to render financial aid. According to the Local Public Finance Law, the former is called the 'national treasury obligatory shares', and the latter, the 'national treasury grants-in-aid'. In principle, all the expenses necessary to perform local functions should be borne by the local governments themselves, but for certain kinds of task their costs should partly be shared by the national government. There are three different obligatory shares. 1. Ordinary obligatory shares. The ordinary obligatory shares are granted for expenditures for specific or nationwide programmes which require financing by the national government. They are of interest to both national and local governments, but are conducted by the local governments for the sake of efficiency. From the standpoint of the national government, some grants are disbursed to maintain a certain level of services for specific statutory functions of local governments. Also, they are provided to mitigate the financial burdens of local governments. Examples of obligatory shares include compulsory education and livelihood protection (i.e. cash benefits for the poor). 2. Obligatory shares of public works. These concern payments for public works and other construction works, which are implemented by the local governments in accordance with comprehensive plans to promote the national economy. Examples are obligatory shares of road and harbour construction, and river conservation works. 3. Obligatory shares of disaster-related works. Obligatory shares are also provided for disaster relief which the local governments cannot afford to finance from their general revenues. Although assisted by obligatory shares, local governments must also bear responsibility for some portion of project expenses. In order to secure the necessary expenses
376
Recent Tax Developments
at the local level, the share of local governments is included in the calculation of the basic financial needs under the local allocation tax. Categorical Grants-in-Aid Another type of cost-sharing grant is the categorical grant-in-aid which is given to the local government to encourage and promote specific activities. These grants-in-aid are not obligatory, and thus the national government can create or abolish them with discretion. Unlike the case of obligatory shares, the local governments' shares of the financial burden are not necessarily taken into account in calculating the basic financial needs of the local allocation tax. Given the large role of the specific-purpose grants, the functions of local governments are often separated into subsidized and non-subsidized activities. As is obvious from the above discussion, subsidized activities receive specific-purpose grants, whose costs are partly shared by the national government, whereas non-subsidized activities do not receive such grants. Table 18.8 summarizes the specific-purpose grants to prefectures and municipalities provided by the national government in fiscal 1998. In terms of distribution, 40.0 per cent of total net grants consisted of subsidies for ordinary public works at both the prefectural and the municipal level. In both cases, this was the largest recipient of single, specific-purpose grants. The second-largest grant paid to local governments was the subsidy for compulsory education, and the third went to livelihood protection. They comprise 19.1 and 8.2 per cent, respectively. Among the three sorts of grants discussed, the obligatory shares were the largest. In contrast, payments for agential tasks and grants-in-aid had relatively smaller shares, although they were numerous in kind. Most specific-purpose grants are paid on a conditional, matching, basis. Thus, the matching ratio or the percentage of the grant expresses the degree of support by the national government, which differs from programme to programme. Generally speaking, the matching ratio is determined by various factors such as spillover effects, the national interest, the nature of the project as a local or national function, and the level of financial burden. Table 18.9 compares the percentages of specific-purpose grants. Generally, percentages are higher when heavier financial burdens are imposed on local governments, when the national government is more responsible for the provision of services, and when the national government admits the need to encourage specific services. On the other hand, percentage support is lower for services that are thought of as inherently local functions. Support was reduced for the years 1985-90 with a couple of exceptions because of the large fiscal deficits of the national government, as a temporary measure of specific categorical grant. Usually, a single fixed-percentage grant is determined for each programme which is applied uniformly for all local governments receiving the grant. There are, however,
Table 18.8
Specific-purpose grants from the national government,3 fiscal 1998 Paid to prefectures ¥m.
Compulsory education Livelihood protection Welfare for children Medical expenses for tuberculosis Medical care for the mentally handicapped Welfare for the aged Ordinary public works Disaster restoration works Public works for unemployment-relief Payments for agential tasks Construction works Others Subsidies for financial assistance Others Total a
3011 625 168 733 203717 4256 29289 10062 4 621 104 229 375 5 156 183 838 (27 724) (156 114) 2819 1 693 869 10 163 843
%
29.6 1.7 2.0 0.0 0.3 0.1 45.5 2.3 0.1 1.8 (0.3) (1.5) 0.0 16.6
100.0
Paid to municipalities
Net amounts
¥m.
%
¥m.
1 123 295 388 687 4865
20.1 7.0 0.1
443 627 1 682 750 81544 11 027 168 133 (10414) (157719) 7105 1 670 194
7.9 30.2 1.5 0.2 3.0 (0.2) (2.8) 0.1 29.9
5 581 227
Including special grants for traffic safety measures and municipal charges on government's assets. Source: As Table 18.7.
100.0
3011 625 1 292 028 592 404 9122 29289 453 689 6 303 85 310918 16 183 351971 (38 138) (313833) 9925 3364061 15 745 070
% 19.1 8.2 3.8 0.1 0.2 2.9 40.0 2.0 0.1 2.2 (0.2) (2.0) 0.1 21.3
100.0
Table 18.9
Proportion of support coming from special-purpose grants, selected items, 1984-1999 % of total support coming from special-purpose grants Before 1984 1985
Livelihood protection Aid for the aged Aid for children Retirement pensions for employees of social welfare programmes Subsidies for promoting regional agricultural products Subsidies for kindergartens Road construction National roads Local roads Sewage works Refuse disposal plants River repair Source: Data from the MOP.
1993
1994
1995 1996
1997
1998 1999
75
75 50 50 33.3
75 50 50 33.3
75
75
50
33.3
75 50 50 33.3
50 33.3
50 50 33.3
75 50 50 33.3
1986
1987-88 1989-90
1991 1992
70
50 33.3
75 50 50 33.3
75 50 50 33.3
50 50
80 80 80 33.3
70 70 70 33.3
50 33.3
70 50 50 33.3
50
50
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
33.3
75
66.6 60 60 25 60
60
57.5 52.5 52.5 25 52.5
57.5 52.5 52.5 25 52.5
60 55 55 25 55
60 55 55 25 55
55 50 55 25
55
55 50
55 50
55 50
55 50
55
55 25 50
55 25 50
55
55 50 55 25 50
66.6 66.6 25 66.6
50
55
55 25 55
75 50
50
50 55 25 50
25 50
25 50
Local Taxation
379
some cases in which different percentages are applicable to different local governments, depending upon their financial capability. This is typically the case when grants are paid to Okinawa, where the fiscal ability is poorer than any other area. LOCAL FISCAL P E R F O R M A N C E
Restrictions on Local Government Behaviour. A Summary In many countries, considerable attention is paid to intergovernmental fiscal relations. Since in Japan the fiscal system has traditionally been centralized, the national government can influence the behaviour of local governments through its budget. If a strong national government controls local governments too strictly, the extent of local autonomy can be greatly impaired. This is particularly true in Japan because of the country's great degree of centralization. In fact, the national government regulates almost all the activities of local governments. There are two reasons for the dominant role of the national government. First, in legal terms, no clear division of governmental functions exists between the national and local governments. The second is that the local governments tend to get less favourable access to tax resources. In accordance with the first point, the division of governmental functions is too complicated to establish clear spheres of responsibility among the three levels of government, except for defence and other similar services which are obviously the responsibility of the national government. As for all other functions, the national government tends to exercise strict control over the activities of the lower levels of government. With regard to the second point, poorer financial resources at the local level are in principle compensated for by the equalization payment programme. Given the present system of intergovernmental fiscal transfers, what kinds of factors influence the budgetary behaviour of local governments? Stated differently, how does the national government control the local government effectively? As is evident from the above discussion, there are several instruments which the national government can use to intervene, directly or indirectly, in the fiscal decision-making process at the local level. First, concerning local taxes, the base and rates of general taxes cannot be determined by the independent initiative of local governments. The Local Tax Law prescribes the scope of taxes that local governments may levy, defining the tax base and rates in each case. There are, however, some exceptions sanctioned by the Local Tax Law. For example, the law permits local governments to tax at a rate higher than the standard for some taxes, although it sets a maximum limit. Also, local governments are given discretion in imposing specific non-listed taxes which are not prescribed by the law. Although discretionary power over taxes is admitted to a minor degree, local governments must exercise it with the approval or, if approval is not necessary, the informal agreement of the national government. Obviously, this contrasts sharply with the federal fiscal system in the USA or Canada.
380
Recent Tax Developments
Second, local debts cannot be issued by local governments without restriction. Debt issues are limited to specific purposes by the Local Public Finance Law. Moreover, local governments have to seek permission to float debts by applying to the upper levels of government. Thus, debt issuance is subject to many rules and regulations. Before receiving permission to issue bonds, the financial position of the local government is carefully scrutinized. When local governments have large fiscal deficits, they may not issue debts for construction outlays without the special permission of the national government. Most local governments oppose the permission system, demanding the right to issue local debt more freely at their own discretion. In addition, they insist on simplifying the procedures for debt applications, because of their complexity and the time it takes to obtain the permission. Third, there are some problems with the present intergovernmental transfers, involving the grants system, the local allocation tax, and the specific-purpose grants. The former is working fairly well and is useful in achieving the objective of fiscal equalization. Since the grant is unconditional and is provided on a non-matching basis, there is very little room for it to impair local autonomy. It is considered relatively neutral in intergovernmental fiscal relations. The problem is with the latter. The specific-purpose grants are often used to strengthen the control of the national government over the activities of local governments. Needless to say, such grants are also an important means for securing equal access to local public services throughout the country. Local governments are provided with such grants on the condition that they meet the requirements relevant to the grant policy of the national government. If these requirements are not met, the specific-purpose grants may be cut off. To obtain such grants, local governments must apply to the national government. In the case of obligatory shares—say, aid for children or compulsory education— applications are accepted as a matter of course. On the other hand, categorical grants-in-aid are different, partly because the total amount of such grants is determined by the national budget, and partly because, given a fixed budget, the number of applications must be restricted by the national budget authority. Each application for a local project is screened during the negotiations over the national budget each year. There are political issues in this process. In order to secure these grants, governors, mayors, and other officers at the local level visit the principal ministries or agencies in Tokyo to obtain support for their applications. A number of lobbyist groups and Diet members often press for their individual proposals. Since the specific-purpose grants are important fiscal resources, local governments are very keen on obtaining them. In general, the total applications for such grants by local governments exceed the available funds in the national budget. In the severe competition, applications are often rejected. Failure to obtain a specificpurpose grant means that local governments have to give up the relevant activity. An Empirical Comparison between Japan and the USA I have authored under separate cover an empirical study which reveals the impact of the national government on local governments in quantitative terms (see Ishi
Local Taxation
381
1985; 2000, ch. 13). The following summarizes its main ideas, procedures, and conclusions. The basic aim of the analysis was to investigate the impact of various types of national government fiscal instruments, mainly grants, on local government budgets. At the same time, an attempt was made to compare Japan and the USA, although both countries have quite different fiscal systems. First, a model was constructed to express local fiscal behaviour under the Japanese revenue and grant system, following past attempts developed in the USA.15 Next, estimations were made using Japanese data with rather sophisticated data processing techniques. Finally, policy issues were discussed in light of the US experience. The main results are as follows. An increase in unconditional grants provides local governments with a substantial amount of deliberate budget sources. The more deliberate sources there are in budgets, the more room local authorities have to exercise their own discretion. Thus, discretionary expenditures tend to be increased, and deliberate taxes to be reduced, in response to an increase in unconditional grants. In contrast, categorical, conditional grants substantially affect the determination of the size of local government expenditures, subsidized or nonsubsidized. Obviously, certain items such as social capital, social welfare, and education are more or less controlled by the use of conditional grants-in-aid. These conclusions drawn from the Japanese experience were much more plausible than those derived from the US case. The results were expected at the outset, because the control of the national government via various grant policies is more dominant in Japan. An institutional comparison of Japan and the USA in terms of intergovernmental fiscal relations would make the difference in the empirical results more apparent. In the USA, local governments are endowed with discretionary powers to pursue their fiscal goals to an extent unimaginable in Japan. This is obvious from the fact that there are large differentials in public services (such as social welfare programmes and wages of government officials) between regions, reflecting the different levels of discretionary expenditures. Similarly, local governments can set their own taxes, determined in terms of both tax bases and rates. This is not the case in the Japanese centralized system. Also, local bonds can be issued by local governments without any restriction. Thus, US local governments possess independent sources of revenue and so are subject to much less central government control than Japanese local governments. Note should also be taken of the difference between the USA and Japan in terms of the nature of intergovernmental grants. Generally speaking, both countries have similar categories of grants, although there are no open-ended matching grants in Japan. Control from the central government seems to be more stringent in Japan, however. For example, the central government is empowered to determine the volume not only of the 'mandated' expenditures supported by the matching grant, but also of 'discretionary' expenditures of the recipient governments. This implies that
15 See Henderson (1968), Gramlich (1969), Gramlich and Galper (1973), Galper el al. (1973).
382
Recent Tax Developments
the activities of lower levels of government are controlled by upper levels in conjunction with the execution of the grant policies. Considering these analytical results, attempts should be made in Japan to alter the present structure of intergovernmental fiscal relations so as to create a more decentralized system, in which local governments can achieve their own goals supported by general-type grants. Conditional grants on a matching basis should be converted into unconditional block grants to some extent for the purpose of strengthening local autonomy.
19 Appraisal and Further Reform
In this final chapter, it is necessary to assess the overall performance of past tax reforms which has developed in three stages; that is, the Nakasone and Takeshita reform packages from the mid-1980s, and recent tax reform in the 1990s. From an academic point of view, Japan's tax reforms have been far from satisfactory, and further steps are required to improve the system with regard to equity, neutrality, and simplicity. We must seek how best to achieve these aims. The purposes of this concluding chapter are twofold. One is to evaluate the tax system in terms of basic tax criteria, tracing back to past experience of tax reforms, and to examine possible means of improving the tax system as a whole. The other is to ask where we should go from here to create an ideal tax system in Japan.
ASSESSING TAX REFORM
Appraisals in View of the General Criteria for Tax Reform In undertaking such a fundamental reform of the Japanese tax system, it is important to specify clearly the goals or criteria that should guide such an endeavour. Generally speaking, the essential criteria for assessing a tax reform are equity, neutrality, and simplicity. An equitable tax system is indispensable, not only for attaining economic and social objectives, but also for maintaining a basic respect for the tax system from taxpayers. In initiating the sweeping tax reforms of the late 1980s, Japanese taxpayers have attached great importance to being allocated a fairer share of the tax burden. A more neutral tax system is necessary to improve economic performance in the private sector. Any tax inevitably discourages the type of economic activity that is taxed. A more neutral tax system, however, would interfere with private decisions as little as possible; that is, it would not unnecessarily distort work efforts, consumption, saving, and investment. A simpler tax system is essential so that taxpayers can easily understand how to pay taxes and will do so with a high degree of voluntary compliance. In addition, with a simpler tax system, fewer resources would be devoted to socially unproductive activities such as tax avoidance and tax litigation. It is now necessary to assess the tax reform from these essential tax criteria. Looking back to both the three stages of the reform process from the mid-1980s
384
Recent Tax Developments
up to today, the main tax package consists of four key elements: 1. a simplification of progressive rate structures in the individual income tax and the local inhabitants' tax towards a flatter rate scheme; 2. the repeal of tax preferences on small savings with tax-free interest (effective from April 1988); 3. a cut in the basic tax rate on corporate income; 4. the introduction and further reform of the consumption tax (Japan's VAT).
Equity When we examine the above reform package in greater detail with respect to equity, more importance is placed on horizontal equity than on vertical equity. In fact, it may be protested that the mitigation of the progressive rate structure substantially impairs vertical equity. In particular, the lowering of the top rate may lead to troublesome distributional effects among taxpayers. Although such criticism is understandable, we believe that greater emphasis should be put on the restoration of horizontal equity, because the Japanese tax system has sacrificed fairness among people with equal economic incomes. Reduced progressive rates have major advantages over steeply progressive rates for middleincome-earners because of the wider band of income brackets and the lower marginal rates. Such a tax considerably diminishes the inequality of tax treatment of families receiving equal incomes. In particular, horizontal equity for salaried workers whose income is fully apprehended by tax offices can be recovered by a flatter tax rate scheme. Substantial rate reductions lessen problems inherent in steeply progressive rates, such as bracket creep, the bunching of income at a specific life-stage, and incentives to shift income artificially to family members who are subject to lower tax rates. All of these problems tend to disadvantage wage- and salary-earners, as compared with the self-employed entrepreneurs. Attention should also be paid to the role of improving horizontal equity by introduction of the broad-based indirect tax. We often neglect the question of horizontal equity in relation to indirect taxation, whereas the argument against vertical equity is widely discussed. It is, however, stressed that the broad-based indirect tax is more likely to achieve horizontal equity between wage workers on the one hand and other income recipients on the other than income taxes, the tax bases of which are highly distorted and shrunken. Such a tax is considered to spread the tax burden more evenly among all consumers. Thus, the shift from income taxes to the consumption tax can be justified on the ground that it will contribute to a reduction of horizontal inequity to some extent.1 1 Mathews presents a similar view regarding the reform of the Australian tax system. He concludes: 'The foregoing analysis has been intended to demonstrate that we cannot have an efficient, equitable and effective tax system in Australia without a broad-based consumption tax" (Mathews 1983, 23).
Appraisal and Further Reform
385
The distributional inequity of the tax reform derives not from the simplification of rate structures accompanying the reduced top rate, but from a failure to broaden the income tax base. Flattening tax rates cannot be defended unless they are accompanied by base-broadening efforts. If the two measures are made concomitantly, vertical tax equity can be achieved reasonably even by a flatter rate structure. However, the Takeshita tax reform made no attempt to broaden the tax base of the individual income tax for fear of adverse political repercussions. On the contrary, the income tax base was actually narrowed, by admitting in 1987 a special exemption for spouses for one-earner couples in addition to the existing exemption. Similarly, an additional exemption for dependants aged 16—22 was introduced in 1989 alongside the general exemption for dependants. These new schemes obviously helped to narrow the tax base, and should be severely criticized as hindering the establishment of a comprehensive tax base. They resulted in unfairness to those who were not eligible to take advantage of them. Neutrality The second criterion for tax reform is that taxes should distort private economic choices as little as possible. Traditionally, the Japanese government has used the tax system to achieve specific social and economic objectives. In practice, this traditional rule remains unchanged even now, and the new exemptions described above were adopted as part of the current tax reform. As a result, the additional tax provisions evidently impair neutrality as well as fairness. The special exemption for spouses was doubled mainly to mitigate the horizontal inequity between salaried workers and self-employed entrepreneurs. This provision causes a new distortion, however, in that it acts as a disincentive for spouses to participate in the labour market. Another distortion accompanies the repeal of tax-exempt interest on small savings accounts. A flat 20 per cent withholding tax was levied on almost all savings from fiscal year 1988 by the Nakasone tax reform, but this resulted in distorting the portfolio investment of individual savers. In 1988, capital gains on the sale of stocks were still not taxed except in a few cases ('continuous and voluminous transactions'). Some financial assets, such as life-insurance-type savings (Yard Hoken), discount bank debentures (Waribiki kinyu sai), or investment trusts (Toshi Shintaku), were still treated more favourably in terms of after-tax rates of returns. Thus, a shift from one financial asset to another was stimulated by the termination of tax-favoured small savings accounts in April 1988. Traditional forms of personal savings, such as postal savings and ordinary time deposits, have increased much more slowly on a monthly basis since then. 2 Also, a substantial amount of money for 2 In some sense, these new tax-induced distortions may not reduce overall efficiency in the economy, because other distortions have existed in the form of tax-free saving accounts. Uniform tax rates, say 20 per cent, may gradually offset the existing distortions of savings and help improve economic neutrality in money markets.
386
Recent Tax Developments
portfolio investment flowed into the stock market, reflecting the favourable tax treatment of capital gains on stock sales. In response to complaints among general taxpayers about the unfairness involved in capital gains taxation, the Takeshita tax reform strengthened the tax burden on individuals' capital gains on stocks. Without exception, these capital gains were taxed from April 1989 in one of the following two ways, as explained previously: 1. Security companies withheld 1 per cent of gross proceeds (5 per cent of deemed capital gains ratio x 20 per cent of tax rate). 2. Taxpayers filed final income tax returns and payed 26 per cent (20 per cent of the national individual income tax and 6 per cent of the local inhabitants' tax) of their capital gains. Information returns were submitted to the tax offices by security companies. The selection of which method to use was left to private investors.3 In 1996, the deemed ratio to be multiplied by gross proceeds was raised to 5.25 per cent from 5 per cent. Thus, the taxpayer was required to pay 1.05 per cent (0.0525 x 0.2) of the stock sales price under the withholding system. This device may be praised in that economic neutrality is furthered by weakening the distortions by the tax system on private investors' decisions in markets, although it is far from perfect in reforming capital gains taxation. At least, these methods help prevent individuals from making portfolio decisions based on tax considerations. When the security transaction tax was repealed in 1999 to encourage the stock market, it was decided that capital gains tax on stock sales ought to be limited exclusively to the use of the declaration method. It was proposed that a more generous withholding system should be abolished to secure a more equitable tax burden on capital gains. As a result of postponing its enforcement, a new scheme of capital gains tax will start from April 2001.
Simplicity A good tax system should be as simple as possible. A complex tax system imposes high compliance costs on the community and high administrative costs on the tax authorities. These considerations suggest that, where possible, tax reform should be 3
When stocks are initially put on the market, a sharp rise in the stock price generally occurs in the short term. If stock-holders sell immediately in this situation, they can easily earn huge capital gains reflecting the increased stock price. According to the revised tax bill presented to the Diet in November 1988, the following is treated as an exceptional case. If such stocks have been held for three years and are then sold within one year of being placed on the stock market, the system of withholding 1 per cent is not permitted to apply to gross proceeds. This special treatment is considered too lenient, and is criticized as unfair. Thus, such gains would be subject to income taxation at 26 per cent under the declaration method. However, in the case of shares that have been held for more than three years before the stock becomes listed, the tax rate is reduced to 13 per cent. This revision in the tax law was added in response to the 'recruit scandal' in relation to such capital gains as described above (see Ch. 8).
Appraisal and Further Reform
387
directed towards simplicity. A simpler tax system plays an important role in enhancing the efficiency of tax administration. With a view to simplicity of tax administration, as has already been stressed, the Japanese tax system has traditionally placed greater reliance on withholding taxes at source. One of the most marked features of the Japanese system is that the majority of wage and salary workers are not required to file tax returns. Thus, nearly 80 per cent of individual income taxes are withheld at source, resulting in little direct cost to the tax authorities. In order to support the advantage of such a withholding system, two other devices exist: a year-end adjustment of wage and salary by employers, and the fact that almost all of employees' investment income is subject to separate taxation (see Chapter 4). Tax reforms strengthen the use of withholding, and as a result accelerate the trend towards making the tax system simpler. A separate flat tax (20 per cent) is applied to investment income at source via withholding. Even if a new tax on investment income was introduced, it would neither complicate the existing tax system nor increase the cost of tax compliance and administration. As regards the design of the consumption tax structure, the subtraction method of the value added tax with no invoices was selected because of its simplicity. This was justified by maintaining that the use of ledger entries would give taxable firms less trouble in computing the tax due than the compulsory application of invoices. To be sure, tax reforms may be highly regarded in view of tax simplicity, but it should be noted that this advantage is purchased at the cost of equity. Measures to make the tax system more equitable might require the elimination of withholding and separate taxation. Complexity might also cause economic distortion. Inevitably, the objectives of equity, neutrality, and simplicity conflict during the process of tax reform, and further steps will be needed to make better choices between these criteria.
DIRECTION OF FUTURE REFORM: THREE MAJOR POINTS
Further Necessity of Base-broadening
The process of tax reforms starting from the mid-1980s has not been altogether satisfactory. Many reforms have been proposed, and some of them have already been put in place. However, some necessary reforms have been tactically avoided because the government and the LDP feared strong political repercussions. Thus, great efforts have been made to conceal tax increases from the taxpayers themselves. As a result of political compromises, a substantial number of inadequate and inconsistent tax devices have been left for future reforms to take care of. Without a doubt, the first thing to be done is to broaden the tax base as much as possible. There is wide agreement that an inequitable tax system is unlikely to be remedied without a comprehensive tax base. If certain types of income are omitted from the tax base, or if particular uses of income are treated more favourably than
388
Recent Tax Developments
others, then taxpayers with similar economic situations will not be taxed equally. In addition, since deviations from a comprehensive tax base tend to accrue to higher-income-earners, it is absolutely necessary to broaden the tax base in order to achieve vertical equity. The same arguments are equally important in the pursuit of other objectives of tax reform, such as neutrality and simplicity. The three tax reforms contain very incomplete base-broadening measures. The cut of tax-free interest on small savings accounts is one successful example, but other measures remain to be taken. As has been pointed out in Chapter 6, there are many base-broadening features of the tax revisions which should be enforced. Two points in particular are worth noting relating to the individual income tax. First, there are still too many unnecessary and unimportant exemptions, deductions, and credits. If some of these were eliminated or merged so as to obtain a broader tax base, a more uniform treatment of all sources and uses of income would be achieved and would lead to a more complete reform of the tax system. Several candidates for this purpose can be enumerated from the existing special provisions: e.g. exemptions for aged persons, widows, or working students; special exemptions for spouses; additional exemptions for dependants aged 16-22; deductions for fire and other casualty insurance premiums; deduction for life insurance; and so on. Apparently, it is very difficult to proceed with tax reform further along this line, because these special provisions are strongly supported by specific interest groups. Erosion of the tax base certainly has a heavy political cost; however, further steps should be taken to facilitate future tax reform. Second, the tax treatment of capital gains from the sale of securities has become very controversial among general taxpayers in relation to the elimination of unfair tax burdens. Irrespective of the recent complaints, such capital gains should be taxed more heavily and should be subject to aggregated taxation with progressive lax rates combined with other income sources. The present proposals are insufficient to constitute a genuine income tax reform incorporating a comprehensive definition of income. Opposition parties and labour unions agree to a change from the present separate tax on capital gains with a flat rate into a system combined with ordinary income. Since political compromises have been made to strengthen taxes on capital gains, future reform should move towards a comprehensive income tax, on condition that tax rates ought to be further flattened. Similarly, the reforms to date have insufficient to broaden the base of corporate taxes. The basic rate of the national corporate tax decreased from 42 per cent to 37.5 per cent from 1987 to 1997, but no major efforts have been made to broaden the corporate tax base for a long time. However, the repeal of the provision applying a lower tax rate on dividends may have been of help as a base-broadening measure in 1990, because dividends are fully taxed as a part of corporate taxable income in addition to retained earnings. Except for minor examples (e.g. the increased taxable portion of intercorporate dividends, or the restricted deductions of interest payments on land purchases), tax reformers did not propose any noteworthy ideas for cutting the use of special tax
Appraisal and Further Reform
389
measures in the corporate tax until 1997. As was argued in Chapter 17, however, thereafter tax-free reserves, depreciation allowances, and others have been re-examined to a substantial degree, and special tax measures relevant to the corporate tax have begun to be eliminated at a much greater speed than those relating to the individual income tax. A few words must be added with regard to the full taxation of capital gains from the sale of securities. Heated arguments have taken place for or against introducing the tax identification number (TIN), which, like the US social security number, is intended to assess such gains accurately. No one can deny the necessity of TIN when considering the fair and accurate taxation of capital gains from the standpoint of tax administration. The adoption of TIN in the Japanese tax system is a prerequisite for imposing proper taxes on capital gains from the sale of securities under the unified taxation method. Before reaching a final decision, we need to consider the genuine possibility of introducing the TIN deliberately.4 If the TIN is successfully adopted by the government, other investment income such as interest and dividends can be treated equally, as ordinary income, in a fashion similar to capital gains. Aggregated taxation will be restored, apart from separate taxation on investment income with a flat rate, and such reform should achieve a return to the Shoup proposals. With a view to the 1993 tax reform, the Tax Advisory Commission discussed at length during autumn 1992 the feasibility of aggregating interest and capital gains from the sale of securities with other incomes in view of comprehensive taxation. Of course, the other side of the coin, i.e. the introduction of TIN, was equally taken into consideration. However, a decisive conclusion was not reached on the future direction of tax reform regarding the taxation of investment income. Two reasons behind this are worth noting. First, as has been explained previously, withholding collection at source in taxing interest and capital gains, separate from other incomes, has been firmly built into the Japanese tax system. From the standpoint of tax administration and revenue-raising, the present collection system is highly valued by both the tax authorities and taxpayers. Therefore, there was no strong motivation to change to a new scheme of tax collection which may promote considerable confusion. Second, related to the first reason, it was not the time to seek a national consensus as to which direction should be taken for a more desirable form of taxation; opinion was so divided that any consensus was unlikely. To sum up, to avoid possible big changes to a well-established system, decisions have been put off until the future. The present tax on investment income, therefore, has remained unchanged for practical reasons, although it is far from satisfactory. 4 Despite their rejection of the Green Card system in 1982, opposition parties and labour unions changed their attitude and began to support the introduction of TIN in conjunction with strengthening the tax burdens on capital gains. An ad hoc subcommittee on the study of TIN within the Tax Advisory Commission sent a mission abroad in September 1988 to investigate the experiences of foreign countries. (I participated in this mission.) As a result of the mission's activities, a report was presented in December to provide basic information for our arguments in favour of TIN.
390
Recent Tax Developments Is Further Simplification of Rate Structure Necessary?
Under the Takeshita tax reform, a five-bracket tax rate structure was imposed, covering a wide range of low to middle incomes. Tax rates were sharply reduced from 12 bands in 1988, although the result remained far from the modified flat tax advanced by the USA and the UK. In order to simplify and reform income tax in Japan, a flatter tax rate scheme has become a most important strategy for tax reform. Since then, great efforts have been made to flatten the rate structure, and finally in 1999 a four-bracket rate structure with the top rate of 37 per cent was introduced for the national individual income tax (the combined top rate with local income taxes is still 50 per cent). In theory, the higher the tax rates are, the more taxes tend to distort economic choices in regard to work efforts, saving, investment, etc. Furthermore, the higher the tax rates are, the more valuable any omission from the income tax base becomes. As a practical measure for securing neutrality to some extent, specific incomes, particularly investment incomes under steep progressive rates, tend to be taxed at lower rates, separate from ordinary income: instead, it would be better to develop a more comprehensive definition of income with a flatter rate structure. Thus, an important target of the recent tax reform movement was to keep tax rates as low as possible, while reducing the number of income brackets. Of course, the reduction of tax rates should be accompanied by a broadening of the tax base. It is obviously much better—i.e. more equitable, more neutral, and simpler—to impose low tax rates on all incomes than to levy high rates on a narrower range of incomes. Tax reforms were a long way off from these theoretical requirements. In particular, as often noted, the base-broadening efforts are inadequate; therefore the lowering of tax rates cannot be justified. Lower rates enable the tax base to be broadened, but the government did not fully stress this factor for fear of political repercussions. Reflecting the insufficient attempt to achieve base-broadening, the reduction of progressive tax rates had to be very limited, leaving the flatter rate scheme unfinished, in comparison with the experiences of other countries in the process of world tax reform of the 1980s (see Pechman 1988, 1-14). However, the Japanese government boldly reduced progressive tax rates to a considerable extent in the 1990s, without any base-broadening efforts. Is further simplification of rate structure necessary from now on, given the individual income tax reform in 1999? In order to aggregate all income into a unified income tax, further simplification of the progressive rate structure may be desirable. In so doing, capital gains, interest, and dividends must be taxed so that their after-tax rate of return is reasonably acceptable. In this respect, the top rate might be lowered by as much as 30—5 per cent as a future policy target. At the moment, however, first priority ought to be placed on broadening the tax base, rather than a further flatter rate structure, mainly because of securing an equitable tax burden. Distributional equity will be greatly impaired by a flatter tax rate if the tax base does not include eroded income, such as investment income, which
Appraisal and Further Reform
391
mostly accrues to higher-income-earners. By combining a more comprehensive definition of income with less steeply graduated low rates in a global system, flatter tax proposals are able to achieve gains in equal tax treatment of equals, neutrality, and simplicity with no sacrifice of vertical equity. Further reform in Japan should pursue the above combination in order to achieve the major criteria of tax reform. Alternatively, such an imperfect combination as the present tax package, permitting separate taxation (i.e. a less comprehensive base and a less progressive rate structure), cannot be defensible any more, and further reforms are necessary in order to achieve genuine income tax reform.
Reformulating Japan's VAT Japan's VAT has substantially been amended in recent years, but there still remain two areas which need to be improved further in the future. In comparison with the most common type of VAT used in EU countries today,5 the consumption tax in Japan contains several special measures which are likely to impair the possible merits of such a tax. These measures are acknowledged as falling short of the full standard treatment under the usual type of VAT, but are justified by the existence of administrative and compliance problems. No doubt, political considerations were also involved, since it was necessary to obtain support from opposing groups in the retail and wholesale industries who played a major role in the sales tax controversy noted previously. Special treatments should be phased out in favour of the normal scheme of VAT. First of all, the special simplified procedures for measuring the tax base impair the advantages of broad-based indirect taxes. This special scheme may be justified to some extent in order to simplify the procedure for charging the tax, but the current scheme admits too wide coverage for using the special rule. Taxable traders whose annual sales are less than ¥200 million are eligible to apply the simplified procedures, although the limit was lowered from the original annual sales of ¥500m. The measure intended to simplify the procedures for computing the tax base may be of some help in mitigating the taxpayers' book-keeping burden, and so may be commended in view of tax simplicity. Thus, the coverage of eligible taxable traders should be narrowed to a great extent, say to ¥50 million in terms of annual sales. Second, the consumption tax does not employ the tax-credit method which is used almost universally nowadays, although it is characterized by a very comprehensive VAT. This is one of the special treatments that the consumption tax permits. The invoice system is not required at all stages of transactions. Under the account method with no invoices, a firm's tax base is computed as the difference between the 3 Shoup (1990) points out that the most common—indeed, almost universal—type of VAT contains the following characteristics: consumption type, destination principle, tax credit method, multiple rates that are tax-exclusive, and exemptions rather than zero-rating.
392
Recent Tax Developments
sales of the firm in question and its purchases from other firms. Like the special simplified procedures, the plan to drop the use of invoices has been adopted mainly for political reasons, to help dispel opposition to the introduction of the new tax. As a result, the tax-credit method could not be chosen. The account method has a lot of demerits as compared with the tax-credit method. For instance, without the aid of invoices there is no means to ascertain the chain of transactions from one stage to another. Thus, it is very difficult to assess the value of the tax base accurately because a strong incentive for cheating will remain. In addition, although export sales are zero-rated in accordance with the destination principle, the account method cannot properly exempt exports since a tax return of all prior-stage VAT payments cannot be exactly measured without invoices. If the tax-credit method is employed, the exporter can calculate a tax return based on his purchase invoices, and can get a cash refund of that accumulated from earlier stages from tax authorities. If we seek the normal VAT scheme in the Japanese tax system, the proposed form of consumption tax should be replaced by a tax-credited, invoiced VAT. Third, one of the unique features of the original consumption tax was its extremely low rate. Although the tax rate was raised to 5 per cent in April 1997, even 5 per cent standard tax rate is the lowest among major VAT countries, Canada being second lowest with a rate of 7 per cent. Moreover the consumption tax merely has one single tax rate (plus a zero rate on exports), rather than the multiple-rate structure prevailing in other countries. Such a low tax rate was adopted as a political compromise when the consumption tax was created in 1989; the 3 per cent rate was lower than the 5 per cent sales tax proposed in the Nakasone tax reform, and was chosen so that it would be accepted more readily by taxpayers. However, while it proved successful in achieving acceptance, it could cause other troubles. For instance, in theory, VAT should be passed on to the ultimate consumers, but taxable traders may find it difficult to increase their prices by such a minor amount. There may be the possibility of a failure of forward-shifting at each stage of transactions. In order to pass on the tax burden more fully, the consumption tax rate might be increased, say to 5 per cent initially and 10 per cent later. Now the difficulty of forward-shifting may be mit-igated at the 5 per cent tax rate. In anticipation of increasing the consumption tax rate, it becomes a controversial issue whether or not the earmarked tax should be introduced to be linked to expanding social welfare programmes, although this would be difficult to justify in theory. If the tax rate is to be increased, it would be good to consider a multiplerate structure for the consumption tax. Certain kinds of multiple rates improve administrative performance and enhance compliance. A prompt shift from commodity-differing rates in the existing indirect tax system to a single low tax rate tended to cause difficulties to taxpayers attempting to adjust to the new situation in 1989. Since many still think that the tax rate on food ought to be less than that on motor cars, a uniform 3 per cent rate was likely to be considered unfair because tax
Appraisal and Further Reform
393
burdens on luxuries are greatly reduced from rates as high as 20 and 30 per cent. The multiple-rate structure would aid in the removal of such difficulties. Lastly, attention should be paid to the special treatment of freeing smaller undertakings from the consumption tax. All VAT countries admit problems in applying the normal tax scheme to small traders because of their specific business structure or activities. Thus, some exemption system is usually devised to free certain taxpayers from VAT. At present, the consumption tax fixes the exemption level at ¥30 million in terms of annual sales; that has been unchanged since the beginning, but this is extremely large by international standards. The larger the level of exemption, the more the economy is distorted. For instance, some firms may split themselves into two or three smaller operations in order to escape paying taxes on their sales. To achieve economic neutrality, the tax-exempt level of the consumption tax should be lowered greatly, say, to less than ¥10 million. It is apparent that the Japanese consumption tax departs from the standard type of VAT in several respects. In fact, it may be called an intermediate or imperfect form of VAT, although recent developments of Japan's VAT have made it much better. Since the consumption tax was introduced in such an incomplete form, further reform will be necessary to achieve greater economic fairness.
CONCLUDING REMARKS! WHERE SHOULD WE GO FROM HERE?
A Hybrid Income-Expenditure Tax? A major conclusion that has emerged from the preceding discussion is that the Japanese income tax was, and still is, a hybrid. Since many forms of savings have been allowed full or partial tax exemption, it is frequently pointed out that the income tax is really half-way towards being an expenditure tax. Although interest and capital gains on stock began to be levied at a flat rate from 1988 and 1989, they are still taxed separately from other incomes. We must say that this is far from a genuine income tax. This being the case, where should we go from here? Should we move towards an expenditure tax, or return to a genuine income tax? While the tax reform movement has been ended, it is an opportune time to ask ourselves such questions. There are three alternatives for future tax reform.6 One choice would be to expand the range of tax benefits on all forms of savings in order to take us nearer to the expenditure base. Such a piecemeal approach would avoid the full-blooded switch to a direct expenditure tax, and as a consequence might gradually mitigate the investment distortion among savers. Another alternative would be to get back to a more comprehensive income tax and curtail the area of tax benefits for investment income. If special tax treatments (i.e. non-taxable measures and the application of 6 The following papers provide evidence of the differing views on future tax reform in Japan: Shoup (1990), Kaizuka (1992), Noguchi (1992), and Hatta (1992).
394
Recent Tax Developments
separate taxation) on interest, dividends, and capital gains are removed, the broader tax base would enable the same revenue to be collected at lower tax rates, and would recover equity and improve incentives. Finally, as a third possible solution, we could maintain the existing intermediate position; but this would be the least defensible of the three, and may be justified as a form of scheduled income tax. Judging by the recent tax reforms, it seems that a decision has been made to opt for the second alternative, a return to a genuine income tax. In my view, this is the best choice, and can be justified by its ability to improve the horizontal and vertical equity, and furthermore the neutrality, of the present tax system. Which Taxes are More Equitable? Interestingly enough, in the recent Japanese tax reforms, the expenditure tax approach has not made a practical contribution to reforming the country's existing tax system. Following the long-standing arguments in the literature of public finance, a limited academic debate has occurred among tax experts, but it has not had a major impact on the proposed reform package. In fact, few proponents have insisted on the possibility of developing an expenditure tax. As a consequence, we have selected to return from a half-way to a more perfect income tax, while avoiding controversial arguments over the choice between income and expenditure taxes. The most important reason for this decision seems to be major complaints among taxpayers regarding inequitable taxation, particularly tax preferences on capital gains. Now, most Japanese taxpayers regard a tax that exempts investment income as unfair. Indeed, it is a well-known fact that individuals in the top brackets are given preferential treatment of capital gains from the sales of securities and land, interest and dividends, and other favourable provisions under the present income tax. These special provisions permit them to accumulate large fortunes with little or no payment of income tax. An expenditure tax would theoretically preclude the necessity of taxing investment income. In principle, an expenditure tax does not reach such incomes until they are spent. While the accumulation of savings continues to provide people with great benefits in terms of safety, security, and social prestige, the expenditure tax provides no means of taxing such increased spending power in accordance with the ability to pay. Thus, it leads to an excessive concentration of wealth in the long run. Proponents of an expenditure tax do not explain this to the general public. Why are taxes on interest, dividends, and capital gains not necessary in the Japanese tax system? Even if any attempt were made to justify not taxing investment income when realized but not spent, Japanese taxpayers in general would not support this point of view. Considering the taxpayers' perception of inequality in the context of the recent tax reforms, we believe that the expenditure tax does not satisfy the test of taxation according to the ability to pay; it would exclude truly necessary income, such as capital gains, from the tax base. (See, for a similar argument, Pechman 1987, ch. 6.) In Japan, most taxpayers now put greater importance on strengthening capital gains taxes.
Appraisal and Further Reform
395
As often stressed by proponents, an expenditure tax is theoretically an attractive proposition. We must acknowledge that it possesses a number of advantages over an income tax. In particular, an expenditure tax would obviate the need for inflationary adjustments to measure real income, would avoid the use of unrealized bases for computing capital gains and losses, and would eliminate the complications of accrual account adjustment for depreciation. Moreover, it would have a neutral effect on the incentive to save and invest. Although these relative merits can reasonably be advanced by professionals, the expenditure tax cannot escape the practical problems of perceived inequity among general taxpayers, particularly in Japan. The most conspicuous difference between income and expenditure taxes is in the definition of the time period relevant for measuring the tax base. With income tax, income is defined for a specific time period, currently one year. If we take a longer view and think of an individual's income over his lifetime, the difference between income and expenditure becomes unimportant. It can be deduced under some assumptions that a man's total lifetime income is equal to the total of what he spends on consumption and what he bequeaths to others. Therefore, the effect of raising a tax on consumption on an annual basis and a tax on bequests after death would be the same as imposing a tax on total lifetime income. Based on such a theoretical framework, proponents of an expenditure tax usually argue that an income tax is unfair because it taxes income when it is earned and once again when the amount of income saved earns interest. With the expenditure tax, in order to avoid double taxation, either saving or interest is made exempt. Because of a lesser investment distortion, this may prove the superiority of the expenditure tax over the income tax. However, if we regard income tax as a tax on annual income, then double taxation of savings is not a real issue, because an income tax on savings is regarded as independent of a tax on interest which is earned during a different time period. In effect, whether an income tax is more equitable than an expenditure tax depends on a value judgement. In Japan, the income tax is now preferred because of its greater equity, especially in terms of its capacity to remove unfair tax treatments on investment income. In spite of the fact that income tax distorts the choice between consumption and saving, its superiority would be supportable, and even desirable, in view of this greater equity. Towards an Ideal Tax Mix The tax system in Japan has been reformulated significantly in recent years with consequences that are likely to be far-reaching. Recent tax reforms as a whole can be praised in terms of their effects on improving equity, neutrality, and simplicity. To sum up, the ultimate goal of tax reform should be to alter the relative shares of major tax sources to secure an ideal tax mix. As a result of recent reforms, the relative reliance on tax revenues was expected to be considerably shifted from an income base to a consumption base. In this respect, a new mix of direct and indirect
396
Recent Tax Developments
taxes would be introduced. However, for the early 1990s immediately after the adoption of the consumption tax in 1989, there was no shift from an income base to a consumption base to be observed in national and local taxes. Indeed, the relative share of direct taxes to the total (including local taxes) was 79.3 per cent in 1990, 79.3 per cent in 1991, and 79.9 per cent in 1992. However, as a result of continuous reduction in the individual and corporate income tax in recent years, the relative share of direct taxes in total national taxes has unintentionally been lowered; i.e. 68.9 per cent in 1998, 67.2 per cent in 1999, and 69.0 per cent in 2000. The present tax mix is far from an ideal targeted level (say, 60:40 or 55:45). However, to the extent that consumption taxes are substituted for income taxes and progressive tax rates are mitigated in favour of higher income classes, there is likely to be a corresponding need for strengthening the tax burdens on wealth, in order to maintain distributional equity and to avoid an excessive concentration of wealth. Indeed, reflecting the past trends of sharp rises on land and stock prices, many believe that wealth taxation should be used for redistributional considerations. For this purpose, the tax burdens on wealth and capital transfers (e.g. gift and inheritance taxes, property tax, land-holding tax, etc.) should have been increased more heavily than in the Takeshita tax plan. A response to this criticism was the introduction of the land value tax as a new land-holding tax in 1992, but it was frozen in its enforcement from April 1998. There is widespread criticism that the heavier use of consumption taxes is regressive by nature; that is, the poor thereby pay relatively more of their income in taxes than the rich. If we follow this reasoning, the basic direction of recent tax reform cannot be supported on the grounds that distributional equity can be achieved only by means of a progressive income tax. However, the problem of regressive tax burdens stemming from consumption taxes can substantially be resolved by combining the two other approaches. The first of these is designed to acquire overall progressiveness of the whole tax system by changes in the structure of tax sources, placing heavier burdens on wealth and capital transfers as well as creating a broad-based indirect tax. In addition, the base-broadening measures in the individual income tax must be of great help towards this end. The second approach is directed towards obtaining greater distributional equity via the expenditure side of government budgets, in particular by means of welfare benefits to individuals with low incomes. The combination of these intended effects might achieve better results than the steep progressive income tax. The reason for preferring them to the existing heavy reliance on a progressive income tax is that they can be implemented in such a way that the improvement of distributional effects and neutrality can better be achieved. The proposed tax mix should thus be further biased towards taxes on consumption and wealth, although there is no way of attaining an ideal combination of these three sources.
References Aaron, H. J. (ed.) (1976). Inflation and the Income Tax, Brookings Institution, Washington, DC. and Galper, H. (1985). Assessing Tax Reform, Brookings Institution, Washington, DC. Ackley, G. (1976). 'Fiscal, Monetary and Other Related Policies' (with the collaboration of H. Ishi), in H. T. Patrick and H. Rosovsky (eds.), Asia's New Giant: How's the Japanese Economy Working, Brookings Institution, Washington, DC. Administrative and Management Centre (1982). Rincho and Gyokakushin: Record of Administrative Reform during2000 Days (in Japanese), Tokyo. Amano, A. (1992). Global Warming and Economic Growth—Modeling Experience in Japan (National Institute for Environmental Research). Andrews, W. (1974). 'A Consumption-type or Cash Flow Personal Income Tax', Harvard Law Review, 87 (April). Aoki, T. (1986). 'The National Tax System', in T. Shibata (ed.), Public Finance in Japan, University of Tokyo Press. (1988). 'Tax Reform in Japan', Bulletin for International Fiscal Documentation, March. Atkinson, A. B. (1970). 'On the Measurement of Inequality', Journal of Economic Theory, 2 (September). Australia (1985). Reform of the Australia Tax System—Draft White Paper, Australian Government Publication Service, June, Canberra. Ballentine, J. G. (1986). Tax Reform in Japan and the United States—A Stimulus to New Economic Vitality? Japan Society, New York. Bank of Japan (1966). Hundred-year Statistics of the Japanese Economy (in Japanese), Statistics Department, Bank of Japan, Tokyo. Bayoumi, T., and Collyns, C. (eds.) (2000). Post-Bubble Blues: How Japan Responded to Asset Price Collapse, International Monetary Fund, Washington, DC. Bird, R. M. (1964). 'A Note on Tax-sacrifice Comparison', National Tax Journal, 17 (September). Bittker, B. I., Galvin, C. O., Musgrave, R. A., and Pechman, J. A. (1968). A Comprehensive Income Tax Base? A Debate, Federal Tax Press, Bradford, Conn. Blinder, A. S., and Solow, R. M. (1974). 'Analytical Foundations of Fiscal Policy', in A. S. Blinder et al, The Economics of Public Finance, Brookings Institution, Washington, DC. Boskin, M. J. (1978). 'Taxation, Saving, and the Rate of Interest', Journal of Political Economy, 86 (April). Bronfenbrenner, M. (1950). 'The Japanese Value-added Sales Tax', National Tax Journal, 3 (December). and Kogiku, K. (1957). 'The Aftermath of the Shoup Tax Reforms', National Tax Journal, Parts 1 and 2,10 (September and December). Canada (1987). The WhitePaper, TaxReform, Department of Finance, 18 June, Ottawa. Chelliah, R. }. (1986). 'Changes in Tax Revenue Structure' (mimeo), paper submitted to the 42nd Congress of the International Institute of Public Finance (IIPF). Chiho Jiji Kyokai (1988). Summary of Local Public Finance (Chiho Zaisei Yoran), January, Tokyo.
398
References
Cnossen, S. (1986). 'The Role of Excises in OECD Member Countries' (mimeo), paper submitted to the 42nd Congress of IIPE Cohen, J. B. (1949). 'Tax Reform in Japan', Far Eastern Survey, 18 (December). (1950). 'Fiscal Policy in Japan', Journal of Finance, 5 (March). Council of Economic Advisers (1962). 'Automatic Stabilizers and Fiscal Drag', in Annual Report of the Council of Economic Advisers, January. Deutsch, K. W. (1961). 'Social Mobilization and Political Development', American Political Science Review, 55 (September). Dodge, J. (1949). Letter to L. M. Moss, 30 September 1940, Ministry of Finance, Fiscal History Project Document no. z711-6. EC (1992), Commission Adopts Proposal on Carbon/Energy Tax, 27 May 1992. Economic Planning Agency (1970). Survey of National Wealth, Tokyo. (1975). Actual Performance and Issues on the Distribution of Income and Wealth (in Japanese), Government Printing Office, Tokyo. (1982). Economic Survey of Japan, Government Printing Office, Tokyo. (1990a). 'On Anti-Measures for Price Hike Ensuing from the Consumption Tax' (in Japanese), 28 July. (1990fo). 'On Price Trends after the Introduction of the Consumption Tax' (in Japanese), 7 Oct. - (annual). Annual Report on National Income Statistics, Government Printing Office, Tokyo. Economic Stabilization Board (1949). Reduction of Tax Burden and Slash in Expenditures, 21 July 1949, Ministry of Finance, Fiscal History Project Document no. 7.703-6. Fair Trade Commission (1989), 'On the Results of Monitoring Price Changes and Pricing Methods after the Consumption Tax was Implemented' (in Japanese), 23 May. (1990), 'FTC's Responses in Relation to the Introduction of the Consumption Tax' (in Japanese), 28 July. Feldstein, M. (1976). 'On the Theory of Tax Reform', Journal of Public Economics, 6 (July/ August). Furuta, S. (1965). 'Shifting of the Japanese Corporation Income Tax and the Differential Tax Burdens on Corporations by Size-groups', Keio Economic Studies, 3. (J970). 'The Incidence of Corporation Income Tax in Japanese Industries, 1920-63', Keio Economic Studies, II1. Galper, H., Gramlich, E., Scott, C, and Wignojowijoto (1973). 'A Model of Central City Fiscal Behavior', in H. M. Hochman (ed.), Issues in Urban Public Finance, IIPF, Saarbruken. Gomi, Y. (1985). Guide to Japanese Taxes, Zeikei Shobo Co., Tokyo. Goode, R. (1951). The Corporate Income Tax, John Wiley, New York. (1981), 'Some Economic Aspects of Tax Administration', DM/81/1, International Monetary Fund, 5 Jan. Gordon, R. J. (1964). 'The Incidence of the Corporation Income Tax in US Manufacturing, 1925-62', American Economic Review, 57 (May). Gramlich, E. M. (1969). 'State and Local Government and Their Budget Constraint', International Economic Review, 10 (June). and Galper, H. (1973). 'State and Local Behavior and Federal Grant Policy', Brookings Paper on Economic Activity, 1. Gravelle, J. (1983). Comparative Corporate Tax Burdens in the United States and Japan and Implications for Relative Economic Growth, Congressional Research Service, 6 September, Report no. 83—177E, Library of Congress, Washington, DC.
References
399
Gregory, K., Matthews, A., Newton, A., and Nind, A. (1991). 'The Potential Impact of a $107 barrel Energy/Carbon Tax on UK Carbon Dioxide Emissions' (OECD, Workshop on Carbon Taxes). Harberger, A. C. (1962). 'The Incidence of the Corporation Tax', Journal of Political Economy, 70 (June). Hatta, T. (1992). 'The Nakasone-Takeshita Tax Reform: A Critical Evaluation', American Economic Review, Papers and Proceedings, 82/2 (May). Hayashi, H. (1990). 'Inequities of Income Taxes among Different Taxpayers', (in Japanese), Yokkaichi Daigaku Ronshu, 2/2, Feb. Heller, W. W. (1965). 'The Future of Our Fiscal System', Journal of Business, 38 (July). Henderson, J. M. (1968). 'Local Government Expenditures: A Social Welfare Analysis', Review of Economics and Statistics, 50 (May). Hicks, U. K. (1951). 'The Reform of Japanese Taxation', Public Finance, Parts I and II, 6/3,6/4. Hinrichs, H. H. (1966). A General Theory of Tax Structure Change during Economic Development, Harvard University Press, Cambridge, MA. and Bird, R. (1963), 'Government Revenue Shares in Developed and Less Developed Countries', Canadian Tax Journal, 11 (September/October). Hoeller, P., and Wallin, M. (1991). Energy Prices, Taxes and Carbon Dioxide Emissions (Paris, OECD). Hollerman, L. (1984). Tax Incentives for Saving, Investment and Innovation in Postwar Japan, testimony presented at Hearing of the Joint Economic Committee, 24 September, Congress of the USA, Washington, DC. Homma, M. (1987). Tn Search of the Optimum Tax Mix', Japan Echo, 65 (Spring). et al. (1984). 'Actual Differences of Income Tax Burden among Different Taxpayers— A Micro Approach' (in Japanese), Kikan Gendai Keizai, 59. (1986). The Japanese TaxSystem, Brookings Discussion Paper, June. Horioka, C. Y. (1985). 'A Survey of the Literature on Household Saving in Japan: Why is the Household Saving Rate so High in Japan?' mimeo, April. (1986). Why is Japan's Private Saving Rate so High? International Monetary Fund, DM series 86/44, 26 June. (1987). 'Why Save So Much?' Look Japan, 33 (May). Ihori, T. (1984). Studies on Contemporary Japanese Public Finance (in Japanese), Toyo Keizai Shinpo Co., Tokyo. Ikemoto, Y., Tajika, E., and Yui, Y. (1984). 'On the Fiscal Incentives for Investment: The Case of Postwar Japan', Developing Economics, 22 (December). Institute of Fiscal Affairs (1978). The Structure and Reform of Direct Taxation, George Allen & Unwin, London. Ishi, H. (1968). 'The Income Elasticity of Tax Yield in Japan', Hitotsubashi Journal of Economics, 9 (June). (1973). 'Cyclical Behavior of Government Receipts and Expenditures: A Case Study of Postwar Japan', Hitotsubashi Journal of Economics, 14 (June). (1976). 'The Stabilization of Tax Structure', Japanese Economic Studies, 4 (Spring). (1978). A Reexamination of Generalizations of Tax Structure Development in Light of Japan's Experience', in H. C. Rechtenward (ed.), Secular Trends of the Public Sector, Editions Cujas, Paris. (1979). 'Individual Income Tax Erosion: By Income Class in Japan', Public Finance Quarterly, 7 (June).
400
References
Ishi, H. (1980). 'Effects of Taxation on the Distribution of Income and Wealth', Hitotsubashi Journal of Economics, 21 (June). (1981). 'Inflation Adjustment for Individual Income Tax in Japan', Hitotsubashi journal of Economics, 2\ (February). (1982).'Causes and Cures of Tax Shortfalls', Economic Eye, 3 (September). (1983fl). 'An Overview of Postwar Tax Policies in Japan', Hitotsubashi Journal of Economics, 23 (February). (1983fo). 'International Tax Avoidance and Evasion in Japan', Cahiers de droitfiscal international, 68a, International Fiscal Association. (1985). 'The Budgetary Behavior of Local Governments and Intergovernmental Grant Policies: A Case Study of Japan', Government and Policy, 3. (1986a). 'Overview of Fiscal Deficits in Japan with Special Reference to the Fiscal Policy Debate', Hitotsubashi Journal of Economics, 27 (December). (1986fo). 'Structure and Utilization of Intergovernmental Fiscal Transfers in Japan' (mimeo), paper presented in Expert Group Meeting on Financing Local and Regional Development, 17-20 March, UN Centre for Regional Development, Nagoya. (1986c). 'Moving Toward Tax Reform', Sumitomo Quarterly (Spring). (1987). 'The Impact of the Shoup Mission', in H. de Kar and B. Wolfe (eds.), The Relevance of Public finance for Policy Making, Wayne State University Press, Detroit. (1988s). 'Why Tax Reform Now?' Japan Echo, 15 (Summer). (1988fo). 'Tax Reform: The Takeshita Bill—Changes in Store', Look Japan, 34 (October). (1988c). 'Corporate Tax Burden and Tax Incentives in Japan', in J. B. Shoven (ed.), Government Policy Towards Industry in the USA and Japan, Cambridge University Press, New York. (1989a). 'Individual Income Tax Erosion and Tax Equity in Japan', in A. Chiancone (ed.), Changes in Revenue Structures, Wayne State University Press, Detroit. (1989b). The Japanese Tax System, 1st edn., Oxford University Press. (1990). 'The Land Problem in Japan and the Direction of Land Tax Reform', CaM RI Review, 17 (Dec.), Capital Markets Research Institute, Tokyo. (1991). 'Land Tax Reform in Japan', Hitotsubashi Journal of Economics, 32/1 (June). (1992 o).'Japan's VAT and its Aftermath', Hitotsubashi Journal of Economics, 33 (June). (1992b). 'The Conflict between Efficiency and Equity of Tax Administration', paper presented to the Spanish Conference of International Seminar in Public Economics, 11-12 June. (1993a). 'The Value-added Tax in Japan', Japan Foundation Newsletter, 21 (September). (1993fo). 'Tax Incentives for Export Promotion in Japan, 1953-1964', in R. A. Musgrave etal. (eds.), Tax Policy and Economic Development among Pacific Asian Countries, Westview. (1995«). 'A Design of Environmental Taxes in Japan', Hitotsubashi Journal of Economics, 36 (June). (1995b). 'Rigidity and Inefficiency in Public Works Appropriations; Controversy in Reforming the Budget Process in 1994', Journal of Japanese Studies, 21 (March). (1995c). 'Fiscal Issues of Local Government in Japan', in Proceedings of International Conference on 50 Years of Public Finance in Korea: Retrospection and Perspective. (1996). 'Budget and the Budget Process in Japan', Hitotsubashi Journal of Economics, 37 (June). (1998). 'Japan', in Ken Messere (ed.), The Tax System in Industrialized Countries, Oxford University Press.
References
401
Ishi, H. (1999). 'The Role of Government in the Postwar Growth Process of Japan', Journal of Asian Economics, 10 (65-90). (2000). Making Fiscal Policy in Japan, Oxford University Press, Oxford. Ishikawa, T. (1988). 'The Saving Behavior of Japanese Households: Evaluation of the Recent Experience and Comparison with the US Experience' (mimeo). Ito, H. (1950). 'The Value-added Tax in Japan', Annals of the Hitotsubashi Academy, 1 October). (1953).'Direct Taxes in Japan and the Shoup Report', Public Finance, 8/3. Ito, T. (1992). The Japanese Economy, MIT Press, Cambridge, MA. and Krueger, A. (1992). The Political Economy of Tax Reform, University of Chicago Press, Chicago. Japan Socialist Party (1988). Ultimate Big Tax Increase (Kyukyoku no Daizozei), Bunto-cha. Jichi Sogo Centre (1983), (1988), (1992). Local Public Finance in Japan, March. Kaizuka, K. (1984). 'Recent Developments in the Japanese Tax System', in W. Block and M. Walker (eds.), Taxation, Proceedings of an International Conference, The Fraser Institute, British Columbia. (1992). 'The Shoup Tax System and the Postwar Development of the Japanese Economy', American Economic Review, Papers and Proceedings, 82/2 (May). Kaneko, H., and Gomi, Y. (1988). 'Outlook for Japan', in H. Stein (ed.), Tax Policy in the Twenty-First Century, John Wiley, New York. Kato, K. (1993). 'The Use of Market-Based Instruments in Japanese Environmental Policy', (mimeo). Kay, J. (1987). 'Tax Reform in Retrospect: The Role of Inquiries', in Hans M. van de Kar and B. L. Wolfe (eds.), The Relevance of Public Finance for Policy-making, Wayne State University Press, Detroit. Kay, J. A., and King, M. A. (1986). The British Tax System (4th edn.), Oxford University Press. Kilpatrick, R. W. (1965). 'The Short-run Forward Shifting of the Corporation Income Tax', Yale Economic Essays, 5 (Fall). Kimura, M. (1952). 'Conditions for Direct Taxation', Annals of the Hitotsubashi Academy, 2 (April). King, M. (1986). 'The Japanese Tax Reform Proposals' (mimeo). Komiya, R. (1966s). 'The Supply of Personal Savings', in R. Komiya (ed.), Postwar Economic Growth in Japan (trans. R. S. Ozaki), University of California Press, Berkeley. (1966fc). 'The Levels of Capital Formation and Public Finance in Postwar Japan', in Foreign Tax Policies and Economic Growth, A Conference Report of the NBER and the Brookings Institution, NBER, New York. Kosai, Y., and Ogino, Y. (1984). The Contemporary Japanese Economy, Macmillan, London. Kotlikoff, L. (1984). 'Taxation and Savings: A Neoclassical Perspective", Journal of Economic Literature,22 (December). Krzyzaniak, M., and Musgrave, R. A. (1963). The Shifting of the Corporation Income Tax, Johns Hopkins University Press, Baltimore. Kubouchi, Y. (1984). 'Tax Burden on Corporate Income: An International Comparison', Keidanren Review, 87 (June). Kumon, S. (1984). 'Japan Faces its Future: The Political Economics of Administrative Reform', Journal of Japanese Studies, 10 (Winter). Kuroda, M. (1993). 'Reducing CO2 Emissions and Long-run Growth of the Japanese Economy," paper submitted to Energy Technology System Analysis Program Conference, Oxford, 2 June 1993.
402
References
Land Tax Review Group (1990). Report of the Land Review Group, Government of South Australia (May). Lewis, S. R. Jun. (1963). 'Government Revenue from Foreign Trade: An International Comparison", Manchester School of Economic and Social Sciences, 31 (January). Lewis, W. L. (1962). Federal Fiscal Policy in the Postwar Recession, Brookings Institution, Washington, DC. Lincoln, E. (1988). Japan, Facing Economic Maturity, Brookings Institution, Washington, DC. Martin, A., and Lewis, W. A. (1956). 'Patterns of Public Revenue and Expenditure', Manchester School of Economic and Social Sciences, 24 (September). Mathews, R. (1983). 'The Case for Indirect Taxation' (mimeo), First Annual Endowed Lecture in Taxation Law and Policy, Taxation Institute of Australia, 5 May. Maxwell, A. (1955). Fiscal Policy, H. Holt, New York. Messere, K. C. (ed.) (1998). The Tax System in Industrialized Countries, Oxford University Press, Oxford. and Owens, J. P. (1986). 'Long-term Revenue Trends and Current Tax Reform Issues in OECD Countries', paper presented to the 42nd Congress of International Institute of Public Finance (IIPF), Oxford University Press, Oxford. Minami, R. (1986). The Economic Development of Japan, Macmillan Press, London. Ministry of Finance, Budget Bureau (annual). National Budget (Kuni no Yosan), Tokyo. (annual). Fiscal Statistics (Zaisei Tokei), Tokyo. Financial History Section (1977&, 1977c). Financial History of Japan—The Allied Occupation Period,Vols. 1 and 8, Toyo Keizai Shinposha, Tokyo. Securities Bureau (1975). Material Submitted to Special Committee of Securities Exchange Council (in Japanese), April. (annual) Tax Bureau. Primary Statistics of Taxation (Zaisei Shuyo Sanko Shiryoshu), Tokyo. (1972). Summary of Lectures of Dr Shoup and Professor Surrey (in Japanese), November. (1977a, 1986, 1987, 1991, 1992, 1996, 1998) Tax Bureau. An Outline of Japanese Taxes, Tokyo. (1988). Outline of Tax Reform (Zaisei Kaikaku Taiko), 28 June. Ministry of Health and Welfare (1972, 1984). Studies on Income Redistribution (Shotoku Saibunpai Chosa), Tokyo. Ministry of Home Affairs (MOHA) (2000a). Material of Local Taxes (Chihozei ni Kansuru Sanko Shiryo), February, Tokyo. (2000fo). Performance of Local Public Finance (Chihozaisei no Jokyo), March, Tokyo, Ministry of International Trade and Industry (MITI) (1989). 'On the Shifting of the Consumption Tax' (in Japanese), 28 July. (1990). 'Actual Performance of Implementing the Consumption Tax' (in Japanese), 7 August. Miyajima, H. (1987). 'Tax Reform Gone Away', Japan Echo, 65 (Summer). Mizoguchi, T. (1970). Personal Savings and Consumption in Postwar Japan, Kinokuniya, Tokyo. and Takayama, T. (1984). Equity and Poverty under Rapid Economic Growth: The Japanese Experience, Kinokuniya, Tokyo. Moss, H. (1948). Memorandum to General Market: Subject, Proposed Radiogram Covering Tax Program, 13 December, MOP Financial History Document, no. z702-6.
References
403
(1949). Letters to J. M. Dodge, 23 September, MOF Financial History Document, no. Z711-6. Musgrave, R. A. (1959a). 'How Progressive Is the Income Tax?' Ways and Means Commission, HR, Tax Revis. Comp. (1959b). The Theory of Public Finance, McGraw-Hill, New York. (1969). Fiscal System, Yale University Press, New Haven, Conn. Nagano, A. (1988). 'Japan', in J. A. Pechman (ed.), World Tax Reform, Brookings Institution, Washington, DC. Nakagiri, H. (1978). 'On Tax Equity' (in Japanese), in Reexamination on the Contemporary Public Finance (Gendai Zaiseigaku no Saikento) in Honor of Professor Y. Hayashi, Yuhikaku, Tokyo. National Tax Administration (annual). Annual Report of National Tax Administration (in Japanese), Tokyo. (annual). Statistics on Private Wages and Salaries (in Japanese), Tokyo. (annual). Statistics on the Self-Assessed Income Tax (in Japanese), Tokyo. (1987). Comparison of Tax Administration in Major Countries (in Japanese), July, Tokyo. (1990a). 'A Review on the Filing and Registration of VAT: One Year after the Consumption Tax has Started' (in Japanese), May. (1990fo). 'On Data Collected to Investigate Actual Performance of the Consumption Tax', 7 August. New South Wales Tax Task Force (1988). Review of the State Tax System, NSW Treasury (August). Nikkeiren (1982). Report of Committee on Labor Problems (in Japanese), January, Tokyo. Nishibe, S. (1987).'The Mob Approach to Tax Reform', JapanEcho, 14/2 (Summer). Noguchi, Y. (1983). 'Problems of Public Pensions in Japan', Hitotsuhashi Journal of Economics, 24 (June). (1986). 'The Public Sector in the National Economy', in T. Shibata (ed.), Public Finance in Japan, University of Tokyo Press. (1987o). 'Public Finance', in K. Yamamura and Y. Yasuba (eds.), The Political Economy of Japan, Vol. 1, Stanford University Press. (1987k). 'Tax Reform—The Missing Rationale', Japan Echo, 65. (1990). 'Land Problem in Japan', Hitotsubashi Journal of Economics, 31/2 (December). (1992). 'The Changing Japanese Economy and the Need for a Fundamental Shift in the Tax System', American Economic Review, Papers and Proceedings 82/2 (May). OECD (1972). Guiding Principles Concerning International Economic Aspects of Environmental Policies (Recommendation of the Secretary-General, 26 May). (1976). Income Distribution in OECD Countries, Paris. (1986). Urban Policies in Japan, Paris. (1987). Taxation in Developed Countries, Paris. (annual). Revenue Statistics of OECD Member Countries, Paris. (annual). National Accounts, Paris. (1988). Taxing Consumption, Paris. (1989). Economic Instruments for Environmental Protection, Paris. (1990). Taxpayers' Rights and Obligations, Paris. (1991 a). Environmental Policy: How to Apply Economic Instruments. (1991 b). An Environmental Strategy in the 1990s (Communique of the Environmental Committee at the Ministerial Level, Jan. 1991).
404
References
OECD (1993). Taxation and Environment, Paris. (1994). Environmental Performance Reviews: Japan, Paris. (] 998fl). Revenue Statistics 1965-1998. (1998&). National Accounts. (1999). OECD Economic Surveys: Japan. Office of Management and Budget (1986). Budget of the US Government, US Government Printing Office, Washington, DC. Office of the Prime Minister (1993). Annual Report of Family Income and Expenditure. Ohkawa, K. et al. (1965-88). Long-term Economic Statistics of Japan, Vols. 1-14, Toyokeizai Shinposha, Tokyo. Okuno, M. et al. (1991). 'Structure of Income Tax Burden—Tax Gap by Taxpayers and Income Classes' (in Japanese), Discussion Paper, University of Tokyo (August). Oshima, H. (1957). 'Share of Government in Gross National Product for Various Countries', American Economic Review, 47 (June). (1965). 'Meiji Fiscal Policy and Economic Progress', in W. W. Lockwood (ed.), The State and Economic Enterprise in Japan, Princeton University Press. Ozaki, M. (1991). 'The Role and Significance of the Withholding Tax System', paper presented to the OECD Conference of Economics in Transition, January. Patrick, H. (1970). 'The Phoenix Risen from the Ashes: Postwar Japan', in J. B. Crowley (ed.), Modern East Asia: Essays in Interpretation, Harcourt Brace and World, New York. and Rosovsky, H. (1976). Asia's New Giant: How the Japanese Economy Works, Brookings Institution, Washington, DC. Peacock, A. T., and Wiseman, J. (1961). The Growth of Public Expenditure in the United Kingdom, Princeton University Press. Pechman, J. A. (1956). 'Yield of the Individual Income Tax during a Recession', in Policies to Combat Depression, Conference of the Universities-NBER, Princeton University Press. (1957).'Erosion of the Individual Income Tax', National Tax Journal, 10 (March). (1959). 'What Would a Comprehensive Individual Tax Yield?' US Congress House, Committee on Ways and Means, Tax Revision Compendium, US Government Printing Office, Washington, DC. (ed.) (1980). What Should be Taxed: Income or Expenditure? Brookings Institution, Washington, DC. (1984). 'Comprehensive Income Tax Reform', in J. A. Pechman et al. (eds.), Tax Policy: New Directions and Possibilities, Center for National Policy, Washington, DC. (1986). The Rich, The Poor and the Taxes They Pay, Wheatsheaf Books, Brighton, Sussex. (1987). Federal Tax Policy (5th edn.), Brookings Institution, Washington, DC. (ed.) (1988). World TaxReform, Brookings Institution, Washington, DC. and Kaizuka, K. (1976). 'Taxation', in H. T. Patrick and H. Rosovsky, Asia's New Giant, Brookings Institution, Washington, DC. and Okner, B. A. (1972). Individual Income Tax Erosion by Income Classes, Brookings Print no. 230, Brookings Institution, Washington, DC. Reed, S. R. (1986). Japanese Prefectures and Policymaking, University of Pittsburgh Press. Roth, J. A., Scholz, J. T., and Witte, A. D. (1989). Taxpayer Compliance, vols. 1 and 2, University of Pennsylvania Press, Philadelphia. Royal Commission on Taxation in Canada (1966). Report of the Royal Commission on Taxation, Vol. 3, Ottawa.
References
405
Sandford, C, Godwin, M., and Hardwich, P. (1989). Administrative and Compliance Cost of Taxation, Bath, Fiscal Publications. Securities Transaction Council (1960). On the Stimulus to Increase Equity Capital (in Japanese), July, Tokyo. Shavell, H. (1948a).'Postwar Taxation in Japan', journal of Political Economy, 56 (April). (1948&). 'Taxation Reform in Occupied Japan', National Tax journal, 1 (June). Shinohara, M. (1982). Industrial Growth, Trade and Dynamic Patterns in the Japanese Economy, University of Tokyo Press. Shoup, C. S. (1983). 'Current Trends in Excise Taxation', in S. Cnossen (ed.), Comparative Tax Studies, North-Holland, Amsterdam. (1986). 'Criteria for Choice among Types of Value-added Tax', paper presented at the Conference on Value-Added Taxation in Developing Countries, World Bank, Washington, DC. (1989). 'The Tax Mission to Japan, 1949-1950', in M. Gillis (ed.), Tax Reform in Developing Countries, Duke University Press, Durham. (1990a). 'Choosing among Types of Value-added Tax', in M. Gillis, C. S. Shoup, and G. P. Sicat (eds.), Value-Added Taxation in Developing Countries, World Bank, Washington, DC. (1990fo). 'Tax Reform in Japan', Australian Tax Forum, 7. Shoup Mission (1949). The Report on Japanese Taxation by the Shoup Mission, Tokyo. Shoven, J. B., and Tachibanaki, T. (1988). 'The Taxation of Capital Income in Japan', in J. B. Shoven (ed.), Government Policy Towards Industry in the USA and Japan, Cambridge University Press. Simons, H. (1938). Personal Income Taxation, Chicago University Press. Slitor, R. E. (1948). 'The Measurement of Progressivity and Built-in Flexibility', Quarterly Journal of Economics, 62 (February). Snowborger, M., and Kirk, J. (1973). 'A Cross-sectional Model of Built-in Flexibility, 1954-1969', National Tax Journal, 26 (June). Subcommittee on Land Taxation (1990o). 'Main Issues in the Review of Land Taxation' (mimeo), 29 May. (1990b). 'For the Review of Land Taxation' (mimeo), 22 June. Sumitomo Bank (1988). 'Business Expansion Induced by Sustained Domestic Demand,' (in Japanese), Economic Monthly Report, July/August, 1988. Summers, L. H. (1981). 'Capital Taxation and Capital Accumulation in a Life Cycle Growth Model', American Economic Review, 71 (September). Sundelson, J. W. (1950). 'Report on Japanese Taxation by the Shoup Mission', National Tax Journal, 3 (June). Sunley, E. M. Jun., and Pechman, J. A. (1976). 'Inflation Adjustment for the Individual Income Tax', in H. J. Aaron (ed.), Inflation and the Income Tax, Brookings Institution, Washington, DC. Supreme Commander Allied Power (SCAP)(1949). Teleconference between Tokyo and Washington: Subject, Taxation in Japan, Financial History Project, z703-6. Surrey, S. S. (1973«). Pathways to Tax Reform: Its Concept of Tax Expenditures, Harvard University Press, Cambridge, MA. (197 3 b). A Proposal for Tax Reform in Fiscal 1973 (in Japanese), December. (1981). 'The Concept of Tax Reliefs—Its Relation to Tax Policy and Budget Policy', in K. Hauser (ed.), Subsidies, Tax Reliefs and Prices, Editions Cujas, Paris. and McDaniel, P. R. (1985). Tax Expenditure, Harvard University Press, Cambridge, MA.
406
References
Surrey, S. S., and Sunley, E. M., Jun. (1976). General Report of the 32nd Congress of International Fiscal Association, Kluwer, Amsterdam. Suzuki, Y. (ed.) (1987). The Japanese Financial System, Clarendon Press, Oxford. Tail, A. T. (1988). Value Added Tax, International Monetary Fund, Washington, DC. Tajika, E., and Yui, Y. (1988). 'Cost of Capital and Effective Tax Rate: A Comparison of US and Japanese Manufacturing Industries', Hitotsubashi Journal of Economics, 29 (December). (1990). 'Taxation and Investment: A Comparative Study of Japan and the US' (in Japanese), Financial Review, 18 (December). (1999). Corporate Taxes in Japan (in Japanese), Toyokeizai Nhinpo, Tokyo. Takenaka, H. (1986). 'Economic Incentive for Industrial Investment: Japanese Experience' (mimeo). (1992). The Japanese Economy, University of Michigan Press, Ann Arbor. Tanzi, V. (1976).'Adjustment of Personal Income Taxes for Inflation: The Foreign Experience', in H. J. Aaron (ed.), Inflation and the Income Tax, Brookings Institution, Washington, DC. Tax Advisory Commission (1960). A Proposal for the Current Tax Reform (in Japanese), December. (1961). A Proposal of the Tax Advisory Commission and Its Interpretation of Process for Constructing the Proposal (in Japanese), December. (1962). Materials Relevant to a Proposal of the Tax Advisory Commission (in Japanese), March. (1964o). Materials Relevant to a Proposal of the Tax Advisory Commission (in Japanese), March. (1964i>). A Proposal on the Basic Tax System Suitable for the Socio-Economic Progress of the Future (in Japanese), December. (1968). A Proposal for the Significance of Long-term Taxation and Its Interpretation of the Process for Constructing the Proposal (in Japanese), July. (1970). A Proposal for Tax Reform in Fiscal 1970 (in Japanese), January. (1971). A Proposal for Long-term Taxation (in Japanese), August. (1972). A Proposal for Tax Reform in Fiscal 1973 (in Japanese), December. (1976). Tax Report (in Japanese). (1978). Tax Report (in Japanese). (1986). Tax Report on a Sweeping Tax Reform (in Japanese), October. (1988). Report on Tax Reform (in Japanese). Taylor, L. D. (1971). 'Savings out of Different Types of Income', Brookings Papers on Economic Activity, 2. Thorn, R. (1967). "The Evolution of Public Finances during Economic Development', Manchester School of Economic and Social Sciences, 35 (January). US Treasury (1977). Blueprints for Basic Tax Reform, US Government Printing Office, Washington, DC. (1984). Tax Reform for Fairness, Simplicity and Economic Growth, Vols. 1-3, November. Usui, N. (1987). Actual Performance of Indirect Taxes (in Japanese), Ohkura Zaimu Kyokai, Tokyo. Wada.Y. (1974). Are Tax Burdens Equitable? (in Japanese), Nihon Keizai Shinbunsha, Tokyo. Walsh, C. (ed.) (1990). Issues in the State Taxation, Centre for Research on Federal Financial Relations, ANU, Canberra. Webber, C., and Wibdavsky, A. (1986). A History of Taxation and Expenditure in the Western World, Simon & Schuster, New York.
References
407
Webley, P., Robben, H., Elffers, H., and Messing, D. (1991). Tax Evasion, Cambridge University Press, Cambridge. Williamson, J. G. (1961). 'Public Expenditure and Revenue: An International Comparison', Manchester School of Economic and Social Sciences, 29 (January). Yamamura, K. (1967). Economic Policy in Postwar Japan, University of California Press, Berkeley and Los Angeles. Yamazawa, I. (1975). 'Industrial Growth of Trade Policy in Prewar Japan', The Developing Economics, 13 (March), Institute of Developing Economies, Tokyo. Yashiro, N. (1987). 'Japan's Fiscal Policy—An International Comparison', Japanese Economic Studies, 61 (Fall). Yonehara, J. (1981). Local Public Finance in Japan, Australian National University Press, Canberra. Yoshida, K. (1987). 'Who'll Pay?' Look Japan, May. Yoshikawa, J., and Ohtake, F. (1988). 'An Analysis of Female Labor Supply, Housing Demand and the Saving Rate in Japan' (mimeo).
This page intentionally left blank
Name Index
Aaron, H.J. 91 n, HOn Ackley, G. 125 Amano, A. 314 Andrews.W. 91 n Aoki.T. 4n, 328 Atkinson, A. B. 149 n Ballentine, J. G. 326 n Bayoumi, T. 242 Bird,R.M. 18,24 Bittker, B. I. 97 n Blinder, A. S. 125 Boskin.M.J. 139 Bowen, H.R. 25 n Bronfenbrenner, M. 30,30 n, 31 n, 35 n Chelliah, R. J. 17 n Cnossen, S. 253 Cohen, J. B. 25 n, 26, 26 n Collyns, G. 242-3 Deutsch, K.W. 20 n Dodge, J. 26 n, 32 n Doko, Toshio 321 Feldstein, M. 31 Furuta, S. 185 Galper, H. 91 n, 381 n Gomi,Y. 13 I n , 206 Goode, R. 65,184 Gordon, R.J. 185 Gramlich, E. M. 381 n Gravelle, J. 197, 199 Gregory, K. 314 n Harberger, A. G. 184 Hatfield, R. R. 25 n Hatta.T. 393 n Hayashi, H. 69 n Heller, W.W. 125 Henderson, J. M. 381 n Hicks,U.K. 27, 28n Hinrichs, H.H. 16, 17n, 18,24 Hollerman,!,. 187,198n
Homma, M. 69 n, 325 n Horioka.C.Y. 138n, 139, 142n
Ihori, T. 116n,320n, 322 Ikemoto.Y. 189 Ishi.H. 16n, 17n, 18n,21 n, 25, 37, 46n, 49, 50, 50 n, 53 n, 56, 69 n, 86 n, 88 n, 94 n, 99 n, 103, 106, 1 l O n , 130n, 142, 157n, 187n, 219n, 242, 268 n, 274, 277, 300 n, 319, 320, 320 n, 322, 323, 323 n, 328, 329, 339, 341, 380 Ishikawa.T. 138 n, 142 n Ito, H. 30 Ito, T. 319,327 Kaizuka, K. 4 n, 36 n, 37, 46 n, 138, 142, 393 n Kato, K. 304,309 Kay.J.A. 8n, 12, 78,81, 84,91 n, 92 n, 110 n, 127, 335 n, 358 n Kilpatrick,R.W. 185 Kimura, M. 28 n King.M.A. 8n, 84, 91 n, 92n, HOn, 127,197n, 335 n, 358 n Kirk, J. 86 n Kogiku, K. 30 n, 31 n, 35 n Komiya, R. 37,138 Kosai.Y. 138n Kotlikoff, L. 139 Krueger.A. 327 Krzyzaniak, M. 184,185 Kubouchi.Y. 198 Kumon, S. 321 Kuroda, M. 314 Lewis, S. R. Jun. 21 n Lewis, W. A. 17 n Lewis, W. L. 45,46n Lincoln, E. 50,320,321 MacArthur, General 26 n, 31, 31 n McDaniel, P. R. 95 n Martin, A. W. A. 17 n Mathews, R. 384 n Maxwell, A. 45,110 n Messere, K.C. 271,327,333
410
Name Index
Minami, R. 17n Miyajima, H. 326 n Mizoguchi, T. 138n, 148n Moss, H. 25 n, 32 n Musgrave, R.A. 17n, 19,45,91 n, 184, 185 Nagano, A. 328 Nakagiri, H. 116n Nishibe, S. 326 n Noguchi,Y. 50, 136 n, 221, 320, 393 n Ogino.Y. 138 n Ohkawa, K. 17 n Ohtake, F. 142n Okner, B.A. 103,107, 107 n Okuno, M. 69 n Oshima,H. 17n, 19 n Owens, J. P. 271,327,333 Ozaki.M. 63
Simons, H. 97 Slitor, R.E. 88 Snowborger, M. 86 n Solow, R.M. 125 Summers, L.H. 139 Sundelson, J. W. 32 n Sunley, E.M. Jun. 95n, 112n, 116 Surrey, S. S. 25 n, 27, 95 n, 97 n Suzuki, Y. 131 n Tachibanaki, T. 197n Tail, A. T. 269, 274 n Tajika, E. 187 n, 197 n, 200 Takayama, T. 148 n Takenaka.H. 203,319 Tanzi.V. 111,111-12)1,116n Taylor, L.D. 142,146 Thorn, R. 16 n Vickrey.J.N. 25n,27
Patrick, H. 36, 125 Peacock, A. T. 17 n Pechman, J. A. 4 n, 36 n, 37,46 n, 91, 91 n, 98, 98 n, 103, 107, 107 n, 112n, 116, 138, 142,175, 184, 200, 205, 327, 331, 333, 335 n, 336 n, 357, 390, 394 Reed, S.R. 351, 366 n Rosovsky, H. 36,125 Roth, J. A. 65
Wada.Y. 116n Walsh, C. 234 n Warren, W.C. 25 n Webber, C. 268 Webley, P. 65 Widavsky.A. 268 Williamson, J. G. 17n,22n Wiseman,]. 17 n
Sandford, C. 54 n, 56 n Shavell, H. 26 Shinohara, M. 138 n Shiokawa, M. 240 Shoup, C. S. xiii, 4, 25, 27, 27 n, 28, 31, 253, 269, 273,283, 391 n, 393 n Shoven,J.B. 197n
Yamamura, K. 26 n Yamazawa, I. 18 Yashiro.N. 323 n Yonehara, J. 374 n Yoshida, K. 326 n Yoshikawa, J. 142 n Yui,Y. 187n, 197n, 200
Subject Index
abuses see evasion accelerated cost recovery system (ACRS) 199, 202 accelerated depreciation 190, 202 accession tax 4, 29, 30, 204-6 account method, for consumption tax 281, 387, 391-2 accounting records, small businesses 28, 69 accretion 200 ACRS (accelerated cost recovery system) 199, 202 actual cost method, inventory valuation 168 actually-paid tax rates, corporate taxes 192-6 Ad Hoc Commission for Administrative Reform (Rincho) 321,323 ad hoc relief, capital gains 142 ad hoc remedies, inflation adjustments 111, 127 ad hoc Special Committee of the Lower House 328 n ad valorem taxes: alcoholic beverages 255, 257 tobacco 258 administration 53-71 before Shoup Mission 26 costs 11-12,54-6,386-7 reform 321-2 Shoup recommendations 28 value added tax 291-3 Administrative and Management Centre 321 admission tax 253, 254 age ratio of population 142 n ageing society, effect on savings 146 aggregate savings function 142, 143, 146 aggregation method 79, 84,389 agricultural incomes: redistributive effects 155-7 reporting of taxable income 52, 54,64-5, 70 self-assessment 61 tax gap 68 agricultural land: inheritance tax 212 n, 236, 239-40 land tax 230-2 Urbanization Promotion Areas 230, 236, 239-40 air pollution charge 309 airport construction 306,307 alcoholic beverages, categories 255 n, 256
alcoholic beverages tax 255-8 as revenue source 19, 254 coexistence with consumption tax 284 historical background 3,4 negative externalities 252-3 Shoup recommendations 34 allowances: corporate taxes 190,191, 192,347 see also deductions Annual Household Survey 148 annual tax reductions see reductions Ashrey Committee 12 n assets: business: property tax 362 roll-over scheme 229-30,235-6,239 employees' 132 land value 225-6,233,234,237-8 proportion allotted to financial and real investments 141 revaluation, Shoup recommendations 29, 33 sale see capital gains tax see also depreciation Australia: Ashrey Committee 12 n land-holding tax 234 n sales tax 252 tax equity 384 n tax unit 78 value added tax 270 aviation fuel tax 253, 254 aviation fuel transfer tax 364, 465 avoidance 54, 87 n, 124, 239 backward-shifting, corporate taxes 185 bad debts, tax-free reserves 190, 347 balanced budget policy 50, 320 Bank of Japan 18 bases see tax bases Basic Environment Law 302, 303 basic financial needs, local governments 372-3 basic financial revenues, local governments 373-4 Basic Land Law 222, 235 Basic Law for Environmental Pollution Control 300
412
Subject Index
Belgium, tax burden 92, 93 bequests see gift taxes; inheritance taxes blue returns 69-70 corporations 168,180,181,182 deduction of cost of preparation 77 Shoup recommendations 28 bond dependency ratio 322-3 bonds: discount 134 national 50 proportion of portfolio investments 140,141 bonus effect on savings 139 bonus payments, tax-free reserves 190,196, 347 bonuses, withholding tax table 62 border-tax adjustments, value added tax 270 borrowings, local 365-7 bourse tax 253,254 bracket creep 52,128, 334 bubble boom (bubble economy), effects 215 on fiscal deficits 51, 320 n on income distribution 150—1 on inheritance tax 210 on land prices 220, 221, 243, 244, 245, 328 on relative share of withheld tax 62 on stamp and security transaction taxes 254 on tax reduction policy 42 on tax system elasticity 46 on wealth distribution 328 building see construction built-in flexibility of taxation 124 burdens: alcoholic beverages tax 255-6 business organizations 181-2 capital gains tax 235 capital owners 184 car-related taxes 261-3, 263 carbon dioxide emissions 306-8, 314-15 commodity tax 265, 266 comparison with other OECD countries xxv, 6-8 corporate taxes 197-203,336-8 enterprise tax 273 factor in high savings rate 142 gift tax 213 income tax 92-3, 344 inheritance tax 205, 206, 212-14, 338-9 land taxes 221-5,232-4,235,241-7 property tax 362 reduction 330 tobacco tax 258-9 see also equity; revenues
bureaux 14 business assets: property tax 362 roll-over scheme 229-30, 235-6, 239 business expenses: deduction 52 exclusion from net income 200 exemption from commodity tax 264 interest payments on land purchase loans 338 salaries of family employees 180,181 see also income-splitting self-employed 77 business financing 175,182-4 business incomes: determination 77 income-splitting 52, 80 redistributional effects 155-7 tax collection 60, 61 see also self-employment incomes; small businesses business saving and investment 40,41,188 business tax (enterprise tax) 167, 273, 333 n, 355, 359-60 Canada: Carter Commission and Report 12 n, 27 cost-revenue ratio 48, 55, 56, 58 Royal Commission on Taxation 97 n value added tax 270 capital accumulation, promotion 130-1, 176 capital gains tax: collection 60, 61 corporations 168 neutrality 386 on land 227-30 reforms 238-9,243-5 special tax measures 104-6, 157 unrealized 233 on real estate 60, 77, 84, 142,238 on stocks and securities 30,60, 77,134-5, 173-4,386,388-9 on valuable assets, paintings, jewels 61 redistributive effects 155,157 Shoup recommendations 28, 30, 33 capital owners, tax burden 184 car purchase tax 261 n car-related taxes 261-3,263,309 carbon dioxide emissions 301-2, 306-8, 310-15 carbon tax, carbon/energy tax 310-15 carry-back/carry-forward, operating losses 168 Carter Commission and Report 12 n, 27
Subject Index casual (occasional) incomes 61, 84 casualty insurance premiums, deductions 38, 63 n,135 casualty losses, deductions 77 n, 361 n categorical grants-in-aid 376-9, 380 centralization 10-12, 350 chiho-kofuzei 352 children: incomes 79 see also heirs cigarette taxes 259, 260 see also tobacco taxes cities, designated 352, 371 citizenship 29 City Planning Law 230 n, 362 city planning tax 224, 362 Civil Code, taxable inheritance percentages 207 coal subsidies 306,307 collection: before Shoup Mission 26 business incomes 60, 61 capital gains tax 60,61 centralized xxv, 10-12 costs 11-12,54-6 dividends 60-1 employment incomes 57, 59, 62-4 international comparison 53-7 local governments 357, 359 reform 388 value added tax 291-3 'command-and-control' environmental regulation 304 Commercial Law 170 n commodity taxes 4, 254, 264-7 Shoup recommendations 34 compensation packages, fringe benefits 80 compliance: costs 54,386 value added tax 291-3 'Comprehensive Land Policy Plan' 236 comprehensive tax 4, 28, 84,91, 95,96-103 base 91-2,108-9 computation: corporate taxes 168-9 individual taxable income 75-6 inheritance tax 206-8,217-18 concessions: environmental policy 304 interest income 133 conditional fiscal transfers 352—3 Conference of Parties, third (COP3) 301
413
construction: airports and roads, earmarked taxes 306, 307 houses, tax credits 135 Ministry of Construction, land policy 222 'construction bonds' 320 n consumer price index (CPI) 117n, 286 consumers, corporate tax-shifting 184-5 consumption tax: as value added tax 268, 279-85, 341, 391 in tax mix, 396 indirect taxes a form of 251-2 local 297-9 share of tax revenues 6, 8, 9 see also commodity taxes; value added tax consumption transfer tax 284, 364 co-operatives: corporate tax rates 172 enterprise tax 360 n taxable corporations 169 see also corporations COP3 (third Conference of Parties) 301 corporate net income, calculation 200 corporate rate of return on capital approach 185 Corporate Tax Law 169 corporate taxes 3-4,167-86 before Shoup Mission 26 burdens 197-203,336-8 capital gains tax 224, 228-30,238-9,243-5 enterprise tax 359-60 income tax: basic rate changes 323-4, 336-8 effect of inflation 110 n elasticity 48 share of tax revenues 5,6 Shoup recommendations 4, 33 taxable income 168-9 inhabitants'tax 360-1 profit tax 26,176 reform 345-8,388-9 relative share 167-8,343-4 tax-shifting 184-6, 185 corporations: deemed and quasi-corporations 180,181, 182 taxable 169-71 see also corporate taxes cost-of-living index 117n cost-revenue ratio 48, 55-6, 57, 58-9 costs of collection 11-12, 54-6 Council of Economic Advisers 125,126 n
414
Subject Index
couples: exemptions for spouses 76, 79, 330-1, 333, 344, 385 income-splitting 78,79,331 tax liabilities 78-9 CPI (consumer price index) 117 n, 286 credits: corporate taxes 189,190,191,192 dividends 175-6 effects on equity 94-5,103 erosion of tax base 96,388 foreign 77, 338 individual income tax 77 dividends 134 loans for house purchase 135 inheritance tax 207-9 need for reform 388 Shoup recommendations 33 crowding-out, capital markets 320 n cultural-political factors 24-5 customs duties 252 cuts see reductions data sources: income tax 107-8,116n, 147-9 savings 143 death taxes see accession tax; gift taxes; inheritance taxes debt burden 320n debt charge ratio 366 debt financing: business enterprises 182—4 local government 365-7 debt permit system 366-7, 380 debt service costs 320 n debts: bad, tax-free reserves 190,347 local governments 354,365—7,380 decentralization 350 declining balance method, depreciation 168 deductions: blue returns, cost of preparation 77 business incomes 77,182 capital gains 77 casualty losses 77 n, 361 n dividend-paid 175 donations 77 n effect on income distribution 153-4 employment incomes 76, 77 erosion of tax base 96, 97-8, 102, 388 individual income tax 76-7, 344, 345
at year-end adjustment 63 effects on equity 94-5 insurance premiums 38, 63 n, 135,141 medical expenses 42, 63 n, 77 n, 361 n municipal inhabitants'tax 361 need for reform 388 physically handicapped persons 77 n retirement incomes 77, 335-6 Shoup recommendations 33 timber income 77 'deemed corporate income' 182 'deemed corporations' 77, 333 deferred taxes: agricultural land 230-1,236, 239-40 corporate taxation 190 'deficit-covering bonds' 320 n deficit-operating corporations 170-1, 229, 235, 238-9 deflated taxes 116-24 delinquency 63-4 democratic citizenship 29 Denmark: tax burden 92, 93 tobacco tax 259 density modification coefficients 373 Department of Labour 66 dependants, exemptions for 76, 79, 330-1, 333, 336, 344, 385 deposit-refund system 309 depreciation, corporate taxation 26,168,190,191, 202, 336 n, 347 designated cities 352,371 development: effect on tax structure 16-24 see also economic growth direct taxes: definition 8n reliance on xxv share of tax structure 8-10,18-19, 25 Shoup recommendations 4,27 switch to indirect taxes 271,343 'direct taxpayers' (withholding agents) 62, 63 disaster-related works, obligatory shares 375 disbursements, national government 374-5 discount bonds 134 'disguised'land 231 disguised subsidies, environmental policy 304 distortions: effect of inflation 110-15 tax-induced 385-6, 390 distributed profits, tax rate 174,175
Subject Index distribution of income see redistributive effects 'District for Identifying and Promoting the Utilization of Idle Land' 238 dividends: business costs 183 corporate taxes 168, 174,175-6,182-4, 337 distribution 184 double taxation 172,174-5,177-80 special tax measures 134 Tax Advisory Commission proposals 175 tax collection 60-1 tax credits 77 tax rates 84, 175 withheld taxes 63 doctors, tax benefits 332-3 'Dodge Line' 26-7 donations, deductions for 77 n double taxation: dividends 172,174-5,177-80 savings 92,395 earmarked taxes 253, 254,261, 306, 307, 313, 364, 365 earned incomes see employment incomes; self-employment incomes easy money policy 221 EC (European Community): carbon/energy tax 310 dispute on alcohol tax 256-7 value added tax 268,270 economic development, effect on tax structure 16-24 economic growth: after 1973 oil crisis 319-20 effect of corporate tax burdens 197 promotion through tax system 36 n, 37 relation to environmental policy 301, 313-14 relation to natural increase in tax revenues 48-9 stimulation by tax cuts 339-40 economic income 97, 200-2 economic instruments, environmental policy 303-5 Economic Planning Agency (EPA) 44, 66, 285, 320 Annual Report on National Accounts 143 economic power 200 Economic Stabilization Board 26 n effective tax rates: alcoholic beverages tax 255-6, 257 corporate taxes 192-6, 200, 203, 346, 348
415
effects of inflation 115,122-4 individual income tax 99-103 land tax 223,225 property tax 226 n, 230 n, 362 efficiency of administration 11-12, 55-6, 71 elasticity of tax revenues 44-8 electricity tax 363 n employee compensation, fringe benefits 80 employees'assets 132 employment incomes: special deductions 76, 77 tax burden 328 tax collection 57, 59, 62-4 tax gap 68 Employment Status Survey 66 energy taxes 305-8,310-12 energy-saving equipment 189, 304 n Engels's law 16 enterprise tax 167, 273, 333 n, 355, 359-60 entertainment expenses, fringe benefits 80 entrepreneurs, registration as corporations 170 n Environmental Agency 300, 303, 307 n, 314 environmental quality, promotion by tax incentives 40,41 environmental taxes 300-15 EPA see Economic Planning Agency equalization, fiscal 367-74 equalization coefficient: income taxes 149,151,152-3,159,161, 162 inheritance tax 214-16 equalization grant, Shoup recommendations 35 Equalization Grant Scheme 29 equity: income tax 103-7 erosion 94-6 v. expenditure tax 394-5 policy 49-52 relation to tax administration 53-71 Shoup recommendations 28, 29-30 tax reform 327, 328, 329, 332, 383, 384-5 see also burdens equity financing 182-4 equity securities, proportion of portfolio investments 140,141 erosion: by tax incentives on savings 146 individual income tax 94-109 special tax measures 154n estate tax 204,205,206 estate transfers see inheritance taxes EU (European Union) 257-8
416
Subject Index
European Community see EC evasion 54, 87 n, 124,137-8 excess profits tax 19,26,33 excise taxes 252-67 need for reform 271 repeal 283-4 share of tax revenues 6 Shoup recommendations 34 exclusions: effects on equity 94-5 erosion of tax base 97-8,102 exemptions: corporate taxes 189 effect on income distribution 153-4 erosion of tax base 96, 97-8, 388 individual income tax 76, 330-1, 344, 345 inheritance tax 209-12 land value tax 237 municipal inhabitants'tax 361 non-profit organizations 169 reform 333,388 Shoup recommendations 33 value added tax 240-1,294, 296, 297, 393 expenditure, v. income, as tax base 91-2 expenditure cuts 320-3 expenditure tax 92,251 n, 394-5 see also hybrid taxes expense accounts, fringe benefits 80 expenses see business expenses; medical expenses; R&D (research and development); tax credits exports, promotion 40, 41, 187-8, 189 expropriated land, capital gains tax 239 Fair Trade Commission 289, 290-1 fairness see equity family, as tax unit 78 farm incomes see agricultural incomes farm land see agricultural land farmers: household units, income distribution 158, 159, 160 other occupations 65 n, 160 n farming sector, decline 71 Federation of Economic Organizations (Keidanren) 198,321 FIFO, inventory valuation 168 financial reconstruction bodies 366 fire insurance premiums, deductions for 38, 63 n, 135 fiscal austerity 322, 322-3
fiscal consolidation (reconstruction) 319-20, 328 fiscal deficits 49-51,319-20 fiscal dividend 124-7 fiscal drag 124-7,129 fiscal equalization 367-74 fiscal performance, local 379-82 fiscal reconstruction 319-24,328 fiscal transfers 352-3 fiscal 'window-dressing' 322 n fixed asset tax see property tax flattening of rate structure 333-4, 390-1 flexibility of taxation 124 foreign corporations, taxable 169 foreign investment, promotion by tax incentives 40,41 foreign tax credits 77, 338 foreign trade taxes 18, 19,21 n forward-shifting: corporate taxes 185 value added tax 287-91, 392 France: corporate tax rates 171 corporate tax share 167 military expenditure 25 savings rate 136 tax burden 92,93 tax unit 78 tobacco tax 258 value added tax 268 fringe benefits, income tax 80-1 fuel taxes see energy taxes; liquefied petroleum gas (LPG) tax; petrol tax gains, capital see capital gains tax gas tax 363 n gasoline see petrol tax GDP (gross domestic product): ratio of tax revenues 7-8 social welfare transfers as percentage 25 general account deficits 50 n general consumption taxes 251-2, 268, 274, 276-7, 320 general sales tax 251 generalizations of change 16-17 Germany: avoidance of double taxation of dividends 177 corporate income tax 4 corporate tax rates 171 corporate tax shares 167 military expenditure 25
Subject Index savings rate 136 tax rates 83 tobacco tax 258 gift taxes 204-18 on transferred income 137n postwar system 4 redistributive effects 161 reform 339 share of tax revenues 5, 6 see also inheritance taxes Gini coefficient 149,151,162 inheritance tax 215-16 global warming 301-2, 304,313 n GNP (gross national product): correlation with tax shares 21—2 deflator 117 percentage of military expenditures 24-5 response of tax revenues to change in growth 44-8 tax ratio 18-19 'goal system" of collection 26 goods and services taxes 252-67 environmental protection 309 local 299 New Zealand 270 share of tax revenues 5, 6 see also excise taxes Gordon (G) model, corporate tax-shifting 185-6 government inquiries into tax systems 12 'grace period', value added tax 294,296 grants: equalization grant, Shoup recommendations 35 Equalization Grant Scheme 29 impact on local government budgets 380-2 specific-purpose 353, 354, 374-9, 380 Green Card system 389 n gross domestic product (GDP): ratio of tax revenues 7-8 social welfare transfers as percentage 25 growth see economic growth GS taxes see goods and services taxes Haig-Simons concept of income 97,200 n handicapped persons: deductions and exemptions for 77 n, 330 gift tax 209,339 inheritance tax 339 Hashimoto administration 340-1 health insurance, reform 322 heirs, inheritance tax 207-9,211-12,214,215,338-9
417
heuristic model of tax structure change 18,24 high-income earners 94,148,157, 335, 388,396 hikiatekin 190 Hitotsubashi University 17 Hokkaido-Takushoku Bank 340 Home Ministry 350 horizontal equity 64, 94-5, 335, 384 house purchase: relation to land prices 221 tax credits 77,98n, 135 household expenditure, environment-related 312-13 household sector, portfolio investments 140-2 household survey data 157-60 household units, redistributive effects 157-60 housing: promotion 40,41 subsidized, fringe benefits 80 hybrid taxes 4,91-2,393-4 corporate taxes 177 effect on portfolio decisions 141-2 ideological commitments 24 idle land 233,238 implicit carbon taxes 314 imputation method 175,176,177 incentives: corporate 187-203 effect on savings 138-46 policies 36—42 see also credits; special tax measures incidence of taxation 184,251 income: definition 97, 200 distribution and inequality 147-63 national, tax revenues as percentage 10, 11 taxable see taxable income v. expenditure, as tax base see hybrid taxes income classes, tax erosion 99-103, 104,105, 106 income elasticity of tax revenues and yield 44-8, 115n income-splitting: business incomes 52, 80 couples 78,79,331 income taxes 3-4,73-163,167-86 as revenue sources 19 avoidance and evasion 54, 87 n, 124, 239 before Shoup reforms 26 burden 92-3,344 international comparison 343 land transfer 224
418
Subject Index
income taxes (cont.): progressivity 28, 85-90, 95, 148, 153-4, 333-6 reductions 330-1 reform 344-5,393-5 share of national and local taxes 8-10 share of tax revenues 5,6 shift from indirect taxes 18,19 Shoup recommendations 33 tax delinquency 63-4 see also withholding at source index of tax rates 46 n indexing (inflation adjustments) 111-15,127 indicators, basic financial needs 372 indirect taxes 3-4,251-67 definition 8n, 251 equity 384 increases 324, 325 share of tax structure 8-10, 25 shift from land taxes 18,19 Shoup recommendations 27, 33-4 switch from direct taxes 8,10,271-2, 343 see also environmental taxes; value added lax individual, as tax unit 78 individual income tax 3-4,73-163 before Shoup reform 26 classification of taxable income 57, 59-61 elasticity 48 international comparison 343 reform 330-6,344-5 share of tax revenues 5, 6 Shoup recommendations 28, 33 industrial equipment modernization, promotion 40 industrial policy, tax incentives 187-8 Industrial Structure Council 313 n inefficient expenditures 320 n inequity see equity inflation: adjustments 110-29 effect on corporate tax base 202 inhabitants' tax: municipal 334, 355, 358-9, 360-1 prefectural 334-5, 355, 357-9 Shoup recommendations 34 inheritance taxes 204-18 burdens 205,206,212-14,338-9 land 224,227,234-5,238 agricultural 212n, 230-1, 236, 239-40 postwar system 4 redistributive effects 161,204, 213-17 reform 338-9
share of tax revenues 5,6 tax delinquency 64 initial depreciation 190 innovations, technological, tax incentives 40,188 Institute of Fiscal Affairs 91 n, 205 insurance: deductions for premiums 38,63 n, 135,141 exemptions for payments to heirs 339 proportion of portfolio investments 140-1 integrated system, corporate and individual income taxes 172-80 interest elasticity of savings 139 interest income: inclusion in or exclusion from tax base 91-2 inhabitants'tax 359 privileged savings 131-4 Shoup recommendations 33 special tax measures 96,133-4 tax collection 60 withheld tax 63 interest payments, corporate tax 182 interest rate, impact on savings 143 interest surcharges on retained corporate profits 172,173 intergovernmental fiscal relations 34-5,349-82 intergovernmental transfers 29, 284 International Fiscal Association 95 n International Institute of Public Finance 95 n interventionist approach 52 inventories, valuation 168,203 investment: business 40,41, 188 household sector 140-2 proportion of assets allotted 141 investment income 130-46 aggregation 79 withheld tax 387 invoices, non-use for consumption tax 281, 387, 391-2 Irish Commission 12 n 'irrational' sources of erosion 98 Italy: savings rate 136 n tax burden 92, 93 tobacco tax 258 Japan Tobacco and Salt 321 Japanese National Railroad 260, 321 Jichi Sogo Centre 374 n Jichicho see Ministry of Home Affairs Joint Committee 296
Subject Index joint-stock companies 170 n junbi-kin 190 'Kakuremino' ('hiding behind the government') 14 n 'Kasumigaseki' 142n Keidanren (Federation of Economic Organizations) 198,321 key industries, tax incentives 187-8,189 kokko-shishutsukin (specific-purpose grants) 353, 354, 374-9, 380 Korea, value added tax 274 n Krzyzaviak and Musgrave (K—M) and Kilpatrick (K) models, corporate tax-shifting 185-6 'Ku-ro-yon'phenomenon 64-70, 160n Kyoto protocol 302 labour force, size, effect on income inequality 151 land: agricultural see agricultural land Basic Land Law 222,235 capital gains tax: collection 60 reforms 227-30,238-9, 243-5 special tax measures 104-6,157 unrealized gains 233 'Comprehensive Land Policy Plan' 236 'disguised' 231 idle and underutilized 233, 238 inheritance taxes 224, 227, 234-5, 238 municipal taxes see land-holding taxes; property tax purchase, interest on loans, business expenses 338 land-acquisition tax 224 land-holding taxes 223-7, 232-4, 237, 240-1, 363 'land myth' 220 land ownership 222, 229 land policy 222 land prices 219-22, 226, 243, 244, 245, 328 land problem, nontariff barriers 221 Land Reform Bill 239 land regulation 221 land supply 222-3,239 Land Tax Bill 219 Land Tax Review Group 234 n land taxes: as revenue source 19 historical background 3 property tax 361-3 reform 219-47
419
land transfer taxes 227-8 land use 221 land value tax 234 n, 237, 240-1, 241-3, 396 land values 221, 225-6, 237-8, 245-7, 362 LDP see Liberal Democratic Party levels 7, 187-203,341-4 liabilities: couples 78-9 effect of inflation 110-24 see also taxable income Liberal Democratic Party (LDP): Tax Committee: land tax reform 233, 234, 236, 237, 240 relation to Tax Advisory Commission 13,15 Tax Council: land taxation 219 value added tax 278 tax reform compromises 219, 334 n, 335, 387 value added tax proposals 276,279,293^, 295-6 life cycle hypothesis, effect on savings 139 life insurance: deductions for premiums 38,63 n, 135,141 exemptions for payments to heirs 339 proportion of portfolio investments 140-1 LIFO, inventory valuation 168 light oil delivery tax 261 n, 363 n light vehicle tax 26 I n limited companies 170n liquefied petroleum gas (LPG) tax 253, 261,262, 306 liquidation income, enterprise tax 360 n liquor tax see alcoholic beverages tax loans: for house purchase, credits and exemptions 135, 142 for land purchase, interest as business expense 338 fringe benefits 80 local autonomy 26, 29, 349, 350, 355, 357, 359 Local Autonomy Board 350 local borrowings 365—7 local debts 354, 365-7, 380 Local Decentralization Law 357 Local Finance Law 365 local governments 351-4 finance, Shoup recommendations 29 fiscal deficits 320 n fiscal performance 379-82 see also local taxes local public bodies 351—2 Local Public Finance Law 375, 380
420
Subject Index
Local Public Finance Programme (LPFP) 352 Local Tax Law 357, 379 local taxes 349-82, 354-63 allocation tax 352, 354,367-74 before Shoup Mission 26 business tax 346 n consumption tax 284, 297-9 corporate taxes 171 distribution tax 369 goods and services (GS) tax 299 income tax 34,81 property tax 234 proportion to national taxes 10-12, 343 road tax 261,262 road transfer tax 364, 365 Shoup recommendations 34-5 sources of revenues 8—12 surtax 22 n, 273, 355, 357 transfer taxes 353,364-5 'lock-in' effects, capital gains tax on land 239 long-term gains, taxable income 135 Lorenz curve 149-50,215 losses on returned goods, tax-free reserves 190 low-income earners, tax burden 335, 344 LPFP (Local Public Finance Programme) 352 LPG (liquefied petroleum gas) tax 253, 261,262, 306 luxuries see commodity taxes machinery industry, tax incentives 188 Management and Coordination Agency, Statistics Bureau 66 manufacturers' excises 255, 264 manufacturers' sales taxes 251-2, 278 marginal tax rates, income tax 83-4 'Maruyu' system 131,132,328 Meade Report 27 medical doctors, tax benefits 332-3 medical expenses, deductions for 42, 63 n, 77 n, 361 n Meiji government 255,350 MERGE data 107 middle-income earners 331,384 military expenses, relative to GNP 24-5 Ministry of Construction (MOC), land policy 222 Ministry of Finance (MOF): Budget Bureau 50 consumption tax 298 corporate tax burden 197,198 excise tax 252
fiscal reforms 320, 322 n, 323, 324, 326, 329, 347 land policy 222 land taxation 230 n, 241 local taxes 357 natural tax increases 48 petrol-related taxes 261 relation to Tax Advisory Commission 14 revenue adjustment for tax changes 46 Securities Bureau 178 special tax measures 38,40-2 tax threshold calculation 76 tobacco tax 258 Ministry of Health and Welfare (MOHW) 8n, 56 n, 147,158 Ministry of Home Affairs (MOHA): consumption tax 298 local taxes 350, 352, 357, 361, 362, 365, 366, 367 property tax 245 relation to Tax Advisory Commission 14 Ministry of International Trade and Industry (MITI) 288, 303, 304, 313, 320 Ministry of the Interior 350 mix of taxes 395-6 modification coefficients, basic financial needs 372-3 modified flat tax 334 Monopolized Public Corporation 258 motor fuel taxes 309 see also energy taxes; liquefied petroleum gas tax; petrol tax motor vehicle tax 261 n, 363 n motor vehicle tonnage tax 253,254, 261, 262 motor vehicle transfer tax 364, 365 motoring excise taxes 254 multi-stage taxes 251-2 municipal taxes: income tax 81 inhabitants' tax 167, 224, 334, 355, 358-9, 360-1 land taxes see city planning tax; land-holding taxes; property tax tobacco consumption tax 363 n see also local taxes municipalities 351 governments see local governments Naimusho 350 Nakasone administration: control of fiscal deficits 320 n income tax cuts 331,334
Subject Index savings, withholding tax 385 tax reforms 275, 277, 279, 324-6, 329 value added tax 70, 268,279 national bonds 50 national government disbursements (specificpurpose grants) 353, 354, 374-9, 380 national income, tax revenues as percentage 10, 11 National Land Agency (NLA), land policy 222 National Tax Administration (NTA): Annual Report of National Tax Administration 66 n, 107, 215 blue returns 69 consumption tax enforcement 285 implementation of VAT 291 n, 292 Statistics on Private Wages and Salaries 66 n, 86, 107, 116n Statistics on the Self-Assessed Income Tax 68 n, 107, 116n withholding tax tables 62 national treasury grants-in-aid 375, 380 national treasury obligatory shares 375, 380 natural resources, promotion 40 natural tax increases 48-9 negative externalities 252 n, 253 net annual value, land 221 net income, corporate taxation 168, 200 net lending 141 net worth tax 29, 30, 83 Netherlands, tax burden 92, 93 neutrality 49-52, 383, 385-6 New South Wales Tax Task Force 234 n New Zealand: goods and service tax 270 tax unit 78 Nippon Telegraph and Telephone 321 Nippon Tobacco Product Industry 258 nitrogen oxide emissions 304 n NLA (National Land Agency), land policy 222 ncfukin 362 n nominal tax rates, corporate taxes 192-6, 197-8, 199 non-cash benefits, income tax 80 non-energy taxes 308-9 non-entitlement expenditures 322 non-juridical organizations 169 non-profit organizations 169-70, 172 see also corporations non-receiving local governments 371 non-reporting see under-reporting nontariff barriers, land problem 221
421
Norway, tax burden 92, 93 NTA see National Tax Administration nuclear power 312,357 obligatory shares 375, 380 Obuti administration 341 occasional incomes 61,84 OECD countries: company taxation 177 environmental taxes 301, 303 n, 304 n, 305, 308 tax revenues 5, 6, 7, 8 n value added tax 269-70 withheld taxes 61 n Office of Management and Budget 38 Ohira administration 274,275, 276, 277 oil, crude 306-7,310 oil crises 46,51,301,319 openness of economy, correlation with tax shares 21-2 Ordinary Account, local government 352 ordinary allocation tax 370-1 ordinary obligatory shares 375 overseas investment, promotion 40 over-shifting, corporate tax 185-6 partnership method, integration of corporate and individual income taxes 175 partnerships, registration as corporations 170 n payments in kind, fringe benefits 80 pension funds 135,140-1 per capita GNP, correlation with tax shares 21-2 per capita tax, inhabitants' taxes 359, 361 per-staff revenue 56-7,58-9 personal income, definition 142 n personnel 28 n, 55, 56-7, 58-9,291 see also administration petrochemical industry, encouragement 187,189 petrol tax 253, 260-3, 284, 306 see also excise taxes petroleum gas transfer tax 364,365 petroleum subsidies 306, 307 'piggyback' surtax 273, 355, 357 policy: environmental 301-3 formulation 12-15 individual income tax erosion 106-7 industrial, tax incentives 187-8 inflation adjustments 124-9 link with income distribution 161 personal savings 136-9 postwar strategies 36-52
422
Subject Index
policy (cant.): redistribution of income 161 tax incentives 36-42,187-96 Polluter-Pays Principles (PPP) 305, 309 pollution control 40, 41, 188, 300-1, 309 population: age ratio 142 n designated cities 352 prefectures 351 portfolios, personal savings, tax incentives 140-6 postal savings 131,132,135 postwar growth, effect on tax share 19 postwar military expenditures 24 postwar tax system 4, 6, 36-52 power resources development tax 253,254 PPP (Polluter-Pays Principles) 305, 309 prefectural taxes: consumption tax 297-9, 355 enterprise tax 355,367,368 income tax 81 inhabitants' tax 334-5, 355, 357-9, 367, 368 land taxes 224 see also local taxes prefectures 351 governments see local governments preferences 130-6 see also incentives price escalators 112-13 price-monitoring system 285-7,289-90 prices, effect of value added tax 285-91 privatization 321 privileged savings 131-4, 135-6,137-8 Productive Green Tract Areas 239-40 'productivity', tax personnel 55, 56-7 professional incomes, redistributive effects 155-7 profit share approach 185 profits, corporate taxation 1,19,26,33,167-86 progressive taxes: gift tax 209 income tax 28, 85-90, 95,148, 153-4, 333-6 property income, relationship to savings 143 property tax 6, 29, 232, 245-7, 361-3 agricultural land 230-1 land holding 224,232, 234 land valuation 237-8 share of tax revenues 5, 6 Shoup recommendations 34-5 tax burden 223 Urbanization Promotion Areas 230 property transfers see inheritance taxes protectionism, environmental policy 304-5
Public Enterprise Account, local government 352 public good, land 222 public works, obligatory shares 375 pyramiding, turnover tax 252 'quasi-corporations' 180-2 see also deemed corporations R&D (research and development), tax credits 77, 169,189,190, 191 rates of tax see tax rates rates (property tax), Australia 234 n 'rational' sources of erosion 98 Reagan administration 202,277 real estate taxes 61,361-3 rebuilding 319-49 'Rebuilding Japan's Archipelago' 220 receiving local governments 371 recreation facilities, fringe benefits 80 'recruit scandal' 386 n redistributive effects: of income tax 84,147-63 of wealth taxes 213-17, 396 reductions: annual 42-9 crucial issue for Shoup Mission 26 effect on redistribution of income 153,161 effect on revenue 56 individual income tax 329-30, 344-5 interest income 133-4 stimulation of economy 339-41 tax reforms 339—41 to offset inflation 115-24,125,127 reforms 3-4, 25-35, 319-49, 327, 383-96 regional development, promotion 40 registration and licence tax 224 regressivity: carbon/energy tax 313 value added tax 294, 396 reliefs: capital gains 142 land transfers 229-30 remuneration see salaries rental value, land 221 rents: imputed 135 subsidized 80 repairs, tax-free reserves 190, 347 research and development (R&D), tax credits 77, 169,189,190,191 reserves, tax-free 190-2, 196,202,347
Subject Index retail prices, effect of value added tax 285-91 retail sales tax 278 USA 270 retailers'sales taxes 251-2 retirement allowances, tax-free reserves 190, 196, 347 retirement incomes: collection 57, 59 received by heirs 339 separate tax rates 84 special deductions 77, 335-6 returned goods, losses, tax-free reserves 190 returns see tax returns revaluation see assets, land value; assets, revaluation revenue non-neutral approach 330 revenue-neutral approach 326 revenues: basic financial, local governments 373-4 changing patterns in the 1990s 341 cost-revenue ratio 48, 55-6, 57, 58-9 decrease by tax reform 329-31 effect of inflation 124-9 effect of tax reductions 56 elasticity 44-8 erosion 9 fiscal deficits 49-51 losses, from special tax measures 38-40,188-9, 190-1 Ministry of Finance, adjustment for tax changes 46 national and local 11 OECD countries 5, 6, 7, 8 n per-staff revenue 56-7,58-9 percentage of national income 10, 11 ratio to GDP 7-8 relation to economic growth 44-8 sources 5,6,8-10 alcoholic beverages tax 3,19,254 consumption tax 6, 8, 9 corporate taxes 3,167-8 income taxes 6, 19 inheritance tax 6,212,213 land taxes 3, 19,241-2 local taxes 8-12 social security contributions 5, 6, 8 n, 56 n tobacco tax 3 value added tax 294-5, 297 wealth taxes 8, 9 see also structure 'Rincho' (Ad Hoc Commission for Administrative Reform) 321,323
423
Rio Environment-Summit 302 road construction 261,306,307 Road Rating Assessments 237-8 road tax 253,254,261 road transport 309 roll-over scheme 229-30,235-6,239 Royal Commission on Taxation in Canada 97 n salaries: proprietors'and family employees', deduction 77, 182 withholding tax table 62 see also employment incomes; self-employment incomes salary earners, income inequality 150-1 sales tax with exemption for inter-firm transactions 278 sales taxes 251-2,268,277-9,278 savings 130-46 corporate, stimulation 40,41, 188,189 double taxation 92, 395 flat tax rate on interest 84, 385 inclusion in or exclusion from tax base 91-2 rates, personal 136-8 tax incentives 40,41 SCAP (Supreme Commander for the Allied Powers) 25-6, 26 n, 31 schedular taxes 3,4,28,91 Securities Transaction Council 183 securities transaction tax 253, 254,284 self-assessment 53-4, 61 effect of inflation 117,118-19,124 progressivity 87, 89-90 redistributive effects 151-2, 155-7, 162 Shoup recommendations 28 tax delinquency 64 tax returns 69-70 see also self-employment incomes self-employed entrepreneurs, registration as corporations 170 n self-employed households, income distribution 158, 159, 160 self-employment incomes: income inequality 150-1 income-splitting 52,331 manipulation of taxable income 52, 54 redistributive effects 155—7 tax collection 61 tax gap 68 underreporting 64—5 see also self-assessment
424
Subject Index
self-financing, business enterprises 175, 182-4 SEO (Survey of Economic Opportunity) 107n separate taxation: corporate taxes 176,177 erosion of tax base 96,103 income tax 84 see also special tax measures services taxes 271 see also goods and services taxes shochu, dispute with EU 257-8 short-term gains, taxable income 135 Shoup Mission and Report 4, 25 accession tax 204, 205 aggregation approach 84 commodity tax 264 corporate taxes 172-3, 174, 175,176,180, 189 income tax 91, 95 tax unit 78 turnover tax 273 n, 274 value added tax 273,274 SII (Structural Impediments Initiative) 221,236 simplicity, criterion for tax reform 383,386-7 single-stage taxes 251-2 small businesses 170 value added tax 278, 291, 393 see also self-employment incomes social security contributions: relationship to savings 143 tax revenues 5, 6, 8 n, 56 n social security system: effect on savings 139 reform 322 social welfare: benefits, fringe benefits 80 effect on fiscal deficits 51 enhancement through tax incentives 188 programmes 25, 51 reduction of spending 322 transfers, as percentage of GDP 25 socio-cultural factors, effect on savings 139,146 Sotozei 290-1 Special Account for Road Construction and Improvement 261 special allocation tax 370, 374 Special Council on the Transition to a new tax 285 special depreciation, corporate taxation 190-2 special land-holding tax 224, 232-3, 238, 363 special reliefs: inheritance tax 227 property tax 223
special tax measures 37-8 corporate taxation 187-203,389 effectiveness 139,140-6 erosion of individual income tax 94-109 investment income and savings 130-6 redistributive effects 154,161 see also incentives Special Tax Measures Law 37, 37 n, 38 special tonnage transfer tax 364, 365 special wards 352 specific-purpose grants 353, 354, 374-9, 380 split-rate system, corporate tax 4,174,175,177, 183-4 spouses: exemptions for 76, 79, 330-1, 333, 344, 385 tax burdens 339 staff 28 n, 55, 56-7, 58-9, 291 see also administration stamp tax 253, 254 'starving-out' of government expenditures 321 Statistics Bureau, Management and Coordination Agency 66 statistics, data sources: income tax 107-8, 116 n, 147-9 taxable income 65, 66, 67, 68 savings 143 statutory heirs, inheritance tax 207-9,211-12, 214,215 steel industry, tax incentives 188 stocks, sale, capital gains tax 30,60, 77,134-5, 173-4, 386, 388-9 straight-line method, depreciation 168 Structural Impediments Initiative (SII) 221, 236 structure 6-12,16-25,341-4 environmental taxes 305-9 subsidies: environmental policy 304 fuel 306,307 housing 80 to local governments 35 sugar tax 253,254 sumptuary tax 252 n, 254, 255 Sunley-Pechman formula 116 Supreme Commander for the Allied Powers (SCAP) 25-6,26n, 31 surtax 273,355,357 Survey of Economic Opportunity (SEO) 107 n Suzuki administration 320 n Sweden: savings rate 136 tax burden 92, 93
Subject Index Takeshita administration: capital gains tax 134-5, 386 income tax rates 334, 335, 336, 390 inheritance tax 338 reforms 326, 328,332, 333 n tax base 385 value added tax 70,268, 275-6, 277,279, 285 tangible business assets, property tax 362 tariff autonomy and structure 18 tatewari gyosei 351 n tax administration see administration Tax Advisory Commission 12-15 corporate taxes 174, 175, 176, 181,183-4,348 enterprise tax 360 land taxes 219, 222-3, 232, 233, 234, 235, 236-7, 240, 243 special tax measures 38 n, 139 Tax Identification Numbers 389 tax reforms 325, 326, 332 value added tax 276, 277,278, 279, 298 tax avoidance 54, 87n, 124,239 tax-base elasticity 45 tax bases 75-81 broadening through tax reform 331-3, 385, 387 comprehensive, definition 96-8 corporate taxation 200, 388 definition of time period 395 effects of inflation 110-15 enterprise tax 360 erosion 91,94-109, 109 income v. expenditure 91 inheritance tax 206-7,212 need to broaden 387—9 selective excise taxes 271 tax benefits: effect on personal savings 138-9 see also concessions; incentives Tax Bills 237,296 tax burdens see burdens tax bureaux 14 tax collection see collection tax concessions see concessions tax-credit method, consumption tax 281, 282,392 tax credits see credits tax cuts see reductions tax deferrals see deferred taxes tax delinquency 63—4 'tax design' 31,32 tax equity see equity tax erosion see erosion tax evasion 54, 87n, 124, 137-8
425
tax exemptions see exemptions tax-free reserves 190-2,196, 202,347 tax gap 64-70 tax-GNP ratio 18, 19 Tax Identification Numbers (TIN) 63,135, 389 tax incentives see incentives tax indexing (inflation adjustments) 111-15, 127 tax instruments, environmental policy 305 tax levels 7,187-203, 341-4 tax liability see liabilities tax mix 395-6 tax personnel 28 n, 55, 56-7, 29) see also administration tax policy see policy tax preferences see preferences tax process 12-15 tax-rate elasticity 45 tax rates: adjustments, effect on income distribution 154 analysis of tax erosion 99 capital gains on land 229 commodity tax 265, 266 consumption tax 283, 392 corporate taxation 171-2, 173,192-203, 346-8, 360 dividends 134, 178 enterprise tax 360 flattening of structure 333-4, 390-1 index 46 n individual income tax 81-90, 99-103, 344-5 inhabitants'tax 361 inheritance and gift taxes 209-12, 338-9 interest 133-4 property tax 226 n, 230 n, 362 simplification 390-1 tax reductions see reductions tax reforms 3-4, 25-35, 319-49, 327, 383-96 tax reliefs see reliefs lax returns 61,69-70 corporations 168,180,181,182 deduction of cost of preparation 77 Shoup recommendations 28 tax revenues see revenues tax shares 18, 19,20-2,24 tax-sharing ratio, local allocation tax 369-70 tax-shifting, corporate 185 tax statistics see statistics tax styles 25 tax thresholds see thresholds tax unit 78-80 tax yield 99, 100,109
426
Subject Index
taxable corporations 169-71 taxable income: classification 57, 59-61 compulation 75-6 corporate taxes 200-2 definition 95 technological development and innovation, tax incentives 40, 188 Temporary Tax Measures Law 37 n, 38 n, 189 textile consumption tax 33 thresholds: income tax 76, 128-9 inheritance tax 338 timber income 61,77,84 TIN (Tax Identification Numbers) 63,135, 389 tobacco taxes 34, 252-3, 254, 258-60,284, 363 n see also excise taxes Tokyo: agricultural land 230 n, 231 prefectural population 351 special wards 226,352,371 tradable permits 309 n trade (foreign), taxes 18,19 training, tax personnel 28 n transactions tax (turnover tax) 4, 33, 252, 272-3, 282, 325 'transfer from the local to the national government' 353 transfer payments, erosion of tax base 98 transfer taxes see gift taxes; inheritance taxes transfers: fiscal 352-3 property tax 362 n travel tax 253,254 treasury grants-in-aid 375, 380 treasury obligatory shares 375, 380 turnover tax 4, 33,252, 272-3, 282, 325 Uchizei 290-1 UK see United Kingdom UN (United Nations), Framework Convention on Climate Change 302 unconditional fiscal transfers 352 underground economy 65 n under-reporting 64-5, 148 omission from statistics 149 source of tax erosion 107 underutilized land 233, 238 undistributed profits, tax rate 174, 175 unfairness see equity
unit of taxation 78-80 unit-land price 221 United Kingdom: corporate tax rates 171 corporate tax share 167 cost—revenue ratio 55, 56, 57, 58—9 Meade Report 27 military expenditure 25 PAYE, Schedule E 63 n portfolio investments 140,141,142 savings rate 136 tax burden 92,93 tax rates 83,171 tax unit 78 tobacco tax 258 United Nations, Framework Convention on Climate Change 302 United States of America see USA unrelated business income 169,170 urban land prices 220-1 Urban Promotion Areas 236 Urbanization Control Areas 230 n Urbanization Promotion Areas 230, 238, 239-40 USA: corporate taxes 167,171,199-203 cost-revenue ratio 55, 56, 57, 58 impact of national government on local government 380-2 military expenditure 24, 25 occupation, effect on local government structure 350 portfolio investments 140,141, 142 retail sales tax 270 savings rate 136 tax burden 92,93,199-203 tax rates 83,171 tax revenue sources 5, 6 tax unit 78 tobacco tax 259 Treasury 12 n, 91 n, 92, 327, 331 value added tax 270 see also Shoup Mission and Report valuation: assets 29,33 inventories, valuation 168,203 land 225-6, 234, 237-8, 362 see also assets, land value value added ratios 292, 293 value added tax (VAT) 268-99 alcoholic beverages 257
Subject Index cash-flow benefits 294 'Ku-ro-yon' phenomenon 70-1 merged with commodity tax 264, 267 multi-stage sales tax 252 need for reform 391-3 prefectural tax 273 Shoup recommendations 29, 30, 34, 273-4 simplified procedures 391 tobacco 260 vanishing exemption method 282-3,292-3 VAT see value added tax vehicle-related taxes 261-3, 263, 309 vertical consolidation 351 n vertical equity 94-5, 103-4, 335, 384, 385, 388 Vietnam, value added tax 269 wage earners, income inequality 150—1 wages: family employees', allowable deduction 77 see also self-employment incomes; withholding at source wartime: military expenditures 24-5 taxes 3-4 commodity tax 264 wasteful expenditures 320 n, 322 wealth: as tax base 91n
427
redistribution 84, 396 accession tax 205 income tax 147-63 inheritance and gift taxes 161, 204, 213-17, 396 taxes' share of revenues 8, 9 welfare see social welfare whisky, international disputes 256, 257-8, 271 white returns 69, 70 wholesalers' sales taxes 251-2 windfall revenue gains, value added tax 294-5, 297 'window-dressing', fiscal 322 n withholding agents 62, 63 withholding at source 53-4, 57, 59-64 effect of inflation 120-1,122,124, 127 investment income 387 Nakasone administration, savings, withholding tax 385 OECD countries, withheld taxes 61 n redistributive effects 151, 163 Shoup recommendations 28 World Trade Organization (WTO) 258 Yamaichi Security Co. 340 'year-end' adjustment 63, 387 yield 99,100, 109
![Fushigina Toshokan [Japanese Edition]](https://epdf.mx/img/300x300/fushigina-toshokan-japanese-edition_5adf1863b7d7bc19071a9d07.jpg)
