Spending Without Taxation: FILP and the Politics of Public Finance in Japan

Spending without Taxation

FILP AND THE POLITICS

OF PUBLIC FINANCE IN JAPAN

Gene Park

Stanford University Press

Stanford, California

© 2011 by the Board of Trustees of the Leland Stanford Junior University.

All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.

Printed in the United States of America on acid-free, archival-quality paper

Library of Congress Cataloging-in-Publication Data

Park, Gene, author.

Spending without taxation : FILP and the politics of public finance in Japan / Gene Park.

       pages cm. — (Studies of the Walter H. Shorenstein Asia-Pacific Research Center)

   Includes bibliographical references and index.

   ISBN 978-0-8047-7330-0 (cloth : alk. paper)

   1. Fiscal policy—Japan—History. 2. Finance, Public—Japan—History. 3. Japan—Appropriations and expenditures—History. 4. Taxation—Japan—History. 5. Japan—Politics and government—1945– I. Title. II. Series: Studies of the Walter H. Shorenstein Asia-Pacific Research Center.

HJ1394.P37   2011

336.52—dc22                                 2010043067

Typeset by Thompson Type in 11/14 Adobe Garamond

E-book ISBN: 978-0-8047-7766-7

Contents

Figures and Tables

Acknowledgments

1 Introduction

2 Understanding the FILP System

Part I: FILP and the Postwar Settlement

3 The Common Origins of Budget Restraint and FILP, 1945–1953

4 Balancing Fiscal Policy, Industrialization, and Distributive Politics, 1953–1970

5 The Electoral Logic of FILP Allocations, 1960–1993

Part II: The Limits of FILP

6 Pushing the Limits of the FILP Compromise, 1970–1990

7 The Politics of FILP Reform, 1990–2001

8 The Koizumi Reforms and the Legacy of FILP, 2001 and After

9 Conclusion

Notes

Bibliography

Index

Figures and Tables

Figures

1.1   Government Outlays as a Percentage of GDP for All OECD Countries, 1960 and 1970

1.2 Government Receipts as a Percentage of GDP for All OECD Countries, 1960 and 1970

2.1   The Old FILP System

2.2   The New FILP System

2.3   Growth of FILP, 1953–2001

2.4  FILP relative to the General Account Budget, 1953–2000

2.5  Public Fixed Capital Investment and Government Receipts, 1970

2.6   Public Fixed Capital Investment, 1999

3.1   Factors behind the Establishment of FILP

4.1   Change in Budget and FILP Plan during Revival Budget Negotiations, 1954–1965

4.2   Interest Rates, 1951–1970

4.3   Change in Supplementary Budget and FILP Plan during Negotiations, 1954–1965

4.4   The Structure of FILP

4.5  Considering a Counterfactual: Options If FILP Did Not Exist

5.1   Testing the Causal Pathways

5.2   The FILP Variables

6.1   Budget Spending as a Share of GDP, 1970–1989

6.2   Bond Dependency Ratio, 1967–1989

6.3   Road Financing, 1954–2005

6.4 TFBF Lending to the Local Fiscal Allocation Special Account, 1970–1990

6.5   Repayment Burden for Lending to Local Allocation Special Account, 1975–1990

6.6 Total Local Government Bonds in the Local Government Debt Plan, 1970–1990

6.7  TFBF and JFCME Purchase of Local Government Bonds, 1970–1990

6.8  Absorption of Local Government Bonds Using Public Funds, 1970–1990

6.9   TFBF and Postal

Savings Share of New Government Bond Issues, 1970–1990

6.10 TFBF and Postal

Savings Purchase of Government Bonds, 1970–1990

6.11 Interest Rates, 1970–1990

7.1 Postal Savings Deposits as a Share of Total Household Savings, 1990–1999

7.2  TFBF and JFCME Purchase of Local Government Bonds, 1990–2000

7.3  TFBF and Postal Savings Purchase of Government Bonds, 1990–2000

7.4  TFBF and Postal Savings Share of New Government Bond Issues, 1990–2000

7.5   FILP Allocations, 1990–2001

8.1   Declining Size of the FILP Plan, 2000–2009

9.1   Relative size of budget and policy finance, 2004

Tables

1.1 Government Outlays and Receipts, G-7 plus Sweden

1.2   Social Security Transfers as a Share of GDP, G-7 plus Sweden

2.1   FILP Funds by Source, 1960–1999

2.2   FILP Agencies

2.3  FILP Plan Allocations by Policy Area, 1953–2000

2.4   Budget Expenditures by Purpose, 1960–2003

2.5   Policy Finance in the G-5, Outstanding Government Loan Balance, 2003–2004

2.6   Comparison of Budget and FILP

2.7   Public Finance Goals of Key Actors

3.1   Industrial Capital Investment

3.2   RFB Financing Balance, 1947

3.3   Counterpart Fund, FY 1949–1952

3.4   Establishment of Various FILP Exit Institutions

3.5   Requests from Business Groups

4.1   Estimated Annual Tax Reductions, 1954–1970

4.2   Paying for the National Income Doubling Plan

4.3   Social Security Transfers as a Share of GDP

5.1   Regression Results for Seat Share of LDP in the Lower House

5.2   Dependent Variables for Models 3–6

5.3   Ruling Party Variables

5.4   Control Variables and Expected Sign

5.5   Regression Results for FILP Variables

7.1   Influence of Amakudari

7.2   Special Corporation Reform

7.3   Special Corporation Reform

7.4   Special Corporation Reform

8.1   Reform of the Government Financial Institutions

8.2  Budget Transfers to the Government Financial Institutions, 2006

Acknowledgments

In completing this project, I have incurred many debts to people and institutions that were critical to making it possible. I owe the greatest debt to my former advisors at Berkeley, who helped shape and guide the project from its inception. I initially went to Berkeley to work with Steven Vogel, who provided intellectual guidance and support from my first day of graduate school. I had the great fortune to work with Jonah Levy and T. J. Pempel, who generously read numerous drafts and shaped my ideas at critical junctures. Jonah first suggested the title of this book, which nicely captures the essence of the project. Kevin O’Brien and AnnaLee Saxenian provided advice in the early stages of the project.

I am also appreciative of the many people who helped me with my research. Many individuals in Japan generously offered their time. This work would not have been possible without the many interviews and meetings I had with numerous government officials. Kagehide Kaku provided several key introductions. Mao Guirong helped me navigate the Japanese literature. Iris Hui offered helpful advice on my statistical analysis. Many others provided input at various stages, including John Ciorciari, Eisaku Ide, and Ed Fogarty. The anonymous reviewers will notice the significant contribution they made in improving the manuscript. This project would not have been possible without institutional and financial support. Thanks to Fulbright IIE I was able to spend more than a year in Japan conducting research. The Blakemore Foundation funded my language training. The Japanese Ministry of Finance’s Policy Research Institute (PRI) hosted me for two years. I am very grateful to the Shorenstein Asia-Pacific Research Center (APARC) at Stanford University for providing funding and a very congenial environment to work on my project.

Finally, I would like to thank my family. My parents have always encouraged me to pursue my intellectual interests, and my brother has been a great lifelong friend. Emily Chung made the last several years very happy ones, and she is also a great editor.

Spending without Taxation

ONE

Introduction

Everyone likes spending, although no one likes to pay. Governments are the same. While they would like to deliver popular goods and benefits to voters, paying for such spending requires unpleasant choices, levying taxes or running budget deficits. Because they cannot have high spending, low taxes, and balanced budgets, they have to make difficult political choices. Governments, thus, face a fiscal trilemma. But what would happen if a government found a means of spending without taxation? This book contends that this is precisely what Japan did. The ruling party, the Liberal Democratic Party (LDP), used a system of public finance that did not rely on taxes—the Fiscal Investment Loan Program (FILP)—and allowed it to do the seemingly impossible: keep taxes low and budgets balanced, all without having to restrain public spending. This combination was at the core of a distinctive postwar political bargain: one that eschewed high budget spending and taxation, expansion of the welfare state, and Keynesian-inspired fiscal stimulus. By doing so, though, it was striking a Faustian bargain that eventually undermined the political settlement that it helped underwrite.

By focusing on FILP, this study presents several novel findings. First, it demonstrates that a financial mechanism, FILP, enabled the Japanese government to run a distinctive neoclassical fiscal policy based on low budget spending from the end of the 1940s through 1970. This ran directly counter to the postwar trend in other industrial democracies, where governments increased budget spending and taxation to finance the expansion of the welfare state and in many cases employed fiscal stimulus to maintain full employment. Second, it shows that the government’s policy of budget restraint and pork barrel spending were two sides of the FILP coin. This finding resolves the contradiction between the view that the LDP has stayed in power through profligate public spending and the reality of low budget spending in Japan. Third, this study reveals that, while the government’s early commitment to budget restraint initially delivered economic benefits, it came at a very high long-term cost: heavy state intervention in finance, deferred fiscal burden, and the political challenge of reforming the mechanism that made it all possible.

In comparative perspective, the Japanese case illustrates a larger point about the politics of public spending. While most comparative studies focus on budget spending, taxation, and budget deficits, the experience of Japan demonstrates that governments can finance their activities not only through taxes but also through the allocation of credit and other financial mechanisms. Ignoring the role of such policy finance comes at the risk of underestimating or mischaracterizing the size and scope of the state, a point often overlooked by studies focused on explaining fiscal outcomes. Comparing Japan’s experience with several minor cases, this study finds that three factors contribute to extensive use of policy finance: external budget restraint, strong domestic political support for policy finance, and the centralization of policy finance within the budget-making apparatus.

Fiscal Choices and Policy Finance

Public spending choices are at the heart of how governments attempt to reconcile the competing demands of democratic politics and the market system. After the Great Depression and continuing after World War II, higher taxes, higher spending, and, in many cases, fiscal stimulus through deficit financing were central features of the postwar political settlement throughout the industrialized democracies. The expansion of welfare programs led to a larger and more redistributive state, and Keynesianism-inspired fiscal stimulus dampened the effects of cyclical economic downturns by maintaining employment. Along with the Bretton Woods institutions and a new trading and financial regime, these changes established the foundation for what Shonfield called modern capitalism and what Gourevitch and Hall observed served as an historic compromise between capital and labor.¹ The expansion of the state and Keynesianism were defining features of postwar political economy across the industrialized democracies.

This bargain was not without political trade-offs and costs. Deficits strained budgetary resources, and, as taxation rose, so did opposition to it. Since the 1970s, the renegotiation of the terms of the postwar settlement has been one of the central political dramas unfolding throughout the industrialized world. A new economic orthodoxy challenged the utility of fiscal stimulus and called for reductions in public spending, including retrenchment of the welfare state. Since the 1990s, the acceleration of economic globalization has sharpened the fiscal trade-offs confronting governments. Global financial integration has increased the costs of deficits and certain forms of taxation. This has not, however, eliminated government choices. As Carles Boix has argued, governments can still choose between two supply-side economic strategies, a public investment strategy and a private investment strategy. Governments run by left parties favor the former in an attempt to raise directly the productivity of capital and labor through more expenditure on infrastructures and education and, sometimes, through the creation of a public business sector.² By contrast, right parties attempt to lower taxes to increase savings and private investment, that is, the private investment strategy. Each strategy, though, creates distinct electoral dilemmas. Excessive spending cuts may alienate many middle-class voters. On the other hand, high public investment requires high taxes, which may lead to a political backlash.

This formulation, while parsimonious and useful, overlooks that governments can finance public investment not only through taxes or borrowing but also through financial mechanisms to steer credit and investment, what this study calls policy finance.³ Policy finance is the use of credit and other financial mechanisms to achieve public policy purposes. In a deliberate attempt to avoid making difficult fiscal choices, the Japanese government mobilized and deployed a system of policy finance—the FILP—that did not rely on taxes, providing the government, at least initially, with a form of spending without taxation. While often overlooked, policy finance is an important component of many states’ systems of public finance. Comparatively, though, the size of Japan’s system of policy finance and its structure have given it a particular salience in Japan’s political economy. Until reforms that took effect in 2001, the state-run postal savings system, public pensions, and several other smaller sources provided funds to FILP. Established in 1953, FILP grew rapidly; at its peak, FILP drew on approximately four trillion dollars of funds, and annual allocations reached 80 percent of the size of annual general account budget expenditures. None of the other major industrial countries have had a policy finance system nearly as large nor one as closely connected to the management of the formal budget (see the next chapter for specific comparisons).

The Argument

This study contends that the state’s mobilization of policy finance was central to the postwar political bargain in Japan, first by enabling the ruling party to forge a political settlement that delivered economic growth and political stability and then by sowing the seeds of its own unraveling. Initially, FILP allowed the government to maintain budget restraint without having to sacrifice spending. This combination of otherwise contradictory policies was vital to the ruling party. Low budget spending was a pillar of the government’s economic growth strategy, allowing the government to keep taxes low and budgets balanced through the start of the 1970s. Low taxes would promote savings and private investment. A small public sector would unleash the dynamism of the private sector. Suppressing budget spending would enable the government to maintain a balanced budget, which in turn would prevent private sector crowding out,⁴ help stabilize Japan’s international balance of payments, curb inflation, and allow the government to run a more activist monetary policy.⁵ Limiting budget spending also meant that the Japanese government could deliver popular tax cuts, an often-overlooked yet significant pillar of the ruling LDP’s political strategy, without running up deficits.

Yet contrary to the view that Japan’s fiscal conservatism reflected the social coalition or ideological orientation of the political party, budget restraint was unpopular within the conservative camp and conflicted with its political strategy of delivering political pork to its constituents. FILP allowed the government to square the circle. The Japanese government established and then deployed FILP as a means to limit budget spending and to pay for the priorities of the ruling coalition. FILP not only helped finance the government’s industrialization policies but also kept the ruling party in power and the conservative coalition unified. FILP financed the government’s economic development priorities by providing credit to strategic industries and funds for critical economic infrastructure. The ruling Liberal Democratic Party also used FILP along with the budget to provide generous material compensation to its supporters, channeling funds to its conservative base—farmers, small and medium-size enterprises (SMEs), rural areas, and big business—and over time to new constituencies.⁶ These allocations via FILP translated into electoral support for the LDP, as the statistical analysis presented in this study demonstrates. By serving as a supplement to the budget, FILP also helped keep budget expenditures down, allowing the government to deliver tax cuts and extend tax exemptions to supporters (small business and agriculture and rural workers) without sacrificing budget balance.⁷ FILP thus linked the government’s fiscal policy, industrialization strategy, and the ruling party’s political strategy.

Although FILP helped forge Japan’s stable postwar political settlement, it also embodied the limitations of this arrangement. The government could use FILP to broker a larger political settlement because it is a financial system, which drew on the nation’s large pool of savings rather than taxes. Yet precisely because FILP is a financial system, it could not sustain this compromise over the long term. Unlike the budget, FILP allocates finance capital in the form of loans and investments that must be repaid. The government, however, subordinated the financial management of FILP to two competing goals, minimizing budget spending and funding the LDP’s political strategy. This practice intensified from the 1970s as political pressure on the LDP mounted, the budget fell into deficit, and industrialization declined as a national priority. During the 1980s, the government restored budget discipline by pushing items that should have been funded by the budget onto FILP. While exploiting FILP helped the LDP stay in power and solved the political problem of balancing competing interests in the conservative camp, it weakened the finances of FILP as the quality of its investments and loans deteriorated.

Over time, FILP’s capacity to paper over differences between fiscal hawks and the pork-barrel wing in the LDP deteriorated. The government was forced to use budget funds to cover losses from failed FILP-financed projects. Moreover, reform of the FILP system emerged as a highly divisive issue within the ruling party. The expansion of FILP conflicted with the government’s goal of financial liberalization and drew opposition from private banks that argued that the system competed with them. The rapid growth of FILP also fed into the perception of a state apparatus that had grown out of control. Finally, failures of FILP-financed projects and stories of mismanagement and corruption increased public opposition, leading to calls for reform. In short, the features of FILP that had made it so useful to the ruling coalition caused FILP to become a political liability over time.

The issue of FILP reform came to a head with Prime Minister Koizumi, who launched an attack on the entire FILP apparatus after coming to office in 2001. Battling opposition from his own party and the bureaucracy, Koizumi passed numerous reforms by, among other methods, even expelling members of his own party. Despite the political drama, the reforms are relatively modest in aim, rationalizing the FILP system rather than abolish it. Reforms have limited abuse of the FILP system by making it harder for the government to use FILP as a substitute for the budget. As a result, the reform will force the government to confront its fiscal trade-offs more squarely. In the end, FILP ironically exacerbated the fiscal options confronting the government. FILP not only has left behind significant debt, but it has created powerful constituents that rely on public largesse. Balancing Japan’s need for fiscal reconstruction, budget spending, and taxes will be one of the central constraints as political parties, both the DPJ and LDP, attempt to build an alternative to the postwar settlement that FILP had made possible.

Public Spending and Japanese Political Economy

This book clarifies several issues central to understanding Japan’s political economy. First, by focusing on the role of FILP, it helps resolve a central debate in the study of Japanese political economy: how the economic and political sides of Japanese political economy fit together. On the one hand, liberal public spending is widely cited as one of the primary means by which the ruling LDP has stayed in power.⁸ The LDP is often portrayed as a pork-barrel machine that lavishes spending on its supporters, an image bolstered by its high level of spending on public works, which was five times higher than other Organisation for Economic Co-operation and Development (OECD) countries.⁹ Yet this has never squared well with the reality of low budget spending in Japan, which formed a critical element of its economic growth strategy. Even in recent years, despite large deficits, Japanese spending has remained low despite very high public investment and increasing welfare commitments.

The role of public spending is also tied to two very different views of the nature of Japanese politics. In the work on the developmental state, the Ministry of Finance (MOF), along with the former Ministry of International Trade and Industry (MITI),¹⁰ is cited as the paradigm of state autonomy, and MOF’s supposed ironlike grip over the budget and ability to limit budget spending one of the best expressions of this autonomy.¹¹ Yet those focusing on the clientelist aspects of Japanese politics have pointed to the ruling party’s liberal use of state spending to reward supporters and extend their political base.¹² According to Scheiner, pork-barrel spending combined with clientelist politics and high fiscal centralization undermines the development of a viable opposition and thus supports ruling party domination. Others have pointed out how under Japan’s former single non-transferable vote electoral system (SNTV), the LDP used pork-barrel spending to split the vote share in electoral districts allowing them to win a higher number of seats.¹³ As this book shows, FILP was the critical link between the developmental and clientelist sides of Japan’s political economy that made it possible for the LDP to spend liberally and restrain budget outlays.

Second, this book helps better situate Japan’s spending choices in comparative perspective. Japan has long stood out among the advanced industrial democracies for its low budget spending, a point noted and commented on by others.¹⁴ Japan’s budget restraint was particularly striking until the early 1970s. During this time the government kept budget spending and taxes low, and with only minor exception budgets balanced. In 1960, Japan’s government outlays were the second lowest in the OECD. Only Switzerland, whose constitution imposes limits on the government’s power to tax, had lower outlays.¹⁵ By 1970 Japan had the lowest level of taxation and budget outlays of all OECD countries, including countries at a lower level of economic development as well nondemocratic ones (see Figures 1.1 and 1.2). The government deliberately suppressed budget spending as part of a distinctive economic growth strategy. By contrast, governments of other advanced industrial democracies embraced higher taxes and higher public spending; many governments also employed fiscal stimulus to maintain full employment. More than just a fiscal policy, a larger and more activist state represented a political accommodation that helped balance the competing demands of the market system and democratic demands for social protections.¹⁶ The Japanese government, however, eschewed the expansion of the state and activist fiscal policy. To keep spending low, the government limited welfare spending until the early 1970s, which helped it keep taxes low and budgets balanced (see Tables 1.1 and 1.2).

Perhaps even more striking, the LDP, despite limiting budget spending, managed to forge a lasting and stable political coalition. Since the 1970s, both public debt and budget spending have grown, but Japan’s budget spending remains below the OECD average, even accounting for Japan’s low defense spending and special account budgets (see nondefense outlays in Table 1.1).¹⁷ In comparative terms, the combination of low budget spending and the LDP’s political dominance are remarkable. The LDP ruled virtually without interruption from its formation in 1955 until 2009. Among the advanced industrial democracies, parties with long periods of dominance have used generous budget spending to build political coalitions. Sweden’s Social Democrats (SAP) and Italy’s Christian Democratic party (DC) both increased budgetary spending sharply after World War II and embraced Keynesian stimulus. Although they had very different political economies, both quickly became big spenders. In Sweden, the SAP built on its Red-Green alliance by winning over white-collar voters through the expansion of its universalist welfare state.¹⁸ Italy’s conservative DC, an ideologically diverse catchall party with similarities to the LDP, used patronage and pork-barrel spending to win support and enhance its electoral appeal. By the end of SAP’s period of electoral dominance, Sweden had become the highest-spending country in the OECD as measured by government outlays as a share of GDP. By the same measure, at the time of the DC’s collapse in the mid-1990s, Italy was the highest-spending of the G-7 countries and the fourth highest in the OECD; only Sweden, Finland, and Denmark were higher. Japan’s experience is only exceptional, however, if one ignores Japan’s heavy reliance on policy finance. The characterization of Japan as a low spender, as this study shows, is incorrect. Rather, Japan supplemented its formal budget with liberal use of policy finance, a choice that proved to be highly consequential.

Third, the book illuminates why an otherwise esoteric institution—FILP—was the center of such divisive contemporary political battles in Japan. On the face of it, Japan’s efforts at cutting spending and virtually constant administrative reform since the 1990s seem puzzling. Japan’s level of budget spending is low, and its state apparatus is trim. Much of what became targeted for reform, though, was part of a sprawling extrabudgetary, parastate apparatus. Many of the most contentious and high-profile reforms centered around various parts of the FILP system from the post office, which provided funds to FILP, to the panoply of institutions that lent and invested FILP funds, including the infamous Road Corporation. Scholars have analyzed the politics of many of these reforms.¹⁹ Several studies have suggested that electoral reform, passed in 1994, pushed reform onto the agenda by forcing political parties to attend less to particularistic concerns and more to broader national issues.²⁰ While electoral reform certainly accelerated the reform agenda, what is missing is a discussion of why FILP itself came into existence and how it was embedded in a larger political bargain. As this study will show, although electoral reform clearly mattered, the issue of reforming FILP had deeper, partly endogenous origins.

Explanations in the Literature

Both comparativists and Japan specialists have noted Japan’s pattern of low budget spending, and some have offered explanations that differ from the argument this book presents. While these alternative arguments offer plausible or at least partial explanations, none alone are fully satisfactory. This section reviews these arguments, while the following chapter discusses the existing literature on FILP.

THE ECONOMIC GROWTH ARGUMENT

Several scholars have argued that rapid economic growth allowed the ruling party to keep budget spending low while also attending to its political needs. The argument is as follows. Rapid economic growth increased tax revenues. Thus, increases in budgetary spending could be financed through natural tax revenue growth rather than an increase in tax burden or government debt, allowing the government to avoid difficult trade-offs. Gerald Curtis, who argues that the LDP’s control over budget resources was the single most powerful tool for perpetuating its dominance, explains how growth ameliorated the need for hard political choices: A rapidly growing economic pie made possible the steady expansion of the national budget, enabling the LDP to respond to public demands without making any hard choices about priorities.²¹ Yasusuke Murakami similarly points to the role of growth in holding together the LDP’s coalition: Especially the LDP and the bureaucracy obtained, thanks to rapid economic growth, sufficient funds to subsidize agriculture and other declining industries liberally so as to appease potential discontent in these sectors.²² In a slight variation on this argument, Richard Katz contends that Japan could grow more rapidly during its phase of industrial catch-up and that this growth allowed the government to promote winners and compensate losers; he writes, The government didn’t choose between these two options. It did both.²³

An increase in the supply of tax revenues from rapid economic growth, however, shapes the context in which choices are made, not the actual choices made. Although growth is correlated with budget balance, there is no clear evidence that increased economic growth actually leads to lower levels of public spending or lower tax burden. Wagner’s Law in fact posits that increased wealth leads to higher levels of public spending, although empirical studies have reached contradictory findings.²⁴ During the long boom of thirty years of rapid economic growth after World War II, governments in advanced industrialized democracies increased state spending and taxation, often more rapidly than the rate of economic growth. Consequently, both increased relative to GDP. Economic growth, thus, did not correspond to lower levels of state spending and taxation relative to the size of the economy, although it did make it easier for most governments to limit budget deficits.

Thus, while growth certainly eased the choices confronting the government by increasing revenues, it is not a sufficient explanation of how the Japanese government managed the political challenge of limiting budget spending. A government must have the intent to limit spending and the political ability to enforce it. The Japanese government, at least through the early 1970s, has been singular in both regards. Yet there has been inadequate consideration of the emergence of the goal of fiscal discipline or how the ruling party reconciled this goal with its other priorities. Fiscal hawks faced the challenge of pursuing their goals without alienating other parts of the coalition, particularly pork-oriented politicians within the ruling LDP. Growth helped alleviate the conflict, but unless fiscal hawks were in a strong enough position to enforce their preferences or accommodate others in the coalition, it is doubtful that growth alone would have sufficed to limit spending growth. Moreover, although economic growth generates growing tax revenues, it also has the effect of creating additional demand for public investment. This is precisely what happened during Japan’s so-called high-speed growth years; growth led to surging demand for expensive industrialization-related infrastructure.

THE CONSERVATIVE COALITION AND PARTISAN POLITICS ARGUMENTS

Other scholars have pointed to the nature of the ruling party’s social coalition and government’s dominance by a conservative party to explain Japan’s high level of budget restraint. T. J. Pempel argues that the LDP forged a conservative social coalition between big business and the less competitive sectors such as small business and farmers.²⁵ The exclusion of labor allowed the LDP to avoid an expansion of welfare spending, thereby keeping budget spending low. This argument follows the research in comparative political economy that shows that conservative governments spend less than left ones.²⁶ According to this partisan politics argument, low budget spending is the consequence of the dominance of the Japanese government by the conservative LDP, which ruled from 1955 until 2009 with only a brief spell out of power.

At first glance the conservative coalition and partisan politics arguments seem persuasive. In Japan, labor was excluded from the ruling coalition; welfare spending was suppressed, at least until the 1970s; and a center-right party governed virtually without interruption. Nonetheless, there are several reasons why neither explanation is fully satisfactory. First, a conservative coalition does not necessarily make it possible for a government to limit public spending. As other scholars have noted, France’s conservative postwar parties, which dominated the presidency until 1981, relied on the support of a very similar social coalition that excluded labor and included farmers, small business, and big business.²⁷ Yet the French government oversaw a very large expansion of the state after World War II and also presided over large periodic budget deficits.²⁸ Likewise, Italy’s DC party relied on the support of a similar social coalition, but its government also exhibited little budget restraint.

Second, it is problematic to assume that Japan’s conservative coalition actually had a preference for low spending and fiscal balance. Japan’s prewar and wartime coalitions were essentially the same as the postwar coalition, yet the government exercised little of the same postwar budget restraint.²⁹ In fact, budget spending in Japan in 1913 and 1938 was higher than in the United States, the United Kingdom, or France.³⁰ Even after the war, there was no clear dominant preference within the conservative coalition for limiting spending, and the government ran large budget deficits. It was the occupying U.S. military authorities who imposed budget restraint. Within the conservative coalition there was deep resistance to it, and, after the Americans departed, opponents, including conservative ones, denounced it as a legacy of political subjugation and demanded that the policy be abandoned. Thus, rather than assuming a preference for budget restraint, it is necessary to explain its emergence and how the government could make it palatable to the broader conservative coalition. As Curtis has noted, this postwar coalition was fragile.³¹ Budget restraint only made it more difficult to hold together the conservative coalition.³² Even between the ministries that oversaw economic policy, particularly the Ministry of International Trade and Industry (MITI) and the Ministry of Finance (MOF), there were deep tensions over priorities, with MITI emphasizing the need for rapid industrialization over budget restraint.

Third, with regard to the partisan politics argument, the record of the LDP suggests that it was a party that successfully oversaw both extreme budget restraint and rapid increases in spending and government debt after 1970. While the partisan politics argument might seem to apply to the period prior to 1970, it does little to explain the subsequent increase in spending, upgrading of welfare policies, and rising government debt afterward. As Calder, Woodall, and others have shown, in the face of growing electoral pressure the LDP often responded by using compensatory spending to solidify support and broaden its appeal.³³ During the very brief period when the LDP was voted out of power from