This Selected Issues paper analyzes the key features of the Japanese business cycle, and investigates whether the current recovery differs from past recoveries. In particular, this paper poses the following questions: what are the main characteristics of Japanese business cycles since 1980, and what happens to output, expenditure components, and prices over the cycle? The paper reviews the recent performance and policy issues with respect to the banking sector, insurance sector, and public financial sector in Japan. The stability of the financial sector is also assessed.


This Selected Issues paper analyzes the key features of the Japanese business cycle, and investigates whether the current recovery differs from past recoveries. In particular, this paper poses the following questions: what are the main characteristics of Japanese business cycles since 1980, and what happens to output, expenditure components, and prices over the cycle? The paper reviews the recent performance and policy issues with respect to the banking sector, insurance sector, and public financial sector in Japan. The stability of the financial sector is also assessed.

IV. Reform of Local Government Finances in Japan1

A. Introduction

1. The local governments’ fiscal position has deteriorated significantly over the last decade. Net local government debt increased from 5 percent of GDP in 1990 to 24 percent of GDP in 2002, while gross debt more than doubled to reach 36 percent of GDP. Moreover, transfers from the central to local governments have grown much faster than overall tax revenues, which has contributed to the rise in central government debt.


Local Government Debt

Citation: IMF Staff Country Reports 2004, 247; 10.5089/9781451820577.002.A004

Source: National Income Accounts, Cabinet Office.

2. The framework of inter-governmental fiscal relations in Japan is characterized by strong transfer dependence and limited local autonomy. These institutional features appear to have contributed to the weakening of government finances. Local governments have limited control over their revenues and a significant portion of their expenditure is legally mandated. Transfers from the central government constitute a large share of revenues for most local governments. Nonearmarked transfers (the local allocation tax or LAT) are effectively used to cover any gap between expenditure and other revenues, creating little incentive to improve expenditure efficiency at the local level.

3. This paper describes the nature and incentive effects of inter-governmental relations in Japan and discusses recent reform efforts. A reform of local finances initiated by the Koizumi government in 2003 seeks to limit the growth of expenditure at the local level and to encourage a more efficient allocation of resources.

B. Local Government Finances and Intergovernmental Fiscal Relations

4. Local governments account for a substantial share of public spending, although they have limited discretion in the allocation of expenditure. The local governments’ share of total public expenditure is about 63 percent (19 percent of GDP), while their share of tax revenue is about 43 percent (7 percent of GDP). The local share of total public expenditure is very high compared to other countries; however, the local governments’ spending autonomy is limited (see table). The main categories of local expenditure are public investment, education, and health and welfare. There are centrally determined minimum standards for most public services provided at the local level (including education, child care, and health care). The amount and type of public investment spending is also broadly determined at the central government level as part of the general direction of fiscal policy. Local public investment expanded in the first half of 1990s as part of government policy to stimulate the economy, but it has been contracting both in absolute terms and as a share of total expenditure since 1996.

Local Expenditure Composition, FY2001

article image
Source: White Paper on Local Public Finance, 2003, MPHPT.

Includes both interest and principal payments.

5. Local taxes account for less than half of total local revenue. The main sources of local revenues are local taxes, LAT (nontargeted transfers from the central government), and earmarked central government subsidies. The share of local taxes in total revenue has declined over the last decade, while the shares of LAT and bond financing have increased. Most of the important local tax sources, tax bases, and standard tax rates are set by a national law and local governments have somewhat limited flexibility to impose new taxes or change the tax rates.2

Local Revenue Composition, FY2001

article image
Source: White Paper on Local Public Finance, 2003, MPHPT.

Proceeds from bond issuance are registered as revenue in the cash-based Japanese fiscal accounts.

6. The gap between local taxes and expenditures is filled largely by various transfers from the central government. In addition to providing nontargeted transfers (LAT), the central government subsidizes a fixed share of certain expenditure items—for example, medical and nursing care insurance, education, and some public investment projects. The sum of LAT and subsidies exceeded ¥34 trillion (about 7 percent of GDP) in FY2001. Any remaining deficits are financed by local debt issuance.

Budget process

7. Total local expenditure and the amount of transfers are set through an annual negotiation process. In the first step, the overall local government expenditure plan for the coming fiscal year is discussed jointly by the Ministry of Public Management, Home Affairs, Posts and Telecommunications (MPHPT)3 and the Ministry of Finance. Once the expenditure side has been agreed upon, the amounts of local taxes, ear-marked subsidies, and construction bonds issuance are projected. “Construction” bonds can be issued by local governments to finance a certain share of local public investment expenditure. Finally, the amounts of LAT and deficit-financing bonds are determined. The resulting budget is referred to as “the local finance program.”

8. The main goal of the LAT is to equalize the capacity of local governments to provide a standard level of services, assuming that all governments raise local tax revenues using the same tax bases and the same rates. The LAT transfers compensate for the uneven distribution of per capita tax resources among local governments. The standard amount of LAT is equal to a certain percentage of five national taxes.4 However, in practice, the actual allocation of LAT often significantly exceeds the standard formula-based amount since LAT transfers are used as a de-facto financing item in the budget (together with special deficitfinancing bonds).5 Estimation of the deficit that needs to be financed by LAT allocations entails an assessment of both the revenue-raising capacity and standard expenditure needs for local governments.6 LAT has increased rapidly relative to overall tax revenues in recent years (from 14 percent of consolidated general government revenue in 1990 to 20 percent in 2002).

9. Local governments can issue bonds mainly to finance public works projects (“construction” bonds). In addition, the issuance of special deficit-financing bonds has been allowed since 1999. These bonds are used in addition to LAT to finance the gap between local expenditures and revenues as explained above. Bonds can also be issued occasionally for other special purposes. The amount of bond issuance is subject to approval by the central government (through the system of “local bond permits”). The upper limit on local bond issuance is determined by a complex formula that takes into account the repayment capacity of each government.

10. The central government has powers to enforce fiscal discipline at the local government level in certain cases. The main channel is by directly controlling the amount of local bond issuance. In addition, if the annual deficit of a local government exceeds a certain threshold, the respective government has to revise its budget under the oversight of the central government (under the so-called “financial reconstruction system”). The revisions include streamlining of expenditure and a review of the scope for increasing local revenues. In exchange, the central government provides funds to cover the deficit and guarantees continued debt repayment for the respective local government.

C. Policy Issues

11. There are wide differences in the design of intergovernmental relations across the world, reflecting both idiosyncratic historic developments and different preferences with respect to the provision of public services and redistribution. Federal systems, like that in the United States, are characterized by high level of financial independence at the state government level, lack of uniformity in the provision of public services, and a low level of central government transfers. Unitary systems, like those in Australia, Germany, and Japan, are characterized by preference for equal access to public services, greater centralization of decision making, and high transfer dependence.

12. However, well-functioning systems of intergovernmental fiscal relations all have in common an incentive structure that promotes fiscal discipline and efficient resource allocation.7 Subnational governments would behave in a fiscally responsible manner if they have to bear the cost of their actions. Typically, that would require a certain degree of autonomy in deciding expenditure priorities and the authority to own-finance locally provided services at the margin. To promote adequate voter and market discipline, local borrowing should not have any explicit or implicit central government guarantees. Transfers from the central government have to be allocated in an objective manner that does not discourage efficient resource allocation nor create soft-budget constraints at the local level. Finally, data on fiscal outcomes at the local level should be collected and made available on a timely basis to improve accountability and facilitate planning. These conditions are not fully satisfied in the case of Japan:

  • Expenditure priorities for a significant share of local spending are decided at the central government level and financing at the margin relies largely on transfers (the LAT), instead of local revenue;

  • Local debt is perceived as effectively guaranteed by the central government;

  • The allocation of public works subsidies is not based on explicit objective criteria;

  • Data on local government finances is released with a delay of about one year after the end of the fiscal year, which makes it difficult to monitor trends in expenditure and revenue on an ongoing basis.

13. The gap-filling function of the LAT creates little incentives for local governments in Japan to improve the efficiency of expenditure. With large share of expenditure mandated by the central government and financing ensured through the LAT, there are limited incentives to reduce the cost of different services or eliminate wasteful spending. Indeed, studies have shown that the costs of public services at the local level are generally higher than the costs of comparable services provided by the private sector.8

14. Local debt is effectively guaranteed by the central government and is not subject to market discipline. The main channel of limiting local debt issuance in Japan is through direct administrative oversight by the central government (the system of local bond permits). In other countries, market discipline has proven effective in controlling local government finances and has complemented any quantitative restrictions imposed by the central government. This channel of control is largely nonexistent in Japan. Since the amount of local bond issuance is subject to approval by the MPHPT, the bonds are perceived as effectively guaranteed by the central government.9 A large share of local debt is placed with regional banks and with the Fiscal Investment and Loan Program (FILP).10 The local bonds traded on the market have almost no premium over JGBs, even though they are much less liquid; and with few exceptions, local bond prices are the same across all prefectures.

15. The allocation of earmarked subsidies is only partially based on objective criteria. While subsidies for standard mandated services are largely determined by costsharing formulas, public works subsidies are allocated in a discretionary manner by the central government. This has the potential to create a classic “common pool” problem and encourage rent-seeking behavior. Some studies suggest that there are strong political influences in the distribution of public work funds.11 That may result in an inefficient allocation of resources and could account for the observed decline in returns to public investment in rural areas.12 Having more transparent rules for allocation of public works subsidies, based on cost-benefit analysis, could improve the efficiency of public expenditure.

16. High per capita public works subsidies to certain provinces do not appear to stimulate private sector activity and may not be justified from an allocative efficiency standpoint. Some provinces have benefited disproportionally from central government transfers and subsidies—per capita fiscal revenue after transfers is generally higher for the provinces with the lowest own per capita tax revenues. The correlation between per capita local tax revenue and total revenue (after transfers) was -0.26 in FY2001. This is rather unusual and could be a sign of inefficient allocation. In other countries, per capita fiscal revenue differentials among local governments narrow after transfers, but are typically not reversed. Temporarily higher capital transfers to less developed areas could be justified for development purposes. However, analysis suggests that higher government transfers to poorer areas have not stimulated self-sustaining growth; instead, the areas have become heavily dependent on public construction. Braun and Kubota (2000) find that although there has been significant convergence in capital per worker among provinces over the last 30 years (including public capital), there is no evidence of convergence either in per capita income or in GDP per worker, so the accumulation of public capital has not translated into higher income or higher labor productivity.


Prefectures’ Revenue per Capita, FY2001

Citation: IMF Staff Country Reports 2004, 247; 10.5089/9781451820577.002.A004

Source: Ministry of Finance.

17. A number of experts argue that a review of both current and capital expenditure could help reverse the deterioration in local government finances:13

  • The scale and scope of guaranteed public services could be reviewed with an eye to attuning them better to local needs and realizing expenditure savings where possible. The needs for certain services (child care and elderly care, for example) vary across regions and local governments could be given more discretion in setting expenditure priorities. Costs also differ significantly across localities and the provision of uniform levels of services may not be desirable from an efficiency standpoint. Privatization of select services could result in higher quality and lower costs for taxpayers;

  • A careful cost-benefit analysis of public work projects, based on conservative assumptions, could be used as the main criterion for allocating public investment funds. In recent years, the government has acknowledged that the level of public investment is too high and that some investment projects are inefficient (especially in agriculture and rural infrastructure) and has made efforts to reduce public investment in these areas and redirect funds to urban infrastructure.

18. Reduction in transfer dependence and removal of implicit central government guarantees on local debt could strengthen the accountability of local governments. The gap-filling role of the LAT should be reduced to align the incentives of local governments with those of the local taxpayers. Under the current system, the local electorate has little incentive to enforce fiscal discipline on its government, since the accumulation of deficits does not translate into higher tax rates (some local tax rates are fixed by law and a large share of local expenditure is subsidized).14 Reduction in the gap-filling role of the LAT would involve a review of the scope of expenditures covered by LAT in the local governments’ budgets and financing expenditure at the margin with local tax revenue. However, increasing the share of own revenue and allowing greater discretion in expenditure allocation would only work if a hard budget constraint is imposed through the removal of any implicit central government guarantees on local debt.

D. Recent Reform Initiatives

19. A reform of local government finances, initiated in 2003, aims to achieve fiscal consolidation and increase the efficiency of local government expenditure, by allowing local authorities greater discretion in allocating funds, while limiting their overall resources. The so-called “trinity” reform of local government finances entails reductions in subsidies for specific local projects and in nonearmarked transfers (LAT), partially compensated for by transfer of tax resources from the central to local governments. In FY2004, the reduction in subsidies would be ¥1 trillion; the LAT would decrease by ¥1.2 trillion; and the local tax resources would increase by ¥0.4 trillion. The overall decline in subsidies by 2006 would be ¥4 trillion (about 5 percent of total local revenues), compensated for by the transfer of about ¥3 trillion in additional tax resources to local governments. No target has been announced for the overall reduction in LAT.

20. The ongoing reforms of the pension system and government financial institutions could strengthen market discipline on local government financing. Since 2001, the requirement that the assets of the pension system and postal saving system should be automatically deposited with the FILP has been abolished and an increasing share of these assets have been invested in the market. The resulting reduction of FILP liabilities has been mirrored by a reduction in its assets, including holdings of local government bonds. Therefore, the share of local bonds placed on the market would gradually increase, which could strengthen market discipline on local government finances.

21. The system of central government permits for local bond issuance will be relaxed starting in 2006, which could reduce the perception that local debt is guaranteed by the central government. There will still be formal limits on how much bonds each local government can issue based on its repayment capability, but explicit permission from the MPHPT will not be required. The repayment of local bonds issued without central government approval will not be covered by the local finance program and therefore will not be financed by the LAT. To the extent that these changes reduce the perception that all local debt is implicitly guaranteed by the central government, they could be effective in imposing greater market discipline on local governments.15 However, if the perception of implicit guarantees remains in place and spreads over JGBs stay the same for all prefectures, debt could actually accumulate at a faster pace.

E. Conclusions

22. The system of local finances in Japan is characterized by a low degree of tax and expenditure autonomy, high transfer dependence, and implicit central government guarantees on local debt. A large share of local expenditure is either mandated or encouraged by the central government. In combination with the gap-filling function of the local allocation tax, this creates little incentive for local governments to increase the efficiency of public expenditure. Due to the low degree of subnational autonomy, market discipline and voter influence on local finances are relatively weak. The main mechanism for disciplining subnational finances is strong central government influence over the overall local budget and the amount of local debt issuance.

23. The trinity reform seeks to increase the autonomy of local governments and thus encourage a more efficient allocation of public resources, while limiting the growth of expenditure. The reduction in central government transfers and subsidies and the increase in independent local tax revenues would increase the flexibility of local governments to determine expenditure priorities. However, the scale of the reform is modest relative to the total amount of transfers and further efforts may be needed to achieve a steady improvement of subnational finances.

24. Further reforms recommended by a number of academics specializing in local government finances include the following:16

  • Review of expenditure commitments. The necessity and affordability of standard public services could be reviewed. Public investment could be downscaled further and its efficiency improved through greater use of cost-benefit analysis;

  • Reduction of the gap-filling role of the LAT. The scope of expenditures whose financing is ensured by the LAT could be reduced. That, combined with increased discretion in imposing taxes at the local level, would allow spending to be financed by local revenue at the margin;

  • Reforming the framework regulating subnational borrowing to reduce the perception of implicit central government guarantees and increase market discipline;

  • Improving the quality and timeliness of local government financial data and strengthening budget planning and implementation. Better and more timely statistics would improve transparency and accountability, facilitate fiscal planning, and allow informed credit risk assessment by lenders.


  • Braun A. and Kubota, K., 2000, “The Effect of Government Capital on Labor Productivity in Japan’s Prefectures,Working Paper presented at the University of Tokyo.

    • Search Google Scholar
    • Export Citation
  • Doi, T., 1998, “Panel Analysis of Public Capital in Japan,Kokumin Keizai 161, p.2752.

  • Doi, T., 2002, “The System and Role of Local Bonds Permits in Japan,” in Government Deficit and Fiscal Reform in Japan, edited by Ihori and Sato.

    • Search Google Scholar
    • Export Citation
  • Fiscal System Council, 2004, Heisei 17 Nendo Yosan Hensei No Kihonteki Kangaekata Ni Tsuite (Proposal for FY2005 Budget Formulation).

  • Ihori, T. and Sato, M., 2002, Government Deficit and Fiscal Reform in Japan.

  • Kobayashi, Y., 1997, Gendai Nihon no Seiji Katei (The Political Process in Modern Japan), Tokyo: Tokyo University Press.

  • Kondo, H., 2002, “Public Works,” in Government Deficit and Fiscal Reform in Japan, edited by Ihori and Sato.

  • Konishi, S., 2002, Chiho Zaisei Kaikaku Ron (Reform of Local Government Financing), Nihon Keizai Shinbun Sha, Tokyo.

  • MPHPT, 2003, White Paper on Local Public Finance.

  • Society of Local Government Management (Chiho Jichi Keiei Gakkai), 2000, Koritsu to Minkan tono Cosuto to Sabisu Hikaku (Comparison of the Cost of Services Provided by the Public Sector and the Private Sector), Tokyo.

    • Search Google Scholar
    • Export Citation
  • Ter-Minassian, T. (editor), 1997, Fiscal Federalism in Theory and Practice, IMF, Washington D.C.

  • Wada, J., 1985, “Seiji Katei no Keizaiteki Bunseki (An Economic Analysis of the Political Process),Hermes 36: 75115.

  • Yoshida, Y. K., 1998, Chihou Bunken no Tame no Chihou Zaisei Kaikaku (Reform of Local Government Finances Towards Decentralization), Yuhikaku, Tokyo

    • Search Google Scholar
    • Export Citation
  • Yoshino, N. and Nakano, H., 1996, “Interregional Distribution and Productivity Effect of Public Investment,Financial Review 41, p.1626.

    • Search Google Scholar
    • Export Citation

Prepared by Dora Iakova (ext. 35365) and Takuo Komori (ext. 37613).


Local governments have discretion to vary the tax rates on their main tax sources within certain limits. For example, prefectures are allowed to raise the tax rate for corporate income tax up to 1.1 times the standard rate and for property tax up to 1.5 times the standard tax rate. There is no upper limit for the personal income tax rate. Local governments are also allowed to impose some minor taxes, subject to approval by the central government.


This Ministry oversees the operations of local governments.


Based on the formula, LAT should be equal to the sum of 32 percent of the individual income and liquor taxes, 35.8 percent of the corporate income tax, 29.5 percent of the consumption tax, and 25 percent of the tobacco tax.


Since 2001, half of the fiscal deficit that remains after the standard LAT amount has been allocated is financed by additional LAT allocation. The other half is financed through local deficit bond issues. Prior to 2001, the gap was filled entirely by additional LAT allocations and special account borrowing. In FY2004, for example, the standard LAT amount based on the formula was ¥11.2 trillion, while the actual allocation was ¥16.9 trillion.


The standard expenditure need is estimated separately for each expenditure item, taking into account its average cost and the cost differentials among different regions (based on factors such as population density, area and geography, and industrial diversification). The local governments’ revenue raising capacity is estimated as the local tax revenue that would be obtained under the standard tax rates determined by the central government.


There is a complex formula that determines the shares that need to be repaid by the central and local governments respectively, but since the LAT covers interest and principal payment for most local debt, full repayment is practically guaranteed by the central government.


The FILP is a government program that provides financing for public policy purposes. Doi (2002) argues that the central government implicitly supports financially weaker local governments with limited market access by allocating a larger share of FILP financing to them (according to his analysis, FILP borrowing is at interest rates slightly lower than that of privately placed funds for the same maturities).


A number of studies have shown that returns on public investment have declined over time (see Doi (1998) and Yoshino and Nakano (1996)). Kondo (2002) finds that rates of return differ significantly for different types of public capital and are relatively low in road construction and agriculture. EPA(1997) estimates that the return to social capital in cities is roughly twice the rate of return in rural areas.


See, for example, Konishi (2002) and Yoshida (1998).


Even the limited freedom to set higher local tax rates has been rarely used. In part, this has reflected policy preferences as fiscal policy has been stimulative for most of the last decade. However, the incentives to reduce local deficits through raising additional revenue are relatively weak, since debt repayment is effectively guaranteed by the central government.


Some credit rating agencies have already started rating local government bonds.


See Konishi (2002) and Yoshida (1998) and references therein. See also the “Proposal for FY2005 Budget Formulation” by the Fiscal System Council of the government (May 2004).

Japan: Selected Issues
Author: International Monetary Fund