Book
Share
Chapter

11. Why is Japanese Banking Sector Profitability so Low?

Author(s):
Alessandro Zanello, and Daniel Citrin
Published Date:
November 2008
Share
  • ShareShare
Show Summary Details
Author(s)
Alexander Wolfson

Introduction

Significant progress has been made in stabilizing the Japanese banking system in recent years. Efforts to address the bad debt problem in the banking system, including public capital injections and heightened supervision by the Financial Services Agency (FSA), have borne fruit. Major banks have lowered their nonperforming loan (NPL) ratios dramatically and key indicators such as capitalization and bank ratings have also improved, although progress has lagged among the regional banks (Table 11.1 and Figure 11.1. Fears of a financial meltdown have eased, and the near-term stability of the banking system is not in doubt.

Table 11.1Key Indicators of Japanese Banks
Major banksRegional banks
FY2001FY2006FY2001FY2006
NPLs (¥trillion)26.84.114.87.8
NPLs (ratio to total loans)8.41.58.04.0
Capital ratio10.813.19.210.4
Sources: FSA; IMF staff estimates.
Sources: FSA; IMF staff estimates.

Figure 11.1Net Operating Income

(in % of assets)

Source: OECD Bank Profitability Statistics.

Despite these improvements, bank profitability remains low by advanced country standards. Japanese banks have much lower profit margins than their counterparts in other G7 countries, and this situation has persisted over the past two decades. As a result of this low profitability, Japanese banks are more vulnerable to shocks and less able to absorb losses, which constrains their ability to perform effective financial intermediation and leaves them at risk of future capital shortfalls (retained earnings are a key potential source of bank capital). Going forward, raising core profitability is key to developing a healthy banking system that can support higher economic growth.

This chapter explores the factors behind the low profitability of the banking system. The next section will review developments in bank profitability among different sectors of the Japanese banking system, with a careful consideration of underlying core profitability. A number of structural factors that may affect banking profitability, including the role of government financial institutions and Japan Post are then explored. The chapter concludes with a consideration of the main priorities for revitalizing the banking system.

Developments in bank profitability

The Japanese banking system is characterized by a small number of large, internationally active major banks and a larger number of regional banks. As of mid-2005, there were eleven major banks, down from 21 in the mid-1990s.1 The major banks concentrate mostly in commercial and corporate lending. The 113 regional banks (in two tiers) emphasize local retail business as well as small and medium-size enterprise (SME) lending, and also participate in corporate loans. In addition to these private banks, government financial institutions play a dominant role in the financial system, especially Japan Post which is the country’s largest deposit-taker by a wide margin.

Overall profitability has improved in recent years, as credit costs stemming from the “bubble” period have declined (Figure 11.2). The collapse in asset prices following the 1980s boom and subsequent economic slowdown left banks saddled with bad debts. Accordingly, over the past decade, Japanese banks have recorded over ¥60 trillion in loan losses (due to both write-offs and higher provisions), offset to some extent by gains on their holdings of bonds as interest rates declined. After large loan losses through fiscal year (FY) 2001—driven in part by heightened scrutiny by the FSA—credit costs have declined steadily and the banking system has begun to turn a profit. Most observers believe that this trend is set to continue, and the “stock problem” of bad debts has to a large extent been resolved.

Figure 11.2Japanese Bank Profits

(in % of total assets)

Sources: Japanese Bankers Association; IMF staff estimates.

However, underlying core profitability has remained weak, as it has been for decades. Excluding credit costs and capital gains or losses from holdings of securities and other assets, core profits in the banking system have averaged under ¥2 trillion per year since FY1994, or under 0.25 percent of total assets (Figure 11.3).2 The ratio has improved in recent years to about 0.5 percent of assets and 0.4 percent for the major banks alone; core profitability tends to be slightly higher among the regional banks, reflecting the higher interest margin achieved on SME lending. Even after this recent improvement, profitability remains low by international standards, with Japanese banks’ net interest margins, noninterest revenues, and return on equity much lower than among their U.S. or U.K. counterparts.

Figure 11.3Breakdown of Core Profits

(in % of total assets)

Sources: Japanese Bankers Association; IMF staff estimates.

Factors limiting bank profitability

Background

Bank profitability is affected by a wide range of factors, and it is difficult to point to a single cause of low profitability. Broadly speaking, as in other industries, bank profitability is driven by managerial decisions (particularly the deviation from best practices, which evolve over time) and external factors (Berger and Mester, 1999). The inefficient use of resources and limited success in pursuing profit-maximizing strategies could be a reflection of poor management, or an inability or unwillingness for shareholders to hold management to account. External factors can also affect bank profitability. Such factors include both cyclical economic developments which affect overall business conditions, as well as the institutional and regulatory environment facing banks. With such a wide variety of factors at play, it is unlikely that low bank profitability in Japan can be explained by a single cause.

A number of recent studies have examined the poor performance of Japanese banks. Much work has focused on the limitations of the relationship-banking system, which has discouraged banks from recognizing losses to large borrowers (Peek and Rosengren, 2003). Excessive reliance on real estate collateral, rather than forward-looking cash flow assessments of borrowers, has exacerbated these shortcomings. More generally, a number of researchers have focused on weaknesses in credit allocation, based on evidence that banks maintain loans to companies that will be unlikely to repay (Sekine and others, 2003). This work has been complemented by recent efforts to discern the underlying strength of Japanese bank balance sheets, in order to gain a more accurate understanding of their financial condition. Recent papers by Fukao (Fukao, 2004; Fukao and others, 2004) have reclassified Japanese bank accounts and determined that their underlying health and profit potential is much weaker than headline numbers would suggest. After accounting for one-off factors such as valuation gains and loan loss charges, Fukao concludes that Japanese banks will be unable to generate sufficient profits to repay past capital injections (the analysis ends in FY2002, before the recent improvement in bank profits). The calculation of core profits in this chapter is inspired by Fukao’s methodology.

Cyclical factors

Poor bank profitability could be a reflection of Japan’s weak economic performance over the past decade. Economic stagnation would tend to reduce the ability of borrowers to service their debts, with ongoing deflation exacerbating the situation. A vicious cycle could ensue, with weakened banks unable to support healthy lending activities, thereby contributing to a further slowdown. By contrast, a recovering economy would improve credit quality and strengthen the banks, potentially leading to a virtuous cycle. On this view, movements in bank profitability would mostly reflect broader economic developments.

However, core profits do not appear to be very sensitive to cyclical economic developments. As documented above, core profitability has been low over the past two decades, even as the economy experienced episodes of booming growth and sharp contractions. Headline operating profits reported by Japanese banks (which includes credit costs and losses or gains on security holdings) generally fluctuate with indicators of the economic cycle in the expected manner, reflecting the strong association between the level of loan losses and indicators of economic health. However, there is no strong relationship between core profitability and indicators of the economic cycle, and the relationship with the level of loan losses is much weaker.

Econometric analysis generally supports the view that low banking profitability is not well explained by cyclical factors (Tables 11.2 and 11.3). In a regression framework including both major and regional banks, core profits are affected by the health of the corporate sector (represented by the debt-to-sales ratio) and the level of loan losses, but are not directly related to cyclical indicators. Core profits are not significantly affected by the level of interest rates, and the relationship between core profits and capacity utilization, while significant, is of the incorrect sign.3 Interestingly, in some specifications there is a negative relationship between core profits and Japan Post’s share of total bank deposits although the level of significance is low (see below). The results suggest that weak core profitability is not just a reflection of the weak economic environment or high levels of loan losses, and that an improvement in the economic outlook in itself will be insufficient to improve bank profitability. Rather, fundamental structural factors appear to be at play (Figure 11.4).

Table 11.2Regression Results: Core Profitability, FY1979-2003a
Constant3.333.411.71
(3.54)(3.58)(3.20)
Time trend0.0220.0210.023
(1.89)(1.77)(1.85)
Major bank dummy-0.249-0.263-0.249
(-6.04)(-6.49)(-5.84)
Debt/sales ratio-0.630-0.622-0.600
(-4.83)(-4.71)(-4.46)
10-year JGB rate0.0110.0200.010
(0.35)(0.63)(0.30)
Japan Post deposit share-0.011-0.0170.001
(-0.92)(-1.66)(0.13)
Capacity utilization-0.011-0.011
(-2.06)(-1.99)
Loan losses-0.174-0.167
(-1.46)-0.167
Memorandum items
Number of observations505050
R-squared0.6570.6400.623

t-statistics in parentheses.

Source: IMF staff estimates.

t-statistics in parentheses.

Source: IMF staff estimates.
Table 11.3Correlation Between Profits and Economic Indicators, FY1979-2003
Core ProfitsOverall Profitsa
10-year JGB rate0.280.70
Unemployment rate-0.02-0.68
Capacity utilization0.120.65
Nikkei index-0.370.15
Loan losses-0.34-0.91

Includes loan losses and gains on securities holdings.

Source: IMF staff estimates.

Includes loan losses and gains on securities holdings.

Source: IMF staff estimates.

Figure 11.4Bank Operating Costs

(in % of total assets)

Source: OECD Bank Profitability Statistics.

Cost and income structure

Broadly speaking, Japanese banking profitability has been driven by weak revenues, rather than high costs. Banks’ operating costs average just 1 percent of total assets (of which about half are staff costs), much lower than in other countries. This reflects the major banks’ relatively small branch networks, as well as a lower and flatter salary scale, with more limited use of performance bonuses. But it also reflects that Japanese banks have largely avoided high-margin products which also involve higher costs. While there may be some scope to achieve additional cost savings (e.g. through the consolidation of banks or branches), the key to raising bank profitability will be to raise revenues.

Japanese banks earn narrow interest margins and limited noninterest income (Figure 11.5). Thin net interest income (about 1–1¼ percent of assets) is a reflection of the low returns associated with core banking business in Japan. The low interest margin is not merely a reflection of the current low interest rate environment, as it has been low for the past decade—even during periods when lending rates were more than double the sub-2 percent level observed recently. Noninterest income has also been very low, and even negative on a net basis in some years. In contrast with profitability. While such practices could persist, the improved health of the corporate sector in recent years, together with heightened oversight by the FSA, is likely to have reduced the extent of zombie lending. Indeed, estimates from private analysts suggest that new loans to distressed firms have fallen by over 80 percent since FY2001.4

Figure 11.5Lending and Deposit Rates

Sources: CEIC database; IMF staff estimates.

Bank governance

The persistence of low bank profitability suggests that bank management has not been driven to raise profitability. Japanese banks have only gradually moved into more profitable lines of business, and have been slow to cease lending to distressed companies. This suggests that bank managers have not felt shareholder pressure to recognize losses, clean up balance sheets, and raise profits (the pressure to take action has been largely external, from the FSA). Observers have pointed to the prevalence of cross-shareholdings (whereby banks hold shares in some of their main borrowers, who in turn hold bank shares) and the lack of independent boards of directors as key elements of bank governance in Japan (e.g. Hoshi and Kashyap, 2001). While it is difficult to assess the quality of bank governance directly, governance should improve as cross-shareholdings unwind and bank boards become more independent.5

The role of government financial institutions

Competition from government financial institutions (GFIs) could affect the profitability of private banks. Although the government aims to consolidate the GFIs, they continue to hold a central place in the Japanese financial system. Japan Post, the biggest GFI, is the country’s largest deposit-taker, with postal savings accounting for almost 25 percent of total deposits, and a network of almost 25,000 branches, compared with under 13,000 for major and regional private banks (Figure 11.6). Postal insurance accounts for a large share of the market. The Government Housing Loan Corporation (GHLC) holds about 30 percent of all home mortgages, and other GFIs are active in corporate and SME lending.

Figure 11.6Bank Deposits, End 2007

(in ¥ trillion)

Source: CEIC.

Special advantages have allowed the GFIs to offer products on more favorable terms than private banks. While facing certain restrictions and obligations, Japan Post has been exempt from many of the costs faced by the private banking system, including deposit insurance premia, income taxes, and certain other taxes. In addition, Japan Post has faced a separate regulatory environment, with operations overseen by the Ministry of Internal Affairs and Communications, and the balance sheet reviewed by the FSA. These advantages have given Japan Post access to a large pool of low-cost deposits, including through ten-year teigaku deposits which can be redeemed early at no penalty. Japan Post does not appear to cause large distortions in the lending market, as its assets mostly consist of Japanese government bonds and bonds issued to finance public works projects (which might not be privately financed otherwise). However, the GHLC’s dominant position in the home mortgage market may have limited the scope for private banks.6 Competition from GFIs in mobilizing deposits and lending activities could put pressure on interest margins and profits at private banks. Indeed, as noted above, there does appear to be a negative relationship between Japan Post’s share of deposits and core profits in the banking system (Figure 11.7).

Figure 11.7Core Profits and Japan Post’s Deposit Share

Sources: Bank of Japan; Japanese Bankers Association; IMF staff estimates.

On balance, the impact of the privatization of Japan Post on the private banking sector is likely to be positive. Steps to level the playing field will likely encourage an inflow of deposits to the private banking system, especially if private banks are given access to Japan Post’s branch network. This could reduce funding costs and provide a boost to net interest margins. There could also be gains as this additional funding base is redeployed away from the sorts of activities currently funded by postal savings deposits. The reallocation of resources towards more productive uses over time should increase overall economic efficiency, although its impact on bank profitability is uncertain, as loan pricing and standards could come under pressure as banks try to make use of the new deposit inflows.

Conclusion

With banking system stability now achieved, policies are increasingly focused on raising bank profitability. While no single factor accounts for the longstanding low banking profitability in Japan, weaknesses appear to relate more to low revenues, rather than high costs. Raising revenues is likely to require a change in focus of the banking system, with a move away from reliance on traditional low-yielding corporate finance toward new higher-revenue products. Policies increasingly need to focus on promoting a more profitable environment for private banks.

The FSA’s Program for Further Financial Reform (PFFR), announced in December 2004, could provide a useful framework for these reforms (the details of the plan are still being formulated). Key priorities would include the following:

  • Ensuring that past balance sheet problems do not re-emerge. This will involve continued regulatory scrutiny to ensure that loans are properly classified. The FSA’s plans in the PFFR to assess banks’ risk-management systems will also help ensure that loans are properly priced.

  • Broadening the range of products and services which banks can provide. Banks have recently been allowed to offer mutual funds and insurance products to their customers, and the PFFR will extend further the scope of products offered through bank windows. Investor protection will be enhanced through a comprehensive Investment Services Law.

  • Taking steps to enhance bank governance. Continuing to unwind cross-shareholdings and increasing the role of independent directors are important steps in this regard.

  • Leveling the playing field with GFIs. The planned privatization of Japan Post will play a central role, especially as exemptions from fees and taxes are phased out.

References

    Bank of Japan2003Overview of Japanese Banks: Observations from Financial Statements for Fiscal 2003.

    BergerAllen N. and Loretta J.Mester1999What Explains the Dramatic Changes in Cost and Profit Performance of the U.S. Banking Industry?Finance and Economics Discussion Series 1999–13Federal Reserve Board.

    CaballeroRicardo J.TakeoHoshi and Anil K.Kashyap2004Zombie Lending and Depressed Restructuring in Japanmimeo.

    FukaoMitsuhiro2004The Profitability of Japanese Industries: Non-financial Sectors, Banks and Life-Insurance CompaniesJapan Financial Review No. 5Japan Center for Economic Research.

    FukaoMitsuhiro and others2004Banks and the Safety NetJapan Financial Report No. 11Japan Center for Economic Research.

    HoshiTakeo and Anil K.Kashyap2001Corporate Financing and Governance in Japan: The Road to the Future (Cambridge, Massachusetts: MIT Press).

    HoshiTakeo and Anil K.Kashyap2004Solutions to Japan’s Banking Problems: What Might Work and What Definitely Will Failmimeo.

    JerramRichard2005Zombie Lending Update: Getting No CreditMacquarie Research EconomicsJune2005.

    PeekJoe and Eric S.Rosengren2003Unnatural Selection: Perverse Incentives and the Misallocation of Credit in JapanNBER Working Paper 9643.

    SchaedeUlrike2004The ‘Middle Risk Gap’ and Financial System Reform: Small Firm Financing in JapanInstitute for Monetary and Economic Studies Discussion Paper Series Discussion Paper No. 2004-E-11Bank of Japan.

    SekineToshitakaKeiichiroKobayashi and YumiSaita2003Forbearance Lending: The Case of Japanese FirmsMonetary and Economic StudiesBank of Japan.

    SmithDavid C.2003Loans to Japanese BorrowersJournal of the Japanese and International Economies Vol. 17 pp. 283304.

The number of major banks declined to ten with the merger of UFJ Holdings and Mitsubishi Tokyo Financial Group.

Core profitability data are calculated from detailed balance sheet and profit and loss data prepared by the Japanese Bankers Association, and cover the period from FY1979 (ending March 1980) through FY2003.

Specifications with alternative cyclical indicators such as the unemployment rate or the stock market index also show no evidence of a positive relationship between the cyclical strength of the economy and the level of core profits.

According to one calculation, new loans to distressed firms are at the lowest point in over a decade (Jerram, 2005).

The public bailout of one major bank involved the adoption of an independent board of directors.

Going forward, direct competition from the GHLC should be reduced as its activities are rationalized under a reform program.

    Other Resources Citing This Publication