IV. Revisiting Japan’s Lost Decade

International Monetary Fund. Asia and Pacific Dept
Published Date:
May 2009
  • ShareShare
Show Summary Details

This chapter discusses Japan’s experiences with its banking crisis in the 1990s and the potential implications for resolving the current global crisis. Drawing on insights from a recent IMF seminar (Box 4.1), the chapter revisits the Japanese fiscal, monetary, and financial sector policy responses through the lens of the present turmoil. In doing so, it draws possible lessons for policymakers today, including in Asia, as they attempt to stabilize their economies and orchestrate a recovery.

Box 4.1.IMF Seminar on Japan-U.S. Parallels: Summary of Proceedings1

Through most of the 1990s and early 2000s, Japan grappled with a financial crisis in many ways reminiscent of the turmoil affecting the United States today. Both crises originated in the bursting of asset bubbles fueled by excess liquidity, lax financial regulation and irrational exuberance. The asset collapse spread to other financial markets, raising liquidity and solvency concerns for systemically important institutions and weakening growth. Addressing these concerns required unprecedented intervention to stabilize financial markets, while cushioning adverse feedbacks through supportive macroeconomic policies. Motivated by these parallels, the IMF organized a seminar in March 2009 to discuss Japan’s experience and the potential implications for resolving the financial difficulties facing the United States today.

In his opening remarks, Anoop Singh (IMF) discussed some key similarities between the two crises—in terms of their origins, evolution, and policy challenges—while noting that the present crisis was moving much faster, was global in scale, and involved more complex assets. He stressed that while a strong policy response had so far allowed the United States to fast-forward its way through the first half of Japan’s lost decade, some of the more difficult challenges still lie ahead. In particular, Japanese history suggests that a post-bubble recession is much harder to combat than a cyclical downturn and that the detrimental effects on growth can be long lived. In this context, a speedy and strong U.S. recovery was likely to hinge on continued success in financial and macroeconomic policies, as well as longer-term reforms to raise productivity.

Credit Easing: The Bank of Japan’s Approach

Participants acknowledged that the precise transmission channels of unconventional monetary policies are considerably uncertain, but warned against falling behind the curve:

  • Although policy rates have approached the lower bound in many advanced economies, Hiromi Yamaoka (IMF Office of the Executive Directors) pointed out that a number of unconventional policy tools can still be used. The Bank of Japan’s experience suggests that these include influencing expectations by committing to keeping policy rates low for an extended period, expanding the size of the central bank balance sheet, or changing its composition by purchasing financial assets with longer maturities or credit risks. However, the extent to which credit easing worked in Japan remains highly uncertain and it was not a substitute for policies to fix the financial system.

  • Vincent Reinhart (American Enterprise Institute) suggested that there were four channels through which credit easing could work: (1) supporting the prices of certain assets; (2) complementing fiscal policy; (3) encouraging the expansion of bank balance sheets through reserve creation; and (4) influencing expectations. There could also be an exchange rate channel, although there was little evidence about the effects on currencies, especially when credit easing was being attempted globally. Unconventional policies are also much harder to explain to the public, are open to political interference, and can be difficult to unwind. However, these limitations were not an excuse for inaction, and he urged the U.S. Federal Reserve to continue taking a range of aggressive measures.

Japan’s Fiscal Stimulus in the 1990s: Did It Work?

Participants were split on the impact of Japan’s fiscal stimulus, but generally agreed that some degree of activism was needed to support the economy while the health of the financial system was restored:

  • Takatoshi Ito (University of Tokyo) discussed the empirical difficulties of assessing the effectiveness of fiscal policy in Japan, including quantifying the fiscal impulse and the unobservability of the counterfactual. Some empirical work suggests that fiscal policy helped to prevent a complete meltdown of the economy but was constrained by packages containing relatively small “real water.” However, others point out that fiscal policy—hampered by low multipliers and Ricardian effects—could not prevent the economy from sliding into stagnation. Despite these controversies, he noted that the scale of the present crisis calls for a response in targeted areas to increase productivity and long-term growth—such as the environment, medical care, and education in Japan—with attention to debt sustainability.

  • Adam Posen (Peterson Institute) argued in favor of fiscal activism, particularly given that conventional monetary policy was close to its limits. However, if private demand was not supported by resolving the problems of the financial system and a restructuring of the economy toward productive sectors, the effects of fiscal stimulus were likely to be short-lived. In addition, he noted that most of the current debates on the effectiveness of fiscal policy—specifically on tax cuts versus public works spending, temporary versus permanent tax cuts, and Ricardian effects—were overly simplistic. For a productive dialogue, he felt that there was a need to move beyond clichéd labels and focus on the nature of the policies themselves, as in recent work by IMF staff.2

Resolving Japan’s Banking Crisis: Strategies Adopted and Fiscal Cost

Participants agreed that the United States may be repeating some of Japan’s early mistakes, arguing for forceful actions to encourage loss recognition, restructure distressed assets, and recapitalize viable institutions:

  • Takeo Hoshi (University of California, San Diego) argued that by providing money to banks without assessing their financial conditions and capital needs, the United States was guilty of some familiar mistakes. Japan had adopted many of the same strategies that the United States is considering now, but it eventually needed to force banks to clean up their balance sheets and dispose of bad assets.

  • Jonathan Fiechter (IMF) stressed that, like the United States today, Japan’s policies initially involved ad hoc responses to capital injections before a more rigorous assessment of banks was finally implemented. It would therefore be instructive to understand how Japan overcame public resistance to bank bailouts and the stigma that banks attached to accepting public capital. To the extent that the Japanese approach resulted in a banking sector that has been more resilient to the current crisis, strategies adopted could also provide useful insights for reforming the global financial system.

Concluding Roundtable Discussion: Parallels with the United States

Panelists agreed that the U.S. response has generally been swift but, with the current crisis more daunting in some ways, difficult steps could still lie ahead:

Olivier Blanchard (IMF) stressed that as long as measures were chosen carefully, fiscal multipliers are likely to exceed unity and likely to increase output as the U.S. output gap is large. Hence, the only relevant constraints on fiscal policy are the impact on interest rates and market perceptions of debt levels. In addition, understanding why deflation in Japan did not become more severe despite the large output gap has important implications for whether the United States will also manage to avoid a deflationary spiral. Professor Ito welcomed the close cooperation between the Federal Reserve and the Treasury, but warned that this should not be at the expense of violating the independence of the central bank. Krishna Guha (Financial Times) hinted at the possible trade-off between fiscal stimulus and bank recapitalization, given political resistance to using public funds. The limited funds available implied a further case for setting a high benchmark in assessing the potential effectiveness of fiscal policy measures. Adam Posen pointed out that the much weaker global environment presents an added challenge during the current crisis, by limiting the scope for an export-led recovery. Daisuke Kotegawa (IMF Office of the Executive Directors) suggested that by influencing public perceptions, the mass media could play a useful role in generating support for needed bank recapitalization.

1 IMF seminar on “Japan’s Policy Response to its Financial Crisis: Parallels with the U.S. Today” (March 19, 2009). See for supporting materials.2 See Spilimbergo and others (2008).

Japan’s crisis was successfully resolved and most of the public funds deployed were recovered, but not before a “lost decade” of stagnation and prolonged deflation. While today’s policy challenges are compounded by the complexity of the distressed assets involved and the global scope of the crisis, they are in many ways similar to the problems Japan had to confront. Therefore, Japan’s eventual success—and early difficulties—in overcoming its crisis could provide useful insights.

The results are likely to be of interest to a range of economies. They apply most directly to the United States and other advanced countries whose financial sectors are at the epicenter of the crisis. However, given unprecedented real and financial spillovers, various aspects of the policies discussed are becoming increasingly relevant more broadly, including in Asia. Fiscal policy has assumed center stage in the region, with most economies planning significant stimulus packages to combat large output gaps. On the other hand, with deflationary pressures emerging and credit markets impaired, traditional monetary policy may be reaching its limits and central banks may need to resort to unfamiliar credit easing measures. And although financial and corporate sectors in Asia have been generally resilient, contingency plans to deal with heightened distress may be called for given the significant downside risks that remain. In all these areas, Japan’s experiences provide a rich array of lessons that could guide policy responses.

Accordingly, the chapter asks:

  • Based on Japan’s experience, can fiscal policy be used to stimulate the economy and, if so, what measures are likely to work? Looking ahead, how concerned should policymakers be about risks to medium-term fiscal sustainability?

  • How successful was Japan in easing credit conditions and fighting deflation using unconventional monetary policies?

  • In the financial sector, what strategies did Japan use to clean up bank and debtor balance sheets and to restore lending? How were the fiscal costs of these interventions managed and what were the key lessons in designing an exit strategy?

Background: Stylized Facts from Japan’s Lost Decade

Following unprecedented run-ups, Japan’s stock and real estate markets collapsed in the early 1990s. After an extraordinary bull market that saw share prices rise almost threefold in only four years, the stock market plunged in 1990. Notwithstanding some intermittent upswings, the gains during the bubble were given up entirely over the next 12 years, with most of the decline occurring in the first two and a half years after the crash. In 1991, the property market also started to falter: after tripling between 1985 and 1990, prices gradually slid back to their initial level by the early 2000s (Figure 4.1).

Figure 4.1.Japan’s Twin Bubbles: Stock Market and Real Estate


Source: Haver Analytics.

The bursting of these twin bubbles interrupted Japan’s long post-war expansion, but the immediate effects were not devastating. After growing by about 4 percent in the early 1990s, the economy stagnated until the middle of the decade, with average growth of about 1 percent (Figure 4.2). Unemployment ticked up and inflation fell gradually from highs of about 3½ percent, although credit growth remained relatively resilient and official nonperforming loans (NPLs) were low. The economy was generally expected to emerge relatively quickly from what was seen as a typical cyclical downturn, obviating the need for aggressive policy action. Indeed, a recovery appeared to be taking hold between 1994 and early 1997, as growth and inflation picked up and the stock market staged a rally (Figure 4.3).

Figure 4.2.Japan: Growth and Unemployment

(In percent)

Source: IMF, WEO database.

Figure 4.3.Japan: Inflation and Output Gap

(Year-on-year percent change)

Source: IMF, WEO database.

The full scale of Japan’s problems became evident when a systemic banking crisis erupted in 1997. The bursting of the asset bubbles left Japan’s financial system saddled with large problem loans and rising risks from a weak economy. Financial vulnerabilities arose from the absence of a sustained recovery, continued high corporate leverage, and significant market and credit risk that placed mounting pressure on bank capital. However, it was a full six years after the property market bust before mounting losses on failed real estate loans and falling share prices led to the interbank market freezing up and a wave of failures in the financial sector, featuring some of the country’s largest banks. The situation was compounded by the deterioration in the external environment induced by the Asian crisis. The economy contracted for two consecutive years, the first time growth had fallen into negative territory since the oil shock of the early 1970s.

Subsequently, the economy seemed to be on the mend between 1999 and 2000, helped by the global IT boom. However, following the collapse of the IT bubble in 2001, the situation took another turn for the worse as deteriorating corporate profits damaged the still fragile banking system and resulted in a renewed phase of financial stress. The economy barely grew in 2001 and 2002. A large output gap opened up again and deflation worsened significantly, as credit contracted in the face of long-delayed but much-needed deleveraging in the financial and corporate sectors. Table 4.1 provides a chronology of the crisis.

Table 4.1.Key Events in Japan’s Banking Crisis
1989Stock market peaked
Crisis Phase I: Slowdown
1990Land prices peaked
1994Hyogo Bank failed
1996Series of housing loan companies failed
Crisis Phase II: Escalation and fledgling recovery
1997Yamaichi Securities Co. Ltd. and Hokkaido Takushoku Bank failed
1998First injection of public funds into banks Long-Term Credit Bank of Japan and Nippon Credit Bank nationalized
1999Second injection of public funds into banks
Bank of Japan implements zero-interest rate policy
Resolution and Collection Corporation (RCC) starts purchase of NPLs from healthy financial institutions
Series of mergers among major banks
2000Bank of Japan lifts zero interest rate policy
Crisis Phase III: Renewed systemic stress
2001Bank of Japan lowers interest rates and implements quantitative easing policy
2002Full deposit protection terminated
Financial Revitalization Program implemented
Bank of Japan starts stock purchases from banks
2003Resona Bank nationalized
Industrial Revitalization Corporation of Japan (IRCJ) established
Ashikaga Bank nationalized
Sustained Recovery
2004Full protection for deposits payable on demand terminated
2005Outstanding balance of banks’ lending trends upwards
2006Bank of Japan lifts quantitative easing and zero interest rate policy

Moreover, the usual policy defenses against a slowdown appeared to lose their effectiveness as monetary policy hit the zero rate bound and deficit spending could not reverse the economy’s slide. In 2002, unemployment rose to a post-war high of 5½ percent and NPL ratios peaked at almost 9 percent. Meanwhile, net public debt continued to escalate, doubling to nearly 75 percent of GDP in net terms between 1997 and 2002, by far the highest among advanced economies.

The corner was finally turned in 2002. A more aggressive approach to dealing with problem loans and capital shortages—together with the rescue and nationalization of two major banks and the resolution of small regional institutions and credit unions—helped to reduce systemic stress. A virtuous cycle began to take hold as the health of the banking system improved and corporates made progress in redressing the underlying imbalances of the bubble period by shedding the triple excesses of debt, capacity, and labor (Figure 4.4).

Figure 4.4.Japan: “Three Excesses” of Corporate Sector

(In percent)

Source: Ministry of Finance, Corporate Survey.

Successful resolution of the financial crisis laid the foundation for the longest uninterrupted expansion in Japan’s post-war history on the back of surging exports amid strong global growth, rising corporate profits, and expanding employment. Growth picked up, averaging a healthy 2 percent through 2007 and the stock market surged. Price pressures remained mild, with headline inflation only edging into positive territory from 2006.

Japan’s crisis was eventually resolved successfully, but not before a “lost decade” characterized by an extended period of sluggish growth, commodity and asset price deflation, banking failures, and persistent NPLs. By 2000, GDP was nearly 40 percent lower than if growth had continued at the same rate as during the 1980s, and prices were locked in a downward trajectory. At more than 20 percent of GDP, bank losses were eventually much larger than first envisioned, and about 10 percent of GDP in public funds was needed to dispose of NPLs and recapitalize banks.

Fiscal Policy: Did Stimulus Work?35

The effectiveness of fiscal policy in Japan during the lost decade has been the subject of much debate. Fiscal stimulus was used to combat the downturn, but growth remained weak and stagnant tax revenues and increased spending createdaverage deficits of more than 5 percent of GDP between 1993 and 2000. As a result, net debt rose to 60 percent of GDP. While some argue that expansionary fiscal policy was effective but not tried consistently, to others the combination of rising deficits, mounting debt, and stagnant growth points to strong Ricardian effects, mistargeted stimulus, or constraints from a dysfunctional banking system. The evidence itself is mixed, although the effectiveness of fiscal policy appears to have been dampened by weaker fiscal multipliers during the crisis as well as some mistimed consolidation efforts.

Japan’s Policy Response

During the 1990s, Japan introduced a number of fiscal stimulus packages. These packages were in the form of supplementary budgets, which are typically used to address unforeseen events during the year.36 While these packages had large headline numbers—altogether totaling ¥140 trillion, including credit guarantees and public investment—actual spending was considerably smaller—about ¥40 trillion (8 percent of 2000 GDP). Stimulus measures mainly took the form of public investment, support for small and medium-sized enterprises (SMEs), and employment assistance on the spending side, as well as tax measures.

On average, public works accounted for about 40 percent of Japan’s fiscal stimulus measures, and were particularly important in the packages of the early 1990s. They included spending on roads and bridges. Although the returns from such public investment projects may have been low,37 they appeared to have served a safety-net purpose, by creating jobs during the downturn. In the late 1990s, public investment shifted toward arguably more productive spending, including IT-related infrastructure.

Another important element of the stimulus packages was an expansion of credit guarantees on SME lending. When the credit crunch became more pronounced in the late 1990s, Japan introduced a special credit guarantee program that provided 100 percent coverage to banks against losses.38 These guarantees reached nearly ¥30 trillion (6 percent of GDP) by 2001.

At the same time, the stimulus packages of the late 1990s attached greater weight to employment support, given the sharp rise in unemployment, and the share of social security spending, including support for the elderly, also increased. In addition, cash vouchers (¥0.7 trillion) were distributed in 1999 to households that were potentially liquidity-constrained.39

The government also implemented sizable tax cuts, primarily on income, but with mixed results. In 1994, a tax cut of about ¥5.5 trillion (1.1 percent of GDP) was enacted (Table 4.2). However, in 1997, in response to rising government debt and growing concerns about the fiscal implications of population aging, the government changed course and passed a budget that aimed for a substantial down payment on medium-term consolidation (Figure 4.5). The budget raised the consumption tax rate by 2 percentage points and abolished the temporary part of the earlier tax cut, raising the overall tax burden by some ¥7.0 trillion (1.4 percent of GDP). In the wake of the sharp economic contraction that followed, the government again changed course and reintroduced a temporary income tax cut of about ¥4.0 trillion in 1998, followed by another tax cut of ¥6 trillion in 1999.

Table 4.2.Japan: Major Tax Cuts during the 1990s(In trillions of yen)
Income taxationCorporate taxation
Sources: Ministry of Finance; and Cabinet Office.

Introduced as semi-permanent tax cut and fully lifted in FY2007.

Sources: Ministry of Finance; and Cabinet Office.

Introduced as semi-permanent tax cut and fully lifted in FY2007.

Figure 4.5.Japan: Fiscal Situation of the General Government

(In percent of GDP)

Sources: Cabinet Office; and IMF staff calculations.

Importantly, the fiscal stimulus measures during the 1990s were not framed within a medium-term strategy. Although the Fiscal Structural Reform Law—which aimed at a reduction in fiscal deficits over the medium term—was formulated in 1997, it was scrapped a year later in light of the sharp economic contraction.40

Despite the seemingly significant fiscal stimulus, the economy remained largely stagnant until the early 2000s (Figure 4.6). Between 1993 and 2000, average growth was slightly above 1 percent, and tax revenue remained almost flat, leading to larger fiscal deficits. Over this period, the general government deficit averaged more than 5 percent of GDP. As a result, net debt increased sharply to 60 percent of GDP in 2000 from about 15 percent of GDP a decade earlier.

Figure 4.6.Japan: GDP Growth and Supplementary Fiscal Stimulus Package

(Supplementary budget basis; in percent of GDP)

Sources: Ministry of Finance; and Cabinet Office.

Assessment of Japan’s Experience

While deficits appeared large, the actual fiscal impulse was modest, with the cyclically adjusted deficit (the “structural” deficit) increasing only modestly between 1994 and 1998 (Figure 4.7). It was only after 1998 that fiscal policy became truly expansionary, with a more significant widening of the structural deficit. As discussed below, the limited fiscal impulse may have reflected several factors.

Figure 4.7.Japan: Structural Balance of the General Government

(In percent of potential GDP)

Source: IMF staff calculations.

First, actual public investment was smaller than the deceptively large headline numbers, as public investment in the central government initial and supplementary budget did not increase much after the mid-1990s (Figure 4.8). The economic impact may also have been limited by the large share of land purchases, which were as high as 30 percent of the project size in some cases (Kalra, 2003). Finally, about 15 percent of budgeted public investment remained unused partly because local governments were unable to obtain matching funds.41 As a result, public investment remained flat after the mid-1990s, as reflected in the national accounts data, where real public investment (including by local governments) started to decline as early as 1995 (Figure 4.9).42

Figure 4.8.Japan: Central Government Public Investment1

(In percent of GDP)

Source: Sadahiro (2005).

1 Includes land purchases.

Figure 4.9.Japan: Real Public Investment1

(In trillions of 1995 yen)

Source: Cabinet Office.

1 SNA basis.

Second, the limited fiscal impulse also reflected the stop-start nature of Japan’s early stimulus efforts, in particular, the premature tax reversals. With the economy appearing to recover in 1997, the government proceeded to reverse the income tax cut and raise the consumption tax. The larger-than-expected fall in household spending that followed stymied the short-lived recovery, plunging the economy back into recession.

Third, stimulus may have been hampered by low fiscal multipliers.43 Estimates of fiscal spending multipliers cover a wide range (0.4–2.0) (Table 4.3), but there is general consensus that these declined over time. For example, the Cabinet Office’s estimate for the public investment multiplier declined to 1.1 in 2004 from 1.3 in 1991. As laid out below, possible factors behind the declining multipliers include:

Table 4.3.Japan: Estimated Multipliers
SpendingTax cutEstimation period
Kalra (2003)0.40.4-0.51981-2000
Bayoumi (2000)0.650.21981-98
Murata and Saito (2004)1.110.51985-2003
Kuttner and Posen (2002)2.02.5FY1976-99

Multiplier for public investment.

Multiplier for public investment.

  • Lack of private sector response. Private spending may not have responded to the stimulus because the banking sector was not able to play an effective intermediary role given its weak balance sheet and bad loan problems (e.g., Kuttner and Posen, 2001). This view is supported by empirical evidence of a credit crunch during the late 1990s (Motonishi and Yoshikawa, 1999). Heavily indebted corporates were also not in a position to increase spending, as they were deleveraging. Indeed, flow of funds data suggests that the corporate sector’s financial surplus was on an upward trend until the end of the 1990s (Figure 4.10).

  • Shift to lower multiplier spending. The share of central government spending on social security, which is typically thought to have a smaller multiplier than capital spending, increased to 3.5 percent of GDP in 2000 from 2.6 percent in 1990 (Figure 4.11). The disbursement of cash vouchers in 1999 also had a limited impact, with an estimated multiplier of at most one-third, perhaps due to substitution effects (Cabinet Office, 1999).

  • Ricardian equivalence. Although the evidence for Ricardian effects is mixed, some have argued that private demand could have been suppressed by concerns over future tax increases and the rapid rise in public debt (e.g., Bayoumi, 2000).

Figure 4.10.Japan: Financial Surplus of Nonfinancial Corporate Sector

(In percent of GDP)

Sources: Bank of Japan, Flow of Funds statistics; and Haver Analytics.

Figure 4.11.Japan: Share of Central Government Spending

(In percent of GDP)

Sources: Ministry of Finance; and Cabinet Office.

Potential Implications for Fiscal Stimulus in the Current Crisis

Amid today’s concerted push for a global fiscal stimulus, Japan’s experiences provide some useful guidelines to policymakers. The need for discretionary actions is particularly acute in Asia, given generally weak automatic stabilizers and large output gaps. There is also ample fiscal space following years of prudent policies in many economies (as discussed in Box 1.5):

  • Successful fiscal stimulus requires identifying spending with high multipliers. Within the G-20 economies, stimulus packages announced by Asian countries are, on average, more heavily weighted to spending—with particular emphasis on investment in infrastructure. To justify spending against debt accumulation and its potential negative effects on interest rates, measures must be well-targeted, e.g., transfers should be aimed at lower-income households with a higher marginal propensity to consume (Spilimbergo and others, 2008). On public investment, the priority should be projects that are more likely to stimulate private demand.

  • While getting the timing right can be challenging, fiscal stimulus should be withdrawn only after clear signs of an economic recovery. Japan’s experiences highlight the difficulty in deciding the timing of tax and spending changes. Tellingly, the Japanese economy quickly fell into a recession in FY1997 when the temporary tax cut was reversed. Policymakers must strive to ensure that actions are sustained as long as needed and guard against premature withdrawal of stimulus in the face of false dawns.

  • Equally, a delay in restoring tax increases can be costly. In Japan, the income tax cut introduced in the late 1990s was fully lifted 10 years later (in 2007), partly contributing to persistently large deficits and a continued rise in public debt even during the recovery period after 2002.

  • To maximize the impact of fiscal stimulus, attention must be paid to restoring the credit function of the banking sector. The effects of fiscal stimulus could be short-lived unless the financial system is in good health. In Japan, large-scale capital injection into banks took place only in 1999, with fiscal stimulus packages before that date proving ineffective in generating a sustained recovery.

  • Finally, it would be useful to outline at an early stage a concrete and credible medium-term strategy for returning to a sustainable fiscal position. It was only in 2002, after net debt had reached nearly 75 percent of GDP, that the Japanese authorities announced a target for achieving a primary balance (excluding the social security fund) by the early 2010s (Figure 4.12).

Figure 4.12.Japan: Fiscal Balance and General Government Debt

(In percent of GDP)

Sources: Cabinet Office; and IMF staff calculations.

1 Excludes social security.

In the current setting, deliberations on near-term fiscal stimulus provide a good opportunity to build agreement on a medium-term plan for achieving fiscal sustainability. Any costs of government intervention in the financial sector (including by the central bank) should be accurately and transparently recorded to properly assess the balance sheet risks to the public sector.44

Monetary Policy: The Bank of Japan’s Approach to Credit Easing45

As Japan’s crisis unfolded, the Bank of Japan (BoJ) faced an unprecedented set of challenges. After some delay, monetary policy was gradually loosened over the 1990s, but the impact was dampened by banking sector weaknesses. Unable to lower rates past their zero bound, the BoJ took some innovative steps from 2001, centered on exceptional measures to provide liquidity, including expanding the range of collateral, direct asset purchases, and quantitative easing under a zero interest rate policy. However, through most of this period, monetary policy appeared to “pushing on a string” as demand for credit shriveled. Each time measures were taken, the economy seemed to be unresponsive, as growth deteriorated and deflationary pressures became more entrenched. Ultimately, fixing the financial system was needed to end deflation and usher a return to a more normal monetary policy framework, with the BoJ managing a smooth exit.

Japan’s Policy Response

At first, the BoJ responded to the crisis with standard monetary policy tools, through successive interest rate reductions. Except for a hiatus in 1994, the BoJ gradually cut its target interest rate: between mid-1991 and mid-1995, the discount rate was lowered eight times, from 6 percent to fractional levels (Figure 4.13). The BoJ then changed its target to the overnight interest rate but, with rates already very low, the initial target was set at just 0.5 percent. In any case, some pickup in economic activity and inflation, together with increased bank lending, seemed to obviate the need for easing over the next few years.

Figure 4.13.Japan: Interest Rates

(In percent)

Source: CEIC Data Company Ltd.

In 1997, the collapse of key financial institutions revealed the full scale of the crisis and called for more forceful actions to ease credit conditions. The macroeconomic environment deteriorated significantly, with credit conditions also tightening (Figure 4.14). However, the scope for further conventional easing was extremely limited—a rate cut of only a quarter percentage point to 0.25 percent was possible in the fall of 1998—necessitating a radical change in the monetary policy framework. A major change in the institutional environment was also enacted, as the BoJ gained formal independence from the government.

Figure 4.14.Japan: Bank Lending

(Year-on-year percent change)

Source: Haver Analytics.

To better provide liquidity support and substitute for the impaired interbank market, the BoJ expanded the range and flexibility of its monetary instruments. These measures evolved over time in response to changing market conditions and focused primarily on (1) broadening the range of eligible collateral to include corporate bonds, loans on deeds, asset-backed commercial paper (ABCP) and other forms of asset-backed securities (ABS); (2) providing liquidity at longer terms by extending the maturity of bill purchases and Japanese Government Bond (JGB) repos from six months to a year; and (3) increasing the number of counterparties for JGB purchases and commercial paper repo operations. The BoJ’s balance sheet swelled as a result.

In February 1999, the BoJ formally shifted to a zero interest rate policy (ZIRP). Following the announcement of the BoJ’s intention to encourage the policy rate to move “as low as possible,” the policy rate was lowered to 0.15 percent, succeeded by further reductions to rates as low as 0.02 percent. The BoJ had felt that the downturn reflected problems in the financial and corporate sectors so that the onus rested with fiscal and structural policy. However, in hitting the zero lower bound on nominal interest rates, the BoJ decided to move beyond conventional monetary policy. Even then, the BoJ’s public pronouncements suggested that it saw ZIRP as an extraordinary move, with uncertain channels of influence.

Some signs of a pickup in activity prompted an early termination of the policy, but this had to be reversed. In August 2000, the BoJ lifted ZIRP and raised rates to 0.25 percent, on some tentative evidence of a pickup in growth and a decline in risk premiums (Figure 4.15). The move was also prompted by fears that excess liquidity could fuel another bubble and unhinge inflation expectations. The government requested a postponement of the decision, particularly because deflation had not abated and bank lending was still contracting. In addition, the recovery appeared fragile, with unemployment still on an upward trend and corporate bankruptcies increasing. In fact, with the economy falling back into recession soon after, the BoJ had to lower the policy rate back to zero within seven months.

Figure 4.15.Japan: Credit Spreads1

(In basis points)

Sources: Haver Analytics; CEIC Data Company Ltd.; and IMF staff calculations.

1 Spread between 10-year A-rated corporate and 10-year government bond.

In March 2001, the BoJ introduced its “quantitative easing” policy.46 The policy instrument was changed, with the BoJ targeting the outstanding balance of banks’ current accounts at the central bank (consisting of required and excess reserves). The initial target was set at about ¥5 trillion, aimed at pushing the overnight call rate to zero, and was increased in a series of steps to about ¥35 trillion by 2004 as credit growth remained lackluster.

The BoJ also strengthened its commitment to quantitative easing through its communication strategy, making it increasingly clear that a more expansionary stance would be maintained until deflation ended. This helped to better manage market expectations about the future path of interest rates (the so-called policy duration effect). In October 2003, the BoJ clarified its commitment by announcing two necessary conditions for ending quantitative easing—that core CPI be non-negative for a few months and that a majority of the Policy Board members forecast positive core CPI inflation.

In addition, greater coordination with fiscal policy was in evidence, with the BoJ gradually increasing its purchases of long-term JGBs from ¥400 billion to ¥1.2 trillion per month, and such purchases were generally regarded by market participants as helping to place a cap on long-term yields. Over time, assets that could be purchased by the BoJ under quantitative easing were expanded to include commercial paper, corporate bonds, equities, and asset-backed securities, although actual amounts were relatively limited. The quantitative easing policy saw the BoJ’s balance sheet increase from ¥91 trillion in 1998 to a high of about ¥155 trillion in 2006 (Figure 4.16).

Figure 4.16.Bank of Japan: Assets and Balance of Banknotes in Circulation

(In trillions of yen)

Sources: Bank of Japan; and Haver Analytics.

At the same time, the BoJ took unprecedented steps to address the capital shortage in banks. Banks’ large equity holdings (¥32 trillion or nearly 150 percent of their Tier 1 capital) constrained their ability to extend credit and take on new risk. To help reduce banks’ market exposure, the BoJ introduced a program in 2002 to purchase equity rated BBB–or higher directly from banks at market prices. In addition to stabilizing the banking system, such operations may have bolstered the asset price channel of monetary policy by reinforcing economic activity through wealth effects. During 2002–04, BoJ purchases of equities reached ¥2.1 trillion (US$18 billion), representing about 6 percent of banks’ total equity holdings. Although significant, the amount was tiny compared to the BoJ’s holdings of JGBs (¥65 trillion).

The BoJ also resorted to unconventional measures to support corporate lending. In 1998, to help firms with their end-of-year funding, the BoJ established a temporary lending facility to refinance 50 percent of the increase in loans provided by financial institutions during the fourth quarter of the year. In 2003, the BoJ initiated a program to assist SMEs by purchasing ABS and ABCP backed by SME loans rated BB or higher.

Meanwhile, the Ministry of Finance (MoF) undertook large-scale foreign exchange interventions in 2003 and early 2004, helping to stabilize the yen during a period of dollar weakness. These operations could have helped to activate the exchange rate channel and prevent an undue tightening of monetary conditions. Amounts were large, with the monetary authorities selling ¥20 trillion in 2003 and ¥15 trillion in the first quarter of 2004.

Quantitative easing did not immediately arrest deflation or lead to an expansion in bank credit, partly reflecting the unwillingness of banks to make loans and the subdued demand for credit from corporates amid deleveraging pressures. These weaknesses disrupted the normal transmission channels of monetary policy (Figure 4.17).

Figure 4.17.Japan: Monetary Aggregates

Source: CEIC Data Company Ltd.

To its credit, the BoJ was able to exit from quantitative easing relatively smoothly, aided by transparent and open communication. After the economy recovered, it took some time before deflation was ended and the preannounced conditions for ending quantitative easing were met. In March 2006, the BoJ orchestrated a smooth exit to a more normal monetary framework, indicating that it would gradually drain liquidity while keeping overnight interest rates effectively at zero until excess reserves were drawn down. By August, current account balances at the BoJ had fallen to less than ¥10 trillion.

Assessment of Japan’s Experience

Loosening of monetary policy in the early 1990s, while arguably appropriate given general expectations of future economic developments at the time, proved to be too slow in light of subsequent declines in output and prices.47 Traditional central bank concerns about inflation, the belief that fiscal policy should take the lead, the lack of precedent—as well as the need to devise new operating procedures—for a zero-rate environment may all have been important constraints. In fact, by the time quantitative easing was introduced, prices were already on a sustained downward trajectory and long-term yields had fallen to low levels.

Premature tightening and lack of coordination with fiscal policy may have hampered the effectiveness of monetary actions. Some have faulted the rate hike in August 2000 as a key policy error.48 After the fact, the decision certainly appears to have been premature, with the choking off of the fledgling recovery and worsening of deflationary pressures prompting an abrupt reversal. Others have suggested the presence of an “independence trap,” with a newly formally independent BoJ resisting a more expansionary policy that could have jeopardized its credibility (see, e.g., Cargill, Hutchison, and Ito, 2000).

A clear communication strategy, together with a transparent objective, may have helped shape inflation expectations and build credibility. Initially, some market analysts questioned the BoJ’s commitment to the ZIRP, viewing its pledge to keep rates at zero “until deflationary concerns were dispelled” as vague. In its defense, the BoJ may have been reluctant to commit to a zero rate target for an extended period, given concerns over potential asset bubbles and future inflation. Due to the inherent uncertainty regarding the potential benefits and costs of unconventional measures, it is also much more difficult for central banks, including the BoJ, to explain whether and how these would work, compared with conventional tools such as rate cuts. In the end, clearer communication with the public and more transparent exit conditions helped the BoJ to manage financial market expectations of future monetary policy actions and avoid an inflationary spike after the recovery.

However, zero interest rates and quantitative easing came at a cost and were not a final solution. Unconventional monetary policy actions have significant negative side effects, notably in the form of compounding the breakdown in money markets, reduced market discipline, compressed credit spreads, pressure on bank profits, as well as reduced incentives for restructuring.49 This may be a necessary price to pay for maintaining financial stability and preventing deflation from worsening. However, the costs—particularly in terms of delayed restructuring and disruptions to the monetary transmission mechanism—increase the longer the zero interest rate policy is in place, necessitating rapid progress to restructure all affected balance sheets.

Potential Implications for Monetary Policy in the Current Crisis

With interest rates and inflation expectations falling sharply in most economies, while banks scale back intermediation, central banks may need to adopt unconventional policies. For those approaching the lower bound and where large risk premiums are disrupting the normal transmission of monetary policy, some consideration of quantitative easing measures may be warranted. Even those that still have scope for lowering policy interest rates to stimulate their economy may have to be prepared to take some of these actions in the future, given significant downside risks.50 In addition, credit easing under way in major economies could have spillover impacts—in particular, economies with fixed or pegged exchange rates could be importing quantitative easing from abroad. Japan’s experiences suggest that:

  • When faced with a marked slowdown and potential deflationary pressures, central banks must not shy away from bold and unconventional monetary policy actions. Despite some negative side-effects associated with credit easing, central banks should be willing to take a range of aggressive measures in light of the severity of the crisis.

  • While private markets remain dysfunctional, direct measures to ease credit conditions that aim to jump-startcredit could be considered. In Japan, the weak condition of the banking system led to a significant decline in financial intermediation, severely limiting the effectiveness of traditional monetary policy. In such cases, policies aimed at strengthening the development of capital markets, or even bypassing the dysfunctional banking system, may be helpful.

  • Some coordination between fiscal and monetary authorities may be considered. Increased government bond purchases by the central bank could help to stimulate the economy by lowering long-term yields and alleviating crowding-out. However, risks to the balance sheet and independence of the central bank must be carefully balanced. In particular, default risks on private debt could be significant while losses on treasury securities could mount as interest rates rise with a recovery. Such vulnerabilities could undermine the independence and credibility of the central bank. In cases where operations have a fiscal nature and credit risk is significant, it seems more appropriate for fiscal authorities to take the lead.

  • Effective communication with markets and the public is vital, particularly when unconventional tools are being used. As the BoJ found, it is important to convince markets and the public that the central bank is committed to sustained expansion until the economy recovers. In taking unconventional actions, Asian central banks could improve transparency by clarifying their near-term objectives.

  • A smooth exit strategy from unconventional operations once the crisis abates needs to be conceived at an early stage. The Japanese money market, which had withered during the late 1990s, recovered, as institutions relied less on the BoJ for funding that had been made available at penal rates relative to normal times and the opportunity cost of idle balances rose. With the recovery drawn out, the BoJ was also able to avoid losses and yield spikes by holding JGBs to maturity.51 As in Japan, the most desirable exit scenario would be for investors’ risk appetite to recover and credit markets to normalize. However, while the natural maturing of government bond holdings presents few risks, other acquisitions may be more problematic, such as long-lived assets for which there may no longer be a market.

In the final analysis, however, credit easing can have costly side-effects and is not a substitute for balance sheet restructuring. This places a premium on timely steps to restructure bank and debtor balance sheets, which would stimulate private credit while creating a plausible exit strategy for central banks. Japan’s experiences in this area are discussed in the next section.

Financial Sector Policies: Resolving Japan’s Banking Crisis and Fiscal Costs

As the crisis intensified and its roots became clearer, the Japanese authorities turned more forcefully to financial sector policies. Their strategy centered on restructuring banks, pushing them to recognize problem loans and raise new capital, and in some cases seek out public funds or exit the sector. At more than ¥100 trillion, bank losses were much larger than first envisioned, and about ¥47 trillion in public funds was needed to dispose of NPLs and recapitalize banks. In the final analysis, tighter supervision, judicious use of public funds, and a sound framework for restructuring distressed assets helped restore health to the financial system and support a sustained economic recovery. To date, nearly three-fourths of the public funds used in the financial sector interventions have been recovered.

Japan’s Policy Response

Starting in 1991, the Japanese government embarked on a series of attempts to address the problems in the banking system (Box 4.2). Beginning with the problems with the jusen mortgage financing companies and credit cooperatives, the government organized joint rescues by private banks (based on the “convoy” approach) centered around loan concessions and liquidity support. But as property prices continued to fall, losses in the loan portfolio increased. By 1995, around three-fourths of jusen loans were nonperforming, forcing the government to liquidate the failed jusen and create a public asset management company to handle their bad assets (Hoshi and Kashyap, 1999,2001).

Box 4.2.Japan: Key Financial System Reforms, 1996–2003

1996: The “Big Bang.” Removal of the remaining legal barriers separating ownership of banks, trust banks, securities firms, and insurance companies; removal of the long-standing ban on holding companies, also allowing the creation of financial groups. Safety net enhanced including temporary comprehensive deposit insurance.

1998: Banking law reform. Prompt corrective action (PCA) procedures established. Financial Supervisory Agency established under Financial Reconstruction Commission (FRC) to oversee rehabilitation of the financial sector and improve supervision. Inspection manual prepared and published, designed to promote more effective loan valuation and provisioning practices (introducing so-called self-assessment process). Securities and Exchange Surveillance Commission (SESC) moved from the Ministry of Finance (MoF) to the Financial Supervisory Agency.

Bank of Japan (BoJ) law passed, establishing an independent central bank. BoJ’s right to examine counterparty financial institutions explicitly confirmed.

1999: Insolvency law reformed under Civil Rehabilitation Law. Disclosure regime enhanced. Banks required to disclose more information on asset quality and unrealized gains/losses on securities’ holdings. The Resolution and Collection Corporation (RCC) created to collect bad loans from failed housing loan companies, banks, and credit cooperatives.

2000: Safety net enhanced. New deposit insurance law codifying the safety net, including a crisis management framework. PCA procedures strengthened. Accounting reforms introduced, including consolidated accounting and mandatory use of market values for securities. Financial Supervisory Agency renamed Financial Services Agency (FSA).

2001: FSA takes over functions of the FRC. Position of Minister for Financial Services within the Cabinet set up. Accounting Standards Board of Japan established to complete task of bringing accounting standards into line with international best practice. Special inspections by the FSA leading to more realistic loan loss provisioning.

2002: Comprehensive deposit insurance withdrawn; large time deposits no longer insured. Government and BoJ establish schemes for purchasing bank equity holdings. Program for Financial Revival published; key elements include (1) new inspection of major banks’ loan classification and provisioning; (2) introducing discounted cash flow (DCF) methodology for provisioning loans to large “special attention” borrowers; (3) harmonizing loan classification for large borrowers across banks; (4) disclosing the gap between major banks’ self-assessment of problem loans and FSA assessment; and (5) external auditing of capital adequacy ratios, starting in FY2003.

2003: Industrial Revitalization Corporation of Japan (IRCJ). Set up to promote more effective corporate restructuring. Another round of special inspections leading banks to raise external capital and set up asset resolution companies, often in conjunction with international investors.

Source: IMF (2003).

The tendency early on toward regulatory forbearance and an ad-hoc, case-by-case approach reflected to some extent a lack of understanding on the size of the NPL problem and the initial belief that an economic recovery would soon take hold. Public resistance to bank bailouts coupled with deficiencies in the deposit insurance scheme and legal framework for resolving a large-scale banking crisis may have also limited the authorities’ ability to act (Kanaya and Woo, 2000).52 As problem loans were allowed to fester, funding costs for Japanese banks continued to rise during the mid-1990s (the so-called Japan premium), making it difficult for banks to simply “grow out of their problems” (Figures 4.18 and 4.19).

Figure 4.18.”Japan Premium1

(3-month LIBOR, in percentage points)

Sources: Bloomberg LP; and IMF staff calculations.

1 Average of 3-month US$ LIBOR of Bank of Tokyo, Fuji, Sumito minus overall LIBOR; from 1999, includes Bank of Tokyo, Fuji, and Norinchukin Bank.

Figure 4.19.Cumulative Loans Losses of Japanese Banks since 1992

(In billions of yen)

Source: Bank of Japan.

As strains in financial markets heightened in 1997, the BoJ was forced to intervene to stabilize the system. Successive failures of several banks and securities houses beginning in the mid-1990s paralyzed the financial markets, requiring the BoJ to step in with emergency assistance. Such assistance ranged from providing lender-of-last-resort liquidity support to the interbank market to directly injecting capital into failed banks through the Deposit Insurance Corporation (DIC).53 Despite these efforts, financial strains intensified, leading the BoJ and the MoF in November 1997 to announce a blanket guarantee on all deposits and interbank transactions to safeguard the system.

The BoJ’s interventions helped to stabilize the credit markets, but they did not solve the problem of banks’ capital shortage. LIBOR spreads for Japanese banks came down starting in 1998, and the volatility and level of short-term interest rates were reduced, but banks’ weak capital position limited their ability to extend new credit or take on risk, raising concerns over a credit crunch. Regulatory forbearance also reduced management and shareholders’ incentives to take action, either by raising new equity or writing down bad loans. To resolve this impasse, the BoJ and others pushed strongly for the government to inject public funds as a means of freeing banks’ capital constraints and reviving the credit channel.

Early attempts at public recapitalization came with few conditions. A new Financial Crisis Management Committee was created to identify banks with capital shortages and the amounts to be injected. By defining conditions under which regulators were obliged to take remedial actions, the scope for regulatory forbearance was narrowed. Under this new framework, public fund injections took place in three stages.

  • In February 1998, the government made ¥30 trillion in public funds available, of which ¥13 trillion (about 2½ percent of GDP) was for capital injection and the rest for deposit insurance. To minimize the stigma associated with public funds, banks were encouraged to apply together for public funds; by end-March, ¥1.8 trillion had been disbursed almost equally to 21 large banks but without a comprehensive examination or clean-up of bank balance sheets.

  • As financial market conditions deteriorated, the Diet in October 1998 doubled the pool of public funds earmarked for strengthening the banking sector to ¥60 trillion (12 percent of GDP), of which ¥25 trillion was set aside for capital injection into solvent banks, and the rest for resolving failing banks, and supporting deposit insurance. Despite these efforts, two sizable banks—Long-Term Credit Bank of Japan and Nippon Credit Bank—failed and were temporarily nationalized.

  • In March 1999, an additional ¥7½ trillion was injected into 15 major banks. To qualify for the capital injection, each bank had to submit a restructuring plan that included raising new capital from the private sector, which was reviewed quarterly.54

Although these attempts helped to recapitalize the system, NPLs continued to rise, and ultimately, a more comprehensive strategy to clean up banks’ balance sheets was required. This strategy, which complemented previous capital injections, adopted a more forceful approach to using public funds, concentrating on four key elements:

  • Ensuring realistic valuation of bad assets. The strategy began with so-called special inspections by the Financial Services Agency (FSA) focusing on large borrowers at the major banks and then later extended to regional banks. The results confirmed that self-assessments of asset quality were overly optimistic and that nonperforming loans had been significantly understated. Starting in 2002, prudential norms were strengthened by introducing mark-to-market accounting, stricter loan classification and loan-loss provisioning. In particular, the introduction of discounted cash flow methodology to value loans and the cross-check of loan classification across major creditors helped to improve provisioning and raise banks’ incentives for restructuring.

  • Accelerating the disposal of nonperforming loans. Under the so-called Program for Financial Revival, major banks were required to accelerate the disposal of NPLs from their balance sheet within two to three years by selling them directly to the market, pursuing bankruptcy procedures, or rehabilitating borrowers through out-of-court workouts. Remaining loans were sold to the Resolution and Collection Corporation (RCC) charged with disposing of bad assets of failed banks. In contrast to the ineffective warehousing of bad jusen loans in the early 1990s, the RCC and banks looked more to restructure nonperforming assets.

  • Improving bank capital. About ¥12½ trillion of public funds (including past injections) was used to recapitalize both major (except for the Mitsubishi Tokyo Financial Group) and regional banks, mainly through preferred stock or subordinated debt. In the later stages, in exchange for public funds, banks were required to write down the capital of existing shareholders, replace senior management, and submit a reorganization plan to be reviewed regularly by the FSA.55 Banks were also required to undertake governance reforms consistent with Basel Committee guidelines, such as appointing outside directors and establishing a board audit committee.

  • Strengthening supervision. In 1998, the FSA was created (later renamed the Financial Services Agency), consolidating supervision from the MoF and other government agencies into a single entity. A new law was also passed, authorizing the FSA to prescribe prudential rules and apply prompt corrective action when rules were breached or when institutions were viewed as unsafe or unsound.

At the same time, the government took steps to facilitate the restructuring of distressed borrowers. In 2003, the government established the Industrial Revitalization Corporation of Japan (IRCJ) to purchase distressed loans from banks (up to about ¥1 trillion) and work with creditors in restructuring. To support private-sector-led restructuring, the government also reformed the insolvency system (introducing a faster and more efficient “Civil Rehabilitation Law”), introduced guidelines for outof-court corporate workouts, and upgraded the accounting and auditing framework. These measures helped to create a market for restructuring distressed assets, drawing in private capital and expertise, including from overseas.

In the end, the government injected public funds of nearly ¥47 trillion (10 percent of 2002 GDP) to recapitalize the banking system and dispose of problem loans. In 2003, banks’ share prices started to recover, as banks’ NPLs began to trend down and capital ratios stabilized (Figure 4.20). At the same time, the banking system underwent significant consolidation, with several large banks and many smaller institutions either closed or merged. To date, nearly three-fourths of the ¥12.5 trillion of public capital has been repaid, and about 80 percent of total funds are expected to be recovered.

Figure 4.20.Japan: Nonperforming Loans

(Risk management loans, in trillions of yen)

Source: Financial Supervisory Agency of Japan.

Assessment of Japan’s Experience

Delays in recognizing problem loans exacerbated Japan’s financial crisis and postponed a sustained recovery. Weak accounting practices and regulatory forbearance masked the NPL problem for many years and limited incentives for remedial action by both the government and the banks themselves. The delay in recognizing the losses proved costly, both in terms of taxpayer funds but also in holding back a recovery, as insolvent “zombie” firms were allowed to linger and constrain investment by sound firms.56 The result was ultimately a “lost decade” of growth, wasteful pump-priming spending, and a large buildup of public debt. At a minimum, earlier action to recognize problem loans and raise adequate provisioning would have helped identify the capital shortage and jump-start the process of restructuring.

Liquidity provision helped forestall an immediate systemic crisis, but could not adequately address the fundamental problem of an undercapitalized banking system. In Japan, exceptional liquidity was required to stabilize the system, but without accompanying steps to recognize losses and address the capital shortage, its effectiveness diminished over time. As noted earlier, if left for too long, exceptional liquidity can also generate negative side effects by distorting the functioning of the markets and delaying needed restructuring.

Public funds that were conditional on equity writedowns and steps to dispose of bad assets ultimately proved effective. In Japan, the injection of capital into viable institutions, together with the orderly resolution of nonviable ones, helped support credit and bolster capital ratios, but only after they were linked to strong steps to clean up balance sheets and undertake restructuring. Such steps were supported by close monitoring by the FSA under an agreed reorganization plan. Public funds also helped to promote needed financial consolidation, with several large banks and many smaller institutions either closed or merged.

A centralized asset management approach helped accelerate the clean-up of banks’ balance sheets. Government purchases and sales of NPLs through the RCC and the IRCJ facilitated a market for restructuring by enhancing price discovery, resolving credit disputes, and providing legal clarity and accountability.57 They also allowed bank management to concentrate on extending new loans and restructuring their business operations. With asset prices recovering, these interventions ended up costing taxpayers far less than their original price tag—indeed, the IRCJ even managed to generate a small profit before it shut down in 2007.

On the borrower side, a sound private-sector-led framework was needed to assist in such restructuring. Although a public asset management company can quickly remove distressed assets from banks, recovery values are likely to depend on the private sector taking the lead in restructuring. Getting the incentives right hinged on proper valuation of distressed assets and a sound prudential framework. Bankruptcy reforms and improvements to the accounting and governance framework also provided the private sector with useful tools to restructure distressed firms.

Finally, to restore market discipline and minimize moral hazard, an exit strategy for divesting public shares in the banking system and other interventions in the financial system needed to be developed. In the case of Japan, the shift from a blanket guarantee to partial deposit insurance and the gradual repayment of public funds were fairly orderly and smooth. However, the BoJ has not managed to unwind fully its purchases of equities held by banks, while some banks are still struggling to repay their public funds. The gradual withdrawal of public support of the SMEs, such as through credit guarantees, may have also held back the restructuring of smaller firms that continue to suffer from excess leverage and low profitability.

Potential Implications for Financial Sector Policies in the Current Crisis

Economies facing similar acute banking distress should be wary of repeating Japan’s early mistakes of the 1990s and be prepared for forceful actions to recognize bank losses, restructure distressed assets, and recapitalize viable institutions. Faced with a quickly deteriorating outlook, Asian authorities have taken a range of steps to promote financial stability (Table 4.4). The degree of intervention has varied across Asian economies, mainly reflecting the relative funding needs and balance sheet strength of banks. Temporary guarantees to boost confidence have been put in place in many countries, including deposit insurance and blanket guarantees on other bank liabilities. Thus far, there has been less need for direct capital support or measures to remove or guarantee bad assets, although Hong Kong SAR, Japan, and Korea have set up funds to bolster bank capital. However, there may be a need for a broader range of Asian economies to shore up capital to limit adverse fallout from the crisis, and preemptively prepare plans to deal with distressed debt and potential corporate failures (as suggested in Chapter 3). Japan’s experiences suggest some key priorities:

Table 4.4.Summary of Bank Support Measures in Asia
Measures adoptedEconomies
Direct liquidity and funding supportAustralia, China, Hong Kong SAR, India, Indonesia, Japan, Korea, Philippines, Singapore, Thailand
State guarantees for bank obligations
Deposit protectionAustralia, Hong Kong SAR, India, Indonesia, Malaysia, New Zealand, Philippines, Singapore, Taiwan Province of China, Thailand
Debt issuanceAustralia, Korea, New Zealand
Removal of bad assetsJapan, Korea
Insurance of bad assetsKorea
Capital supportHong Kong SAR,1 India,2 Japan, Korea
Source: Fitch.

Facilities set up but yet to be used.

Purely a policy intention at this stage.

Source: Fitch.

Facilities set up but yet to be used.

Purely a policy intention at this stage.

  • Recognizing bank losses early. In Japan, regulatory forbearance and pricing gaps, particularly on illiquid properties and multicreditor loans, held up the disposal of NPLs. The introduction of discount cash flow methodology and mark-to-market accounting and the cross-check across banks helped to clarify the true extent of banks’ losses and strengthened the incentives for restructuring. For failed banks, the transfer of bad assets was more straightforward, suggesting that a rigorous inspection of bank asset quality should be a prerequisite for using any public funds to remove bad assets.58 If left unaddressed, uncertainty over the value of the nonperforming loans can spill over to affect sound banks, making it difficult to raise private capital.

  • Using public funds to clean up balance sheets. Japan had adopted many of the same strategies that advanced economies are considering now—setting up asset management companies, protecting bank liabilities, and injecting public capital. Nevertheless, the financial system remained dysfunctional until the Japanese government in 2002, under Prime Minister Koizumi and Minister Takenaka, finally forced banks to clean up their balance sheets and dispose of bad assets. Encouragingly, the ultimate fiscal cost was significantly lower than the up-front expenses because a significant portion was recovered once the economy stabilized.

  • Overcoming resistance to temporary nationalization. The Japanese experience demonstrates that there is no silver bullet—crisis responses are inevitably messy and invariably involve a learning curve. In the mid-1990s, public backlash over the ineffective injection of public funds into the failed jusen companies made it very difficult for the authorities to consider additional public funds for some time, limiting their policy flexibility. Japan’s ability to eventually overcome public resistance to bank bailouts and the stigma attached to public capital proved crucial in forging a final resolution to the problem.

  • Measures to restructure distressed borrowers can also help support bank restructuring. In Japan, financial and corporate restructuring went hand in hand and proved mutually reinforcing. Bankruptcy reform, out-of-court workouts, and debt-equity swaps were useful tools for the private sector to rehabilitate distressed, but creditworthy, firms.


Japan’s experiences following the collapse of its asset bubbles in the 1990s speak, in varying degrees, to key dilemmas facing policymakers in different parts of the world today, including in Asia. In economies undergoing—or likely to encounter—acute financial stress, a systemic solution that addresses both sides of the balance sheet will be needed for a sustained recovery. In this context, a comprehensive approach that addresses both solvency and liquidity issues may be needed, including recapitalizing the banks and restructuring the debts of distressed borrowers.

In the meantime, to support growth, in those economies with room for discretionary action, fiscal stimulus should be sustained, centered on high-impact areas, and only reversed when clear signs of recovery emerge. At the same time, it will be important to articulate a concrete strategy for returning to a sustainable fiscal position over the medium term. To help restore credit markets and combat deflation, monetary policy actions will need to be bold, innovative, and wide-ranging. Where credit easing measures are taken, potential losses to the central bank’s balance sheet and credibility will need to be carefully managed. In this context, clear and transparent communication and a considered exit strategy are desirable.

Importantly, Japan did not recover until the three excesses of labor, debt, and capacity built up during the bubble period were sufficiently addressed. Where such imbalances exist in the banking system or household and corporate balance sheets, the goal of policymakers should be to facilitate the required adjustment, without which sustained growth may not be possible.

The challenges posed by the global crisis may be daunting, but ultimately Japan’s experiences inspire confidence in the ability of informed policymaking to lay the foundations of a lasting recovery and a more dynamic—and resilient—financial system. It may seem curious that in order to chart a way forward, this chapter has looked back. But as the crisis unfolds, policymakers in Asia and across the globe are likely to find that Japan’s experiences provide valuable guidance in their search for a sustained economic recovery.


    AguiarMarkGopinathGita2007Emerging Market Business Cycles: The Cycle Is the Trend,” Journal of Political Economy, Vol. 115 pp. 69102.

    Åsberg SommarPerShahnazarianHovick2008Macroeconomic Impact Expected on Default Frequency,” Sveriges Riksbank Working Paper Series Number 219 (Central Bank ofSweden).

    BaigTaimur2003Monetary Policy in a Deflationary Environment” in Japan’s Lost Decade: Policies for Economic Revival ed. by Tim Callen and Jonathan Ostry (Washington: International Monetary Fund).

    Bank for International Settlements2006International Convergence of Capital Measurement and Capital Standards: A Revised Framework Comprehensive Version” (Basel).

    BayoumiTamim2000The Morning After: Explaining the Slowdown in Japanese Growth” in Post-Bubble Blues: How Japan Responded to Asset Price Collapse ed. by Tamim Bayoumi and Charles Collyns (Washington: International Monetary Fund).

    BergAndrew1999The Asia Crisis: Causes, Policy Responses, and Outcomes,” IMF Working Paper 99/138 (Washington: International Monetary Fund).

    BernankeBenGertlerMark1999Monetary Policy and Asset Price Volatility,” Economic Review, Federal Reserve Bank of Kansas City (Fourth quarter), pp. 1751.

    BryGerhardBoschanCharlotte1971Cyclical Analysis of Economic Time Series: Selected Procedures and Computer Programs,” NBER Technical Working Paper No. 20 (CambridgeMassachusetts: National Bureau of Economic Research).

    CaballeroRicardo J.HoshiTakeoK. KashyapAnil2008Zombie Lending and Depressed Restructuring in Japan,” American Economic Review (December) Vol. 98 pp. 194377.

    Cabinet Office1999Demand Impact of Cash Vouchers” (Tokyo).

    CardarelliRobertoElekdagSelimLallSubir2009Financial Stress, Downturns, and Recoveries,” forthcoming IMF Working Paper (Washington: International Monetary Fund).

    CargillThomas F.; M. HutchisonMichaelItoTakatoshi2000Financial Policy and Central Banking in Japan (CambridgeMassachusetts: MIT Press).

    CarrollChristopher D.OtsukaMisuzuSlacalekJirka2006How Large Is the Housing Wealth Effect? A New Approach,” NBER Working Paper No. W12746 (CambridgeMassachusetts: National Bureau of Economic Research).

    CastrénOlliDéesStéphaneZaherFadi2008Global Macro-Financial Shocks and Expected Default Frequencies in the Euro Area,” ECB Working Paper Number 875 (Frankfurt: European Central Bank).

    ChengArnold C. S.K. FungMichael2008Financial Market and Housing Wealth Effects on Consumption: A Permanent Income Approach,” Applied Economics, Vol. 40No. 23pp 30293038.

    EdelsteinRobert H.Kim LumSau2004House Prices, Wealth Effects, and the Singapore Macroeconomy,” Journal of Housing Economics, Vol. 13 (December) pp. 342367.

    EnglishWilliamTsatsaronisKostasZoliEdda2005Assessing the Predictive Power of Measures of Financial Conditions for Macroeconomic Variables” in “Investigating the Relationship between the Financial and Real Economy” BIS Papers No. 22 (Basel: Bank for International Settlements,February).

    FujikiHiroshiOkinaKunioShiratsukaShigenori2001Monetary Policy under Zero Interest Rate: Viewpoints of Central Bank Economists,” Monetaryand Economic Studies, Institute for Monetary and Economic Studies, Bank of Japan, Vol. 19 (February) pp. 89130.

    GauthierCélineGrahamChristopherLiuYing2004Financial Conditions Indexes for Canada,” Bank of Canada Working Paper 2004–22 (Ottawa: Bank of Canada)

    GoodhartCharlesHofmannBoris2001Asset Prices Financial Conditions and the Transmission of Monetary Policy” paper prepared for the conference on “Asset Prices Exchange Rates and Monetary Policy” Stanford UniversityMarch 2–32001.

    GrayDale F.MaloneSam2008Macrofinancial Risk Analysis (John Wiley & Sons).

    GrayDale F.TamirisiaNatalia2009How Vulnerable Are Nonfinancial Corporates?” in World Economic Outlook (Washington: International Monetary FundApril).

    GuichardStéphanieTurnerDavid2008Quantifying the Effect of Financial Conditions on U.S. Activity,” OECD Economic Department Working Paper No. 635 (Paris: Organization for Economic Cooperation and Development).

    Guimarães-FilhoRobertoHoriMasahiroMinianeJacquesN’DiayePapa2008Can Asia Decouple? Investigating Spillovers from the United States to Asia” in Regional Economic Outlook: Asia and Pacific (Washington: International Monetary Fund, April).

    HarriganJamesN. KuttnerKenneth2005Lost Decade in Translation: Did the United States Learn from Japan’s Post-Bubble Mistakes?” Chapter 3 in Reviving Japan’s Economy: Problems and Prescriptions, ed. by T. Ito, H. Patrick, and D Weinstein (CambridgeMassachusetts: MIT Press) pp. 79106.

    HeytensPaul J.KaracadagCem2001An Attempt to Profile the Finances of China’s Enterprise Sector,” IMF Working Paper 01/182 (Washington: International Monetary Fund).

    HongKiseokLeeJong-WhaChink TangHsiao2009Crises in Asia: Historical Perspectives and Implications,” ADB Economics Working Paper Series No. 152 (ManilaPhilippines: Asian Development Bank).

    HoshiTakeoKashyapAnil1999The Japanese Banking Crisis: Where Did It Come From and How Will It End?” NBER Working Paper No. 7250 (CambridgeMassachusetts: National Bureau of Economic Research).

    HoshiTakeoKashyapAnileds.2001Corporate Financing and Governance in Japan (CambridgeMassachusetts: MIT Press).

    International Monetary Fund2003Japan: Financial System Stability Assessment and Supplementary Information” IMF Country Report No. 03/287 (Washington).

    International Monetary Fund2006Asia’s Investment Decline” in Regional Economic Outlook: Asia and Pacific (WashingtonMay).

    International Monetary Fund2008aThe State of Public Finances: Outlook and Medium-Term Policies After the 2008 Crisis (Washington: International Monetary Fund, Fiscal Affairs Department,March).

    International Monetary Fund2008bWorld Economic Outlook (WashingtonOctober).

    International Monetary Fund2008cWorld Economic Outlook (WashingtonApril).

    International Monetary Fund2009aGlobal Financial Stability Report Chapter 1April (Washington).

    International Monetary Fund2009bWorld Economic Outlook (WashingtonApril).

    ItoTakatoshi2004The Yen and the Japanese Economy 2004” in Dollar Adjustment: How Far? Against What? ed. by C. Fred Bergsten and John Williamson (Washington: Institute for International Economics).

    ItoTakatoshi2006Japanese Monetary Policy: 1998–2005 and Beyond,” in Monetary Policy in Asia: Approachesand Implementation, BIS Paper No. 31 (Basel: Bank for International Settlements) pp. 10532.

    JinnoNaohikoKenekoMasaru2000Stopping Fiscal Collapse” (Tokyo: Iwanami-shoten).

    JinushiToshikiKurokiYoshihiroMiyaoRyuzo2000Monetary Policy in Japan Since the Late 1980s: Delayed Policy Actions and Some Explanations,” in Japan’s Financial Crisis and Its Parallels to the U.S. Experience, ed. by R. Mikitani and A.S. Posen, Special Report No. 13 (Washington: Institute for International Economics) pp. 11548.

    JonesMatthew T.KarasuluMeral2006The Korean Crisis: What Did We Know and When Did We Know It? What Stress Tests of the Corporate Sector Reveal,” IMF Working Paper 06/114 (Washington: International Monetary Fund).

    KalraSanjay2003Fiscal Policy: An Evaluation of Its Effectiveness” in Japan’s Lost Decade: Policies for Economic Revival ed. by Tim Callen and Jonathan Ostry (Washington: International Monetary Fund).

    KaminskyGracielaM. ReinhartCarmenA. VeghCarlos2004When It Rains, It Pours: Procyclical Capital Flows and Macroeconomic Policies,” NBER Working Paper No. W10780 (CambridgeMassachusetts: National Bureau of Economic Research).

    KanayaAkihiroWooDavid2000The Japanese Banking Crisis of the 1990s: Sources and Lessons.” IMF Working Paper No. 00/7 (Washington: International Monetary Fund).

    KangKenneth2003The Resolution and Collection Corporation and the Market for Distressed Debt in Japan” in Japan’s Lost Decade ed. by Tim Callen and Jonathan Ostry (Washington: International Monetary Fund).

    KimSunghyun HenryAyhan KoseM.G. PlummerMichael2003Dynamics of Business Cycles in Asia: Differences and Similarities,” Review of Development Economics, Blackwell Publishing, Vol. 7 pp. 46277.

    KuttnerKenneth N.S. PosenAdam2001The Great Recession: Lessons for Macroeconomic Policy from Japan,” Brookings Papers on Economic Activity: 2, pp. 93185.

    KuttnerKenneth N.S. PosenAdam2002Passive Savers and Fiscal Policy Effectiveness in Japan” paper presented at the conference of CEPR-CIRJE-NBER “Issues in Fiscal Adjustment”TokyoDecember.

    LackCaesar P2003A Financial Conditions Index for Switzerland” in “Monetary Policy in a Changing Environment” BIS Paper No. 19 (Basel: Bank for International SettlementsSeptember).

    LaevenLucValenciaFabian2008Systemic Banking Crises: A New Database,” IMF Working Paper 08/224 (Washington: International Monetary Fund).

    MatsuuraKatsumiHoriMasahiro2003Special Credit Guarantee and Restructuring of Small Firms” Cabinet Office ESRI Discussion Paper Series No. 50 (Tokyo: Cabinet Office).

    McCallumBennett T.2003Japanese Monetary Policy, 1991–2001,” Economic Quarterly, Federal Reserve Bank of Richmond (Winter), pp. 131.

    MotonishiTaizoYoshikawaHiroshi1999Causes of the Long Stagnation of Japan during the 1990s: Financial or Real?” Journal of the Japanese and International Economies, Vol. 13 (September) pp. 181200.

    MühleisenMartin2000Too Much of a Good Thing? The Effectiveness of Fiscal Stimulus,” in Post-Bubble Blues: How Japan Responded to Asset Price Collapse, ed. by Tamim Bayoumi and Charles Collyns (Washington: International Monetary Fund).

    MurataKeikoSaitoTatsuo2004The ESRI Short-Run Macroeconometric Model of Japanese Economy (2004 version) - Basic Structure Multipliers and Economic Policy Analyses” ESRI Discussion Paper No. 122 (Tokyo: Cabinet Office).

    NakasoHiroshi2001The Financial Crisis in Japan during the 1990s: How the Bank of Japan Responded and the Lessons Learnt” BIS Paper No. 6 (Basel: Bank for International Settlements).

    N’DiayePapaZhangPingZhangWenlang2008Structural Reform, Intra-regional Trade, and Medium-term Growth Prospects of East Asia and the Pacific,” Hong Kong Monetary Authority Working Paper 17/2008 (Hong Kong: HKMA, December).

    OhashiKazunariSinghManmohan2004Japan’s Distressed Debt Market,” IMF Working Paper 04/86 (Washington: International Monetary Fund).

    OrphanidesAthanasios2004Monetary Policy in Deflation: The Liquidity Trap in History and Practice” Finance and Economics Discussion Series 2004-01 Board of Governors of the Federal Reserve System (Washington).

    OuraHirokoTopalovaPetia2009India’s Corporate Sector: Coping with the Financial Tsunami,” forthcoming IMF Working Paper (Washington: International Monetary Fund).

    PeltonenTuomas A.M. SousaRicardoS. VansteenkisteIsabel2009Wealth Effects in Emerging Market Economies,” ECB Working Paper No. 1000 (Frankfurt: European Central Bank).

    PesaranM. HashemShinYongcheol1998Generalized Impulse Response Analysis in Linear Multivariate Models,” Economics Letters, Vol. 58 (January) pp. 1729.

    ReinhartCarmen M.S. RogoffKenneth2009The Aftermath of Financial Crises,” NBER Working Paper No. 14656 (CambridgeMassachusetts: National Bureau of Economic Research).

    SadahiroAkira2005Macro Economic Analysis in Japan’s Post-War Period” (Tokyo: Toyokeizaishinposha).

    SellonGordon H. Jr.2003 “Monetary Policy and the Zero Bound: Policy Options When Short-Term Interest Rates Reach Zero,” Federal Reserve Bank of Kansas City Economic Review, Vol. 88 (fourth quarter) pp. 543.

    ShirakawaMasaaki2002One Year Under ‘Quantitative Easing’” IMES Discussion Paper No. 2002 - E - 3 (Tokyo: Institute for Monetary and Economic Studies, Bank of Japan).

    SpilimbergoAntonioSymanskySteveBlanchardOlivierCottarelliCarlo2008Fiscal Policy for the Crisis,” IMF Staff Position Note 08/01 (Washington: International Monetary Fund).

    Standard and Poor’s2008Japan’s Lost Decade Offers Lessons for Current Global Turmoil” Ratings DirectDecember7.

    SwistonAndrew2008A U.S. Financial Conditions Index: Putting Credit Where Credit Is Due,” IMF Working Paper 08/161 (Washington: International Monetary Fund).

    TaylorJohn B.2001. “Low Inflation, Deflation, and Policies for Future Price Stability,” Monetary and Economic Studies, Institute for Monetary and Economic Studies, Bank of Japan, Vol. 19 (February) pp. 3551.

    TettGillian2004Saving the Sun: How Wall Street Mavericks Shook Up Japan’s Financial World and Made Billions (New York: Collins Business).

    UedaKazuo2005.The Bank of Japan’s Struggle with the Zero Lower Bound on Nominal Interest Rates: Exercises in Expectations Management” CIRJE F-Series Paper No. CIRJE-F-375 Faculty of Economics University of Tokyo.

    United Nations Industrial Development Organization2009Industrial Development Report (Vienna).

Note: The main authors of this chapter are Kenneth Kang, Murtaza Syed, and Kiichi Tokuoka.

Figures presented in this section are on a fiscal year basis.

Typical examples of unforeseen events are natural disasters, but stimulus measures can also be included.

For example, little-used roads that were constructed in rural areas likely carried small multiplier effects.

Although this measure was aimed at mitigating the credit crunch, it may also have delayed necessary restructuring. For instance, there is some evidence that the SMEs that used this program were more heavily indebted and faced a higher risk of default (Matsuura and Hori, 2003).

The inability to verify incomes forced the government to seek out proxies, such as the presence of children or the elderly.

The original law targeted a reduction in the general government deficit to below 3 percent of GDP by 2003.

However, unused funds are carried over to the next year’s budget.

Analyses by the Cabinet Office also confirm that the rise in the fiscal deficit and debt during the 1990s was largely due to nondiscretionary factors: a sharp decline in revenues and an increase in social security spending owing to the prolonged slump rather than rising public investment associated with countercyclical policy. Indeed, the Cabinet Office’s estimates indicate that public capital formation contributed positively to the fiscal balance over the period 1990–2002. However, this may largely reflect a drastic cut in pubic investment after 2000.

For instance, the injection of capital into banks, provisions of directed loans to financial institutions, credit guarantees, and purchases of illiquid assets may not entail an upfront rise in net debt or the deficit, but their fiscal impact eventually depends critically on the recovery value of acquired assets.

The terms “quantitative easing” and “credit easing” are used interchangeably in this chapter. The BoJ’s “quantitative easing” policy focused on government bond purchases and featured an operating target on the liabilities side of its balance sheet. By contrast, the Federal Reserve’s current “credit easing” features a more targeted approach by intervening in markets that appear stressed and focusing on the asset side of its balance sheet without an explicit operating target. Despite their separate focus and modalities, however, the potential channels of influence of the two approaches do not appear to be significantly different.

For instance, the BoJ’s initial policy actions fell far short of the level of easing seen during comparable episodes in Sweden and the United States during the 1930s (Baig, 2003) and a number of more formal studies based on Taylor reaction functions suggest that monetary policy was slow to respond to deflationary developments in the 1990s. See, for example, Bernanke and Gertler (1999); Jinushi, Kuroki, and Miyao (2000); McCallum (2003), and Taylor (2001).

See, among others, Harrigan and Kuttner (2005) and Ito (2004). Orphanides (2004) likens the rate hike to that of the U.S. Federal Reserve in 1937, believed by some to have contributed to choking off an incipient recovery from the Great Depression.

For example, ample liquidity and low interest rates can delay the recognition of problem loans and undermine market discipline by making it easier for essentially insolvent borrowers to remain current on their interest payments. The flattening of the yield curve also made it more difficult for banks to raise their core profitability and “grow out” of their problems (see Box 3, in IMF, 2003).

For instance, if credit markets remain unresponsive to lower interest rates or the central bank needs to engage in lender-of-last-resort operations in a systemic banking crisis, credit easing measures might be needed.

The BoJ also began selling large numbers of its share acquisitions from 2007, targeting a 10-year period for complete divestment.

As a result, the BoJ was forced to use its balance sheet to rescue two banks in 1994, later suffering losses.

The BoJ extended US$35 billion in lender-of-last-resort assistance at its peak in December 1997 and some US$74 billion in loans to the Deposit Insurance Corporation for recapitalizing banks. In 2001, the irrecoverable amount was estimated to be US$900 million. See Nakaso (2001) for a discussion of the early policy responses.

If the FSA was not satisfied with progress, it could convert its preferred stock holdings to common stocks after a certain grace period, and demand management changes as the largest shareholder.

At the same time, limits were placed on the amount of deferred tax assets (tax credits based on expected future profits) banks could count toward their Tier 1 capital ratio. Deferred tax assets, which in 2003 accounted for nearly one-half of Tier 1 capital in major banks, generated market concerns over the quality and ability of bank capital to absorb further losses.

See Caballero, Hoshi, and Kashyap (2008) for an empirical analysis of the impact of such “zombie” firms on investment and employment growth of sound firms.

See Kang (2003) and Ohashi and Singh (2004) for an analysis of the development of a market for distressed debt in Japan.

In some cases, such as for Shinsei Bank, where uncertainty over loan valuations was high, partial insurance through “put options” on NPLs was used to encourage investors to take over failed banks. However, insurance must be designed carefully to avoid the risk of “cherry picking” and selling back the worst assets (Tett, 2004).

    Other Resources Citing This Publication