Chapter 10: Restoring Financial Stability: Japan’s Perspective
- International Monetary Fund. Legal Dept.
- Published Date:
- February 2013
The author thanks Takafumi Sato, Masamichi Kono, Nobuchika Mori, and Satoshi Ikeda for helpful guidance and comments on this article. He also thanks Yasuhiro Ishimaru, Yu Nishioki, and Yukiko Kamagawa for research assistance.
The financial world has gone through a historic period since the bankruptcy of Lehman Brothers, and there have been a number of significant policy developments. The worst time may have passed, owing to bold, coordinated policy response by national authorities and international institutions such as the Financial Stability Board (FSB) and the International Monetary Fund (IMF). Accordingly, the focus of the discussions seems to have shifted from emergency response to medium-term reform of financial regulation.
While this shift is appropriate, we cannot say that the crisis is already over, because it is not. The de-leveraging of the financial and household sectors has just started; central banks continue to provide ample liquidity to support bank funding; and public money injected to recapitalise banks has not been repaid. The global economy may have started to recover, but large uncertainties remain. For regulators in Japan, the current crisis seems to have a number of commonalities with the banking crisis the country already experienced in the 1990s. Therefore, we cannot easily espouse the optimism now prevalent in financial markets of the United States and Europe.
In this Chapter, I wish to talk about two issues. The first is what I think remains to be done by regulators or the industry in order to get the world’s financial markets back to normal. I will discuss this by comparing the global financial crisis this time with Japan’s experience in the 1990s and early 2000s. Secondly, I would like to present my views on the discussions underway at international fora on how to strengthen the financial systems around the globe, with focus on bank capital requirements and supervision of what are called “systemically important financial institutions (SIFIs).”
Before starting, let me say in advance that my views may differ from the “mainstream” views of the Group of Twenty (G-20), the FSB, or the Basel Committee on Banking Supervision, where the discussions have been skillfully led by “major powers” in the world of financial regulation. They may also differ from the official views of my institution, Japan’s Financial Services Agency (FSA). Although I used to be member of the Basel Committee, I am currently not directly involved in the international discussions underway.
What Remains to Be Done to Normalize Financial Markets?1
The scale of the current global financial crisis is enormous, and has often been characterised as “once-in-a-century” or “the most severe since the Great Depression.” Since the impact of the market turmoil has been serious and has spread globally, it is not without reason that the current difficulties are labeled in that manner. Yet for many Japanese financial regulators, the current crisis is rather a “second-in-a-decade” event. As mentioned earlier, this is because Japan experienced its own banking crisis in the late 1990s.
The current crisis does have its unique features, such as its cross-border and market-oriented nature. However, if you compare these two crises, you can find that the manner in which the current crisis evolved has a number of commonalities to what happened in Japan in the 1990s:
First, irresponsible lending had been widespread prior to the crisis, on the assumption that real estate prices would continue to go up;
Second, the financial market turmoil was triggered by a decline in real estate prices;
Third, the adverse effect of the market turmoil spilled over to the real economy; and
Fourth, the turmoil resulted in a system-wide financial crisis, necessitating intervention by governments and central banks.
The issue now is whether the global financial markets can recover quickly this time. The adverse effect of Japan’s banking crisis lingered over a decade, until its major banks got rid of their non-performing loans. Are there the conditions in place that will enable a quicker recovery this time? Some people say that the current situation of the US markets is similar to that of Japan’s in 1999, when a sense of security had returned following capital injection into major banks with public funds. However, that sense of security turned out to be a false one in Japan’s case, and its economy was plunged into a “triple dip” in 2001–02.
Based on this experience in the 1990s, we have argued in the course of the current global financial crisis that the five lessons Japan learned then provide useful suggestions as to how our fellow regulators should respond to the ongoing difficulties.
First, prompt and accurate recognition of losses is essential. Lack of it would give financial firms incentives to postpone the disposal of their non-performing assets, which could cause a negative spiral of credit crunch and the weakening economy.
Second, toxic assets need to be taken off the balance sheet. Otherwise, it would be difficult to restore full market confidence as additional losses on those assets could be incurred later.
Third, undercapitalization of financial firms resulting from the disposal of bad assets should be addressed promptly, by injecting public funds if necessary.
Fourth, exceptional measures, such as full protection of bank deposits and temporary nationalization of banks, can be effective options in times of serious crises.
Fifth, regulators need to strike the right balance in implementing short-term stabilization measures and medium-term reforms of financial regulation. Crisis management measures should not remain in place over a prolonged period to prevent moral hazard, but too hasty exit from these measures and implementation of reforms could impede stabilization and economic recovery.
The recent developments in the United States and Europe show that these lessons have been relevant in the context of the current global crisis. Such developments include exceptional measures taken by the authorities of these countries, and the stress tests conducted by the US authorities to determine the additional capital required of the major banks.
It seems to me, however, that the work related to the second lesson has been lagging so far. That is, not much progress has been made in taking toxic assets off the balance sheets of US and European banks. It is partly because the securitization market is still inactive and the banks would suffer huge losses if they are to sell those products immediately.
Notwithstanding such difficulties, restoring market confidence is vital to normalize the situations. Increasing the amount of capital alone would not save banks from failing, if banks were not able to convince the markets that they had come to terms with uncertainties in both on- and off-balance sheet risks. Bad assets remaining on balance sheets would continue to be a source of uncertainty, because of concerns about possible further losses. Taking these assets off balance sheet would help break a vicious feedback loop between the financial sector and the real economy.
Yet the price developments of the underlying assets of the securitized products are hardly encouraging (Figure 1). In the early 1990s, we thought that the adjustment of real estate and share prices in Japan would be over soon after a few years of decline. As it turned out, the decline in land prices lasted almost fifteen years and the prices are getting further lower after a brief pick-up. The share prices are hovering around the level that is about only a quarter of its peak at the end of 1989.
The delinquency rates for various kinds of loans in the US are also alarming (Figure 2). The rate for commercial real estate loans is approaching the level recorded at the time of the savings and loans crisis, whereas those for residential real estate and credit cards are above the previous record. All of these are happening at the same time.
Figure 1.Commercial Property Prices in the US and the UK
Source: MIT’s Center for Real Estate, Bloomberg.
Under these circumstances, my concern is that the situation of the US and European banking sectors could become a drag on global economic recovery, as their credit capacity is constrained. This is exactly what happened in Japan throughout the 1990s, where the banks were crippled with huge amounts of nonperforming loans. It was not until the uncertainties about banks’ soundness were removed in the mid-2000s that Japan’s economy started to pick up.
Figure 2.Delinquency Rates on Loans and Leases at US Commercial Banks
Source: Federal Reserve.
And in my view, the FSA’s prudential policy measures did make at least some contribution to removing these uncertainties. In the early phase of the crisis, Japan did not have in place effective frameworks for disclosure and provisioning with respect to non-performing loans, giving financial firms incentives to postpone the disposal of their non-performing loans for fear of the resulting losses. Based on this bitter experience, the FSA improved disclosure requirements, clarified the rules on write-downs and provisioning, and put in place a prompt corrective action scheme.
Admittedly, this lesson cannot apply directly to solve the difficulties currently faced by the authorities of advanced financial markets, as markets and products have become much more complex. Nevertheless, I think Japan’s this experience does provide some food for thought to the current discussions on ways to move forward.
How Should the Global Financial System Be Strengthened?
In parallel with short-term measures, the world’s financial regulators are advancing medium-term reforms to strengthen financial regulation. As the global financial markets seem to be bottoming out, the focus of policy discussions has shifted to regulatory reforms with a view to avoiding the recurrence of the same kind of crises.
Reflecting the cross-border nature of the current crisis, active discussions are underway internationally, led by the G-20, the FSB, the Basel Committee and others. An international consensus seems to be emerging gradually, as indicated in the G-20 Leaders’ Statement in Pittsburgh in September 2009 and the Communiqué issued by the G-20 Finance Ministers and Central Bank Governors in November 2009.
Next, I will touch on a couple of main issues under discussion: bank capital requirements and how to treat what are called Systemically Important Financial Institutions (SIFIs).
Building High Quality Capital and Mitigating Pro-Cyclicality
“Excess is as bad as shortfall”
“Never go to excess, but let moderation be your guide”
—Marcus Tullius Cicero
Since Lehman collapsed and Western government injected public money into a number of major Western banks to stabilize the global financial system, the perception among policymakers and market participants has been that both the level and quality of bank capital was too low and therefore the existing capital requirements must be strengthened. Thus the Pittsburgh Summit statement says that stronger capital standards are at the core of the G-20’s multi-faceted financial regulatory reform (Group of Twenty (2009)). In response, the FSB reported to the G-20 that the Basel Committee was working to develop new rules by the end of 2009, and conduct impact study and calibration work in 2010 to establish a credible minimum. They are also committed to ensure that capital requirements operate in a counter-cyclical manner, so that financial institutions would be required to hold capital buffers above the minimum requirements during good times that can be drawn down during bad times. The Basel Committee agreed to develop concrete proposals to this end, including forward-looking provisioning and a counter-cyclical buffer to contain excessive credit growth in boom times (Financial Stability Board (2009a, 2009b)).
Japan is member of the G-20, the FSB, and the Basel Committee. We are committed to take necessary measures to make the global financial system more resilient. Naturally, they would include those to enhance the quality and quantity of capital of financial firms.
Nevertheless, I am concerned about the recent discussions at international fora because the proposed measures seem to be strongly biased to increasing the level of bank capital or, more simply, the capital ratio. Let me summarize these concerns into the following three points:
First, I am skeptical that increasing the overall capital ratio would fix the deficiency of the existing regulatory framework. It is essential that regulatory strengthening address the root causes of the current crisis. From this standpoint, we do think that some strengthening in capital requirements is necessary, particularly with regard to the trading book and securitization. However, raising the capital ratio itself would not be able to correct the misaligned incentives. Without addressing the problems of insufficient risk capture, and lack of transparency and market integrity, higher capital could end up merely seeding the next crisis. In the absence of the incentives for strengthening risk management, financial firms could be induced to undertake much riskier businesses to meet the demand for higher returns from investors who supply additional capital. The regulatory tightening should focus more on strengthening risk capture, not on the number of the ratio.
It is not appropriate to impose, across the board, a new regime that is structured by assuming a specific business style employed by only a segment of banks. The losses incurred from securitized products greatly differ between large and complex banks in the US and Europe on the one hand, and other banks, such as Japanese banks, on the other (Figure 3). Given that the crisis has severely affected the former banks, it may be inevitable that the discussions on reforms centre on how to cope with the business models taken by these banks. In fact, however, few other banks take that route. In these circumstances, simply increasing the capital ratio regardless of the risks associated with different kinds of business could impair the banks’ capacity to lend to ordinary companies in need of financing.
Figure 3.Losses Incurred from Securitized Products
Source: Banks’ published financial results.
Also, it is debatable whether the markets actually see the amount of overall capital as the determinant in assessing the soundness of financial firms. Admittedly, the Tier 1 ratio of three Japanese megabanks is on the lower side. If you look at CDS spreads together, however, there are a lot of large banks that set aside much more capital but are deemed by the markets as more risky than Japanese banks (Figure 4).
Second, the strength of the regulatory package must be calibrated to an appropriate level. This is critical to ensure that these measures, seen as a whole, will not adversely affect the nascent economic recovery.
To this end, a quantitative impact study is essential in assessing the cumulative effect of the comprehensive package of regulatory measures not only on soundness of the financial system but also on sustainability of economic growth. So far, discussions on how to strengthen bank capital are made in various subgroups under the FSB and the Basel Committee. For example, there is a subgroup on the definition of capital; there are some groups working on pro-cyclicality; and yet a few others on systemically important banks, and so on. Each individual proposal made by these groups may have sound rationale; but there may be a fallacy of composition. Thus, it is absolutely essential that the Committee (or the FSB) ensure that these measures combined are sound as a whole package. The cumulative effect of these measures on the level of capital requirements could turn out to be too high. Then, this could seriously affect bank behavior and the real economy.
Figure 4.CDS Spreads and Tier 1 Ratio
In addition, the timing of implementation needs to be carefully judged in terms of whether “financial conditions improve and recovery is assured,” (Group of Twenty, 2009) depending on national circumstances.
Third, the design of counter-cyclical capital buffers warrants careful consideration. Some regulators argue that a higher rate including the buffer could be set as “target,” as opposed to the minimum, implying that meeting the target is not an absolute requirement. However, such a target may not be appropriate and may even be undesirable. Regardless of such a rhetorical difference, markets would deem the target rate as if it were the new minimum. This would require banks to set aside ever more capital, even during difficult periods, which could merely exacerbate the problems I just mentioned.
Reducing Moral Hazard Posed by SIFIs
“The awareness of the ambiguity of one’s highest achievements (as well as one’s deepest failures) is a definite symptom of maturity”
“Intolerance of ambiguity is the mark of an authoritarian personality”
—Theodor W. Adorno
The bankruptcy of Lehman and, especially, the bailout of AIG (American International Group, Inc.), brought about an enormous negative effect on the global financial system and the world economy, leaving the impression that such large financial firms are too big to fail. This could cause serious moral hazard among large financial firms, giving them incentives for further excessive risk taking.
In light of this experience, arguments have been put forward at the G-20 and the FSB that such firms need to be regulated so that the negative externality they can bring about is internalized (Group of Twenty (2009), Financial Stability Board (2009a)). The proposed measures include tighter capital and liquidity requirements, preparation of contingency and resolution plans (called “living will”), separation of some types of trading and investment banking activities from commercial banking, shrinking of the firms’ size, and funding of resources needed to preserve financial stability in case of failure.
These arguments are theoretically understandable and indeed appealing. In my view, however, careful consideration is warranted if the regulations from this standpoint are to be introduced, and regulators should not rush to conclusions under pressure. This is because the measures contemplated could well have unintended effects on credit flows and the broader economy, if not appropriately designed.
Here, I will touch on three issues in turn, that is, identification of SIFIs, a living will, and surcharges or cost-sharing of externalities.
Identifying “Systemically Important” Financial Institutions
The first issue is whether we can identify and, if we can, should specify SIFIs in an articulated manner. On this issue, a report has been prepared recently by the IMF staff and the secretariats of the BIS and the FSB and submitted to the G-20 Finance Ministers and Central Bank Governors (IMF, BIS, and FSB (2009)).2 The report defines systemic risk as a risk of disruption to financial services caused by impairment within the financial system that has the potential to have serious negative consequence for the real economy, and identify (i) size; (ii) substitutability; and (iii) interconnectedness as the three key criteria in identifying the systemic importance of financial institutions.
It is expected that further considerations will be made at the FSB and the Basel Committee to put this concept into actual regulation. In light of the experience just more than a year ago, and to avoid causing moral hazard among major financial firms, one cannot deny that something must be done to address this issue. At the same time, we must also recognize that the matter is not straightforward, and that a mechanistic approach just for the sake of simplicity could be detrimental to achieving financial stability. The IMF-BIS-FSB report already mentions this, but let me highlight three points that regulators should keep in mind.
The assessments of systemic importance necessarily involve a high degree of judgment. As the report says, the assessments will be conditioned by a number of considerations, including the robustness of financial supervision and infrastructure. Although we usually say “too-big-to-fail,” size alone cannot justify systemic importance. If we reflect on our experience in the 1990s, it was the default of a medium-sized securities house in the interbank market that triggered Japan’s banking crisis. At the height of the crisis, the failure of even a small local bank could have resulted in a system-wide meltdown. Determination of systemic importance would therefore be situation-dependent.
Domestic and cross-border banking should be separated in the assessment. A financial institution that is systemically important in the domestic setting may not be so in the global context. This distinction should be relevant in part because a robust framework for cross-border crisis management and bank resolution is almost non-existent at this point in time. We should therefore distinguish the former from the latter, e.g., in terms of their composition of assets.
Constructive ambiguity could be more desirable than explicit identification of SIFIs with a view to preventing moral hazard. Identifying SIFIs risks being interpreted by the markets as granting a “too-big-to-fail” status to these financial firms. What is called “cliff effect” is another concern, as clear distinction between SIFIs and non-SIFIs could induce regulatory arbitrage.
The idea of a “living will” in the context of financial reform seems to have been floated initially by Lord Turner, Chairman of the UK’s FSA (Financial Services Authority), earlier this year.3 As I understand it, a “living will,” namely a contingency plan and a resolution plan, would aim to reduce the probability of failure of a SIFI and, in case of a failure, enable the authorities to resolve it without causing systemic disruption and without injecting taxpayers’ money. This concept has been incorporated into the FSB’s work program to address the problems associated with SIFIs, and the FSB plans to submit final recommendation in October 2010 (Financial Stability Board (2009b)).
My concern is that expectations for a “living will” seem too high. Contingency planning and raising preparedness in anticipation of time of stress is highly important. This is the hallmark of risk management and supervision, and there is no need to come up with a separate plan. The issue here is the extent of detail in such planning exercise. Our experience suggests that a financial institution’s will for survival often leads to last-minute deals which could easily make a pre-determined plan completely out of date. Then the plans would turn out to be completely useless, even if they had been prepared in detail.
A reasonable target may consist of more practical efforts, such as: (i) identification of impediments to emergency operations; (ii) enhancement of information sharing; and (iii) identification of jurisdictions where transactions are booked. Without such examination and common understanding of the extent of details, “living wills” could end up as excessively time- and resource-consuming exercises without commensurate expected benefits.
In this respect, I think that the Lehman bankruptcy offers an excellent case study to identify what the crucial issues are and what sort of measures are needed. Drawing from that experience, possible challenges for orderly resolution of SIFIs would include: the booking of transactions managed cross-border; the scope of application of domestic insolvency laws, the international consistency of these laws, and their effects on supervision of overseas subsidiaries; and the extent of information sharing and the nature of decision-making process in resolution of these firms. In fact, Lehman’s transaction in Japan were booked mostly in London, and that had significant legal implications when Lehman collapsed, given the difference in bankruptcy laws between the UK and Japan.
Capital Surcharge and Cost-Sharing
A prevalent argument goes that SIFIs should be subject to stricter capital requirements so that they would be able to withstand stress on a going concern basis, as it is difficult to make such firms to fail. One idea could be to impose a surcharge on these institutions in the form of an added capital adequacy ratio according to their systemic importance.
I acknowledge that this argument has a point. However, I would also say that a capital surcharge is, even if it is a solution, only part of a package of solutions. We should recall that a number of failed major institutions, including Lehman, once boasted their high capital adequacy ratio. As it turned out, the level of capital they set aside was too low. But this is mostly because the denominator of their capital adequacy ratio, namely their risk-weighted assets, was calculated as a very small amount. It may be the result of regulatory arbitrage or shortcomings in their risk management systems. This means that static compliance with strengthened capital requirements is neither a replacement nor a substitute for good, proper, and dynamic risk management. It should also be noted that financial firms often fail because of illiquidity rather than insolvency. This suggests that considering introduction of liquidity surcharge would be no less important.
I also wonder whether it is unrealistic and even undesirable to make each financial firm prepare for any future crises by building capital upfront. As mentioned earlier, requiring more capital could induce banks to take on risks to achieve greater return on equity while curtailing provision of less profitable but less risky credit to the real sector. Demanding too much capital on banks might increase dead weight cost to the economy and reduce social welfare. Financial stability is a public good: therefore, public support to sustain financial stability should not be ruled out categorically. The issue is how to share the burden of huge tail losses and its associated systemic impact between the financial sector (and borrowers) and the public sector (and taxpayers).
As we have seen, the problems of SIFIs are not simple and straight-forward. It is not the issue of identifying such institutions in a mechanical manner and imposing just a capital surcharge. Dealing with SIFIs would require a much more holistic approach, which would also involve more intensive supervision of risk management practices, more focus on liquidity, a robust framework for resolution of financial firms and realistic burden-sharing of the cost of financial crises.
Financial Services Authority (2009) Turner Review Conference Discussion PaperOctober. (Available at http://www.fsa.gov.uk/)
Financial Stability Board (2009a) Improving Financial Regulation: Report of the Financial Stability Board to G-20 Leaders 25 September 2009. (Available at http://www.financialstabilityboard.org/)
Financial Stability Board (2009b) Progress since the Pittsburgh Summit in Implementing the G-20 Recommendations for Strengthening Stability 7 November 2009. (Available at http://www.financialstabilityboard.org/)
Group of Twenty (2009) Leaders’ Statement: The Pittsburgh Summit September 24-25 2009. (Available at http://www.pittsburghsummit.gov/)
International Monetary Fund Bank for International Settlements and Financial Stability Board (2009) Guidance to Assess the Systemic Importance of Financial Institutions Markets and Instruments: Initial Considerations.
SatoTakafumi (2008) Global Financial Crisis and Japan’s Experience in the 1990s. Keynote speech at the “Symposium on Building the Financial System of the 21st Century” co-organized by Program on International Financial Systems at Harvard Law School and International House of Japan October 25. (Available at http://www.fsa.go.jp/en/)
This section, particularly the reference to commonalities between the global financial crisis and Japan’s banking crisis in the 1990s and the five lessons from the latter crisis, is largely based on Sato (2008), to which I contributed during its drafting.
The report covers also assessing the systemic importance of markets and instruments, but this presentation focuses on that of financial institutions only.
Lord Turner’s idea has been elaborated further in Financial Services Authority (2009).