What needs to be fixed?
It is now widely recognized that in the run-up to the crisis, there was a significant under-appreciation of systemic risk, so much so that many viewed policymakers as having established an era of sustained and stable expansion, labeled the “Great Moderation.” With the benefit of hindsight, low nominal interest rates, abundant liquidity, and a favorable macroeconomic environment encouraged the private sector to take on everincreasing risks. Financial institutions provided loans with inadequate checks on borrowers’ ability to pay and developed new and highly complex financial products in an attempt to extract higher returns. Many financial regulators and supervisors were lulled into complacency and did not respond to the building up of vulnerabilities.
As a result, financial systems and transactions became distorted along several dimensions:
1. The financial system grew highly complex and opaque. Lack of transparency and limited disclosure of the types and locations of risks made it difficult to assess the extent of exposures and potential spillovers. This opacity magnified the shock to confidence as the crisis unfolded. As the financial sector expanded as a fraction of GDP (Figure 1), an increasingly large portion of financial activity did not seem to serve the needs of the real economy.
2. The financial system became over-leveraged and heavily interconnected. Short-term incentive structures undermined good governance and encouraged excessive risk taking. Actual leverage was even greater than was apparent, in part because it was embedded in instruments in ways that were not transparent and in part because regulatory requirements did not capture key risks. This meant that capital was inadequate as a buffer against the drop in asset prices. The interconnectedness of institutions meant that the shocks were propagated across the system, both domestically and globally.
3. Liquidity risk was also higher than recognized. Financial firms and key markets relied increasingly on shortterm, wholesale funding and took on excessive maturity mismatches while failing to build adequate liquid asset buffers.
4. Large complex institutions enjoyed the benefits of being “too important to fail.” The lack of market discipline allowed them to borrow at preferential rates, operate with higher levels of leverage, and engage in riskier activities.
5. In addition to traditional capital market instruments, financial intermediation has increasingly shifted to the “shadow” banking sector. Relatively unregulated nonbank financial institutions and markets thrived in large part because they avoided the more stringent requirements imposed on banks(Figure 2).
Some of these distortions are being unwound as part of the de-leveraging process. Financial institutions have been re-building capital and liquidity buffers, and have been required to bring some of their off-balance sheet activities back onto their balance sheets. The concern remains, however, that many of the structural characteristics that contributed to the build-up of systemic risks are still in place today. Perhaps most worrisome is that the large-scale public support provided to both large institutions and markets, a contingent liability equivalent to about one fourth of advanced economies’GDP, has exacerbated the moral hazards and perceptions that certain institutions and markets are “too important to fail.”The challenge, therefore, remains to establish a policy framework that can both sustain growth and reduce the severity of boom-bust cycles. Some argue that less-volatile economic growth is likely to come at a cost in terms of risk-taking opportunities and innovation in the finan-
cial system and, therefore, be associated with a lower growth path. But, encouragingly, recent empirical work suggests that the trade-off is nearly absent if the large output costs of financial crises are taken into account. While this may lead to consistently lower risks and lower returns in the financial system, these need not be associated with substantially lower output in the short term and may yield significant net output gains as well as stability in the medium term.
The financial oversight framework should be strengthened to help reach these end-goals in terms of depth, breadth, and global consistency, and comprise five key goals:
1. Strong micro-prudential regulation that is globally coordinated. It should strengthen the resilience of financial institutions, ensure as much as possible a level playing field of regulations, and minimize regulatory arbitrage that could be damaging to global financial stability.
2. Effective supervision. The IMF’s work on assessing financial sector standards suggests that countries often lag behind in meeting good practices of supervising key risks, taking corrective action in a timely manner, and enforcing

and sanctioning noncompliance.
3. A robust and globally consistent cross-border resolution framework. An enhanced international coordination framework for cross-border resolution is essential. Such a framework is also needed to ensure that financial institutions that are “global in life” do not become“national in death.”
4. A macro-prudential dimension. Such an approach is needed to reduce the systemic risk contribution of indi-

vidual institutions and markets, and to encourage the build-up of strong buffers of capital and liquidity in good times, which can be run down during periods of stress. An effective macro-prudential framework will depend critically on addressing the flaws in the Micro-prudential regime.
5. A larger regulatory perimeter. The perimeter should be enlarged to cover banks and non-banks alike, so that weaknesses in the entire financial system can be addressed. Consistency in the application of regulations across different types of financial institutions producing similar products is critical to avoid risk being shifted into the shadows.
Making individual firms more resilient
The remainder of this paper examines the current regulatory reform agenda against the goals just laid out for the regulatory framework and what needs to happen to bring it about. Focus of this section is on micro-prudential measures that aim to make individual financial institutions more resilient, or allow them to fail smoothly. The next section discusses macro-prudential policies that aim at making the overall financial system more resilient.
Banks entered the crisis with inadequate capital buffers and suffered severe losses, some of which only became evident as events unfolded. From the start of the crisis, the IMF has been providing objective assessments of the size of write-downs in global banks in an effort to keep the international agenda focused on reforming the capital framework for banks.
At the core of the reform program endorsed by the G-20 are measures aimed at making individual banks less likely to fail through actions to reduce leverage, build more robust capital and liquidity buffers, and limit maturity mismatches. Key measures proposed by the BCBS include:
1. Improving the quantity and quality of capital, so that it can absorb losses more easily;
2. Ensuring that capital requirements are more closely associated with the risks they are meant to protect against, and, in particular, capture more fully market risk, counterparty credit risk, and risk in securitized portfolios;
3. Introducing a leverage ratio as a credible supplementary measure to the risk-based requirements;
4. Introducing measures to protect against liquidity shortages by holding more assets that could be liquidated rapidly, and lowering rollover risks by limiting asset/liability maturity mismatches and less-secure forms of funding.
The IMF has been supportive of BCBS proposals. In particular, the adoption of an enhanced market-risk framework for internal models is critical to reduce incentives for regulatory arbitrage between banking and trading books. Banks are expected to comply with the revised trading book requirements for better risk recognition and capital coverage by end-2011 (originally proposed for end-2010, but subsequently revised).
A key concern with the proposals has been whether the reforms would lower the availability, or raise the cost, of credit and, hence, harm economic growth before the recovery is well established. Work on the macroeconomic impact of the reforms, recently published by the BCBS and the FSB and conducted in collaboration with the IMF, suggests that higher bank capital and liquidity requirements would have only a modestly adverse temporary impact on aggregate output and clear net longterm economic benefits.10 According to the study, a phasing-in period of the reforms over four years would minimize the transitory impact of the reforms on

output.
We welcome the recent proposals of the BCBS which represent a substantial improvement in the quality and quantity of capital in comparison with the pre-crisis situation. Common equity will represent a higher proportion of capital and thus allow for greater loss absorption. In particular, the required minimum will increase to 4.5 percent from 2 percent under existing standards and will be complemented by an additional 2.5 percent capital conservation buffer (composed of fully loss absorbing capital) which would restrict distributions as banks approach the minimum. Also, the amount of intangibles and qualified assets that can be included in capital will be limited to 15 percent.11 Phase in arrangements have been developed to allow banks to move to these higher standards mainly through retention of earnings.
As the global financial system stabilizes and the world economic recovery is firmly entrenched, phasing out intangibles completely and scaling back the transition period should be considered. This will raise further banking sector resilience to absorb any future shocks that may lie ahead. In our view, it would have been desirable to provide for the eventual exclusion of all intangible assets from capital, and, under the baseline scenario of the World Economic Outlook, shorter phase-in periods would not have placed undue pressure on the banking system and the economy. The longer financial institutions remain with lower buffers, the higher the burden will be on supervisors.
Since the outset of the crisis, the focus of near-term policy action has been on strengthening the regulatory framework. But regulations are only part of the solution, and it is through supervision that the authorities enforce compliance with the rules.
Good supervision requires the ability and the will to act, both of which had often been missing in the run-up to the crisis. In no jurisdiction will this ever be an easy task and may require forcing the board of a financial institution to direct management to cease an activity or to replace key managers. Proactive supervision is adaptive to changing conditions and can observe when activities are taking place on the fringe of the regulatory perimeter. The supervisory mandate needs to carry over to systemic concerns, supervisory bodies must be given the authority and mandate to act not just when individual institutions pose undue risks, but also when the entire system is behaving in a manner that jeopardizes systemic stability.
So far there has been little progress on this front. It is thus encouraging that the Toronto G-20 Summit declared supervision a key pillar of the financial reform agenda and gave an explicit mandate to develop it. This focus on strengthened supervision is very important, not just for banks but for the broader financial system. Indeed, evalu- ations of national oversight frameworks as part of the Financial Sector Assessment Program (FSAP) show that countries often do not meet good practices in supervising key risks, taking timely corrective action, or enforcing and sanctioning noncompliance (Figure 3). Thus, it is critical that supervisory agencies be provided with the mandate; resources; and authority, along with accountability, to carry out their tasks. Adopting guiding principles for supervision would be helpful in this respect, and would support supervisors carrying the burden of preventing a new cycle of leverage and excessive risk taking while the new Basel rules are being phased in.
Assessments of systemic importance are also instrumental in determining the appropriate boundaries of regulation. Enlarging the regulatory perimeter will help avoid a repeat of build-up of systemic risk outside the boundaries of official oversight. Consistency in the application of regulation across financial sub-sectors producing similar products is equally critical. For instance, money market mutual funds proved to be of systemic importance during the crisis and, to the extent they provide bank-like services and perform maturity transformation like banks, they should be overseen in a manner that is consistent with that of banks.
One of the key lessons from the crisis is how much damage can be inflicted on the system when the market infrastructure either breaks down or is insufficient, and information on which to base financial decisions is absent. The market disruptions, in the unsecured inter-bank market, the repurchase market, and the over-the-counter (OTC) derivatives markets, for instance, caused by the failure of Lehman Brothers show the importance of resilient markets. The transactions associated with Lehman that were unwound easily were those that had been placed in formal clearing facilities, whereas bilateral contracts took months, in some cases, to sort out. The inability of market participants to see the build-up of risks in the estimated US$600 trillion OTC derivatives market, most specifically in the smaller credit default swaps market, was in part due to the bilateral nature of the trading and the absence of transparency, even to those in the official sector.
Hence it is equally important for regulatory reforms to tackle system-wide problems that emerge in markets. This requires a close look at the functioning of afflicted markets, what information is provided to participants, and when and how trading, clearing, and settlement are conducted. Again, there has been some movement forward in this area of the reform agenda, but it remains centered on fixing identifiable problems in each market, without a holistic approach.
The likelihood that vulnerabilities in institutions or markets reach a level where a systemic event can occur is heightened by the amplification of cycles—credit cycles, and more broadly, business cycles. This is another critical element that must be considered in reform efforts, particularly because some of the pro-cyclicality arises from financial regulation, accounting standards, and business practices.27 Work is underway to design and calibrate specific macroprudential tools that will address procyclicality, but more analysis is needed. The BCBS has requested comment on the basis for computing countercyclical risk weights and more generally, how to construct countercyclical capital charges. As well, further work is needed to calibrate micro-prudential measures such as loan-tovalue ratios so that they can effectively counter real estate booms and busts. Accounting standards also need to be reexamined (and converged internationally) to reduce the pro-cyclicality of loan-loss provisioning and fair value accounting for financial instruments. The FSB has introduced principles and standards to address the pro-cyclicality of compensation, but more efforts are needed by institutions and national authorities to effectively align pay with long-term, risk-adjusted returns.
More generally, there is also a need to develop guidance on the governance and institutional arrangements that will be needed to effectively integrate monetary and macro-prudential policies into coherent frameworks. How to organize institutions will have a bearing on whether central banks should use interest rate policy or other monetary measures to contain the build-up of financial imbalances, especially those related to excessive credit growth or asset-price bubbles. A number of principles can help guide and frame the debate: (i) the financial stability objective is not always aligned with the price stability objective and thus requires a separate set of macro-prudential policies and instruments; (ii) the central bank will need to play a key role in the development and use of macro-prudential policies, whether or not it is the main financial regulator; (iii) financial stability considerations need to be better incorporated into the monetary policy decisionmaking process; and (iv) official interest rates can lean in a non mechanistic manner against financial imbalances when pursuing price stability so as to render policy more symmetric during the business cycle and thus reduce the likelihood of boom-bust cycles.
Using interest rates to counter financial imbalances may risk increasing macroeconomic volatility and thus impose collateral damage to the real economy and, in some cases it may even lead to an increase in capital inflows. Still, the high cost of systemic financial instability shown by the crisis strengthens the case for “leaning against the wind” as a supplement to macro-prudential policies oriented towards preserving financial stability. The lack of understanding of transmission suggests that, for now, central banks should best utilize judgment. The combination of rising asset prices and rapid credit growth may warrant a higher policy rate than otherwise.