Remarks by Philip S. Bates
Global Ratings Development, Standard & Poor's Rating Services

While not as heated and well documented as the debate on the role of major international financial institutions, such as the IMF and World Bank, the Asian financial crises and the ensuing stress in other emerging markets have also drawn some attention to international credit rating services. As in the case of major multilateral financial institutions, conflicting views have been expressed about the influence of credit rating services on the course of the Asian crises, and in many instances these conflicting views reveal some misconceptions by about the role of credit ratings. In addition, like the international institutions, credit rating services have learned some lessons from the Asian crises, which in the case of Standard & Poor's have led to some refinements to our sovereign rating methodology. More importantly, the Asian crises have highlighted the need for an improved financial architecture, including better transparency and information disclosure, something that is critical not only for the effectiveness of credit ratings, but for international capital markets more generally. Finally, and perhaps most importantly, as in the case of many past booms and busts, the build up and fallout of the Asian crises point to the need for adequate credit risk differentiation in the pricing of capital flows.

As background for our discussions, I would like to take a few minutes to cover the following three areas:

1. the role of credit ratings and their impact on the evolution of the Asian crises and more generally emerging markets;
2. a summary of some methodological refinements emanating from recent volatility of private capital flows, and the closely related financial crises that have enveloped a number of major Asian economies;
3. from a credit rating service perspective, a few comments on some of the critical areas that need to be addressed in devising an improved international financial architecture to cope with volatile capital flows and the contagion effects that are likely to be an enduring aspect of the international financial landscape; and
4. some final remarks on why the current environment reinforces the importance of credit risk differentiation.

The Role of Ratings

Beginning with the role of ratings during this crisis, some market participants, especially those most vulnerable to the current global flight to quality, have complained that credit ratings have been destabilizing to global capital flows. These criticisms have largely been of two contradictory camps. First, some observers have stated that sovereign credit ratings provided too little warning of risk, encouraging capital to flow where it was neither efficiently deployed nor adequately compensated for risk. Second, other have charged that rating changes on sovereign governments and financial institutions have precipitated economic crisis or the failure of firms by providing warnings that, in turn, have caused economically debilitating capital flight that squeezed liquidity and induced default. Such criticism, while acknowledging the influence of credit ratings, reflects an inadequate understanding of the role of credit ratings in capital market development and the behavior of financial markets.

Credit ratings are opinions of the probability of default on a company's or government's debt. They speak to only one aspect of the financing and investment process, credit risk. The predictive value of Standard & Poor's credit ratings to differentiate creditworthiness is well documented in our annual default surveys, which confirm the high correlation between ratings and the incidence of default. This predictive value is further confirmed by secondary bond yield spreads, which reflect, sometimes to a significant degree, the credit risk differentiation inherent in ratings. Reflecting these benefits, securities market and other financial regulators in many developing economies are encouraging credit rating services to play a more active role in facilitating the formation of efficient domestic financial markets. Consistent with this trend, Standard & Poor's has established branch offices and affiliates in a dozen emerging markets over the past 5 years.

The notion that sovereign and bank credit ratings have been destabilizing to global capital flows is not apparent from the recent Asian experience, nor is it seen in the capital flows to and from Latin America over the past decade. Estimates of private capital flows to the five Asian crisis economies during 1997 indicate that the largest flow reversal pertained to commercial banking activities. Banks withdrew $21 billion from the five countries last year, compared with net inflows of approximately $55 billion in 1996 and $ 50 billion in 1995. Outflows in 1997 would have been even higher had the G-7 central banks not orchestrated the acceptance of Korean bank deposit rollovers, thus averting a liquidity induced default by the Korean government in late December. Equity flows were the second most important reversal for the Asian five. Outflows of $12 billion last year followed inflows of $12 billion and $11 billion in the previous two years. Neither bank or equity portfolio flows are heavily dependent on independent credit ratings. Banks, in the normal course of business, conduct their own credit analysis, and make independent judgements about the extension of new credits or the rollover of existing credits. Equity portfolio flows may occur for reasons similar to those considered in making credit decisions, but these flows occur independently of credit rating decisions. In contrast, net flows of non-bank private creditors, including bond issuance which is mostly heavily influenced by credit ratings, remained a positive $14 billion last year, slowing from $18 billion in 1996.

The Latin American experience is perhaps more telling, especially when viewed over the past 5-10 years. Latin sovereign credit ratings have been largely stable and predominately at speculative grade levels of 'BB' category or lower during the 1990s, reflecting substantial external debt burdens, but generally improving fundamentals. In contrast, capital flows have fluctuated dramatically as market participants responded to events, particularly the Mexican peso crisis. Indeed, Standard & Poor's rating of the Mexican government's foreign currency debt was reduced only one notch to 'BB' from 'BB+' during the crisis, but net private capital flows sharply reversed from a positive $12 billion in 1994 to a negative $14 billion in 1995.

Methodological Refinements

The volatility of private capital flows and its critical role in the Asian and other emerging market crises has resulted in some refinements to Standard & Poor's methodology for evaluating sovereign risk.

1. Vulnerability to Confidence Sensitive Capital Flows

First, Standard & Poor's is placing greater emphasis on the maturity structure of public and private sector external debt and reliance on confidence sensitive private capital flows in situations where deteriorating private sector asset quality can result in substantial pressure on central bank reserves. The Asian experience has indicated that high domestic leverage, together with reliance on short-term external debt, increase the vulnerability of the banking system and, in mm, the sovereign government to changing investor sentiment, which can pose especially severe pressures for countries with inflexible exchange rate regimes. Under such conditions, Standard & Poor's will give greater analytical emphasis to the vulnerability and cost of maintaining the exchange rate regime, and will undertake sensitivity analysis of the degree of corporate and banking sector financial stress in the event of a large exchange adjustment.

2. External Debt and Reserve Management

Second, Standard & Poor's has increased the analytical weight assigned to certain aspects of external debt and international reserve management policies, controls and information systems. With regard to external debt, lax oversight and poor disclosure of private sector borrowing heightened the vulnerability of many countries to funding and exchange rate stress. Accordingly, the timeliness, coverage and level of detail available on private external indebtedness will receive more attention. In the case of international reserves, imprudent use of reserves and poor disclosure aggravated the financial crises in Thailand and Korea.

3. Banking Sector Contingent Liabilities

Last, but perhaps most important, Standard & Poor's has intensified its analysis of the private sector, especially the banking system, as a contingent liability of the sovereign government, in particular, S&P has undertaken extensive research to develop a series of leading indicators of systemic banking crises and has, in turn, devised a technique for estimating the potential cost of financial stress to the government under a reasonable worst case scenario. Based on the analysis of banking crises in 29 countries over the past two decades, Standard & Poor's research identified four major indicators with the greatest predictive value:

1. the trend in credit growth to private and public sector non-financial enterprises;
2. the level of corporate and household sector indebtedness;
3. the pace of asset-price inflation; and,
4. the trend in external funding by financial institutions.

Based on these indicators, Standard & Poor's divides banking systems into five broad risk categories based on their vulnerability to asset quality pressures during periods of economic slowdown or recession. The rankings are expressed in terms of the potential level of gross problematic assets (or GPAs) that the system may accumulate during an economic downturn. The estimated range of potential GPAs are considered the risk assets of the financial system and, in turn, a proxy for the level of direct and indirect costs to the government and the economy under the worst case scenario. The larger the size of the financial sector, all else being equal, the greater the contingent liability for the government.

Not surprisingly, the analysis reveals a substantial financial burden for the five Asian crisis economies, with the upper bound estimates of gross problematic assets indicating a contingent liability that could more than double the government's debt burden. Perhaps more interesting is the fact that the analysis shows that the banking sector poses a large and growing potential liability for China and Hong Kong, revealing a potential liability that could increase government debt burden by four to six times current levels.

Improving the Financial Architecture

Many of the analytical refinements I have just touched upon are closely inter-related with many of the same issues that are driving the current efforts to improve the international financial architecture. In this regard, from a rating service perspective, perhaps the most important issue is the need to improve transparency and disclosure.

Credit ratings are dependent upon sufficient levels of disclosure and the timeliness of such disclosure to be effective. In some emerging markets, including all of the countries that have endured financial crisis over the past 18 months, inadequate disclosure and lack of transparency in relationships within the private sector and between the private and public sectors not only contributed to the creation and intensification of financial pressures, but also hindered the effectiveness of credit ratings.

For example, Standard & Poor's believes that regulatory and moral hazard issues contributed to the build-up of short-term, external debt by the banking systems in Indonesia, Korea, Malaysia and Thailand where historical precedents and the regulatory environment indicated a high likelihood of government support for banks. Moreover, to the extent that anticipated government support was a credit factor in such lending, the conditions for IMF support, particularly measures to instill greater market discipline in the banking sector, contributed to an acceleration of the capital outflows and the intensification of liquidity pressures.

Clearly, there is a commonality of interest between rating services, international financial institutions, investors, and governments to improve disclosure and transparency. Efforts are needed on many fronts, but the priorities from Standard & Poor's perspective would cover such broad fields as improving bank supervision and regulation, auditing and accounting standards, bankruptcy laws, corporate governance, fiscal transparency and data dissemination.

Risk Matters

In concluding my remarks, I would to briefly comment on the relationship between credit ratings and interest rate spreads for affected borrowers over the course of this evolving crisis.

Prior to the onset of the Asian crises, interest rate spreads for borrowers from the five Asian crisis economies had narrowed to levels well below the normal spreads for issuers with credit ratings in lower investment grade and speculative grade range, as was the case for Standard & Poor's ratings of the vast majority of active private and public sector borrowers in these countries. In general, while spreads did progressively widen as you moved down the credit rating spectrum, the interest rate distinctions were very small relative the differences in credit risk.

At present, the flight to quality has widened spreads to levels not witnessed in over a decade, but with little distinction between borderline investment grade issuers and highly speculative issuers with a significant near-term risk of default. As is often the case, both during the boom and the bust, spreads have not appropriately reflected the differences in credit risk signaled by rating opinions, encouraging excessive and indiscriminate inflows during the boom and the reverse during the bust. Clearly, one lesson from the crisis, and alas not a new lesson, is that credit risk matters and capital flows need to be priced to reflect differences in risk in order to ensure efficient allocation of capital, and to help lessen the frequency and amplitude of future financial crises.

The importance of this lesson is underscored by the fact that today S&P has speculative grade credit ratings on the sovereigns that govern more than 50% of the world's population. Accordingly, market participants should expect sovereign defaults, like the recent case of Russia, to continue to happen.