Print this
Three leading credit analysts explain why credit models failed to warn us of trouble—until it was too late.


Diane Petan
sovereign fixed-income analyst,
Goldman Sachs

I've spent most of this year working in Asia—in many of the crisis countries—analyzing the situation and advising the governments on credit marketing strategies to help them regain access to the international capital markets. Drawing on this experience, I'd like to present some thoughts about why many analysts and market participants missed the warning signs that led to the crisis. Some of these factors are quantifiable, and some of them are not. One of the main reasons the crisis was missed was that, as late as October 1997, most Asian countries were highly rated by the rating agencies. I'll use Korea as an example. Before the October 1997 crisis, Standard & Poor's rated Korea AA-, Moody's rated it A1 and Fitch IBCA rated it AA-. In the depth of the crisis, the S&P rating dropped to B+, Moody's to BA1 and Fitch IBCA to B-. The current ratings have now recovered somewhat, but are all still speculative grade.

In all fairness, the rating agencies were not alone. If you read the Street research from economists and fixed-income analysts, or high-profile reports and articles from individuals such as Paul Krugman, a well-respected international economist and authority on balance-of-payment crises, and even reports from the IMF itself, you would have heard and concluded that many of these countries had a clean bill of health through mid-to-late 1997.

Another reason the crisis was missed was that the risk premium on emerging-market bonds was quite low. Bond spreads in most emerging-market economies were tightening through mid-to-late 1997. At the time, investors believed this was consistent with credit fundamentals. Sovereign credit indicators for most of these countries were generally quite good, and the imbalances that did exist were not severe enough to indicate that a crisis of this magnitude was around the corner. For example, Korea was running large fiscal surpluses and, along with most other Asian economies, was reporting strong positive growth combined with low public-sector debt. In fact, Korea had the lowest government debt/GDP ratio compared with all of the G-7 countries, as well as extremely low external debt ratios relative to exports. These strong credit variables fueled investor confidence.

This leads us to one of the most important factors that was missed. We have to recognize that the Asian crisis was very much a private-sector problem, centered around fragile banks and over-leveraged corporates. Investors did not have the information needed to analyze the weakness in the banks and the corporates, and, quite frankly, it seemed that few investors were demanding it as a condition of purchasing the debt securities of these entities. Now it has become clear that the debts of many of these institutions are a contingent liability to the sovereign. This was made worse by the fact that many of the Asian banks and corporates had borrowed heavily in short-term foreign currency debt. With many of them operating in fixed-exchange-rate regimes, there was a sense of confidence that they would not be exposed to currency volatility—and thus believed it was safe to borrow in dollars and yen. We now know the serious negative repercussions of this liability structure during periods of extreme financial market volatility.

“In all fairness, the rating agencies were not alone. If you read the Street research and even reports from the IMF itself, you would have concluded that many of these countries had a clean bill of health.”

Poor governance was also a source of the crisis, but was not easy to measure and the risks were therefore underestimated. Many of these countries have had a long-standing tradition of alliance capitalism or government involvement in the private sector, combined with weak transparency and disclosure. This prevented many market participants from seeing the extent of the problems.

Creditor behavior was also unanticipated. The best example to illustrate this is Japan, which was the largest creditor to the region. The market was not able to predict the extent and costs of Japan's retrenchment on its lines of credit to Asian corporates and banks. The withdrawal of foreign credit has contributed to the serious contraction now evident in many Asian economies.

Another factor that is extremely difficult to measure is the vulnerability of the region to financial contagion. As market sentiment collapsed, the international investment community panicked, the contagion spread and the cost of the crisis grew. In addition, the market believed that sovereigns would not default, which also boosted investor confidence toward emerging markets.

A final complicating factor was the fact that capital continued to flow into these emerging economies up through the first phase of the crisis. This is explained by the fact that weaker growth and low interest rates in Europe and Japan made investments in those regions less attractive. Investors searching for higher yields were driven to emerging markets.

What is needed to correct these problems? I do not think there are easy answers. We could spend hours debating possible solutions. At a basic level, however, we must first restore confidence and investor sentiment to the regions in crisis, which will enable the stabilization of financial markets, restore access to the capital markets and ease the present liquidity crunch.

Investors and lenders need to continue to impose conditions on the funds and financing made available to these countries, so that they continue with aggressive reform programs. The ongoing efforts to strengthen their financial systems are key to restoring creditworthiness and investor sentiment. In the future, as capital flows back into these economies, stronger financial systems will allow for more efficient allocation of capital to productive areas of the economy.

I also think that investors must demand improved transparency and disclosure. No longer can we tolerate inadequate disclosure of financial statements from private-sector entities. The market cannot tolerate the lack of debt information in emerging-market economies, such as maturity and currency structures of debt. Why should the standards of reporting for foreign entities be significantly different from that of U.S. corporates and banks?

This leads me to the opinion that investors should demand risk premiums commensurate with the risks involved. If information is not available, emerging-market bond yields should reflect this. However, I appreciate that this is sometimes difficult to do when the market for these securities is rallying and total-return investors feel compelled to invest or face underperforming their benchmark indices. During such times, it is common to see investors feeling pressured to jump into credits they may not fully understand.

I also think it's important that we do not underestimate the importance of qualitative variables in risk models. That's particularly true in the case of Russia. Observing the events in that country over the last year, it is clear that the Russian system has suffered from a tremendous amount of corruption, inefficiency and opaqueness—these qualitative variables were behind the crisis that occurred in the middle of August. The market's expectations were much too high and were supported by thin strands of economic data such as modest increases in tax revenues, the stabilization of inflation and the ruble, and a strong stock market.

Bob Strong
executive vice president and chief credit officer,
Chase Manhattan Corp. and Chase Manhattan Bank

Iwould like to take a moment to drill down a little bit into more tactical issues. Certainly, the transparency of financial statements was poor, but it wasn't just that “the ratings agencies missed it.” A lot of people were having trouble dealing with it. As late as last August, a little more than 15 months ago, the IMF was citing Indonesia as a model of how the emerging-market economies should be run. Six months later, I was sitting in a conference room in Frankfurt trying to work out a restructuring. There were a lot of people who missed it along the way.

I think a big part of the problem in Asia was simple euphoria—the tendency over the last number of years for asset prices to escalate fairly continuously. But if you looked carefully, you would have seen some leading signs coming out of asset prices. The stock market in Thailand was giving indications 18 months before the crisis that people should be taking a more prudent course there. In fact, the external market-based prices were probably a better indicator on a leading perspective than some of the lagging indicators like the ratings of the country.

Everybody's increased ability to distribute fed the problem. There was a tendency on the part of a lot of players to say, “If I can sell it, I can originate it.” There were a lot of investors who had not dealt with some of these asset classes before. There was a lot of mutual fund money being thrown at emerging markets because, historically, those rates of return looked great.

“I don't think anybody envisioned the speed with which the contagion would sweep around the world. It was a lot easier for people to simply call up and say they wanted to move out of somebody's mutual fund.”

It's interesting to note that U.S. banks had less than 10 percent exposure in these countries. The traditional view is that the U.S. banks have really driven this problem by building up their balance sheets. But it was not so in this case. U.S. banks, however, were big in origination and distribution.

I don't think anybody envisioned the speed with which the contagion would sweep around the world. A big part of that was that we were dealing with a different investment class than we've historically had. It was a lot easier for people to simply call up and say they wanted to move out of somebody's mutual fund.

I also think the institutional view of the historical precedent was flawed. By that I mean a lot of multilateral institutions seemed a little nonplussed by the fact that you couldn't simply call six U.S. banks together and sit down in a room and solve the problem. In the LDC crisis, that was a lot easier to do. Now you had literally hundreds of creditors. A lot of them were European and Asian banks, and a lot of them weren't even banks. It was a lot harder to get your hands around the problem and try to drive a solution through. The solutions had to be much more market-based than they had been in the past, and since there was no market, that caused a bit of a problem.

The problems in the Japanese economy were also an important contributing factor. There was an awful lot of Japanese money that was flowing out of Japan in search of higher apparent risk-adjusted returns, or just plain higher returns. We had Japanese banks with 30 percent and 35 percent market shares in the credit to certain countries. We even had Korean banks with huge market shares in other Asian countries, so the knock-on effects were quite dramatic. Finally, I think the political forces that are now active could have responded more quickly and more effectively. I continue to believe that if we had focused a bit earlier on Thailand, and the United States had been more committed to stopping that early, it could have been effective. It was not.

It's hard to come up with lessons we can take away from this experience. I guess I'd start back up at the top of the list and say that transparency of financials and a weak banking system are key issues for anybody thinking about being a creditor to these areas.

Michael Pomerleano
senior capital markets specialist,
World Bank

For the past 18 months, this whole crisis has been my profession and possibly my obsession. My affinity for the region dates back to 1992–93, when I lived in Indonesia for two years and worked on restructuring the banking system. I believe the international financial system never appreciated the underlying risk in the region. Having lived there, I probably benefitted from “insider” knowledge—I have always felt there were great risks and vulnerabilities there. I'll try to summarize the experiences of the last six or seven years and explain the views I formed over this time.

A lot of people say, “Nobody saw it coming.” My sense is that all the signs were there. What happened was a failure in judgment in reading all the information. Clearly, Diane is right—there were deficiencies in the data. But I submit to you that there was sufficient cause to know that there were underlying weaknesses throughout Asia.

Part of the reason the crisis was missed by everybody is that everybody was preoccupied with government data, the fiscal balance, the money supply and the external balance. There was euphoria of “growth,” and there was vested interest in rapid growth. Rapid growth was hiding a lot of sins, and nobody looked at the corporate and banking sectors and recognized the risk embedded in the high leverage and low profitability of the corporate sector.

I'll be happy to share with you the findings of an article I wrote [“The East Asia Crisis and Corporate Finances: The Untold Micro Story,” Emerging Markets Quarterly, Winter 1998], which benchmarked Asian corporates against other corporates throughout the world. It used the publicly available 1992–96 financial statements of 1,500 corporates from around the world.

To do this comparison, I generated an Altman Z, which is an indicator of financial fragility developed by professor Edward Altman. In Figure 1 (next page), you can see the marked differences between countries. The threshold for financial vulnerability is a score under 3. By 1996, you can see that the Republic of Korea and Thailand were outliers. What the cross-country data presented here don't capture is the marked deterioration in these countries from 1990 until 1998—which is documented in the article.

In Figure 2 (next page), I do a simple EVA analysis, taking the return generated by a corporate and deducting the domestic lending rate. In other words, how long can a corporate borrow at 12 percent or 15 percent in a domestic market or in foreign exchange—incurring the foreign exchange risk—and generate a 4 percent return on capital? This was the return on capital on the Korean corporate for the last 10 years—the average return. Something has to give.

“My sense is that all the readings were there. What happened was a failure in judgment in reading all the information.”

So why did we fail to see the Asian crisis? My answer is simple. It wasn't a lack of transparency. Throughout the crisis, people who wanted to look could have seen the story. It was a simple lack of due diligence, and a lack of realistic scrutiny, assessment and interpretation of the underlying risks by the market, the credit rating agencies and the investors.

It reached almost comical proportions. Figure 3 (next page) is a graph of the emerging-market bond index spread. By August 1997, the spread was down 325 basis points. On July 2, 1997, Thailand went bust. Instead of the market reacting and demanding a higher premium, this spread continued to decline.

Bill Klein, the chief economist at the Institute for International Finance, tried to warn the financial community. He didn't say, “Look, you're taking risks that you are not being paid for.” What he did was publish a paper that translated this spread into the implied default rate. The message was clear—a spread like that implies almost a negligible default risk. Everyone knew Russia was an extremely tenuous situation, and Thailand had gone bust. But the market didn't listen.

A lot of people blame the multilaterals for not providing a warning, but in May 1997 the IMF in its subtle, tactical and diplomatic language suggested that spreads were not particularly reflective of underlying risks. The lesson to me is that data can help improve risk systems, and models can help, but that the failure in Asia was a failure of judgment.

Granted, there are other elements that didn't help. The lack of transparency is clearly a big culprit. In emerging markets, the data have a lot of pitfalls—not just in terms of the integrity or consistency of the accounting and disclosure standards, but also in the failings of the auditing profession.

You can have the standards in place, but if they're not properly implemented or if people are not penalized for not implementing them, that's a different story. Part of the problem is the franchise nature of the Big Five accounting firms, which have to retain a local partner in emerging markets. People sign off on things they shouldn't be signing off on. Sometimes an audited financial statement in a developing country doesn't have the representative ability of a signed financial statement in the United States.

Figure 1

Diane is absolutely right—there was a disregard of the qualitative element. There was no realistic assessment of the institutional capacity in supervisory systems for financial institutions, or of the corporate governance of companies that are 70 percent closely held, with only 30 percent of the market capitalization free-floating.

The question is: Where do we go from here? I don't propose to deal with how we get out of this crisis. I want to focus more on the long term. Obviously, I think credit derivatives are a good thing. They can put a market price on the underlying credit risk, and make it more transparent. Let's see what the market thinks about this or that credit and whether the credit is priced right. Any transparency will be extremely helpful.

The crisis was a twin corporate and banking crisis. And to some extent, what happened to the banks was just a reflection of what happened to the corporates. The credit-rating agencies are now looking at these things. They have changed their methodologies to examine the corporate and banking sector, as well as the liquidity of the individual countries.

Figure 2

In terms of constructive ideas, I want to suggest a methodology I've developed. You have to consistently review the condition of the “corporate country”—meaning the corporate sovereign. If you tally the top 20 corporates of the country, you probably already get 60 percent of the system. If you start monitoring their condition over time, it may give you an early warning signal.

We also need to devote more attention to intangibles, such as the institutional fabric, the financial sector and the corporate sector. In the financial sector, we need to examine the condition of the financial institutions, the strength of prudential regulatory supervision, the extent of depth in those markets and the extent of leverage. In the corporate sector, we have to examine the insolvency regime, the state of corporate governance and so forth.

Figure 3