Swimming naked: Rethinking risk management after the crisis

July 21, 2009

As Warren Buffet famously said: "When the economic tide goes out, you find out who is swimming naked." Certainly the financial upheaval of the last two years has revealed a number of inadequately clad investors.

Just as every crisis prompts soul-searching about assumptions and standard procedures, banks and other financial institutions are taking a serious look now at how they measure, price and monitor risk in the capital markets. As Washington policymakers debate reforms to address regulatory gaps revealed by the worst crisis since the Great Depression, there is growing discussion about what are the right lessons to be drawn about managing financial risk.

L. Wendell Licon, a W. P. Carey Clinical Assistant Professor of Finance, says many of the lessons learned are the same as those gleaned from previous crises. The problem, he says, is that investors get caught up in the "irrational exuberance" of rising markets and forget fundamental risk management principles. "There are basic points: don't ignore remote possibilities; do your own risk analysis: if you don't know how to price a security, then you probably shouldn't buy it; and understand both the risks and benefits of the herd mentality in investing."

Respecting the black swans of investing

As an example, Licon says that risk models should not rely on historical data alone to predict future market movements. He adds that risks should not be calculated in isolation, but rather take into account the interdependencies of markets and regions.

The extraordinary events of the last two years have produced a new respect for remote possibilities in financial markets. Nassim Nicholas Taleb, author of "The Black Swan: the Impact of the Highly Improbable," described how our brains are not wired for statistical uncertainty. As a result, we tend to underestimate the possibility of unusual events. "The fact that extraordinary events occurred with greater frequency over the last two years highlights the need for more robust scenario analysis and stress testing in financial risk management," Licon says. "Understanding and modeling so-called tail risk events has become more important than ever."

One of the main shortcomings of the mortgage-backed securities market was underestimating the risk of housing price declines.

Subprime mortgages were considered such a narrow segment of the overall real estate market that the spillover risk to the rest of the credit markets was severely miscalculated. Many of the risk models used to price these securities, or the collateralized debt obligations (CDOs) created from them, involved stress tests and scenario analyses that assumed declines in housing prices based on limited historical data. "As it turned out, those assumptions fell far short of what actually transpired," Licon says.

Even recognized market leaders in risk management underestimated how serious the crisis could become. Goldman Sachs, for example, conducted what it called its "wow" stress test -- "worst of the worst" -- before the crisis. That analysis looked at the most negative market events since 1998 and assumed they could get 30 percent worse and occur simultaneously. That still did not adequately capture the severity of events.

The domino effect in market meltdowns

Professor Licon says many of the painful risk management lessons learned in the aftermath of the collapse of Long-Term Capital Management (LTCM) in the late 1990s were forgotten during the housing bubble.

LTCM, the large hedge fund whose board of directors included Nobel Laureates Myron Scholes and Robert C. Merton, illustrated the danger of miscalculating multiple risk factors and the linkages, or correlations, among what may appear to be unrelated assets during a crisis. LTCM's quantitative risk models, for example, concluded that Russian government bonds and Mexican bonds were only minimally related. But because a relatively small number of investors dominated both markets, the default crisis in Russia touched off panic selling in Mexico, setting off cascading losses for the hedge fund.

The fall of LTCM was one factor leading to the creation of instruments such as credit default swaps (CDS) meant to insure against some of the credit risks that led to the fund's collapse. By dispersing credit risk to a wider range of investors, it was hoped that the dramatic concentration of risk in single entities such as LTCM could be avoided and risk to the overall financial system would be diminished.

Drawing an example from recent headlines, the credit default swaps global insurer AIG entered into appeared to offset the various types of credit risk it assumed. However, market participants failed to appreciate the limits of applying those hedging instruments when liquidity dries up. Once again, a single institution threatened to bring down the entire system unless outside intervention could forestall a crisis.

These links between markets and the institutions that participate in them highlight the need for risk models that take into account these relationships rather than viewing single-factor risk in isolation.

Credit derivatives also underlined the risks of financial innovations outstripping the industry's operational capacity to manage them.A credit derivative is an asset that derives its value from the likelihood that a particular debt instrument will one day be in default. Already in 2005, when credit rating agencies downgraded General Motors and Ford to below investment grade, brokerage houses admitted to back-office chaos in trying to sort out how various investors would be affected by the huge volume of credit default swaps traded in the aftermath of the downgrade.

Know your risk

The AIG crisis illustrated the challenges of understanding the credit risk that arises from derivative instruments, and reinforces the importance of comprehensive credit analysis. Throughout the last 20 years, many financial institutions overlooked the basic contradiction in the credit rating agencies' business model that allowed them to earn revenue from the same entities whose securities they were assessing. Those rating agencies are now under intense scrutiny in light of the triple-A ratings they assigned to CDOs that quickly turned toxic once the subprime crisis ensued. Investors now appreciate better the need to conduct their own in-house credit analysis.

Credit derivatives have always had their strong supporters and critics. Supporters argue credit derivatives allow investors to express their credit more efficiently and flexibly, and mitigate credit risk by spreading it among a wider group of investors. Critics, on the other hand, claim these same circumstances magnified systemic risk, especially given the difficulty of identifying counterparties and pinpointing where credit risk ultimately resided.

Some complain that fair-value accounting requirements exacerbated the credit crisis for many financial institutions. But some industry leaders counter that if banks and other institutions had properly valued their risk exposures at the outset, they would have been in a better position to manage and reduce those exposures when the crisis hit.

Goldman Sachs CEO Lloyd Blankfein told Financial Times that the daily marking of the firm's positions to current market prices was the key indicator for reducing the firm's exposure early on to markets and instruments that were fast losing value. While he concedes the process can be difficult, he argues that it should be a discipline followed by every financial institution.

The risks and benefits of herd mentality

Finally, the difficult events of the last two years provide an interesting case study for the risks and benefits of the herd mentality in investing. Despite long-standing warnings of a housing bubble, concerns about ever higher levels of leverage in the financial system and the growing operational risks of credit derivatives, many investors felt duty-bound to continue to pour money into CDOs and other mortgage-related assets. Citigroup's former CEO Chuck Prince summarized this herd mentality: "As long as the music is playing, you've got to get up and dance. We're still dancing."

As long as interest rates remained low and housing prices continued to appreciate, the wisdom of crowds seemed convincing. And, the entire compensation system of the financial services industry was geared towards encouraging higher short-term risks without regard for their long-term consequences.

But, that same herd mentality magnified the pain of deleveraging since everyone was seeking to unwind similar positions at the same time, leading to the credit crunch and liquidity squeeze that froze markets in 2008. Repairing that misalignment of incentives and risk is likely to be one of the industry's biggest challenges in the years ahead.

Bottom Line:

  • When it comes to financial risk management, don't underestimate remote possibilities. Risk models need to incorporate a wider range of possible outcomes and responses.
  • Do your own risk analysis: if you don't understand how a security is priced, then don't invest.
  • Assets need to be fairly valued at the outset and adjusted to market price changes.
  • The herd mentality in investing is double-edged: it can help identify opportunities on the way up but magnifies risk on the way down.