Statement No. 151

For Information Contact:

Richard J. Herring
(215) 898-5613
Hal S. Scott
(617) 495-4590
Kenneth E. Scott
(650) 723-3070

The spectacular near collapse of Long-Term Capital Management LP (LTCM) last week will be the occasion for upcoming hearings in Congress and broader discussion at the IMF/World Bank Annual Meetings. According to information currently available in the press, LTCM was a hedge fund with over $100 billion in assets and notional off-balance-sheet positions many times larger. These highly-leveraged positions were supported by an equity position that began the year at $4.8 billion and dwindled to a reported $600 million, perhaps only sufficient to meet margin calls for a few more days. This rapid decline led to a hastily improvised rescue package, from fifteen of the world’s leading financial institutions, of $3.5 billion in new equity. Strikingly, this rescue package was put together under the auspices of the Federal Reserve Bank of New York.

Although the precise nature of the losses has not yet been disclosed, they appear to have stemmed from positions taken in anticipation of a narrowing of credit spreads among debt instruments, based on highly complicated mathematical models. In fact, in the wake of the Russian default, yield spreads widened throughout world markets, as investors fled to safer securities. Because of the high degree of leverage, that widening of credit spreads produced very large losses of capital for LTCM.

What went wrong? There are a number of hypotheses, which are not mutually exclusive, that Congress should explore:

• Bad luck. The models, analyses, and positions may have been basically sound, but LTCM may have succumbed to an extremely low probability financial disaster.
• Bad model. The analytical model being used to detect these trading opportunities, though sound under most conditions, may have failed to allow for the degree of correlation across markets and instruments that was revealed in the aftermath of Russia’s default.
• Bad disclosure. The managers of LTCM may not have fully informed their lenders, counterparties, and investors about the amount of risk they planned to take. Had these parties understood the nature of the risks being undertaken, they might have exerted greater market discipline.
• Bad incentives. LTCM and its investors and creditors may have been more prone to take large risks because they thought that, being involved in a very large institution, there was a distinct possibility that they would be rescued in the event of a catastrophic loss. This moral hazard concern is most acute when funding comes from banks that have access to the federal safety net.

What concern does the public have in these events? If these are simply private losses shared by large and experienced investors, lenders and counterparties, with no taxpayer exposure, they represent merely an extreme example of the kinds of risk inherent in a market economy. LTCM required a minimum investment of $10 million and dealt only with highly sophisticated parties. So long as the losses are limited to such parties dealing with the firm on a voluntary basis, these risks are better controlled by the market than regulated by government agencies. They do not constitute a basis for enactment of the general regulation of hedge funds.

The Fed claims that its participation in devising the rescue package has been no more than that of a facilitator. If so, its role may have been valuable. Cooperation among creditors in dealing with an institution facing possible insolvency is a familiar phenomenon; it addresses the losses that may be created by a scramble for assets and forced short-term liquidations. Cooperation is, however, sometimes difficult to achieve, because free-rider problems may inhibit attempts to get additional capital from numerous existing creditors for a troubled firm. This particular action, facilitated by the Fed, was accomplished quickly without the costs of a formal bankruptcy reorganization.

Under what circumstances should there be public concern about this near collapse and the way it was handled? Did the Fed play a greater role than that of a mere facilitator? A Fed spokesman stated and the Treasury Secretary confirmed that "there’s no public money." If there is no explicit or implicit commitment of public funds or guarantee of private positions or Fed pressure on creditors to put in additional capital, then there need be no taxpayer concern. These are points that Congress should probe and the Fed should address in some detail.

But, let us suppose that the Fed had committed to some type of public guarantee. Would it be justified? Clearly there should be no bailout, no subsidy, and no guarantee for LTCM or its investors because that would be an extreme example of the kinds of bad incentives that accentuate moral hazard. But are there circumstances under which federal assistance should be extended to creditors and counterparties of a firm such as LTCM? It is sometimes suggested that such assistance might be warranted on the grounds that it is necessary to avoid "systemic risk." If the Fed invokes this rationale before the Congress, it should be asked to spell out in precise detail the analysis it made of the losses in LTCM’s positions, the exposure of other institutions to those losses, the capital positions of those other institutions and the extent to which there would be any likelihood of failures spreading to the rest of the financial system. In the Committee’s view, systemic risk is evoked far more casually and often than it is analyzed and supported. Further, if systemic risk were a reasonable concern, it should be limited to banks that fall within the federal safety net, and not extended to other financial firms. And if applied to banks, assistance should be applied in the manner that Congress specified in the FDIC Improvement Act of 1991, which was designed to impose significant constraints on its easy invocation.

It is the Committee’s policy that members abstain from participation on policy statements in which they have direct personal or professional involvement. Accordingly, Richard C. Aspinwall abstained from voting on this statement.