Long Term Capital Management principals pondered what should the fund’s going concern, including orderly unwinding of fund for purpose of total winding up starting August 1997.

In Sept.1997, Long Term Capital Management principals debated whether or not the fund had excess capital. What course of action should they take? Unwinding?


I not attempt to provide a solution to the LTCM debacle as it is answering with the benefit of hindsight; but tries to put ourselves in the shoes of the principals to predict the market going forward, so as to decide what course of action they should take at end of September 1997.

We shall look at the period from Mar 1994 to September 1997, where data is available for exhibit 1. Market data is obtained from http:\\finance.yahoo.com\. The graphs below are constructed from available open market data like VIX, S&P 500 (contains a wider assortment of companies listed in the US stock market), US 30 years treasury bonds (long term debt), US 5 years treasury bills (medium term debt) and US 13 weeks treasury bills (short term debt).

LTCM also engaged in markets where information is not publicly available like collateralizations, swaps, arbitrages and other derivative and financial services. This resulted in LTCM portfolio containing approximately 100 different strategies and 7,600 positions, of which 6,700 were separate contractual agreements.

Sample market data from Mar 1993 to Sept 1997

The graph below is a representation of historical S & P volatility as a proxy to market risk direction. Data for this table is obtained from http:\\finance.yahoo.com on the monthly close adjusted value. The population mean and standard deviation was calculated.


The table is adjusted to reflect mean at 15.842 and the standard deviation is indicated with symbol, σ. The VIX value is 24.76 in month of August 1997. Note that the period of operation of the fund from Mar 1994 to Dec 1996, the market volatility remained quite predictable within the range of 1 standard deviation from population mean, where 1 standard deviation from the mean is about 3.69.
LTCM’s actual and estimated volatility had been much less than 20% reported in article, which is approximately within one standard deviation from the VIX graph shown above as a proxy to the market.

Table below is an indicative direction of S&P 500 stocks available in the American market. It is safe to assume that market is generally behaving in a calm and rising manner. It undergoes healthy corrections occasionally, and sometimes remained range bound while on the way up. The index movement experiences an accelerating uptrend towards end of august 1997.


Table of S&P 500 stocks index

It is also graphically significant to observe that corrections appears to be steeper as index accelerated upwards.


Table of 30 years US treasury bonds


Table of 5 years US treasury bills

The above yield graphs of 5 years and 30 years treasury allows us to construct a sample of market where LTCM has participated in equivalence. The sample yield curve that LTCM works on for relative value trades appears to behave normally, where shorter term yields are lower than longer term yields.

13 weeks US treasury bills

In a normal market, short term debt is more expensive than mid-term debt which is in turn more expensive than long term debt. This rising yield curve reflects the liquidity is expensive in the short term where financial dislocation does not occur. However, after July 1997, cost of long term debt continues to climb which is reflected by increasingly depressed yield of long term debt. This resulted in a flattening of the yield curve.

Principal’s Directional view of the market

The investment principal of LTCM strongly believed that the market will restore equilibrium in the long term as dislocations are temporary. LTCM seeks out these market dislocations and take opposing directions to the market movement as it seeks out financial opportunities in with these dislocations.

LTCM may even continue to add positions if the dislocations continue to mount, as the principals believed that the more obvious the dislocations, the more market participants will arbitrage away these peculiarities, thus safer will be their positions. LTCM’s strategy of loss averaging could cause it to be stressed with short term liquidity. However, LTCM managed to pull through during the period.
LTCM investment strategies as powered by prize winning strategies from the latest financial thinker, who are also winners of noble prizes. LTCM is backed by heavily invested back office technology which enables LTCM to operate with the most advanced trade algorithms, monitor and maintain trades to minimize risk.

In September 1997, LTCM principals would have concluded that information at hand represented only a 10.7% of risk level which is lesser than the fund’s long term goal of 20%. Based on historical information, only a thirty sigma event would bankrupt the fund in a month.

Long term perspective

LTCM hedge fund employs trading strategies that would exploit market dislocations on the short term so that it could make money over long term. It requires investors to leave their money with the fund with lock in period of three years for every tranche of investment. LTCM also structured the equity movement so that there is no large movement in equity within a short term. Where the market is orderly, LTCM continues to experience inflows and outflows that continue to fund the fund.
Trades are also structured to make profit, if not in the short or mid-term, by holding the position to maturity, in the long term. The experience of the principals told them that it is possible to exit positions by trading counterparty some time later with profit.
LTCM performed regular stress tests and identified its value at risk. It has managed to ride through short term liquidity squeeze even with loss averaging trades.

Available Solutions to Options

Leverage and Risk Management Strategy

LTCM made use of leverage to make money in the market. The leverage allows LTCM to make counterparty trades, provide market liquidity and sometimes becomes a market maker of financial product. LTCM is aware that its size in the market would occasionally make it difficult for the fund to exit in an orderly manner in the fastest possible time. This is a cost to the fund’s liquidity.
This leverage also magnifies downside risks to fund equity although the principals made use of the latest scientific knowledge to operate the fund. The fund could not maintain return without increasing fund size using the current leverage structure will make it too cumbersome.

Unwinding / Dissolve fund and return money to shareholders

Given the track record of the fund and investment in knowledge, it would be painful for principals to dissolve the fund and return money to shareholders.

Investment Strategy and Risk appetite

The principals had pondered moderating the risk strategy to the fund but would produce lesser returns. The principals concluded that given such returns, the fund may not be operating in synch with its investors as they believed that principals and employees of LTCM may withdraw investment capital in the fund, similarly applies to current investors.
However, the fund will attract a new group of investors seeking different kinds of returns.

Recommendation and Conclusion

LTCM principals come to realize that the fund capital could not be maintained if it continues to seek returns using current strategy. Increase the capital could cause it to “explode” as there might not be sufficient counterparties to take up its offered positions. Change the investment strategy while maintaining current capital base would risk LTCM becoming a mediocre fund. Unwinding / Dissolve the fund is not preferred given that September 1997 market appears to present great opportunities for the fund in its field of specialization due to dislocations and increasing volatility.

The sensible solution would be for LTCM could seek orderly unwinding of trades with the purpose of dissolving the fund over time, but it would be quite impossible complete unwinding in the short time due to nature of its trades and sizes.