The company had developed complex mathematical models to take advantage of arbitrage deals (termed convergence trades) usually with U.S., Japanese, and European sovereign bonds. The basic idea was that over time the value of long-dated bonds issued a short time apart would tend to become identical, and by a series of financial transactions (essentially amounting to buying the cheaper 'off-the-run' bond and short-selling the more expensive, but more liquid, 'on-the-run' bond) it would be possible to make a profit as the difference in the value of the bonds narrowed when a new bond came on the run.
Because these differences in value were minute, the fund needed to take highly-leveraged positions in order to make a significant profit. At the beginning of 1998, the firm had equity of $4.72 billion and had borrowed over $124.5 billion with assets of around $129 billion. It had off-balance sheet derivative positions amounting to $1.25 trillion, most of which were in fixed income derivatives such as swaps.
The fund also invested in other derivative security products such as equity options and mortgage securitisations.
The downfall of the fund started in May and June 1998 when net returns fell to -6.42 and -10.14 per cent reducing LTCM's capital by $461 million. This was further aggravated by the exit of Salomon Brothers from the arbitrage business in July 1998.
The scheme finally unraveled in August and September 1998 when the Russians defaulted on their sovereign debt (GKOs). Panicked investors sold Japanese and European bonds to buy U.S. treasury bonds. The profits that were supposed to occur as the value of these bonds converged became huge losses as the value of the bonds diverged. By the end of August the fund had lost $1.85 billion in capital.
The company was providing returns of almost 40% up to this point, and a "flight to liquidity" the company lost a possible $100 bn and needed an Federal Reserve Bank of New York organised bail-out of $3.625 bn, apparently in order to avoid a wider collapse in the financial markets. The fear was that there would be a chain reaction, as one company liquidiated its securities to cover its debt leading to a drop in prices which would force other companies to liquidate its debt creating a vicious cycle. The total losses were found to be $4.6 billion.
Ironically, in the end the basic idea of LTCM was correct, and the values of sovereign bonds did eventually converge after the company was wiped out.
However, the downfall brought to notice an important lesson to the financial community - the need to keep in mind liquidity risk while making Value-At-Risk calculations - one of the primary reasons for the downfall.