The solution
Long-Term Capital Management was founded in February 1994 by former Salomon Brothers vice chairman John Meriwether, with a team that included two Nobel laureates. Its arbitrage book earned 20% in 1994, 43% in 1995, 41% in 1996 and 17% in 1997. The strategy earned thin margins on each dollar, so LTCM leveraged heavily: at the end of 1997 it held about $30 of debt for every $1 of capital, after returning capital to investors without shrinking its positions.
In August 1998 Russia defaulted and devalued. Investors fled to safe, liquid assets, and almost every spread LTCM had bet would converge diverged instead. The fund lost 44% of its value in one month and disclosed the losses in a September 2 letter to investors. On September 18 its officials called New York Fed president William McDonough; a Fed team visited two days later and grasped the scale: counterparties exiting at once would force a fire sale into already turbulent markets.
On September 22 the Fed convened a core group of firms, then thirteen firms that evening. Talks stalled over who would pay how much. On September 23 a rival bid — Warren Buffett's group offered $250m for the partners' stake plus $3.75bn of new capital — lapsed at its 12:30 p.m. deadline. The consortium resumed and by about 6 p.m. agreed: fourteen firms put up $3.625bn for 90% ownership, while LTCM's partners kept running the book on limited salaries and no bonuses, reporting to a creditor oversight committee.
The fund unwound in order rather than in a panic. It sold most of its positions and repaid the consortium in full by the end of 1999, according to the GAO. The Fed never spent a dollar, and the old investors' claims were cut to 10%.
What it achieved
Make the creditors fund the rescue — they price the risk
Why it worked
The creditors had the most to lose from LTCM's bankruptcy, so they were the only bidders who would pay a fair price to avoid one.
Public money would have rescued parties with no exposure to LTCM; the bail-in charged exactly the firms whose books a fire sale would have burned.
The systemic threat was the fire sale itself: LTCM's parallel positions meant one distressed exit would have repriced every comparable book at once.
Keeping Meriwether's partners in charge, unpaid and supervised, meant the unwind was run by the people who understood the positions best.
What can be applied
The party with the most at stake prices a rescue best. Forcing exposed creditors to fund it keeps markets whole and keeps the expectation of public bailouts out of the system.
Aftermath
LTCM repaid the consortium in full by the end of 1999, and the rescue became the reference point for handling a failing institution without taxpayer funds. In April 1999 the Treasury, Federal Reserve, SEC and CFTC published 'Hedge Funds, Leverage, and the Lessons of Long-Term Capital Management', and the GAO reviewed the rescue in 2000. Federal Reserve historians note the episode also showed the need for a standing mechanism against fire-sale risk, rather than ad hoc coordination assembled overnight.
FOLLOW THE EVIDENCE