Toothless Fed, Part 3 (The Ghost of LTCM)

Most sophisticated financial people I know take great comfort from the happy resolution of the LTCM debacle. As you may know, LTCM (Long Term Capital Management) was a hedge fund created by John Meriwether, a star trader from Salomon who headed its highly profitable bond arbitrage group, and included two Nobel prize winners among its partners (Robert Merton and Myron Scholes, for their role in creating the Black-Scholes model for option pricing). It played on its reputation of being the best and the brightest, and for the first three plus years of its existence, it lived up to its reputation. For example, in 1996, it returned 57% before fees, with not a single month of losses.

To make a very long story short (told extremely well in Roger Lowenstein’s “When Genius Failed,” which my trader buddies say got the story 98% right, which is about as good as it gets in reporting-land), LTCM faced more competition, saw declining returns, and started applying the trading practices it used in the bond market to markets in which it had no experience. Its most common strategy was “short vol,” which meant that they ascertained what typical price movements over time were (price volatility) and when derivatives traded as if volatility was higher, it would enter into trades that would have them betting that volatility would return to its normal levels.

LTCM made two very large mistakes. One was the partners never thought about liquidity. If you are a trader, you never want to take a position so large that if you get out of it, your selling pushes the market down (obviously, there are exceptions, but the point is you look at liquidity and figure out how much liquidity risk you can afford to take). The second is that they took bets in lots of markets and figured they were diversified. But they forgot that in a crisis, as traders put it, “all correlations move to 1.” Formerly uncorrelated markets suddenly move together as everyone dumps risky assets and moves into cash or very safe instruments.

The Russia default came in 1998. This on top of the earlier emerging markets crisis sent investors into a complete panic. LTCM saw all of its supposedly uncorrelated positions suddenly hemorrhaging all at once. Its prime brokers started issuing margin calls because LTCM no longer had enough collateral behind various margin loans and repos. But the last thing LTCM wanted to do was to sell into s market that was in free fall, particularly given the massive size of its positions.

Things got worse and worse over a roughly five week period. LTCM was sufficiently large that everyone was worried that a collapse and a dumping of its huge positions would not only disrupt the markets, but could severely damage the banks and brokerage firms that had lent to it. So the Fed got the big creditors into a room and kept on them to work out a bail-out plan. It was messy and not entirely equitable, but it kept the financial system intact.

So why is this story so reassuring? Because everyone believes it was a unique event. You had one very large institution making massive bets in all sorts of crazy places, and the finance equivalent of a perfect storm took them down. Nothing like that will ever happen again.

Narrowly, that is true. Market participants point to the fact that we now have many many hedge funds, each pursuing their own strategies. The risks are well diversified.

But that completely misses the point. We have a system that is much more highly leveraged than before (I can’t readily pull a statistic, but with considerable growth in derivatives and the widespread use of structures like CDOs that often use leverage, it has to be true). Going “short vol” is a standard trade by derivatives firms and hedge funds. That particular trade, with its attendant risks in a crisis, has not gone away.

So what could happen now? Let’s say we have another day like Feb 27, but Bernanke isn’t able to reassure markets. We see further ratcheting down and panic spreading to other markets. Anyone who is short vol is getting killed.

Now what happens? If it hits enough markets, we could have a repeat of LTCM but spread among many players. Enough firms are getting hit by margin calls that they are forced to unwind positions they don’t want to because conditions are adverse. And COLLECTIVELY these trades could be disproportionate to the liquidity of the various markets.

So the key isn’t that you need an LTCM, a single big player. That is faulty thinking. You need a lot of leveraged players short volatility at the wrong time, and an event that has a number of normally uncorrelated markets moving together.

In fact, the lack of an LTCM in a that kind of meltdown is vastly worse. You can’t get people into a room, knock heads together, and force a solution. The problem is too big and spread out for that to work. Geithner in his speech last Friday, “Credit Markets Innovations and Their Implications,” alluded to how messy even having a single large-ish institution melt down would be:

The dramatic growth in the volume of over-the-counter derivatives and the growth in the number and size of leveraged funds inevitably complicate the resolution of the failure of a large financial institution that is active in these markets. The sheer number of financial contracts that would have to be unraveled in the context of a default, the challenge that a former colleague of mine likes to refer to as “unscrambling the eggs,” could exacerbate and prolong uncertainty, and complicate the process of resolution.

Now imagine that it isn’t a single institution, but even a handful of medium sized ones that add up to the market presence of a big one. Or a large institution plus a few middling ones. You get the picture. The diversification of the current system in most cases will reduce the risk of systemic failure. But with derivatives, with the right (meaning wrong) conditions, like another 1998, I am not at all confident that we are better off now.

Print Friendly
Tweet about this on Twitter0Digg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone