Here's a dirty little secret of Wall Street: Risk management isn't a science. Who exactly is this a secret to? To the professionals practicing it, and the innocent people who believe them. Wall Street blew itself up a couple of years ago with excessive risk and irresponsible leverage because practitioners and their “quants” either didn’t know or didn’t care about this secret. And Main Street paid the price, and part of the tab.
As the dust still settles from the subprime mortgage crisis, maybe most of us now accept that risk management isn’t a science. But we’re no better off today than we were two years ago just because Wall Street somehow totally revamped its models. Things are stable and growing again because Washington provided a “get out of jail free” card, and some pocket money, to Wall Street.
So, why does it matter that “risk management isn’t a science”? Because those who act like it is can get in lots of trouble with their capital … and yours. And we had lots of warnings before 2007 that this was true. The best example was 1998’s blowup of quantitative hedge fund Long Term Capital Management (LTCM for short).
Myron Scholes, a Nobel Prize-winning financial theorist, was a partner in LTCM. Along with his equally quantitative partners, he made sizable leveraged bets on interest rates in the late 1990s that got creamed when Russia defaulted on some debt and devalued the ruble, sending shockwaves through global markets.
The Federal Reserve decided to bail out Long Term Capital Management to the tune of about $3.5 billion because it believed that not paying LTCM’s derivatives obligations would create “big” losses for too many Wall Street firms. How ironic that the phrase “too big to fail” may have originated with this now-tiny fund collapse. Roger Lowenstein’s 2000 book on LTCM, When Genius Failed, chronicles the missteps in strategy and risk management that brought the fund to its knees. Moral hazard is a dangerous business. Perhaps because the Fed bailed out LTMC this led others to believing the Fed would do the same for them, but thats a different subject for a different day.
Now before anyone thinks I am completely discrediting volatility as a risk management and trading tool, let me clarify where it is useful and powerful. In options trading, you need a baseline for comparing the perceived risk of positions. It works for options because you are dealing with two elements where the variability of inputs is strictly limited.
First, you are valuing options and their risk on one security. Second, you are doing so to a specific forward date, usually less than one year. Mortgage-backed securities were infinitely more complex than this. By reducing the complexity of your risk analysis, you quickly make volatility a more useful measure.
But option traders don’t pretend they are doing science here. They know it’s still all about probability and that conditions change constantly. They go beyond recognizing that today’s implied volatility number is still only a 68% chance "guesstimate" at what is likely to happen. And they know that this is not a poker game with finite outcomes. When a company or market event elevates the risk measurement (implied volatility) by a factor of two in one day, they take it in stride -- because they never promised themselves or others with massive bets that it couldn’t happen.
There’s hope for Wall Street to learn to use financial modeling and quantitative strategies for good, as long as we learn from our mistakes and don’t let ourselves become seduced by the latest equation from a math whiz promising great returns with low risk. Probably the best teachers here will not be the universities, but the firms that survived the crisis with better, more robust models that took all the money from the illusionists and fools.
0 comments:
Post a Comment