On crashes, briefly

May 9, 2010 in Finance

Congress will be looking into Thursday's stock market plunge, with the likely outcome that we are forced to admit that we didn't actually learn much from twenty years ago (1987) or two years ago (2008). In all three cases, sudden selling was aggravated by algorithmic trading of one form or another. In only one (2008) was there an easily-identifiable catalyst for the initial selling. It just goes to show that you don't need a hard piece of news to trigger behavior - something which should be obvious to anyone who has studied market dynamics.

This is all in line with a thesis I've proposed in the past, that every innovation which expands the ability or volume of trading ultimately leads or contributes to a market crash. Margin trading, portfolio insurance, online trading, HFT... all associated with crashes. That doesn't make them bad, it just illustrates that bubbles can form in infrastructure itself.

The real mystery is not how the index dropped so far so fast - it's how it jumped back up again! Alas, the world may never know.

Leave a Comment

{ 1 trackback }

Previous post:

Next post: