Friday, October 17, 2008

Cross-Talk: Volatility and Stock Market Returns

Here's a great blog post from Market Sci that builds upon an idea I wrote about a while back. Michael Stokes researched what happens in the market after relatively narrow range days versus wide range days. What he found was that markets historically have yielded superior returns following those wide range days compared with the narrow days. Another way of looking at this is that expansion of volatility has generally led to bullish returns going forward.

Michael also notes, "This approach defended well against the bear market of the early 2000’s and our most recent beginning in late 2007 (interestingly though, it crashed and burned in the most recent October death march)."

This gets back to the notion that "this time *has* been different". Many historical patterns--particularly ones in which extreme weakness have led to bullish prospective returns--have not panned out during this decline. Expanded volatility has led to even greater volatility and then record volatility, as prices have moved steadily lower.

Michael's research found that the high volatility/bullish returns relationship was noteworthy from 1993 through 2008, but less convincing when looking at history going back to 1970. This, too, fits with my experience in market research: rarely do patterns hold intact over all markets, across all market history. Rather, we see regimes in which patterns will manifest themselves for months and years, only to give way to different patterns and regimes--usually when market conditions change. I suspect that the crashing and burning noted by Market Sci with respect to the volatility pattern is one piece of evidence that regimes have shifted--and may stay different for a considerable period.

Thanks to Michael for an excellent site; check out his posts on market nuts and bolts and coping with abnormal markets.