The twitchy, volatile stock market

By Felix Salmon
August 12, 2010
stock market fall today, but it was certainly important enough to grab the attention of Mohamed El-Erian, who has an interesting theory of why it happened:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

I did my best to ignore the stock market fall today, but it was certainly important enough to grab the attention of Mohamed El-Erian, who has an interesting theory of why it happened:

As Rich Clarida and I recently argued in the FT, sharp risk-on/risk-off swings in markets are to be expected given the reality of today’s macro context.

A couple of weeks, Fed chairman Ben Bernanke called the outlook “unusually uncertain“. We went further, arguing that expectations of outcomes have evolved in an interesting manner – from the more familiar bell curve (a dominant mean and thin tails) to a much flatter distribution with fatter tails. In such an expectation world, short-term news can have a disproportionate impact on market valuations.

Essentially, El-Erian is saying here that the markets have gone from driving a boring family sedan to driving a twitchy, ultra-responsive sports car. And that’s nothing to do with high-frequency algorithmic trading, or anything like that — such things only serve to exacerbate the underlying fundamentals.

Think about it this way: In the 1950s, or even in the 1980s, the range of possible outcomes was (or was perceived to be) pretty narrow. As the macroeconomic situation evolved in a relatively simple and continuous manner, markets evolved to reflect the new realities. Even the bout of high interest rates and inflation in the 1970s was something that equity markets, in particular, could take in their stride — stocks are, after all, one of the world’s better inflation hedges.

Today, however, we live in a much more discontinuous and uncertain world. If stocks are pricing in a 10% chance of a devastating bout of deflation and that perceived probability rises to 20%, the risk-adjusted present value of those stocks can plunge dramatically — much more than 3% — even if everything else remains the same. We don’t really have a model of how the economy works (or doesn’t) — but insofar as we do, it’s one of those models where small changes in initial assumptions can and do result in very large changes to outputs.

As a result, the long-term volatility of equities is going to continue to rise, I think, as it has been rising of late. (Does anybody have a dataset here? I’d love to see a chart of at-the-money implied 10-year or 15-year volatility for the S&P 500, and not just because it has important implications for Warren Buffett’s equity puts.) The implication of that is that the “long run”, in the concept of “stocks rise over the long run”, is going to continue to get longer and longer. It’s clearly not 10 years, any more, or even 15. But how long is it? 35 years? 50? And at what point is that long run so long that it’s of practical purpose to pretty much no one?

7 comments

Comments are closed.