Join our community of smart investors
Opinion

Volatility's misleading message

Volatility's misleading message
August 23, 2016
Volatility's misleading message

I say this because volatility is an emergent process. It is the unintended effect of thousands of individual actions which are taken for different reasons. Low volatility is in fact a sign that investors disagree. Prices are relatively stable because buyers can find sellers and vice versa without prices moving much. High volatility, on the other hand, would be a sign of consensus; prices fall a lot when most investors want to sell and rise a lot when most want to buy. With hindsight, stock markets were most complacent at the peak of the tech bubble in 1999, and volatility then was high.

All this might sound trivial. But it has an important implication. It implies that low volatility needn't be a sign of complacency but of the opposite. It might mean that there's a significant fraction of investors who are nervous.

This might help explain an otherwise curious fact: historically, low volatility has tended to lead less to falling prices than to rising ones. Since it began in December 1989 the correlation between the Vix index and the change in the All-Share index in the following month has been minus 0.28. That tells us that, on balance, low volatility has more often led to rising prices than falling ones.

So, does this mean we can safely buy now?

No. The relationship between the Vix and subsequent returns is unstable. Sometimes, low volatility is associated with high returns, as we saw in the mind-1990s and mid-2000s. Sometimes, high volatility is accompanied by high returns, as we saw during the tech bubble of the late 1990s and the recovery from the 2009 recession. At other times, though, we get high volatility and bad returns, such as during the bursting of the tech bubble and the financial crisis.

This warns us that one big risk we face is that of a regime change - a worsening relationship between risk and return. We can't say how great this risk is, though at least one valuation measure suggests the probability of it is around normal levels. What we can say, though, is that the Vix index does not measure risk.