A frequent point of conversation that we have at Newfound – both internally and with advisors – is the role of volatility as a measure of risk. On the one hand, from a statistical point of view, higher volatility implies a greater potential magnitude of loss. On the other hand, investors are loss averse, not volatility averse; upside volatility is viewed as a good thing.
We frequently encounter models that tend to view volatility as an indicator of downside risk: increasing volatility is a sign to shore up risk. This probably arises from the empirical evidence that for many asset classes and in many economic cycles volatility and downside returns are positively correlated. But many is not all.
"Crash up" asset classes as the biggest exception. What are "crash up" asset classes? Those for which increasing prices are viewed as a negative (e.g. commodities) or those that tend to benefit from economic turmoil (e.g. Treasuries). We see this "crash up" aspect in both the volatility smile (derived from options) but also in historical realized volatility versus historical price moves.
In the graph above, we can see that volatility tends to be a positive indicator for long-dated Treasuries, which tend to crash up in flight-to-safety scenarios.
Of course, this is a very, very limited view of long-dated Treasuries in an economic environment largely governed by GDP growth. But what if we look towards a different environment – one of high inflation and rising interest rates? Using 10-Year Treasury yields, we re-create a constant maturity Treasury index to evaluate this scenario from the 1970s.
The point here is that volatility is not a straight forward indicator that many make it out to be. From a naive statistical stand-point, higher volatility implies a greater return dispersion around the mean – but the relationship between volatility and returns is largely determined by the economic environment and the asset class we are examining.