Gary:

Thank you for posting your thoughts on risk vs. volatility. This discussion seems to be the focal point of active investing these days.

I can’t help but notice that the discrete risk factors mentioned in your article, are already embedded in the standard individual IPS framework as time horizon, liquidity needs and the willingness to take risk. Maybe we all need to re-frame these every now and then to understand the nature of risk…

Much like you, I do not think that volatility should be used as the sole measure of risk. Volatility or any derivation thereof has some utility as a concept, but I don't accept it as a valid measure of risk. Doing so would assume that (i) markets always efficient and that (ii) correlations between asset classes are stable. This may be a convenient assumption for a mathematical model to work, but it is not necessarily true, especially when panic/euphoria drive the market sentiment and when the contagion breaks correlation barriers.

Under these conditions, being tied to volatility, creates a host of problems, not the least of which is excessive trading to stay within the portfolio mix and risk constraints.

While I see how using individualized risk measures (as opposed to volatility) can help individual portfolios, I think the greater danger of equating risk to volatility is in the institutional portfolio side.

In the institutional portfolios the constraining factors that affect individuals are less pronounced: time horizons are usually long, willingness to take risk (or having a "stomach") isn't usually a factor. Liquidity may be the only factor that may affect both individual and institutional portfolios somewhat equally.

In the individual portfolios, behavioral constraints provide some form of protection (albeit imperfect) against the tendency to using volatility as a sole risk factor. In contrast, within institutional investment mandates, it is not unusual to see portfolio risk coded as some quantifiable measure such as standard deviation.

Re-defining risk as the probability of loss (an intellectually honest thing to do) is the first step in the process of finding the next risk framework. The challenge is finding risk factors that works across all asset classes.