Asset allocation strategies with volatility target mechanisms have become increasingly popular in recent years as practitioners have been exploring dynamic asset allocation strategies that can potentially deliver volatility-buffered returns matching passive buy-and-hold strategies. Market volatility is itself volatile, and the risk associated with a traditional fixed-weight strategic asset allocation policy can be highly variable over time. Dynamic asset allocation differs from conventional fixed-weight strategic asset allocation, which is generally set by an investment committee and reviewed infrequently—typically, once a year.

Historical and implied volatility have received increased attention as indicators of future asset class performance. This rise in interest has been coincident with the introduction of investment products based on the Chicago Board Options Exchange Market Volatility Index (VIX) and the discovery of the so-called low-volatility anomaly, which posits that portfolios of low-volatility stocks have produced higher risk-adjusted returns than portfolios of high-volatility stocks.

In a rational expectations model, investors should be compensated for taking on more risk by receiving a higher expected excess return, but many researchers have found the contemporaneous relationship between excess market return and volatility to be negative. Although this finding is unexpected, the author’s conclusions are ultimately intuitive: A positive unexpected change in volatility leads to an upward revision in predicted volatility and increases future expected risk premiums.

The policy prescription is that practitioners should reduce their exposure to stocks when unexpected volatility increases and increase their exposure when unexpected volatility decreases. For investors who are sensitive to volatility, a more consistent outcome can be achieved—both in terms of the volatility of returns and in terms of how volatile that volatility itself is—by adopting a dynamic, or volatility-responsive, approach.

The puzzling negative relationship between market risk and return can be explained by decomposing total market volatility into expected and unexpected components. The author uses trailing historical volatility as a forecast for future volatility and tests two dynamic allocation strategies that weight the stocks somewhere between 0% and 100%, depending on the value of the unexpected volatility. The objective is a reduction in volatility and/or an increase in excess returns.

In the first strategy, a 50/50 allocation between stocks and bonds is used as a passive benchmark. When the value of the unexpected volatility equals its historical mean, the weight of stocks is 50%. An increase of one standard deviation in the value of unexpected volatility leads to a decrease in the weight of stocks in the portfolio to 16%. By contrast, a decrease of one standard deviation in the value of unexpected volatility below its mean leads to a 34 percentage point increase in the weight of stocks, to 84%.

The second active strategy tested is a “switching strategy” that prescribes investing all in stocks or all in cash depending on whether the unexpected volatility is below or above its historical average. This approach is essentially a market-timing strategy: When unexpected volatility is negative, the entire portfolio should be invested in stocks.

Asset allocation is not about risk alone but about the trade-off between risk and expected return. The introduction of volatility to market-timing decisions is particularly interesting because a good time to buy an asset class (e.g., equities) appears to be after the asset class experiences a meaningful increase in volatility (e.g., 1987, 1999, and 2008 for equities) because an increase in volatility is often associated with large declines in price. Many successful asset allocation strategies with volatility target mechanisms (e.g., risk parity), however, sell asset classes that have experienced increases in volatility, so the relationship is not as simple as it appears.