The authors show that, even at historically low levels of implied volatility, put option portfolio protection is still expensive when measured against subsequent realized volatility. Even protection against “black swan” events is expensive in terms of the frequency of tail events required to make the strategy break even.

The authors succeed in challenging a common perception that successful option trading is about timing the purchase of protection to when current volatility is perceived to be low compared with its historical average. The evidence points to the existence of a substantial volatility risk premium, even in times of relative calm. Moreover, in environments where implied volatility is rising, realized volatility (the variance in prices from the time of purchase through to option expiration) is lower, on average, than the level of implied volatility at the time of purchase.

The authors also show that buying put protection against tail events, such as the 1987 stock market crash or the extreme market moves following the collapse of Lehman Brothers, is expensive in terms of the frequency of events required to make such a strategy break even. This expense is partly related to tail events occurring during times when implied volatility is already high.

The authors initially focus on the S&P 500 Index. They define the volatility risk premium as the VIX minus the S&P 500’s subsequent annualized volatility. By dividing the volatility risk premium into buckets by VIX deciles, the authors demonstrate that the volatility risk premium is consistently positive. The same buckets are used to describe the returns to a delta-hedged strategy that goes long a put option that is 5% out of the money. Again, the evidence points to put options being expensive across all levels of implied volatility, with the strategy yielding negative returns for all buckets.

The data for these exercises are gathered from AQR, Option Metrics, the Chicago Board Options Exchange, and Standard and Poor’s, with most of the analysis covering the period from March 1996 to June 2014.

The second part of the article expands the analysis to several global indexes, including the DAX (German stock index), the FTSE 100, the Hang Seng Index, and the Nikkei 225. Consistently positive volatility risk premiums and negative returns to the delta-hedged put option strategy support the earlier S&P 500 evidence.

Finally, the authors address the use of S&P 500 put option purchases as a hedge against black swan events. At average levels of implied volatility, events of the magnitude of the October 1987 crash need to occur once every 10 years for the hedged put option strategy to break even.

The strength of the authors’ work is in the way it challenges a popular misconception that put option insurance is cheap or expensive depending on the current level of implied volatility relative to its historical average. The authors’ technique of bucketing the volatility risk premium and strategy returns by the level of the VIX is a simple but effective way of demonstrating their argument. Future research could revisit covered call strategies in terms of the way they enhance returns by capturing the volatility risk premium.