Volatility is low and that has people concerned. So much has been written about this subject that one columnist has wryly observed "people are worried that people are not worried enough," which implies markets are supposed to move a certain amount. Some analysts have gone further, presenting an analogy between low market volatility and an overcompressed spring. They claim pressure in this spring is building, and when it explodes, the carnage will be massive — particularly for option sellers.
That sounds ominous. It would certainly be concerning if volatility sellers were lulled into a false sense of security during the past few years of market calm. However, we took a look and found there have been some market jolts in this relatively calm period, and these jolts can shine some light on what option sellers might experience if volatility suddenly spikes after a period of calm markets. This analysis helps to demonstrate that investors can earn the well-compensated volatility risk premium over time without taking excessive tail risk, when implemented in a reasonably sized strategy.
The VIX, volatility risk premium and option selling
The most common measure of implied volatility is the CBOE Volatility index which measures the near-term volatility (30 days) implied by S&P 500 stock index option prices. The VIX generally tracks, but is usually higher than the S&P 500's recent realized volatility.
As a rule of thumb, selling index options tends to be profitable when implied volatility is greater than the underlying index's realized volatility. The VIX is typically higher than the S&P 500's realized volatility because options transfer tail risk to sellers from buyers, and sellers are not willing to accept this undesired tail risk without compensation. The compensation that they earn — i.e., the spread between implied volatility and the underlying equity's realized volatility — is widely referred to as the volatility risk premium.
This framework provides the context for the concern that many market participants have expressed about option selling in the current environment. Option sellers generally lose money when volatility spikes, which often coincides with rapid declines in equity markets. With implied volatility at historical lows, many argue a large spike in realized volatility can spell serious trouble for option sellers.
What's been happening in the VIX?
Let's look at the VIX over the past few years. Figure 1 reports VIX levels since 2013, and also shows the VIX's corresponding percentiles relative to its full history going back to 1990. A percentile of zero indicates it is at an all-time low, and a percentile of 100 indicates an all-time high.
There are two insights evident over the last two years. First, the VIX is now near historical lows and, since 2013, on average it has been within the bottom third of its historical levels. Second, while we might have recently seen mostly low volatility, there have been several large spikes in recent years, and one in particular stands out.
The August 2015 China crash
During the "China crash" in August 2015, global equity markets declined rapidly after the Chinese government devalued the yuan. The S&P 500 declined by about 10% and the VIX rose 30 points (in percentage terms, 240%). This was a serious volatility event. For context, the largest VIX spike since 1990 occurred during the global financial crisis, when it jumped 60 points. The other largest VIX spikes cluster around jumps of about 30 points; August 2015 was one of those times.
To see how harmful this volatility spike was for option sellers, we constructed a simple, representative S&P 500 index option selling strategy. We sized option positions to realize average annualized returns that are similar to equity markets (6% per year) and delta hedge the options daily to manage directional risk to the underlying market. Figure 2 shows the strategy's returns and peak-to-trough drawdowns since 2013. The strategy exhibits the characteristics that we expect to see for volatility selling: positive average returns, but heavy losses coinciding with volatility spikes.
The illustrative strategy lost 3.1%, a little more than half of its average one year return, during the China crash. The strategy fully recovered its losses 10 months later. While unpleasant, it would be difficult to characterize this as a catastrophic loss, despite the magnitude of the volatility spike. In fact, the S&P 500 lost about triple the return over the same period, suggesting many equity investors might not have found the option strategy's tail risk to be unpalatable by comparison.
The 2008 global financial crisis
We selected August 2015 as a recent example of a large volatility spike during calm times — an important event that perhaps many have since forgotten. Of course, there have been larger, more memorable, volatility spikes. Take the global financial crisis. From July 2008 to November 2008, the S&P 500 declined around 40% and the VIX jumped about 60 points. The illustrative option selling strategy lost about half as much, almost 20%, over that same period. That's more than three times its annualized return, and six times its losses during the China crash.
This 2008 example is a much bigger deal, but we maintain these losses are survivable. While the options strategy was constructed to have annualized returns comparable to the S&P 500 index, it experienced only half the losses of the S&P 500 during the financial crisis. And the S&P 500 took a year longer to recover: 29 months vs. 17 months.
VIX exchange-traded products
Another volatility-selling strategy that is misunderstood by many is inverse VIX exchange-traded products, which have grown substantially over the past few years. Many commentators have pointed to these products as evidence that selling volatility is dangerous. We believe these concerns might be misplaced. Because VIX ETPs are exposed to percentage changes in implied volatility, they have more than 15 times the volatility of our illustrative option-selling strategy since inception. No strategy, even one with favorable return characteristics, is immune from the risk of blowing up if it is excessively sized.
Oversizing short volatility positions should be discouraged. But avoiding this exposure altogether for fear of monsters under the bed doesn't seem sensible. The volatility risk premium compensates volatility sellers for bearing uncomfortable risk during rapid market movements. There is ample evidence that exposure to this risk at moderate levels has been well-compensated across many markets and across time, including low volatility periods with some big spikes. How much risk should investors allocate to a heavy-tailed, but well-compensated exposure with low correlation to other exposures in their portfolios? Some; not a ton, but clearly not zero.
Roni Israelov is managing director and head of volatility strategies and Harsha Tummala, a vice president and portfolio manager of volatility strategies, at AQR Capital Management LLC, Greenwich, Conn. This content represents the views of the authors. It was submitted and edited under P&I guidelines, but is not a product of P&I's editorial team.