Recent returns on VIX-linked ETPs may be enticing, but their risks are substantial.
While most investors focus on stock market performance, it is also possible to bet on expected market risk using instruments tied to the CBOE Volatility Index, or VIX. Long VIX positions are essentially a bet on higher market risk, which often coincides with poor market returns. Conversely, being short the VIX conveys a view of lower expected market risk and higher market returns.
In recent years, it has paid off to bet against market risk. A dollar invested in December 2010 in a VIX exchange-traded product that bet against market volatility would be worth $9.38 as of this writing. This phenomenal return can be attributed to the stock market climbing steadily higher with lower volatility than expected. Realized volatility is often lower than the volatility assumptions priced into the VIX because that index is a form of insurance against spikes in market risk. And no one would be willing to write that insurance without being compensated for that risk. This expected compensation is known as the volatility premium.
In June, I explored how investors can capture the volatility risk premium with equity index option-writing funds in an article titled "Systematically Harvesting the Volatility Premium." Like systematic equity index option-writing funds, inverse VIX-tracking ETPs profit from the volatility premium, but VIX ETPs are much riskier and their returns are much more correlated to equity market returns because they track the VIX futures curve and reset their exposure daily.
VIX-tracking ETPs do not invest directly in spot VIX; they use a combination of VIX futures contracts to gain exposure. The VIX is a measuring stick of stocks' expected volatility during the next 30 days. The VIX formula calculates the S&P 500's expected 30-day volatility by averaging the weighted market prices of S&P 500 put and call options across a wide range of strike prices. Individual investors could not easily trade the VIX until CBOE launched VIX futures in March 2004, and Barclays launched the first VIX ETPs, iPath S&P 500 VIX ST Futures ETN VXX and iPath S&P 500 VIX MT Futures ETN VXZ, in January 2009. As of mid-September 2017, there are 22 U.S. VIX-tracking ETPs with $5.2 billion in assets. Not only do these funds target VIX futures with different maturities, but they also offer leveraged (magnified) and inverse VIX futures returns. The top 10 VIX-tracking ETPs, as ranked by assets under management, are shown in the table below.
Fund sponsors initially offered long-only VIX ETPs as a means to hedge equity risk because S&P 500 drawdowns have historically coincided with greater volatility. From Nov. 1, 2006, through Sept. 15, 2017, the daily VIX and S&P 500 price returns have exhibited negative 0.73 correlation. Indeed, a long VIX ETP position can serve as a useful hedge for a sharp S&P 500 drawdown. For instance, the S&P 500 fell 3.6% after the Brexit vote on June 24, 2016 and VXX, which targets the one-month VIX futures contract, posted a gain of 23.7% on the same day. But holding a long position in a short-term VIX ETP has been a losing bet over the long haul. From Nov. 1, 2006, through Sept. 15, 2017, VXX has posted an annualized return of negative 70.3%. This is largely attributable to the VIX futures term structure and VIX ETPs' daily reset.
To maintain a long position in VIX futures, a fund has to sell contracts as they approach expiration to purchase longer-dated contracts. When the price of an expiring contract is lower than the longer-dated one, it hurts the investor's returns, even if there is no change in the spot market. This condition is known as contango. The term structure of VIX futures has been in contango for 71% of the trading days from Nov. 1, 2006, through Sept. 15, 2017.
Equity hedging costs may help explain why the VIX futures curve is usually in contango. At its core, the VIX is a formula that calculates expected volatility using S&P 500 option prices, so the volatility premium contributes to its state of contango. The volatility premium is the difference between an asset's implied (expected) volatility and its realized (actual) volatility in a given period. Each VIX futures contract represents the implied 30-day volatility as of that future contract's expiration date. Pricing forward-looking volatility in the future is difficult, and investors have historically overpaid to hedge risk. the graph below shows the one-month implied volatility (as measured by VIX) minus the S&P 500's one-month realized volatility from Dec. 1, 2006, through Sept. 15, 2017. On average, implied volatility has measured 3.4 percentage points higher than realized volatility during this time period.