(This article originally published 12/1/15 on Trading Volatility)
On April 5, 2014 I wrote an article titled Shorting Volatility: Time To Reduce Profit Expectations for 2014 which outlined why I thought it would be difficult for XIV (the fund which tracks the inverse daily return of volatility futures) to finish positive for 2014. When the article was published XIV had returned an impressive +547% over the past 28 months, as shown in the chart below.
Understandably, many people disagreed with my view, yet XIV finished the 2014 year -9%.
Later in 2014 (September) I wrote The End of an Era for the Market, in which I discussed how a choppy market that is no longer a reliable environment for shorting volatility had become the norm as the Fed concluded their QE program. XIV lost 37% of its value over the next four months and is currently down 30% since that post.
If it wasn't already clear, XIV is not a buy and hold security. It has returned -20% over the past year and -7% over the past two years. The environment is not the same as what it was in 2012 and 2013.
How did I make these calls?
Because returns in XIV are heavily dependent on the Weekly Roll Yield (WRY).
A major factor for the lower returns in the past two years comes from a smaller weekly roll yield that is used to propel XIV. The inverse of the XIV WRY is the VXX WRY (which comes from the difference between first and second month VIX futures). We track this important metric on our site, and as you can see in the graph below, the roll yield has moved from -2% to -1% over the past several years and has recently started to move toward 0%.
This is most important graph because the WRY serves as a headwind or tailwind to VIX ETPs. As the WRY approaches zero, trends become less stable and our advantage in trading XIV or VXX diminishes. Conversely, as the WRY moves away from zero trends become stronger.
Right now there is less of a headwind for VXX and less of a tailwind for XIV/SVXY. At this point in our old bull market we are also at high risk for the VXX WRY turning positive, which would give VXX a tailwind and XIV a headwind.
Revisiting our graph which compares the annual returns of XIV to the average Weekly Roll Yield (below), we can see why the WRY is so important.
Looking at the trendline we can see that with a fairly constant 1% WRY, we can generally expect XIV to return ~+25% per year. As the average WRY moves towards 2% we can expect XIV to return ~+120% per year. At 0%, XIV is likely to return ~-60%. The recent shrinking of the WRY from 2% to 1% has an enormous impact on returns and that's why it is so important. A continued move of the WRY toward 0% or higher would be alarming to investors who plan to buy-and-hold XIV.
Because the WRY is such an important factor in the performance of VIX ETPs, our approach is to trade XIV and VXX using the WRY as a primary input for our VXX Bias and ZIV Bias indicators (our strategies are outlined here). While XIV is negative on the year, our VXX Bias indicator has put forth a performance of +33% YTD.
No credible trader will ever say that trading is easy, and that is doubly true when it comes to volatility. It requires a deep understanding of market forces, a solid process-oriented plan, and discipline in execution. In the current environment it will be more important than ever to stay on top of changing conditions.
We have been in the business of helping traders stay on top of the volatility market for three years now. We look forward to a fourth year of providing a variety of free data services throughout our website and are always happy to answer questions submitted through our Contact page. We also offer subscriber services to people looking for additional data, metrics, automated alerts and daily summaries, and access to a community of volatility traders. If you're interested, don't hesitate to drop us a line.
Hypothetical and Simulated Performance Disclaimer
The results are based on simulated or hypothetical performance results that have certain inherent limitations. Unlike the results shown in an actual performance record, these results do not represent actual trading. Also, because these trades have not actually been executed, these results may have under- or over-compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated or hypothetical trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profits or losses similar to these being shown. Additional performance differences in backtests arise from the methodology of using the 4:00pm ET closing values for XIV, VXX, and ZIV as approximated trade prices for indicators that require VIX and VIX futures to settle at 4:15pm ET.
Disclosure: I am/we are long XIV.