By Samuel Lee
A version of this article was published in the February 2014 issue of Morningstar ETFInvestor. Download a complimentary copy here.
The S&P 500 Dynamic VEQTOR Index is a shining example of financial innovation gone wrong. Start with the name: If an investment sounds like it came out of a science fiction novel, it's likely too creative for its own good. Products with futuristic names are targeted at people impressed by the veneer of sophistication. They are rarely good investments.
Why am I picking on this index? Because at least $850 million tracks it through the Barclays S&P 500 Dynamic VEQTOR ETN (NYSEARCA:VQT) and PowerShares S&P 500 Downside Hedged (NYSEARCA:PHDG) despite the index's short history. I fear that the product is sold on the basis of a literally fantastic back-test that seems to offer the Holy Grail of asset allocation: high returns and negative correlation to the market during downdrafts. Had this strategy been implemented prior to the financial crisis, it would have skyrocketed more than 50% when everything else was collapsing.
Before I get into why the back-tested returns are fantastic, I need to devote a fair amount of space on the nature of the strategy and its underlying exposures (yet another bad sign).
The index allocates among the S&P 500, the S&P 500 VIX Short-Term Futures Index, and cash. The critical piece is the VIX allocation. VIX is the options market's estimate of the S&P 500's future 30-day volatility. The measure is derived from the implied volatilities used to set call and put options on the S&P 500. When the market declines, expected volatility almost always spikes. When very bad news comes out, VIX can shoot the lights out. Owning VIX is like taking out an insurance policy on market fear. If you could directly own VIX, you'd have to be crazy not to. Even though VIX is highly mean-reverting and generates little to no long-run return, its insurancelike properties could dramatically improve a diversified portfolio's risk-adjusted returns. Alas, VIX is not directly investable. One can only invest in VIX-linked futures. The gap between the two is big enough to drive a truck through.
Futures are contracts that obligate the buyer (or seller) to purchase (or sell) some quantity of an asset at a set price at a set time in the, well, future. It would be insane to sell VIX futures without demanding some kind of compensation. The market's solution is that far-dated VIX futures prices are persistently higher than near-dated futures prices, a condition known as contango. Sellers of VIX futures expect to collect a big premium as time passes and higher VIX futures prices converge to the lower spot price. An investor who wishes to maintain constant long exposure to VIX futures must regularly "roll over" his contracts by buying more-expensive long-dated contracts after receiving money from his expiring contracts.
Think about the price you'd demand if you were selling insurance--as VIX futures sellers do--that paid off severalfold during falling markets. You'd charge a hefty premium, because you lose money at the worst times possible, when money is scarce. How hefty? A long position in VIX futures from the beginning of January 2006 to the end of January 2014 would have returned negative 41.4% annualized. The VIX over the same period returned 5.4% annualized.
But what if you could avoid much of the nasty contango and selectively boost your exposure to VIX futures when it's most likely to pay off?
Enter the VEQTOR. It aims to ramp up its VIX exposure at times when volatility is more likely to surge. The VEQTOR Index can be thought of as a highly leveraged market-timing strategy that effectively shorts the market at certain times by greatly increasing its exposure to VIX futures.
My colleague John Gabriel nicely summarizes the gory details of how this market-timing system works:
The benchmark dynamically allocates to the S&P 500 Index and S&P 500 VIX Short-Term Futures. It also includes a "stop-loss" mechanism that shifts the entirety to cash if the index falls by 2% or more over the previous five trading days. The index will remain in cash until the five-day return rises above negative 2%. Two signals are monitored daily and determine the index's exact allocations: one-month realized volatility of the S&P 500 Index and the trend of implied volatility (calculated by reference to five-day and 20-day moving averages of the VIX Index). Volatility is either in an up-trend, down-trend, or no trend. An up-trend is when five-day implied volatility is greater than 20-day implied volatility for at least 10 consecutive trading days. In an up-trend, depending on the realized volatility environment, the equity allocation can be 90% (realized volatility less than 10%), 85% (RV 10%-20%), 75% (RV 20%-35%), or 60% (RV above 35%). A down-trend is the opposite condition, and no trend is when there haven't been 10 consecutive days of any trend. When there is no trend, the equity allocation can be 97.5% (RV less than 10%), 90% (RV 10%-20%), 85% (RV 20%-35%), 75% (RV 35%-45%), or 60% (RV above 45%). Finally, in a down-trend, the equity allocation can be 97.5% (RV less than 20%), 90% (RV 20%-35%), 85% (RV 35%-45%), or 75% (RV above 45%). The index allocations can change daily based on the volatility environment and trend.
Got that? Note the strategy's numerous parameters. Why does it go into cash when five-day returns drop below negative 2%? Why not 0%? Why not 10-day returns? And why are trends defined as the crossover of five- and 20-day moving averages for 10 consecutive days? Why are the downward adjustments to the stock allocation in increments of 5%, 10%, and 15% during an up-trend, but in increments of 7.5%, 5%, 10%, and 15% in no trend, and 7.5%, 5%, and 10% in a down-trend? How does changing these parameters affect the back-test?
The VEQTOR looks overengineered. I wouldn't trust its back-test at all, never mind its short history. The back-test checks off none of the criteria that indicate soundness (I provide a framework for assessing back-tests here).
A risk investors often overlook is that the back-tested strategy's returns could be horrible if we simply change some of its parameters. Since the VEQTOR Index went live on Nov. 18, 2009, the strategy hasn't been too harmful to investor wealth, but that's nowhere near enough time to assess its efficacy. Since then, the VIX has experienced two massive spikes, and the VEQTOR only managed to exploit one of them.
When a strategy derives most of its gains from rare tail events, you need a lot of data to assess with a high degree of certainty its basic properties, like the mean and standard deviation. Even then, that data could be useless or even misleading, as the underlying processes could change. Indeed, the rise of VIX-linked products has affected the behavior of volatility markets. I don't think it makes sense to speculate in VIX with this strategy given the uncertainties surrounding the properties of volatility spikes and whether this index's strategy could plausibly exploit them. The index's back-test begins in 2005, a risibly short period by the standards of quantitative investing.
Of course, the VEQTOR Index might work out just fine. There is some economic intuition behind its strategy. Major volatility spikes are usually preceded by rising volatility, perhaps because investors underreact to bad news. However, the allocation to VIX will be a persistent drag on returns. While long VIX exposure can be as low as 2.5%, that's only during exceptionally benign markets. On average, it'll probably be closer to 10%. Assuming a yearly roll cost of negative 40%, that's an annual drag of 4%, high enough to wipe out much of the expected return from its equities exposure. I wouldn't bother.
Disclosure: Morningstar, Inc. licenses its indexes to institutions for a variety of reasons, including the creation of investment products and the benchmarking of existing products. When licensing indexes for the creation or benchmarking of investment products, Morningstar receives fees that are mainly based on fund assets under management. As of Sept. 30, 2012, AlphaPro Management, BlackRock Asset Management, First Asset, First Trust, Invesco, Merrill Lynch, Northern Trust, Nuveen, and Van Eck license one or more Morningstar indexes for this purpose. These investment products are not sponsored, issued, marketed, or sold by Morningstar. Morningstar does not make any representation regarding the advisability of investing in any investment product based on or benchmarked against a Morningstar index.