Fun With Numbers: Revisiting the VXX and XIV Trade

Something about this iPath S&P 500 VIX Short-Term Futures ETN (VXX) vs. the VelocityShares Daily Inverse VIX Short-Term ETN (XIV) short strategy occurred to me. I don't actually have to speculate how it works, I can use actual data to see how it performed over time.

Just to refresh, the plan is to go short both VXX and XIV, and attempt to capitalize on the knowledge that:

1. They move in almost perfect opposition to one another, thanks to the fact that XIV tracks inverse VXX, and,


2. They should both ultimately drift toward zero. VXX will drift thanks to the near-perpetual contango in VIX futures, and XIV will drift thanks to the occasional churn in VXX and the compounding thereof.

So, here's the plan. We go short $100,000 on both VXX and XIV, and we close the position one month later. I will define one month as 23 trading days. And, I'm ignoring the cost to borrow for the moment (more on that later).

XIV started trading on Nov. 30, 2010, and I used XIV and VXX closes through April 2, 2015, which gives us 1,068 separate one-month holding periods. The dual-short strategy turned a profit (again, not considering borrowing costs) 714 times. That's 66.9% of the time. If options are priced correctly, a dedicated delta-neutral options strategy should work 68.6% of the time (one standard deviation), so this certainly suggests that the short-gamma analogy has some real merit.

If options are priced correctly, then the net expected gain from shorting options should be near zero. That's because the loss-side is open-ended, whereas the profit-side is capped at whatever premium you took in. In reality, though, the marketplace modestly overprices risk, so the net-short strategy will turn a small profit over time.

Well, guess what, so does the dual short VXX-XIV strategy. I show that it has an average profit of $1,054.97, and a median profit of $1,386.44 on our $200,000 investment. Those were monthly numbers, so that's a 6.3% annual return if you look at the mean, or an 8.3% return if you prefer the median.

That's good, but not great considering there's more to the picture. The drawdowns are potentially severe. It peaked at nearly $41,000 in August 2011, or 20.5%, if you had the misfortune of coming up with this strategy back then.

And there's also those borrowing costs. It's not a fixed number, as it can vary over time depending upon the availability of shares in one or both of these pups. And not everyone faces the same costs. A big player might get "comped," a small player might get excess charges, and many can't get permission to short either or both anyway. But as @aaljechin shows me here, the borrowing cost has hovered at about 2.5% over the past couple of years. So, if you assume that going forward, and also assume you'll typically earn "about" 7.5% annually from the actual shorting, that lowers the real return to about 5%.

Is it worth tying up capital and risking a potentially large drawdown in order to take in about 5%? That's up to the individual to decide.

Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.