Authored by Jared Dillian of The 10th Man [9],
Ten years ago, when I worked on the equity index derivatives desk at Lehman, we used to sit around and dream things up. Like, we wondered out loud if you could actually create a volatility ETF.
It seemed like a brilliant idea. Who wouldn’t want to buy volatility? There were some studies going around that said holding volatility as an asset class alongside a diversified portfolio could improve the portfolio’s risk characteristics.
Around the same time, the Chicago Board Options Exchange (CBOE), owner of the VIX Volatility Index, was busy thinking up ways to securitize the Index. So it listed futures on the VIX, allowing investors take a long (or short) position in a VIX future and be positively (or negatively) exposed to changes in volatility.
Not long afterward, the CBOE listed options on the VIX, providing leveraged ways to play increases or decreases in volatility. (If you think about it, an option on the VIX actually gives you exposure to the volatility of volatility.)
Rise and Decline of the VXX
I left Lehman in 2008, but my former boss went on to develop an exchange-traded note (ETN) that gave investors direct exposure to the VIX… the very idea we used to kick around during lunch. The ticker was VXX.
People were really excited about VXX. A volatility ETF—now that is a financial innovation. Naturally, everyone piled into it, especially after the financial crisis, which was the biggest bull market in volatility the world had ever seen. Everyone wanted to be long volatility.
As you can see, that hasn’t worked out too well. Why not?
One thing investors have learned the hard way is that volatility has a very high cost of carry. People discovered this a few years earlier with USO, the crude oil ETF. The crude oil ETF holds crude oil futures, but since futures expire, it has to roll from the first month to the second month. If the futures are in contango [10], there’s a cost to roll the position.
VXX holds VIX futures and has to do the same thing. The vast majority of the time, the VIX futures are in contango, and the term structure is very steep. It’s very expensive to hold VXX, just like it’s expensive to hold a plain vanilla option. It “decays” over time.
VXX does go up occasionally when volatility spikes, but then it goes back to being expensive to hold again.
A lot of people got burned on VXX. They thought it was rigged and blamed the issuer, but they just failed to understand the cost-of-carry concept [11].
A client once told me that he thought this was the biggest wealth transfer from retail investors to professional investors in history. And he’s probably right.
Well, it seems there are a lot of bored financial engineers out there, because a lot of other volatility ETFs have popped up in recent years. Like TVIX, which is 2x long volatility. Or XIV, which is short volatility. If volatility is expensive to hold, wouldn’t shorting volatility be much more fun?
As it turns out, it is.
A Mad Rush into Short Volatility
So people have been discovering this over time. And the short volatility ETFs have been accumulating more and more assets. Until the last two weeks—when interest in the short volatility ETFs exploded:
This chart is from my friends at Deutsche Bank equity derivatives. It may be a little hard to understand, but the point is that people have been fleeing long volatility ETFs and piling into short volatility ETFs, to the point where the public is no longer net long volatility, for the first time in history.
This may seem like very arcane derivatives stuff, but it is not. This has huge implications.
Many people think that the V-bottom that we got last week was about dip-buyers or performance-chasers. That is true, but the bigger story is that retail investors (and professionals, too) came in and crushed volatility in a major way by plowing money into short volatility ETFs.
Whatever happened to this whole bit about owning volatility (I often refer to it as just “vol”) to improve portfolio characteristics?
No more. Now we have the volatility equivalent of “buy the dip”…
Smashing the Vol Spike
There are a few implications here:
- There’s really no other phrase for this than “naked speculation.” And naked speculation doesn’t usually happen on the lows. It happens on the highs.
- Retail investors have forgotten that being short volatility means being exposed to unlimited losses, just like if you were naked short a call or a put. If we get a vol spike of the magnitude of 2008—or even 2010 or 2011—“investors” in XIV are going to be very unhappy.
- In my entire career, I’ve never seen the retail community interested in selling volatility as a strategy. If retail investors ever sold volatility, it was usually a byproduct of something unrelated. Retail investors want to short volatility because they think it works all the time—and they probably think that way because since 2011, it has worked all the time.
Another ancillary concern here is that the popular inverse and leveraged volatility ETFs have the same issues that plain vanilla leveraged ETFs do: they have to rebalance daily. As the leveraged vol ETFs get larger, their rebalancing impact becomes more magnified and could lead to instability.
The statement implicit in selling volatility is “I think things are getting back to normal.” But things have been normal for a long time. We’ve gone 1,000-plus days without a 10% correction, and we will likely go even longer, because that last correction wasn’t even 10%. I could probably think of five things off the top of my head that could get the VIX back to 40 or 50.
The equity derivatives market is in many ways so complex that the financial press doesn’t even report on it. They don’t understand much they report on as it is, but with equity volatility, they’re really in over their heads.
What I’m describing here is a sea change in investor attitudes that has profound implications for the rest of the market. What you do with that information is up to you.
