"There's been a lot of talk about this next song. Maybe… maybe too much talk."
- U2, Sunday Bloody Sunday
Volatility, or the lack thereof, has been much in the news lately. The Wall Street Journal, Financial Times, and Bloomberg have each run long features on the continued plumbing of generational lows in volatility measures such as the VIX, with vague yet mildly apocalyptic warnings about what could happen when things finally reverse. The FT, for example, refers to "an unlikely doomsday scenario of how the volatility complex could turn a manageable wildfire into an unholy inferno."
For my money - and yours! - such warnings are, as a general rule, not particularly useful. But we are not talking about averages here. The question today is whether historically low volatility is telling us something actionable, or is yet more sound and fury… signifying nothing.
It is true that the VIX index is at historic lows but also that this, in and of itself, tells us little. The fact that investors are collectively complacent is not prima facie evidence that worry is in order. Maybe things are just going well.
That said, a big part of the reason vol is so low is that as it has continued to fall, a growing number of investors have discovered that selling it is the easy path to instant riches (more on that below). As a result, every time vol has risen of late, e.g., before the first round of the French election - it has almost immediately been met by waves of selling.
Well, now we may be getting somewhere.
When everyone is on one side of a trade, there is often opportunity. I believe we are close to reaching that point, if we haven't already, with selling vol strategies. The "best" trades in this area are horribly overcrowded, and as a result are like a lit fuse.
Before we discuss how to play this, however, a brief primer.
How To Measure (And Trade) Volatility
While most investors have heard of the VIX index, far fewer understand what it actually measures. It is often referred to as the "fear gauge," with high readings suggesting investors are worried about volatility, and vice versa. Thus, the reason people are today concerned about the low levels of VIX is that it indicates investors are too complacent, particularly in the face of high valuations, political upheaval, geopolitical concerns, etc.
The purpose of VIX is relatively straightforward - it seeks to provide the expected, or implied, volatility for the S&P 500 over the next 30 days by measuring the prices of S&P 500 put and call options. Said differently, the VIX measures how volatile investors expect the market to be over the next month or so. Importantly, while many investors equate a high VIX with worries about stocks falling, volatility can of course be in either direction. This is perhaps best seen around earnings release dates for high-vol stocks such as Tesla (NASDAQ:TSLA), which will often have extraordinarily high levels of implied vol not because traders are worried specifically about the stock falling, but because they expect it to move strongly in one direction or the other.
That said, a big down move in the market tends to coincide with a rising VIX and vice versa; many are the stories about options traders who correctly called a large up move, only to see the value of their calls decline as implied vol fell.
The VIX has been around since the mid-to-late 80s, but because of the way the index is constructed, you cannot actually buy it. Futures and options on the index (yes, the volatility of volatility…) have traded since the mid-2000s, while most ETFs that offer exposure to the VIX "complex" only came into existence after the global financial crisis (see below for a partial list). This makes backtesting a bit more difficult, but hardly impossible.
In sum, the VIX is a non-investable index that has been around for a few decades, during which time a growing number of strategies to capitalize on it have evolved.
How Do Volatility ETFs Work?
One of the first questions people typically have when they stumble on this area is why the returns of volatility ETFs do not necessarily track the day-to-day moves in VIX. For example, on May 3, the VIX fell by a little over 1%, but the XIV ETF, which provides inverse exposure to volatility futures, also fell…by 2%. The UVXY ETF, meanwhile, which gives 2X long exposure to VIX futures, rose 4%.
What is going on?
The answer is that, since VIX itself is not investable, vol ETFs use the VIX futures curve to construct a synthetic one-month volatility exposure. Thus, day-to-day returns are driven not by VIX, but by movement along the curve.
For example, here is what the curve looked like on May 3, with VIX trading at 10.45:
As you can see, the curve is in contango (later months higher than closer months), which is pretty typical for this market, and a big part of what drives the stellar returns on short vol ETFs. As mentioned, these ETFs are constructed by buying and selling a mixture of VIX futures to construct that one-month exposure, and thus changes in those futures prices drive their returns. However, it is the shape of the curve that is responsible for the relentless push lower in prices.
Put simply, so long as the VIX is below the curve - historically about 75% of trading days - futures prices will inexorably move lower as time moves forward and they converge with the VIX. On the other hand, periods when VIX is above the curve leads to compounding returns for long ETFs - as prices rise to converge with VIX - and steep losses for short strategies.
In sum, long volatility ETFs - e.g. VXX and UVXY - tend to drip lower most days (and over time), but with periodic, and often violent, spikes higher, and vice versa for short ETFs such as XIV and SVXY.
Money Lying In The Street?
My interest in this area dates to late 2010, when I and a colleague began work on a paper colloquially titled, The Benefits of Selling Volatility. The premise was, and is, disarmingly simple: there has always existed a gap between implied volatility - what people expect volatility to be - and realized volatility - what it actually is. This gap, with implied vol being higher than realized, is robust across time periods, developed markets, most asset classes…in Zen terms, it just is.
Thus - and it is difficult to overstate this point - you can boost returns, more or less for free, by systematically selling volatility against a long equity portfolio.
My first reaction to this apparent fact was skepticism. What could possibly explain such an obvious mispricing… and free lunch? Clearly, we must have missed something…
In fact, the answer, like the strategy itself, is almost ridiculously simple.
Options function like insurance.
As anyone who has traded options can attest, you tend to be far more price sensitive as an options seller than as a buyer. The reasons are pretty straightforward: the seller gets only the premium and has more or less unlimited risk (more with calls than puts, but in both cases, it is a large potential liability), while the buyer is often betting on some sort of event that, if it occurs, will drive the price of the options far past the strike price.
Or as Bloomberg columnist Matt Levine once put it: "Traders should overpay for options, versus actual realized volatility, because they're buying insurance, and insurance costs money."
As shown below, implied volatility tends to be higher than realized volatility most of the time… even during crisis periods.
As noted, one way to capitalize on this is through a plain vanilla strategy of selling options against an equity portfolio. But there are far more juicy returns to be had through more direct strategies. The basic (ahem) options for selling vol are:
- Sell short a long volatility ETF (e.g. VXX or UVXY) or buy a short volatility index (e.g. XIV or SVXY).
- Sell long-dated calls on short volatility ETFs, or sell long-dated put on long ETFs.
There are also a variety of strategies for implementing these positions. Some traders wait for a vol "event" - e.g. the 2015 China mini-crisis - and then load up on short vol ETFs (or sell short long ones). Some use a systematic approach - e.g. a friend of a friend who "sells 100 shares of VXX every Friday." And some simply take big short positions in long-dated calls and hope for continued calm.
Leveraged ETFs such as UVXY have a similar makeup to the leveraged ETFs discussed in prior issues (e.g. DUST) in that the boosting of daily volatility will lead to steeper losses over time… but at the cost of higher short-term risk.
Everybody In The Pool?
As noted, selling volatility has become very popular of late - see below for the run-up in short interest on VXX - as even the historically low levels of VIX have been well above the levels of realized volatility.
Such extreme popularity in this type of strategy is worrying to say the least. Bluntly put, at some point there will be an "event" that causes vol to spike…and stay high for more than a day or two.
Or as a friend of mine recently said about short vol traders - riffing on the old adage about nickels and steamrollers - "it's like they're standing in the street saying: 'Hey look… another nickel!'"
Well, perhaps gold coins would be a better moniker. Consider the following:
The SVXY ETF, which tracks the inverse return of the VIX futures index, rose from 121 on April 13 - the last trading day before the French election - to 150 on May 1, for a two-week return of 24%. UVXY, which is a 2X leveraged long strategy for VIX futures, fell from 21.46 to 13.1 over the same period, for a whopping 37% return. (Don't even ask what that is annualized…)
Well… since you did ask, over the past year, SVXY has nearly tripled ($51 to $147), while UVXY has fallen by… wait for it… 97% ($420 to $13)! Returns for selling long-dated options on such strategies have been even better, as premiums have collapsed along with vol.
Longer term, here is a chart (which I swear I am not making up) of UVXY since its 2011 inception:
Yes, those prices are correct; it currently trades for about $13.
Meanwhile, here is SVXY since its 2011 inception:
The starting price is about $13, and today it trades for around $150.
The obvious question at this point is… assuming you have the fortitude (and capital) to ride out the inevitable spikes, why not just lever up and buy SVXY, short UVXY, and be done with this whole investing business? Or, given that long-term options exist on both products, why not sell puts/buy calls on SVXY, or vice versa on UVXY?
The answer is that such strategies do in fact make a lot of sense. (Well, perhaps except for the levering up bit.) But they require patience and timing… or, as we like to say in this publication, waiting for the fat pitch. Investing while sitting still necessarily involves a lot of… well, sitting.
Here, for example, is a chart of UVXY for the six months starting just before China's 2015 near-implosion.
And here is SVXY for the same period:
Hopefully, the point is clear: these strategies can be very profitable over long periods of time… but can also ruin you in the short term. Or even the not-so-short term.
Here is backtested data for UVXY for the entirety of 2007-09:
And here is SVXY:
That represents about a 15X max return for UVXY, and a more than 90% peak-to-trough decline for SVXY. I don't care who you are… nobody survives being on the wrong side of that trade.
So What To Do?
Well… what if you could get on the right side of that trade?
Look, trying to call the bottom in vol is - much like calling a market top - typically a fool's errand. And anyone thinking of putting on a long vol trade - whether by buying long vol products or options (my personal preference), shorting short vol products, or any of the various permutations therein - should only do so with money they are willing to lose.
The most likely result is that you will be stuck holding a worthless option, or be way underwater on your long vol fund whenever the next "event" hits. And to be honest, I don't really enjoy making this type of recommendation. It is not only speculative and highly dependent on unknowable events but also a bet on bad things happening.
That said, the timing feels right. Anytime something is dubbed the "magic money tree" - as the FT just dubbed the XIV - it's time to take the other side of that trade.
The relentless decline in volatility, both implied and realized, is the story in global markets today. It could turn out to be, as the kids say, a "big nothingburger," but I find the circular logic inherent in many defenses of these vol-selling strategies quite troubling. While it may not be the modern equivalent of portfolio insurance in 1987… or Long-Term Capital's epically-leveraged "nickels in front of steamrollers" strategy… or 2007's CDS squared… well, then again, maybe it is.
Again, I don't really enjoy making this recommendation, and please don't confuse this with "rooting for" chaos. But if there is a common theme in markets today, it is that risk (i.e. something bad and unexpected happening) is badly underpriced. As in poker, investors must play the hand they are dealt… and from a traditional risk asset framework, we are collectively staring at some pretty thin-looking rags.
 I'm not actually sure if something can be "mildly apocalyptic," but I really like the way it sounds…
 The underlying mechanics are pretty complex, but the idea is pretty basic.
 This is true of all indexes, but many, e.g. the S&P 500 or Russell 2000, are relatively easy to replicate by buying a representative number of the stocks in the index. A similar strategy for the VIX would require far too much capital and rebalancing to make it cost-effective.
 It is worth noting that many Zen practitioners - not to mention Albert Einstein - believe time to be an illusion created by human perceptions. But for our purposes we will hew to the traditional understanding.
 A summary of the paper is available on the CBOE website.
 There are a variety of ways to do this, but the most common method, as detailed in my paper, is to sell out-of-the-money puts and calls while holding a mixture of T-bills and a broad equity index.
Disclosure: I am/we are long UVXY. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.