Understanding VIX ETFs by Dave Nadig, Director of Exchange Traded Funds – FactSet
With volatility near multi-year lows, my email box has had more than the usual number of requests to discuss the pros and cons of trying to “play” volatility using one of the dozen or so volatility-based ETFs on the market. On the one hand, I get the appeal; if something seems artificially low, it’s logical to want to invest in it before it goes up. On the other hand, volatility exchange traded products are some of the most complex ETPs trading.
The straw that broke the camel’s back was the recent spate of articles with headlines touting the enormous short position hedge funds have in volatility futures, as some sort of evidence that volatility could somehow go even lower. I’ll come back to why that’s wooly thinking, but let’s start with what the heck an investment in a Volatility ETF even is.
At the root of most volatility ETFs is the CBOE Volatility Index (VIX). It’s often called the “fear index” and mislabeled “market volatility” on TV. What it actually is, however, is much more complex (and worth reading, if you’re a deep nerd). The VIX is calculated using the implied volatility of a basket of options on the S&P 500, both those about to expire, and those expiring next month. The idea is to come up with a number that represents the level of volatility the options market is expecting in the S&P 500 over the next 30 days. It’s essentially reverse engineering the math that options traders are using when they decide just how much to pay for a put or a call.[drizzle]
Related: ETFs: Who’s Buying What?
There are a few very important things to note here. The first is that there’s no direct mathematical relationship between the actual volatility of the S&P 500 and the VIX. The second is that it’s a forward-looking guess. The actual volatility of the market can be theoretically very low, while options traders are panicking about next month, and send VIX skyrocketing. Similarly, the market could go bananas and options traders could be sanguine about next month, and have VIX much less affected than you might expect.
In practice, what generally happens is that VIX rises substantially in response to downside surprises in the market. That’s the reason many investors have (for better or worse) seen an “investment” in VIX as a kind of hedge against market risk. It definitely has worked that way in short term crises, like last August:
Practically speaking, you can’t just “buy” the VIX, unfortunately. Instead, the CBOE has futures contracts on the value of VIX that are widely traded. Like any futures contract, the VIX futures are simply a bet on where a particular number is going to land on a particular day in the future. So while an oil future bought today might say “I want the right to buy oil for $40 in a month,” and thus be worth $2 when oil is trading at $42 on settlement day, a VIX futures contract is pegged against a certain value for VIX.
To put it another way, a futures contract for a month out is betting on what options traders, a month from now, will be predicting the volatility of the S&P 500 will be for the thirty days after that. In other words, not only are you not tracking actual market volatility, you’re buying a derivative (the futures contract) on a derivative (the implied volatility of the options).
VIX Futures Indexes and ETFs
Every ETF tracks some form of an index of VIX futures, the most popular ETFs, like the iPath S&P 500 VIX Short Term Futures ETN, track the S&P 500 VIX Short Term Futures Index. That index, rather than simply buying the near term contract and rolling it as it expires, actually rolls every day in order to make it’s notional exposure always 30 days out. So on the day that one month’s contract expires, it will be 100% in the nearest month. The next day, it notionally sells a small portion in order to buy the second month contract, repeating that process every trading day. The idea is to provide some level of “smoothness” in the daily roll.
Why does that matter? Because pretty much since the dawn of the modern VIX futures market, near-term VIX futures have been in contango. Here’s what the curve looks like right now:
As I write this, the VIX futures settling September 21 are $14.70. The October 19 contracts are at $16.92. Spot VIX is at 13.02. That means if VIX remains at current levels, the contango from Spot to front month is 13%, and 15% between September and October. Annualized that’s a 330% or 440% annualized headwind. This crippling contango is persistent in VIX futures for a pretty simple reason—the future is always unknown, and thus an estimation of potential outcomes has a wider variance than the driver of immediate volatility in the market, which is information.
The side effect of this contango is that any long term investment in a VIX futures based products ends up having enormous difficulties. Here, for example, is the actual spot VIX vs the biggest short-term ETP, VXX, and the inverse version, the VelocityShares Daily Inverse VIX Short-Term ETN, XIV:
Over the last year, VIX is down almost 50%, but the ETF tracking it is down almost 65%. You might think that the inverse version would be up 65%, but in fact it’s “only” up 40% in the last year. The reason for this is the way daily rebalancing affects the performance of all leveraged and inverse products; if the path of the thing being tracked (in this case, VIX) is itself volatile, the daily rebalance adds an additional drag on performance (in a smooth trending market, this actually works in your favor, delivering better than expected performance, but VIX is rarely in a smooth trending market).
Nuts and Bolts
All of this brings us to the sticky wicket in these products. Let’s do some quick math on how important these ETFs are in the market for volatility, back of the envelope style. Here are all the VIX ETFs listed in the US, with their assets in mid August:
Those tracking the first and second month futures (the short term products), had $2.3 billion in notional exposure. Some of these products are Exchange Traded Notes, like XIV. For XIV to actually get its exposure, a bank (in this case, Credit Suisse) actually issues debt that will promise the pattern of returns of a short position in the VIX futures index. To offset that risk, the ETN issuer goes out into the market and buys offsetting exposure in the futures market—in this case, they’ll sell a bunch of futures to match the index. If the ETN goes up in value, so does their short position, so it all balances out.
Some of these products, like VXX, are in fact regular old mutual funds under the hood. They own a giant pile of cash, and then a handful of total return swaps. Those swap counterparties also tend to be large money center banks like Goldman Sachs or Bank of New York. Those swap counterparties do the same thing the ETN issuer does—they hedge out the risk using futures.
Importantly, I am assuming that we can net these exposures for the purpose of this analysis. After all, if a big bank was behind both the long and inverse swaps, they wouldn’t hedge both in the futures market, they would likely just hedge out any net exposure.
Now let’s match up this known exposure with the size of the entire market for VIX futures on the day in question:
On the short end, this means that the near two months had about $5.1 billion in notional value outstanding, of which $2.3 billion(45%) was held by the ETFs above. This isn’t actually that high,in previous years I’ve run this analysis and found the number to be close to 100%. On the medium-term front, the number is much less interesting, with the ETF complex owning just $38 million of a $2 billion market for the fourth throughseventh month futures.
By cross referencing this against the CFTCs commitment of traders report, we can actually start drawing a few conclusions about market dynamics. Here’s the CoT report from August 16:
A few notes here: The CFTC considers anyone who actually is in the business of creating and selling futures positions to be a dealer and sell side. These would typically be the folks behind an ETN or providing the swaps underneath a VIX ETF. Asset Managers are pension funds, endowments, and other institutions, including mutual funds who are directly owning futures contracts. Leveraged Funds are hedge funds. I’ve used the weighted average settlement price to impute the notional exposure here, which is the best we can do, as the CFTC doesn’t report which specific contracts are owned by each class here. Hedge funds might be heavily concentrated in the out months, and dealers in the near months, and we’d have no idea.
Related: Another Active ETF(ish) Shoe Drops
Based on this table, we can make a few broad generalizations: Hedge funds are the biggest net sellers of volatility, and dealers are the largest net buyer. Put another way, the folks offsetting the VIX-based ETFs are hedge funds. But why? It’s easy to know why the swap counterparty agrees to provide the swap to the ETF—they get paid a direct fee for doing so. So why is the hedge fund community acting like a market maker to ETFs?
I would argue it’s not, as many articles have been suggesting, that they are making a bet on the VIX actually going lower. Rather, they’re counting on VIX just staying somewhere about where it is. As long as that happens, they’re earning the contango. They’re selling futures at $16, which, all else equal, will crash to $13 in a matter of weeks. If they’re right, they stand to earn not a market rate of return, but literally hundreds of percent returns, annualized.
Of course, if they’re wrong, and we have even a short term spike in actual volatility, they could quickly find their $16 short positions repriced into the 20s, 30s or higher.
So back to the original emails asking me about “investing in VIX.” The short answer is, you can’t. What you can do is enter the fray with the hedge funds that are counting on persistent contango and no surprises. You could choose to bet with them, using an inverse ETF, or to bet against them, but using a long ETF. In either case, however, it’s important to consider what could go wrong, and what being right actually looks like. Consider what happened in the last two weeks of August last year:
Here, the long speculator certainly made money, but nowhere during the few-hundred percent rise in VIX was there an opportunity to make more than about a 50% profit— in a futures market with a few-hundred percent annual headwind. Conversely, the short speculator lost nearly half their investment in a hurry. The presence of contango—and contango-hoarding sellers—acts as a dampener on the actual realizable gains from even the most prescient VIX ETF investor.
These products, in general, do exactly what they promise to do. They give you highly liquid, easy to access exposure to an incredibly narrow, highly volatile corner of the futures market, based on multi-layered derivative pricing and possibly daily resets (if you’re levered or inverse). That’s awesome. I love it when products deliver.
But if ever there was a market to be careful what you wish for, it’s this one.
Receive stories like this to your inbox as they are published. Subscribe by e-mail and follow @FactSet on Twitter. If you are looking to source FactSet data or analytics in your publication, email [email protected].
© Copyright 2000 – 2016 FactSet Research Systems Inc.[/drizzle]