This week we lost a truly special financial product. XIV was an Exchange Traded Fund (technically ETN but who cares) that sought to provide a daily return equal to the inverse of the daily return in the vix index. The vix index is an index constructed from the implied volatility of SPX options. So, basically vix is a measure of how volatile traders expect the stock market to be, or alternatively how nervous people are about stocks. When volatility or expectations of volatility of general nervousness went down then XIV went up and vice versa. That seems pretty boring until you consider the unique features of volatility as a trade-able instrument.
I. Behavior Of Volatility
Unlike a stock or commodity that can go up forever or down to $0, volatility spends most of the time being low and occasionally and temporarily spikes. Over the long run we see that vix tends to hang out around 12 or so. Sometimes it can be higher for an extended period of time, like during the 2008-2009 financial crisis, its floor seemed to be elevated to more like 20. Sometimes it spends a few years never getting much above 10. When there are big market moves or events, it can spike up into the 30s, 40s or even I think 90 at the peak of the financial crisis. But every time it goes up it crashes right back down again in a short period of time. Now, naively you might think that if it never gets much below 10 and occasionally goes way higher then the smart and obvious strategy is just to buy it when it’s low and then wait and sell when it spikes. After all, if you buy for 10 and a year later its 10, you haven’t lost anything right? Actually that’s wrong and that is part of the magic of XIV.
II. Volatility Term Structure
XIV and other volatility-linked ETFs relied on vix futures listed on the CBOE to get the exposure they wanted. Vix futures have the feature that they expire on a monthly basis so if you want to constantly stay short or long vix you have to frequently roll from the future that is expiring to the next future. XIV would move a fraction of its position from the closest to expiration future to the next future every day. So right now if it was still operating it would be moving a few% of its portfolio from the February future to the March future. When February ended it would start moving the March futures to April etc. By moving I just mean it would need to buy some of the February futures that it was short and sell the same number of them in the March contract. The very interesting thing about the vix futures is that when volatility is low the further out contracts are always priced higher than the closest to expiration contract. Why would this be? Well, it has to do with the difference between median and mean. Median is the value that a variable spends half its time above and half its time below. Since volatility is low most of the time, the median value of the vix is pretty close to its lowest values. The mean is just the average of the values that the variable takes on over time, so the mean value of the vix is higher than the median because sometimes you get those big spikes. The longer a future has before it expires the more likely it is to be worth the mean (in a stationary process like the vix). So when volatility is below the mean (which it is most of the time) the further dated futures will be higher than the closer dated ones. When vol spikes up then you see just the opposite. The long dated futures don’t diverge too much from the long term mean but the short dated ones really rip higher to reflect the short-term spike.
III. The XIV Strategy
So when XIV constantly rolled its exposure from the front month to 2nd month future, it would constantly be buying at a lower price than where it sold. The difference between the months was substantial. In 2017 typically the front month was around 10 and the 2nd month was around 11. So each month as long as volatility and the term structure remained constant XIV would earn about a 10% return just from rolling its position. Over the course of the year XIV earned almost 150%. And for any extended period when volatility stayed low it could earn these amazing returns. Overall from 2012 until the day before it blew up it earned nearly 1000%, vastly outpacing the stock market. So you could buy it and hold it and have a great ride but you had to be aware that the ride was going to end with it crashing into a brick wall at some point. Volatility always spikes eventually and eventually it would have a big enough spike to wipe out all its capital. The long term probability of that occurring was 100%. So, what do you do with a product like that? Its not really that difficult. Suppose that you put $5,000 into XIV a few years ago, and then whenever your XIV position reached $10,000 you sold off half of it. Over the course of the last 5 years or so you could have sold half your position at least 3 times. So, even though your original $5,000 investment is gone now, you would still have the $15,000 in cash. That still beats most other investments including the stock market over that period.
IV. The Real Magic of XIV
Now you might say, sure XIV is gone now but there are still vix futures and there is still VXX which is an ETF that positively tracks the vix. You could sell futures or sell VXX. That’s true but its just not as good. Suppose that you are short vix futures on Friday afternoon and they are trading for 10. On Sunday Kim Jong Un’s iphone crashes while he is about the answer the 15th question right on HQTrivia so he decides to Nuke San Francisco. You can imagine any horrible event you wish. On Monday morning the stock market is down 30% or whatever and vix is up to 100. Now you have not just lost the notional value of your future on Friday, you have lost 9 times that much! If you don’t have enough capital to cover your losses you could be forced to get out of other positions. If you had 10% of your money in XIV you could go to bed knowing that no matter how much volatility spiked that was all you could lose. If you had 10% of your money short VXX or vix futures you could potentially lose most or all of your net worth. That is what made XIV a truly magical product.
V. So Who Would Have Ended Up Paying?
XIV had about $2 Billion of exposure when it blew up. Considering the scenario where vix closed at 10 on Friday and opened at 100 on Monday it would have lost $18 Billion. The holders of the ETFs would be out their $2 Billion but who would absorb the rest of the loss? Someone would have to! I’m not 100% sure of the answer, but it seems like it would either have to be the exchange that offers the vix futures or the financial institution which managed XIV, or some combination of the 2. In any case, those companies were making tiny amounts of money in exchange for shouldering that enormous risk. While an XIV holder could make 10%/month in a low volatility environment, the management company (Credit Suisse) was making something like 0.5%/year! The CBOE was only making the transaction fees on the futures trades which are something like 0.01% of the value of each future. Because of this potential for explosive harm, its amazing that this product ever existed. And now that its gone and Credit Suisse will probably spend most of the money it ever made from it defending lawsuits from people who owned it. So, I’m sad to say that I don’t expect it to come back any time soon.