The Search for the Holy Grail of Volatility Trading
Options expert Lawrence McMillan, of McMillan Analysis Corporation, outlines how volatility trading is subjective and highlights the number of products that have been rolled out over the years in various efforts to track the volatility index.
Depending on your viewpoint, the holy grail of volatility trading can take on a different meanings. To traders, it’s a product that tracks VIX closely, if not exactly. To exchanges and market makers, it’s a liquid product that draws a lot of trading interest.
To that end, a number of products have been rolled out over the years. None has matched the VIX options and futures in trading volume and they have been the basis for nearly every other product: VXX, in particular (the Barclay’s Volatility ETN).
As we have pointed out in various recent articles, there are problems with most of these as far as tracking VIX goes. VXX, for example, has lost 99.7% of its value since inception in January, 2009. It keeps reverse splitting, else it would be trading for pennies. But it only mirrors VIX (and even outperforms VIX), when the term structure of the VIX futures inverts and slopes downward. In particular, that is when the front month VIX future (July futures, currently) trades at a higher than the second month VIX future (August, currently). That inversion in the front two months is rare and generally only occurs during fairly steep corrections or bear markets, both of which have been rare commodities for years.
We recently wrote about the new ETFs—VXUP and VXDN—designed to track VIX daily. It’s really too soon to judge how these are going to do in that regard, but from May 19 to June 15, VIX rose from 12.85 to 15.39, a gain of 19.8%. VXUP rose only 2.9% over the same time period. That’s because it was so overpriced on May 19, trading at a 13% premium to Net Asset Value (NAV).
As we’ve often pointed out, the best way to simulate VIX—not duplicate it (because it can’t be duplicated)—is to take a short-term position in the VIX futures and roll them over. A short-term position in VIX options is the next best thing.
But traders have also realized that a product that settles to VIX will track it over the short-term as well. This has led to a somewhat ill-conceived preoccupation with weekly volatility derivatives.
Ironically, the theoretically best product in that regard—the CBOE’s Short-term Volatility Index (VXST) futures and options—have been a terrible bust. In fact, they have been delisted. No new contracts have been added in the last month and the last VXST futures and options expired Tuesday, June 30.
For some reason, the only weekly product that has had successful trading volume is the VXX options. VXX is based on the front two months of VIX futures and does not settle to VIX each week. Furthermore, as we know, VXX has a very negative bias to it, falling 99.7% since inception. No matter, the weekly options on VXX are popular with the public and they have almost all of the trading volume in weekly volatility options.
NEXT PAGE: Good News for Liquidity?|pagebreak|
We also pointed out in the past that the CBOE is now going to list weekly VIX futures and options, but those are going to use the regular VIX 30-day formula, even though the options expire weekly. These new weekly VIX futures are slated to start trading on July 23 and the VIX weekly options should start trading a couple of weeks after that.
This is actually the CBOE’s third attempt at a weekly product. Originally, they had weekly options listed on the mini VIX futures (VM), but they were delisted in January 2014.
The CBOE is searching for a product that will rival the trading volume in VXX weekly options. I thought they had it with the VXST options, but the volume was so bad that they dropped the product.
I personally know that we tried a few times to buy VXST options as protection for our put ratio spreads in our managed accounts, but we rarely were able to get a fill. All of the CBOE VIX-related products are listed solely on the CBOE, so there is just one market. A typical market in a VIX option is the following, taken from current data:
VIX July 20 call: 0.65 bid, 0.75 offer,
with 9,000 contracts on both the bid and offer.
With that deep of a market, it is usually easy to trade some calls at 0.70 - in the middle between the bid and offer.
VXST calls, however, were never like that. Ones similarly out-of-the-money, might have had a market something like 0.40 bid, 0.80 offered, with less than a hundred shown on each side. But we could never get anything done at 0.60 or any other price in the middle.
The new VIX weekly options will be a 30-day volatility estimate, because that’s what VIX is. However, each week, the weekly futures will settle to VIX, using the weekly SPX options that expire 30 days hence. So, in that regard, they will track VIX more closely prior to expiration than regular VIX futures do.
This should be superior to VXX weekly options, although so was VXST and that didn’t matter, because VXX never does settle to VIX at any time. When the front month VIX futures (part of the underlying to VXX) approach expiration, they are dropped from the VXX portfolio and the 3rd month VIX futures replace them.
The word is that the potential market makers in these new VIX weekly futures and options seem excited about the product. That’s good news for liquidity, I would think. Market makers who want to promote the product will usually fill orders between the bid and offer in order to attract more business.
By Lawrence McMillan, Founder and President, McMillan Analysis Corporation