On Monday I created a Pseudo Buy-Write position with VXX as the underlying and selling VIX July 25 options. Even with VIX at 23.98 the options did not trade below 0.10 on Tuesday, the last day to trade them. I thought about just letting the options expire, but two things deterred me: 1. the recent weird behaviours in VIX on their expiration morning, and 2. the potential for VXX to drop dramatically. Number 1 did not happen, the VIX opened at a very reasonable 24.05, number 2 did happen, although if I had sold VXX at opening I would have been ok with a .3 overall credit (excluding commisions). However throughout the day VXX dropped like a rock, closing -4.24% lower, a much more significant drop than the VIX's 1.68%.
On Tuesday afternoon I was thinking about the people that run the VXX. They have stated that their goal is to track a very short term index, the VIX--on the other hand their primary tool as market makers are the two short term contracts of the S&P 500 volatility futures. The literature I have seen claims that the VXX will roll over the VXX position, perhaps daily to decrease their position in the near term contract, while increasing their position in the next month contract. So, the day before July VIX option and S&P 500 voltatility contracts expire you would imagine that the VXX folks would be almost entirely invested in August contracts.
The fly in this ointment, is that the August Volatility futures have been runing considerably higher (~17%) than the July Volatility futures. The VIX/VXX ratio was running at around .3711, which is less than 5% away from its average value of .395 --so the simple rolling position approach doesn't match up to the reality of a full August based position. If the VXX market makers were just selling relatively cheap July Futures contracts and buying expensive August contracts the value of VXX relative to VIX should have been dropping all month--which it wasn't.
I managed to scare myself into thinking that this correction in VXX value might happen on today, on options expiration day morning --the only time ever where the VIX and the July Volatility contracts have to line up--by definition. To avoid taking the risk on VXX I closed out the position for a net loss of 0.02 per contract. A very small price to pay for a good night's sleep.
The VXX did take a big hit, but the supposed underlying, the Aug futures dropped almost the same amount (-3.78%). The next mindbender is whether this drop might have been driven by the VXX folks selling August futures contracts today, either to close out open positions, or to take short positions to hedge VXX against possible drops in August futures relative to the VIX index. With high volume days running in the million share range ( > $60Million worth) there may be enough money to trade a lot of contracts). The futures chart shows a lot of open interest creation in the last few days.
There was certainly nothing noteworthy in the behavior of the $SPX index or VIX that would suggest the market or these indexes drove the major drop-off in the August volatility futures today.
My guess is that there is more going on under the hood of VXX than just rolling futures contracts day by day...
No comments:
Post a Comment