Our old friend iPath S&P 500 VIX Short-Term Futures ETN (VXX) hit new all-time lows again on Friday. I know, I know -- shocking.
But hey, I'm not here to bury VXX (again). I'm not really here to praise it, either. I'm just going to attempt some objectivity on the subject.
VXX first listed 6.5 years ago at a split-adjusted price of $6,400. It's now $16. I'll spare you the math on that one.
It has a less evil -- and very much less publicized -- cousin, CBOE 3-Month Volatility Index (VXV). The difference is in the duration, as it proxies a constant 30-day VIX future. The iPath S&P 500 VIX Mid-Term Futures ETN (VXZ) proxies a 4-7 month VIX future. The term structure is generally much closer to flat about that far out, perhaps very slightly upwards -sloping. While VXX loses value literally every nanosecond thanks to the contango, it's less absolute in VXZ.
And as such, VXZ has done way better -- or, more accurately, less badly. VXZ closed at $10.39 on Friday, vs. a listing price of $100 6.5 years ago. Thus it hasn't even dropped 90%! Wahoo!
So, does that make VXZ a better product than VXX? I would suggest not.
I mean, obviously if you want to buy and hold something, VXZ costs you less money. But I hope we know now to NEVER buy and hold a volatility ETF.
Rather, let's go on the assumption we will only use these pups as trading vehicles and only hold them in the short term. And our goal is to either hedge against a volatility spike, or speculate on one. If that's the case, you should just use VXX for the somewhat intuitive reason that a shorter-term vehicle tracks short-term moves better.
Since inception, day-over-day VXX tracks about 46% of the day-over-day VIX move. In other words, if VIX popped 10% today, you'd get a 4.6% pop in VXX, on average.
What's more, the relationship is fairly consistent over time. Here's how a rolling 20-day VXX Chg to VIX Chg relationship looks since the beginning of VXX time.
It dipped down to 20% midway last year for no apparent reason. But by and large it spends almost the entirety of the time tracking between 30% and 60% of the VIX move.
How about VXZ? It tracks about 20% of the VIX move. The graph looks like this:
Thus, even though VXZ loses value less rapidly, you'd have to buy maybe 2.5x as much VXZ to track VIX just as effectively. And yes, I know this is all basically a visual representation of beta.
If you want to play longer term, then yes, use VXZ. But you really shouldn't play any of these with any duration; they're still going to drift in value and those extra shares you need to own will end up costing.
If you're playing short term, VXX will not kill your portfolio. Just do the sizing right and DON'T hold it.
Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.