In my line of work, I often receive questions about the CBOE Market Volatility Index (VIX), iPath S&P 500 VIX Short-Term Futures (VXX), and all the related offshoots. And every once in a while, I actually answer one. OK, just kidding -- I always try to answer. Anyway, I got this query last week: "I am trying to figure out why VXX was proportionally so much higher when compared to spot VIX values in the past? I know about the VXX reverse split, I think I understand that VXX is comprised of VIX futures attempting to create 30-day forward volatility exposure and everything that goes with that. Simple example of my question on closing price: VIX vs VXX I am just struggling to grasp why in the past VXX was priced proportionally so much higher. There is an increasing dislocation at all four dates. I would have thought that the two would have been more closely correlated on price in the past. Maybe my data is just wrong. The reason for my question is that I am trying to figure out how best to take a long volatility position that I can hold until VIX values inevitably move higher back towards their long term mean. VXX seems like the way to go; however I am uneasy because the past correlations confuse me." ...to which I responded: "VXX tracks a constant duration (30-day) VIX future, not VIX itself. Since futures premiums rise and fall, VXX can (and does) move differently from VIX. The reason you see that ugly relationship over time is that in order to maintain 30 days duration, VXX must roll from a nearer-month future (or swap equivalent) to a further out one. If you look right now, that means that every day VXX has to sell some May futures and buy June futures. Right here right now, May futures trade at $17.60, and Junes trade at $19.25. So VXX effectively loses money simply rolling. A little bit every day. But over time, that causes the disparity between VIX and VXX that you see. Now VIX futures don't always trade in contango, but they do way more often than not -- and always when VIX is at low levels. What's more, VXX listed in January 2009, hence the utterly awful VXX performance. Bottom line is they won't correlate that terrifically over time. That's not to say you can never buy VXX, just think of it like you were buying SPX puts. The SPX puts will decay in value if nothing happens, but will still lift if SPX gets plowed. Same effect in VXX, just different reasons. If VIX explodes, VXX will explode too, just not to the same degree." That all being said....wow, those numbers do look ugly in his table. I mean, think about it. In less than a year and a half, VIX has lost about 33% of its value. Meanwhile, VXX has shed about 85%. This truly highlights a valuable point on VXX: You simply can't hold it for an extended time frame. By all means, use it as a trading vehicle, but don't stay at the table too long. It's built for short-term plays, not long-term hedges. We also have VXZ, which tracks volatility futures of a four-to-seven month duration. VXZ has behaved better over time, as it does not have the same contango problem. The VIX futures curve out that far in time is generally quite flat. The issue, though, is that the premiums to VIX itself often trade quite high, so you always have risk that the "optimism" about future VIX pops will simmer down. Disclaimer: The views represented on this blog are those of the individual authors only, and do not necessarily represent the views of Schaeffer's Investment Research.