Rethinking VXX as an Equities Hedge

20Feb13

A popular trade is to hedge long equities exposure with a long position in the volatility ETF VXX (or VXZ, etc.)

I don’t like this play when VIX futures are contangoed, which they usually are (and are right now), because the futures term-structure is working against VXX a little bit each day like a dripping faucet.

In over-simplified terms, when the market goes up or down, the hedge will mostly do its job, but when the market goes sideways, the VIX futures term-structure acts as a drag on the portfolio.

Another approach is to flip that trade on its head: go long XIV (the daily inverse of VXX) and hedge with a short equities position (short individual stocks, long SH, etc.)

Now, you’re getting exposure to the “market” (because of the high correlation between the VIX and equities), plus the benefit of contangoed VIX futures (i.e. the faucet drips in your favor), and a reasonably tight beta hedge.

When VIX futures move to backwardation, the trade is then flipped back to what most folks are doing today: long equities + long VXX.

Shameless self-promotion: a portion of our own Volatility ETF Strategy follows this same basic logic. Click for actual, verifiable results.

This approach isn’t without its own drawbacks. The two biggest are:

First, the relationship between volatility ETFs and stocks is not perfectly correlated and not static, meaning the hedge won’t always do exactly what we think it’s going to do. But that’s also true for the popular equities + VXX trade, so this isn’t a new problem.

Second, and much more importantly, should volatility suddenly explode (i.e. the market crashes), the losses (gains) in XIV (VXX) will likely accelerate faster than the market, relative to norms.

Oversimplifying things (a lot), imagine that in a “normal” market, the daily move in XIV is 5x the change in the S&P 500. In a market crises, maybe that number jumps to 8x or 10x (warning: totally made up numbers).

That means, to stay well hedged to market crashes, you must carry less exposure to market beta, and rely more on contango for returns.

With the popular equities + VXX trade, you can carry a smaller hedge, and still maintain a lot of exposure to market beta, because that acceleration works in your favor during a market crash.

That’s an important difference.

The popular approach, long equities + long VXX, is a more beta-centric trade. The investor (a) can carry more beta, and (b) relies more on beta to generate returns.

My proposed trade, long XIV + short equities, is a less beta-centric trade. The investor (a) must carry less beta, and (b) relies more on the VIX futures term-structure to generate returns.

Regardless of which approach is “better”, I wouldn’t discount the fact that the popular equities + VXX approach is probably easier on most investors’ psyches. Most investors like beta, even if they don’t realize it.

Happy Trading,
ms

. . . . .

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.



8 Responses to “Rethinking VXX as an Equities Hedge”

  1. A very timely article given what the VIX did yesterday…

    I completely agree with you on VXX. I’ve never liked it as a hedge for long equity exposure. The vast majority of time you’re just bleeding money.

    However, I also don’t like the reverse trade you proposed for the exact reason you mentioned ( which happened yesterday ) I don’t think the payoff to having that hedge up when XIV crashes will nearly make up for all the wasted hedging during the majority of times when you didn’t need it.

    It’s like a person who month after month buys fire insurance for their home, and on that day when their house actually burns down the insurance company is only willing to pay a small fraction of the damages. In that case, why waste your money on monthly premiums in the first place? You can’t fight the volatility risk premium, so the best hedge seems to be a direct one. If you’re trading a VIX product, hedge directly with VIX futures to smooth returns, and cut out the imbalances that are inherent to indirect hedges.

    • 2 MarketSci

      I would suggest actually running some numbers before commenting.

      Hedging vol ETFs with VIX futures or another vol ETF, while a fine (but totally unrelated) strategy, is going to remove part of the benefit of the futures term-structure because the term-structure of the hedge asset runs counter to the primary ETF traded (see XVIX for an obvious extreme example).

      What I’ve proposed, hedging XIV with a short equities position doesn’t have that same drawback. The benefit of the futures term-structure is preserved.

      P.S. I’m not getting sucked into another discussion about whether risk-adjusted returns (as opposed to hoped-for-terminal-wealth) matter – that’s a tired conversation.

      michael

  2. 3 MarketSci

    P.S. to all readers…

    There was a very good post written by another quantitative blog within the last few months or so talking about a similar concept (long XIV + short SPY I believe?)

    I wanted to link to it here but can’t remember where I saw it. If anyone knows, please leave a comment so we can send some eyeballs their way.

    michael

  3. 6 Alex Argyros

    Assuming that you’re long equities, what would be a good hedge that doesn’t involve options? Specifically, I was wondering if you wanted to hedge X dollars using the S&P, would it make more sense to go short X/3 UPRO rather than X SPY, in order to take advantage of the bleed that leveraged ETFs suffer from?

  4. 7 Anton

    On February 25, 2013 implied volatility (VIX) increased by 38%, which is a rare development.
    There was no way one could tell in the middle of the day how big the $SPX drop would end up being and how high volatility would jump…..it could have been another Black Monday….
    Volatility-based position sizing is a key tool for managing drawdowns in many strategies, particularly those that are not based on proper diversification.
    So the question one had to answer this Monday was “Should you scale down intra-day or not?”
    If you did, you would have sacrificed return potential, because as you have shown in the blog, expected return INCREASES after days like this. If you did not, you could have experienced another Black Monday, which would have been a disaster if you were even slightly leveraged.
    Michael, have you run any numbers to obtain evidence with regard to what a prudent capital management tactics should be during days like this? Thank you. Anton

    • 8 MarketSci

      Hello Anton – my own thought is that we should much more aggressive in scaling down unhedged XIV (or other short vol ETP) exposure during market crises than we would with say normal equity exposure such as SPY.

      These short vol ETPs can get out of hand quickly due to the fact that (a) market crisis tends to be accompanied by strong backwardation, and (b) beta to the market is magnified during market crises.

      As far as how to answer that question for “normal” equity exposure such as SPY (that doesn’t suffer from the two issues above) I don’t have a one-size-fits-all answer. Real diversification is one solution. Options “disaster hedging” strategies are another. Beyond that it’s the tradeoff you mention between reducing risk and the fact that the farther a market is stretched, the more likely it is to rebound.

      michael


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s