Accumulating XIV

Monday, August 8th, 2011 | Vance Harwood

As far as crashes go, this one has been pretty orderly.   Volume continues to be heavy, which I see as one of the few encouraging sign.  Ironically Treasuries are up—but lower than investment grade paper is getting hammered.   Obviously the big fear is a recession—but without any real news, similar to the Japanese earthquake/tsunami/reactor meltdown correction.

VIX is at 44 right now.   I wouldn’t be shocked to see it touch 50 before it starts winding its way down.   I’m accumulating XIV right now because I think the heavy action in the volatilty futures is probably over, and it is cheap enough (in the low 10s)  that even in the worst case scenarios the losses would be manageable.   I don’t pretend I can pick the top—or the bottom.  I’m looking for when the odds seem weighted in my favor.


Monday, August 8th, 2011 | Vance Harwood


  • Comment by sjev — August 8, 2011 @ 4:24 pm

    I am with you on this one. Have changed my short VXX position (that was suffering a heavy loss) to an even larger long XIV position. Put/call ratio and VIX/VXV ratios are very high now, these circumstances are usually followed by a rally.

  • Comment by Simon — August 8, 2011 @ 5:38 pm

    You don’t think VIX could reach its heights of 2008?

  • Comment by Vance — August 8, 2011 @ 9:25 pm

    Hi Simon,
    Of course, VIX could go that high, but my opinion is that there is less true bad news in the current panic. The US debt downgrade has been rumored for weeks. The Euro troubles don’t appear to directly threaten US banks, we don’t have hundred of billions of supposedly AAA debt turning into worthless junk.
    What we do have is emotion, and the less sophisticated investors are calling up their brokers and bailing out. Until that wave finishes we won’t get a rally.

    — Vance

  • Comment by Andrew — August 8, 2011 @ 11:45 pm

    Seems I accumulated XIV a little bit early. You make a good point about there not being any strong drivers for this falldown, except for maybe emotion. However in the end, emotion trumps everything, and that’s what XIV/VIX is all about. So overnight I did a 1:1 dollar match of my long XIV position with a short E-mini S&P 500 futures position. I think we’re in the process of trying to catch a falling knife, and I don’t want to stay unprotected until XIV has a positive session.

  • Comment by baruch — August 9, 2011 @ 1:06 pm

    Even if you see a nice mean reversion the backwardation is at about 20% right now. Do you really think this is a good trade? 2: You seem to want to avoid holding xiv in a bear market. So does this mean that you think we’re still in a bull market?

  • Comment by Vance — August 9, 2011 @ 1:32 pm

    Hi Baruch,
    My feeling is that we are still in a bull market. I don’t have that sense of no-upside that I got before the last bear. The other factor is that XIV has already been knocked down from 19 to 10, so it feels like there is a lot less downside–excepting of course a 2008 style meltdown. I need to look at the data, but my guess is that volatility peaks and goes into mean reversion well before the bottom of the market itself–that assumes this is a correction and not the beginning of a bear.

    — Vance

  • Comment by Andrew F — August 9, 2011 @ 6:28 pm

    Somewhat off-topic question: any idea why VXX/XIV tend to only move by about 50% the magnitude of the change in the underlying futures? Is it possible that both of these funds are the equivalent of 0.5x futures funds? Otherwise, I just don’t get why they would be geared lower.

  • Comment by Vance — August 9, 2011 @ 8:42 pm

    Hi Andrew F, For VXX vs futures I looked at the August 2011 volatility futures chart ( and the VXX chart for June through August. They are very similar (range 18 to 36 on futures, 20 to 36 on VXX).

    My guess is that you are assuming that the VIX index is the same as the futures prices–they are not. The VIX index does tend to move about 50% more than the futures / VXX movement. I think this is because the VIX is giving 30 day expected volatility (based on implied volatility of a set of SPX options) whereas the front month futures are giving a 30 to 60 day estimate of expected volatility depending how close they are to expiration. VXX has a constant 60 day estimate of expected volatility because it is a rolling mix of the front, and next to front futures designed to give a constant maturity.

    I can usually estimate the percentage change in the opening VIX value pretty closely by doubling the pre-market VXX percentage move.

    XIV is based on inverse percentage moves in the VXX, so it will be different than a true short. It has compounding / path dependencies. Over the last 6 months its range has been 9.5 to 19.3, so it has had a little wider range than VXX or the futures, but about 2x less than VIX.

    — Vance

  • Comment by Jev — August 10, 2011 @ 3:14 am

    I am puzzled about yesterdays discrepancy between VXX and XIV returns. It seems that XIV is 2% behind. While VXX fell ~10% , XIV rose only ~8%. So basically we missed out 2% on our long positions in comparison to short VXX. Any ideas?

  • Comment by Vance — August 10, 2011 @ 7:57 am

    Hi Jev,
    I noticed the discrepancy also. Today XIV is lagging by 1%. My guess is that the Velocity shares futures traders’ algorithms are struggling with the extreme market conditions–or perhaps the spreads are wide. Volatility futures are in strong backwardation, which is very unusual, and would tend to depress XIV because it is having to buy the near term futures and sell the next month.

    – Vance

  • Comment by Andrew — August 10, 2011 @ 12:41 pm

    Another concern is this heavy volatility in XIV is causing the fund to have to buy and sell more futures contracts than usual for the daily rebalance. This is creating a kind of friction loss that degrades returns. For instance, if 30 day future volatilty were to drop by 40% today from 30 to 18, XIV would only rise 40% to 14. Yet a month ago when futures actually were at 18, XIV was trading at 19. So this churning is BAAAD for XIV holders…

  • Comment by Jev — August 10, 2011 @ 3:20 pm

    I have made a quick analysis of this discrepancy, seems like it is the first time something like this is happening.
    Details are here:

  • Comment by Jev — August 10, 2011 @ 3:21 pm

    @Andrew: 40% down move != 40% up move, this is absolutely normal.

  • Comment by Andrew — August 10, 2011 @ 7:06 pm

    Jev, Yeah I wasn’t talking about the tracking error which is a legitimate beef and thanks for the link. I was just remarking that excessive churning is painful for inverse ETF holders that have to accept slippage (or whatever you want to call it) that occurs due to the need for daily rebalancing by the ETF… Andrew

  • Comment by jones — August 12, 2011 @ 10:29 am

    Guys, glad I found you all . Im really nervous because I also put my eye on the XIV all year and noticed it is a winner. I waited for VIX to go above 40 to enter a huge position knowing VIX will eventually go back to the low 20’s and in the mean while I will enjoy the contango. but what happened this week was extremly bad for XIV holders.
    Just looking on last friday 05/08 VIX daily high is 39.23 and XIV daily low is 10.57.
    We are just one week later and as I write VIX is approaching 34 and XIV doesn’t even manage to get to 10!!! this is hugeeee erosion in only one week, bigger than contango of several months in the XIV. what is going on? another few weeks like that and it will be a dissaster!! please help me and try to explain the situation and what is going to happen here. another turmoil in the markets can take XIV to very low numbers.

  • Comment by Vance — August 14, 2011 @ 11:23 pm

    Hi Jones,
    Remember that XIV is the daily inverse of the same index that VXX uses, NOT the VIX. This index always moves slower than the VIX, both up and down. Historically the volatility futures always go back into contango. Unless the world economy collapses I expect the same thing to happen this time–sooner or later.

    — Vance

  • Comment by jones — August 15, 2011 @ 7:41 am

    Hi Vance,
    I read some stuff about the curve of the vix right now going into backwardation.
    The big question now is what is the right the way to trade the xiv.
    I thought that buying it when vix blowing up to the 40’s and 50’s would be a great hold and buy but now with this huge backwardation it is not that simple any more. 7 points of backwardation compared to 1-2 points of contango in normal times can cause heavy loses in a matter of a month even if vix goes down. maybe waiting for the vix to go back to contango is the right time to enter but it would probably happen only when the vix go doen to 25 or so and then it’s not that tempting any more . but who gaurantees another disaster is not around the corner and then we go back to vix 50 and backwardation and heavy losses.
    What is the best entry point in your opinion to enter xiv?
    I’m also long from low 100’s but although vix went down i’m still losing.

  • Comment by Andrew — August 15, 2011 @ 8:42 am

    XIV is mainly short September futures right now, and the futures prices from September on out are only in slight backwardation, so from a contango/backwardation point of view, XIV doesn’t look that expensive to hold. On the other hand, VIX is dropping, so holding XIV currently gives you the benefits of being short volatility in an environment where volatility is dropping across the board. So at this rate, we should see contango being re-established soon (since volatility drops fastest in the near term), and lower volatility levels in general. So right now I am a holder…

  • Comment by jones — August 15, 2011 @ 9:06 am

    Andrew , correct me if I’m wrong but from what I read the fact that xiv is mostly short september doesnt change the daily huge backwardation until the expiry of aug contract because each day you get 1/23 of the difference between them no matter what time of the month we are.
    Agree with you things seem a little better right now concerning Volatillity but you can never know what is around the corner that can cause vix to rise again above 40 and backwardation between sep and oct contracts to grow as well. it takes me back to my question what would be a good entry point of the xiv in general in times of crisis and backwardation.
    Can anyone caclculate what will be the new value of xiv right now if we go back to vix 16? it wouldn’t be near the previous highs of 194 I quess.

  • Comment by Andrew — August 15, 2011 @ 11:23 pm

    Yes, what you’re saying is strictly true about the rollover still being August to September for another couple of days (roll period ends this Tuesday, since Wednesday is August VIX expiration). However, all of the money invested in XIV today would be exposed to what looks like a declining volatility environment… and then after Tuesday it would also be a lower backwardation environment. Obviously I am doing some directional thinking here… thinking this volatility spike is over, but it’s gamble like everything else. Finally to answer what would happen if vix went back to 16: If it dropped to 16 tomorrow and Sept. futures dropped to 16, then since XIV is almost totally invested in Sept futures, that would represent a 40% reduction in volatility, since Sept futures are at 26.7. So XIV would move from 10 to 14. Obviously this is not 19! Furthermore, it will not do this in one day, and so what happens to XIV is path-dependent. Whether vix returns to 16 anytime soon and what the value of XIV will be on that day is anyone’s guess.

  • Comment by jones — August 16, 2011 @ 5:26 pm

    Andrew, so basicly from your estimation I understand that 3 weeks of backwardation cost xiv almost 30% !! (14/19), heavy loses that erased months of contango.
    So again it raises the question how to decide when to enter , because enetering xiv at vix 70 for example with backwardation could be worst than vix 40 a short time after, very confusing and troubeling….

  • Comment by Andrew — August 17, 2011 @ 12:04 am

    Yes, it looks like exposure to the sharp backwardation during this extended period of volatility was responsible for half of the losses of XIV off the highs. I didn’t expect the volatility ‘spike’ to last so long. Live and learn. Next time volatility is high with strong backwardation, I will be far less likely to buy XIV. Anyway, backwardation is almost gone now, and so I feel ok holding XIV at this time…

  • Comment by Andrew — August 17, 2011 @ 10:27 am

    Jones, I want to clarify that all the losses off the high (XIV 1914) are not all due to backwardation. There is the rebalancing effect that increases as the square of the change in the index that an inverse fund is tracking. Simply put, when an inverse fund like XIV loses value, it must reduce exposure to the underlying index to assure correct day-to-day percentage tracking. When the index returns to the original value, the fund is holding less when the favorable move occurs, and so the fund does not return to the original value. I think when you run the numbers on XIV (using actual VIX futures prices), over half the 1914 discrepancy is due to this rebalance effect, but still a big chunk is due to the exposure to backwardation that we were talking about before.

  • Comment by jones — August 17, 2011 @ 1:39 pm

    Andrew, one thing I dont understand, when I first read about vxx everybody complained it is the worst long term investment possible so I thought what would be great to hold a short position on vxx and then xiv came to play.
    The rebalancing effect was always there and still xiv kept making new highs all the time even with vix staying at place , so how come all those terrible effects come into place so dramitically and before the last spike we didnt felt it so much and why do you keep your position with odds so against us?
    even today with aug contract expired vix is down 3 % and xiv also down 1.5%, this position is killing me:(

  • Comment by Vance — August 17, 2011 @ 1:57 pm

    Hi Jones,
    In my opinion the only real risk in holding XIV right now is the possibility of a bear market. If we stay in a bull market all that is required is patience. VIX has always reverted to the mean, VIX futures spend most of their time in contango. The market has taken a scare, and it will take a while to get over it, but judging by the drop-off in VIX, it is getting on its feet again. Trying to guess the “right” time or entry point in my opinion is a waste of time. No one can predict the future, there will always be risk, and we should not expect our investments to move in our direction immediately. All we can do is look for asymmetrical risks that favor us, and go with that.

    — Vance

  • Comment by Andrew — August 17, 2011 @ 8:27 pm

    Jones, to answer your question about why rebalancing was not a big deal before the big spike and now it became bad, the fact is it is a quadratic effect. If an index goes up by a fractional amount 0.01, you rebalance, and then it goes down by 0.01, the net fractional loss is 0.01 squared, or 0.0001 which is one-hundredth of one percent. But if an index oscillates by 0.1 on subsequent days, the loss is 0.1 squared which is 0.01 (one percent). Finally if the index oscillates over a period of time by 0.4 (40%), then the net loss will be 0.4 squared which is 0.16 which is 16%, which is painful. So using Vance’s terminology, there is an’asymmetric risk’ that volatility of VIX futures will all of a sudden blow up and wipe out months of gains. Since people are very uncomfortable with asymmetric risk, there is a hope that there is a considerable risk premium for people willing to take on those risks.

  • Comment by jones — August 18, 2011 @ 3:59 am

    Thanks for the explaination Andrew. still how would you explain the performance of xiv yesterday after the expiry of aug contract? vix goes down 4.5% and xiv goes down 2% with almost no backwardation. not a big move to relate it to rebalancing.

  • Comment by Andrew — August 18, 2011 @ 10:46 am

    Jones, yesterday VIX Sept futures went up and XIV was mostly into Sept futures, so I am not surprised XIV went down. Of course VIX and VIX futures don’t always move in the same direction…

  • Comment by Scott — August 31, 2011 @ 3:03 pm

    Hello everyone,

    Let me begin by saying that I have found your insight very informative and helpful. On 8/11 I initiated a position in XIV at $9.31 thinking that I would get in and out in a week or two to make some easy money as the VIX drops, however I was a little too early and the VIX spiked again. So on 8/29 I initiated another position at $7.98 thinking this was an opportunity to get in even lower, especially since I had assumed the backwardation in the futures market was mostly finished. Despite my thoughts, I am still struggling to make money. Do any of you still have your positions in XIV or have you given up? Should the VIX continue to decline to a normal level, what do you believe is in the future for XIV? Has it eroded too much?

  • Comment by Vance — August 31, 2011 @ 9:44 pm

    Hi Scott, I’m still holding onto a fair amount of XIV, with an average buyin price of ~$11. The 2nd VIX spike was pretty painful for me. Regarding XIV going forward, I think the key question in overall market direction–if we are headed into a bear market it will be a long time before your positions will be profitable. If the market levels out, or goes up in a fairly sustained manner, then I expect XIV to perform well. Unlike VXX, VIX options, etc. XIV is less about timing and more about patience. In general, in a neutral to bullish market XIV should perform pretty well. Usual disclaimers apply…

    — Vance

  • Comment by jg — September 14, 2011 @ 11:39 am

    My concern about the XIV (as someone who has a considerable position at about $8.25 which is underwater) is that it has plainly not performed well in two ways. 1. is that with the volatility since August 2011 it has not been an exact mirror of the VIX, tending not to spike as much (if you compare the VXX & XIV you will see this), and 2. it has over time drifted to make a less of a return. You can see the latter when you look at a year long chart where the cross over points between the two does not happen at zero but at an ever increasing slope. This means that as the VIX decreases, the XIV will increase less proportionally. This is bad for holding a longer position in this as I gather we folks on this thread have. Fundamentally, I like the idea of owning this now, but time seems to be against us. Does anyone know what this drift is a function of? Is it the mean-reverting nature of the VIX, management fees, or just a defect in the implementation of the XIV ??

  • Comment by Vance — September 14, 2011 @ 10:01 pm

    Hi Jeff,
    I haven’t looked at the last week, but before that XIV was performing exactly as the way it was designed to do. It’s not management fees or a poor implementation. First of all it’s not the inverse of the VIX–I suspect you know that, but your comment was ambiguous on that. It’s not the true inverse of VXX either, although that it much closer to the mark. It does, almost exactly match the opposite of the daily percentage moves of VXX. It is important to look at the percentage moves, not the absolute point moves when monitoring performance. A five percent move with VXX at 44 and XIV at 6 gives dramatically different point moves. There are also compounding or path dependency factors. If VXX goes up 10% on day and then down 9% the next it is back to where it started, XIV on the other hand will be down 2% from the starting point. A true volatility short like the ill fated IVO avoids these percentage style problems, but suffers from variable leverage (high leverage in bad times, low leverage in the good times).

    Obviously this is not a good time for XIV. Volatility has stayed elevated, and the short term volatility futures are trading higher than the longer term futures (backwardation). If history is any guide both of these situations will revert to the usual mean volatility and futures in contango.

    — Vance

TrackBack URL

Leave a comment