XIV and VXX return difference

[ajensen]

Usually the daily return of VXX is close to that of the opposite of XIV. Since December 2010, when XIV started trading, the sum of the returns of XIV and VXX has had mean of 0.05% and standard deviation of 0.33%. For the last few days, a portfolio of long VXX and XIV has done terribly, losing about 2% a day (see below). Daily management fees or differential cash returns cannot explain this. Can anyone explain the greatly diverging returns? I understand that VXX returns will not be the opposite the XIV returns over a holding period of many days, but I thought the two products should track each other closely on a daily basis. VXX and XIV are exchange-traded notes, and a spike in volatility (which generally benefits VXX) could increase the chance of a bankruptcy of Credit Suisse, the issuer. But I don't this explains the underperformance of VXX, because VIXY, an ETF that does not have Credit Suisse credit risk, is doing almost as badly as VXX.

Date VXX XIV VXX+XIV
6/24/2016 23.66 (26.79) (3.13)
6/27/2016 (0.12) (2.45) (2.57)
6/28/2016 (9.58) 7.74 (1.84) as of 3pm

 

[lindq]

I noticed that as well, and at one point yesterday there was such a differential I thought I had a data problem.

My only explanation is that none of the VIX derivatives are perfect instruments, and as VXX and XIV are already quite removed from the actual VIX, extreme moves in volatility impacting the first and second month's futures contracts can just put things out of balance for short periods.

Shit happens.

 

[JackRab]

This is breaking my brain... but there are some issues with both VXX and XIV.

- Firstly, an inverse of an index always loses relative to the index returns...
Long: 100 +50% = 150 then -33% = 100, returns 0 after 2 sessions
Inverse: 100 -50% = 50 then +33% = 66, returns - 33%
So both long VXX and XIV as a strategy gives negative returns... over last year it's -34% (checked historical prices)

- Secondly, the long VXX should give negative returns in general when the volatility term structure is in contango, which it usually is.... the 2-month volatility will drop towards the 1-month and therefore more negative returns so VXX on the long run is losing.

Maybe the bigger difference between the two has something to do with this?

Also, I think these (and ETF's in general) are restricted in shorting, so it's biased up already... so maybe a large move upwards in either one is likely to be less then expected...?

Thoughts??

 

[d08]

Noticed the same thing, typically the differences were minor but after Brexit they are trading almost like completely uncorrelated instruments. I expect them to return to normal within a week though.

 

[Victory5]

This is breaking my brain... but there are some issues with both VXX and XIV.

- Firstly, an inverse of an index always loses relative to the index returns...
Long: 100 +50% = 150 then -33% = 100, returns 0 after 2 sessions
Inverse: 100 -50% = 50 then +33% = 66, returns - 33%
So both long VXX and XIV as a strategy gives negative returns... over last year it's -34% (checked historical prices)

- Secondly, the long VXX should give negative returns in general when the volatility term structure is in contango, which it usually is.... the 2-month volatility will drop towards the 1-month and therefore more negative returns so VXX on the long run is losing.

Maybe the bigger difference between the two has something to do with this?

Also, I think these (and ETF's in general) are restricted in shorting, so it's biased up already... so maybe a large move upwards in either one is likely to be less then expected...?

Thoughts??

This is just how math works...not sure what the confusion is. If you don't understand the relationship between the two just look at their respective charts.

Vol decay is real and something people should be aware of.

 

[ajensen]

Bloomberg has an article on the strange behavior of VXX and XIV this week:

http://www.bloomberg.com/news/artic...avoc-with-volatility-notes-taking-futures-cue
VIX Tempest Plays Havoc With Volatility Notes Taking Futures Cue

by Joseph Ciolli and
Alexander Dzerneyko
Bloomberg
July 1, 2016 — 12:00 AM EDT Updated on July 1, 2016 — 8:21 AM EDT

“I don’t like these things and I don’t trade them,” said Steve Sosnick, an equity risk manager at Timber Hill LLC, the market-making unit of Greenwich, Connecticut-based Interactive Brokers Group Inc. “They’re embedded options, so each layer you get away from the original product means more variability. There are so many moving parts and hidden fees that ultimately make these products fluky and difficult to trade.”

One way to see how strange the products can get in times of stress is to compare daily moves in VXX with a security that should be its polar opposite, the VelocityShares Daily Inverse VIX Short-Term ETN, which is short volatility. On most days, the two securities’ price action is an almost perfect mirror image, which makes sense since one is long and one is short.

Andrew X. Smith, a spokesman for Barclays Plc, the issuer of iPath funds, declined to comment. A spokesperson from Janus Capital Group Inc., which owns VelocityShares, did not respond to a request for comment.

On Monday the securities not only moved in the same direction, but did so to a degree not seen April 2013, Bloomberg data show.

 

[lesserfool]

Also, CS released this note today. I am still trying to make heads or tails of it:

http://finance.yahoo.com/news/credit-suisse-places-conditions-acceptance-204500232.html

 

[ajensen]

Also, CS released this note today. I am still trying to make heads or tails of it:

http://finance.yahoo.com/news/credit-suisse-places-conditions-acceptance-204500232.html

Thanks for mentioning this. Quoting the press release:

Beginning on July 5, 2016, Credit Suisse may issue additional ETNs on a weekly basis and may condition its acceptance of a counterparty's offer to purchase the ETNs on its agreement to sell to Credit Suisse certain hedging instruments consistent with Credit Suisse's hedging strategy, including but not limited to swaps. Any such hedging instruments will be executed on the basis of the indicative value of the ETNs at that time

I think this means that to create new shares of XIV, Credit Suisse now wants its counterparties to do a VIX futures trade or equivalent swap trade with them that replicates XIV in addition to putting up cash equal to the indicative value (plus their fee). With the increased volatility of XIV, CS wants to hedged at the time it creates new shares of XIV, instead of hedging after it creates the shares.

 

[JackRab]

This is just how math works...not sure what the confusion is. If you don't understand the relationship between the two just look at their respective charts.

Vol decay is real and something people should be aware of.

My confusion is about what the OP put down, trying to make sense of it. And I was just pointing out that the relationship is far from a perfect strategy, since both long returns negative.. indeed because of maths. Thanks for the input.