Best and Worst ETFs and CEFs for VIX Exposure
One of the common threads of discussion in the financial community lately has been volatility - where it is now, where it has been, and where it might be soon. Though yesterday's events certainly caused a spike, the futures this morning seem to indicate that the VIX will likely drop again into the weekend.
Regardless of what will happen today, I wanted to cover which ETFs, either by asset choice or by asset management, are most affected by the VIX and what it represents.
Below are the twenty most and least correlated ETFs and CEFs to the VIX.
Twenty Most Negatively Volatility Correlated ETFs and CEFs
| Symbol | Corr |
| JSC | -20.42% |
| MVV | -19.76% |
| SSO | -17.30% |
| DDM | -16.28% |
| KRE | -15.46% |
| JPP | -14.79% |
| DXD | -12.77% |
| SDS | -12.76% |
| MZZ | -12.37% |
| XPH | -11.31% |
| KIE | -10.17% |
| RPV | -9.88% |
| CHI | -9.71% |
| QTEC | -9.67% |
| QID | -9.54% |
| HCF | -9.50% |
| DBC | -9.48% |
| MUA | -9.20% |
| PNI | -9.16% |
| NZX | -8.49% |
Ten Most Positively Volatility Correlated ETFs and CEFs
| Symbol | Corr |
| PTH | 10.61% |
| XRO | 10.62% |
| DPD | 10.66% |
| EGF | 10.69% |
| PIC | 11.36% |
| PIQ | 11.36% |
| RYE | 11.37% |
| FXA | 11.74% |
| DEF | 12.28% |
| PYZ | 12.61% |
| PJB | 12.84% |
| NFO | 13.10% |
| CVY | 13.18% |
| STH | 13.79% |
| PSQ | 14.27% |
| EEB | 14.52% |
| PFI | 14.91% |
| DOG | 15.85% |
| MYY | 16.52% |
| SH | 18.64% |
Strangely there are both long and short ETFs on the negatively correlated side. JSC, MVV, SSO, and DDM are all common cap-specific S&P and Dow trackers. DXD and SDS, however, two funds inverse on some of the exact same assets, are also in the list.
This suggests to me that the options and futures used to manage the leverage or inverse on these funds are too sensitive to volatility. Though I would expect any reasonable fund manager to be able to partially vega hedge, it would seem to me that SDS nor DXD are far too vega-sensitive given that similar short funds are on the exact opposite side of the table.
On the positively correlated side, the top three ETFs are inverse on the cap-specific S&P and Dow indexes. Interestingly, the financial and insurance ETFs and the Claymore BRIC EEB seem to be where money flows when the volatility increases, though it is important to note that regional banking as in the top negatively correlated assets.
As well as these, it's also interesting to note the other funds that have done well. Most seem to be related to dividends or currency, as the Australian dollar fund and the Zacks dividend yield hog are in the top 20. It's interesting to see that the Claymore Defense fund also see positive correlation here, but this is probably not causal but coincidental, given security shocks normally drive money to defense naturally.
It's also nice to see DPD in there, as it is a Dow covered call fund. This shows that the asset is well managed, as holding calls while volatility increases should be profitable.
In conclusion, though there are some surprising results that possibly indicate poor derivative management of inverse funds, the results can largely be used to hedge against volatility changes in non-derivative portfolios.
- Michael J Bommarito II's blog
- Add new comment
- 6405 reads
Similar entries
- See S&P Run: History of Attained S&P 500 Maximum Percentage of Past Sessions Up
- See Dow Run: History of Attained Dow Maximum Percentage of Past Sessions Up
- Dow Jones Industrial Average Continues To Break Record Run Percentages
- First Trust AlphaDEX ETFs Meet Uncertain Attention
- Not-So-Often-Cited Economic Indicators
VIX and Leveraged ETFs
Hi Michael. I think you're seeing a cross-correlation. The VIX is positively or negatively correlated to many other market indicators. For example, its negatively correlated to the overall index. You would have to run a regression with those other factors to see the effect of the VIX alone.
The VIX is fascinating, and poorly understood as far as I can tell. People act like it has predictive power, but the studies show that it has a lot more to do with what happened last week than what happened next week. It's part of some short-term market feedback loop that's probably not healthy (because it contributes to wider market swings).
If the VIX doesn't have any predictive power, than why do we pay attention to it? Because the damn thing makes our options more expensive, which can turn a high-profit opportunity into a barely break-even one. That changes our buying and selling patterns, which *does* give it predictive power. Its a self-fulfilling prophecy.
But isn't volatility supposed to be measured over time? Sort of like a moving average? How can it go up 50% in a single day? How would a few additional data points make this kind of extreme change? See the paradox of the VIX?
Thanks,
Tristan Yates, IndexRoll
Hello Tristan, Yes, you
Hello Tristan,
Yes, you make some good points about that. I understand that there are feedbacks and exogenous factors I should regress out. To be honest, just deciding which primary components to address was complicated enough that I decided I would rather keep it simple.
With respect to the predictive nature of the VIX, you are exactly correct. I've had some good luck with backtesting the VIX combined with other indicators, but as you mention, it's almost a lagging indicator, especially when compared to the Put volume, the Put-Call Ratio, and the Short-Term Trading Index.
DBA
Is there a web site that will tell me the holdings in DBA? Thank you.
Hello, Good question.
Hello,
Good question. The DBA fund is equally weighted on the corn, wheat, soybean, and sugar futures, rebalanced annually. If you're interested in more specific information, check out their prospectus:
http://www.dbfunds.db.com/dba/index.aspx
Tristan - The VIX isn't
Tristan -
The VIX isn't "volatility" (of which there are different types), it's "implied volatility". Which I'm sure you know - it goes up 50% in a single day because SPX options prices imply it.
IndexRoll is an interesting idea, however it's very difficult (read: impossible) to get good prices on LEAPs - it seems your strategy requires passive rolling over a long period of time, and each time you pay an expensive spread to do so.
Volatility
I think that unconditional variance is *measured* over time, but is in fact a fundamental parameter of a distribution. Implied volatility is a measure of conditional variance-- that is, the distriubtion of stock realizations today conditional on yesterday. In this case, you only have one observation (today) from which you are backing out a distriubtion. I think of the fix as the cost of risk in the S&P-- its the product of the underlying risk in the cashflows times the price the marginal investor pays for a unit of risk. Without knowing one of those values for sure, you can't nail down the other one.
Does anyone know of an EFT for the VIX, or any other second moment derivative?
Thanks,
Tim Jones, IMF
Hello Tim, Thanks for the
Hello Tim,
Thanks for the comment. I think, as usual, this word has too many uses depending on context. I know volatility is one that we must be especially careful about, but when it comes to the VIX, it has to be that what we're discussing is the implied short-term volatility, as is described on page 5 of this CBOE VIX white paper. I don't know how many other reasonable metrics we can consider to be meaningful and calculable on a useful scale.
As far as the existence of such ETFs, unfortunately none currently exist. That was something of a primary motivation for this post, though obviously none are strikingly powerful, and the option-writing funds are predictably in the best position. There are of course VIX derivatives offered by the CBOE with decent volume, but I'm still not sure whether any sane ETF issuer could make it work given how Poisson-jumpy the VIX is and how liquid the range of derivatives are.