VIX to VXV Ratio Nears All-Time Lows

By Richard Bloch

A few days ago I pointed out that while the VIX has fallen since late November, it’s certainly not nearing all-time lows (yet). And I agree with some who suggest that may not mean much. After all, those all-time VIX lows were set years ago, before the financial crisis.

But the ratio of the VIX to its sister index, the VXV, was nearing record lows as of the close on December 17.

First some background: While the VIX is calculated to measure a theoretical indexed implied volatility for S&P 500 (SPY) index options that expire in one month, the VXV does the same for options that expire in three months. So the VXV is a representation of what traders expect over the somewhat more longer term.

The chart below shows the S&P 500 index for the year. The bottom shows the levels for both the VIX and VXV. The part in the middle with the purple line shows the ratio between the VIX and VXV.

As you can see, the ratio is normally less than 1.0 because the shorter-term VIX is usually lower than the longer-term VXV – essentially representing a contango in the market for volatility.

More volatility, but when?

Psychologically, this seems to make sense. Even if you’re expecting a future surge in implied volatility, the question is when do you expect it? If the VIX and VXV are any indication, most believe later rather than sooner, at least most of the time.

Or to put it another way, when the VIX is higher than the VXV, that means options traders are expecting more volatility in the near term – generally a sign that fear is returning to the markets.

The CBOE didn’t officially introduce the VXV index until late 2007, so there’s not a lot of actual historical data to measure, but the VIX to VXV ratio of 0.794 on December 17 is the second lowest reading ever (on a closing basis). The all-time record low was 0.790 on October 11 of this year.

However, the CBOE has calculated what the VXV would have been, going all the way back to 2002.

I didn’t find the actual data, but just eyeballing a portion of a chart in the CBOE description of the VXV, I can see two times when the VIX to VXV ratio might have been as low as 0.74 or so – at least in theory. (Note: I added the gridlines and annotations to the chart.)

Quadruple witching and possible rollover distortion

I don’t think you can necessarily conclude that this low ratio represents less fear in the market. After all, December options expired on Friday, it was a quadruple witching day, and the upcoming three-day weekend could impact time decay expectations.

What’s more, as the CBOE notes in its VXV description:

When SPX contract months are sequential, that is, expiring one month apart, the “roll” is a smooth transition from one set of options to the next. Yet, when the expiration dates of the SPX options used to calculate VXV are two or three months apart, there is a “jump” in the option weights by as much as 35%. CBOE has filed a proposed rule change with the SEC to allow CBOE to list four consecutive SPX contract months at any given time, which should eliminate this type of discontinuity in the future.

I’m not sure if that rule change was ever implemented, but if not, that rollover could have had an impact on the VXV index.

It is interesting to note that in 13 trading days, from November 30 to December 17, the VIX has dropped more than 30%, while the longer-term VXV is down by only about 19.5 percent.

But as my first chart shows, back in January 2010, it didn’t take long for the VIX to VXV ratio to surge to its first 1.0+ reading of the year, so it will be interesting to see what happens as 2011 dawns.