VIX Direction and Options Trading Volume: Correlation?

I’m going to interrupt the regularly scheduled program to briefly tackle one of my side interests in finance – options theory – having been inspired by this article from Bloomberg on options trading volumes. From the article:

Average daily trading increased 24 percent to 15.7 million contracts in May, the biggest jump in seven months, as investors boosted the use of equity derivatives, Options Clearing data show. Volume had slowed last year after professional investors such as hedge funds sold positions to pay back clients. Forced sales and the highest prices on record cut trading by 7 percent in December from a year before. November volume fell 21 percent.

Last July, around the time of the first widespread financial panic of 2008 (i.e. post-Bear), I spoke with an equity options trader at one of the major investment banks. The topic of discussion was the divergence between the CBOE Volatility Index (VIX) and the S&P 500, which he attributed to funds getting shorter on the market at the time.

A short time later, the same trader said that institutional trading desks were pulling back on their options market making activities – capital constrained trading would later be enormously important in the stressful times that were to follow.

What I was interested in looking at is the correlation between OCC trading volume and changes in the VIX. Having read a number of papers about that the VIX actually represents, one of the simpler conclusions that I intuitively accept is that it models the balance between supply and demand for options.

How does that simply hypothesis test out? The chart below shows the month-over-month change in average daily OCC trading volume, along with the month-over-month change in the VIX (based on month’s end).

Excel gives a correlation coefficient of 0.53, and an R-squared value of 0.28; the dataset I used goes back to January 2008, so it is a short time period with some extraordinary circumstances.

The thing that jumped out from looking over this was the recovery in trading volume, which accompanied a large drop in the VIX. Is this actually a strong indicator of a return to normalcy by large traders who were skittish with capital in the fall of 2008?

I’m sure that others with more experience analyzing options trends could add to this, so I’m going to stop here and ask for feedback – first, do you agree or disagree that the VIX is best conceptualized as a proxy for the balance of supply/demand for options; and second, what (if anything) should be drawn from the recent trends in trading volume and VIX direction?

Disclosure: None.

Comments

[Drew Arnold:]

The notion that the VIX is a proxy for supply/demand is a very interesting thought, and it makes sense for a few reasons. Surely over the time period of this data set, when the VIX spiked in 2008 due to an extreme amount of uncertainty, firms were less willing to commit capital to the options market and to financial markets in general, driving down demand. Secondly, it makes sense that when the VIX goes higher and premiums go up, put protection is more expensive and therefore suffers lower demand.

Jul 06 09:29 AM