What is VIX telling us about where the market is headed?

Years ago I paid a lot of attention to VIX, the CBOE implied volatility index or "market fear gauge/index", but after they changed its composition a few years ago and people began shifting to other techniques for hedging stock positions VIX has lost a significant amount of its value as an "indicator." I only write this post because I just ran into the following misleading claim about what VIX "indicates":

The VIX has indicated correctly over the past few years when the market would decline, particularly after 9/11, the start of the Iraq War and the movements at the beginning of this year on Chinese economic growth fears.

Be clear on one point: VIX is not a leading indicator. VIX does not tell you what the market is going to do days or weeks in advance. It is a coincident indicator of the anxiety of market participants. It tells you how worried people are right now. VIX tends to spike up when people are selling feverishly, not in advance of a sell-off.

If you had sold your stock in early June when VIX spiked from under 13 to 17, you would have missed out on the latest big run-up of the Dow. And if you had not sold on July 19th when VIX had declined to around 15, you would have missed an opprtunity to avoid last week's sell off. Some "indicator."

To reiterate, VIX indicates the current anxiety level, not what the market will do in the days and weeks ahead.

The best use of VIX is to detect times of maximal anxiety, which is when selling peaks, which is precisely the time to jump in and buy the dip.

Alas, there is no indication in VIX that there might not be an even higher spike tomorrow or the day after or next week or next month.

VIX is not a good indicator of precisely when to buy or sell, but you can use it to judge when you should start considering a shift. Even then, I have to advice extreme caution. If you had listened to chatter about VIX being low over the past few years, you would have missed a huge run-up in the bull market. Since VIX was changed a few years ago (September 2003), it has not gone through even a single major crisis and not a single transition to a bear market. It could take another five years before we have a thick enough portfolio of market transitions so that we can calibrate VIX for the kind of markets and the kinds of investors that are active today.

The other answer I can give to the headline question is that the recent spike of VIX up above 24 suggests that the sell-off may be very close to running its course, with an emphasis on the qualifier "may." Based on experience, I wouldn't be surprised either way if the market fell another leg or began a new up-leg immediately.

Timing of market trend changes is very risky business.

-- Jack Krupansky