Stocks quoted in this article:
I just want to reiterate again that I'm not bearish on the market at this point, but I do side with the camp that volatility is ready to trend higher. Let me do some 'splaining.
I sporadically mention that CBOE Volatility Index (VIX) tends to trade in long-term "regimes," with four to six years of "low" followed by four to six years of "high." I believe this graph highlights that behavior fairly well. Here's the average VIX level for each calendar year going back to 1993 (click chart to enlarge):
By this metric, we troughed for four years (mid-'90s), spiked for about six to seven years (late-'90s into early "aughts"), troughed again for four years (mid-"aughts"), and then spiked for about four years. That brings us to 2012-2014, which is clearly a trough.
If the pattern stays true to form, 2015 looks like somewhat of a transitional year. That's exactly what happened in 1996, and again in 2007. Both of those years see volatility trend higher, but not explosively so … yet. The big pop happened in the following years.
It's only really a sample size of two, so I'd reserve making too many sweeping judgments. But it does have the feel that VIX is slowly climbing out of hibernation. Spikes are becoming more commonplace, which certainly seems consistent with an overall uptrend. So why doesn't this bode very poorly for the market?
Well, as we noted yesterday, if you played it that way in the '90s, you stayed on the sidelines for one of the best markets ever. On the other hand, if it scared you out of the market in 2007, it saved you a fortune. The point isn't that it's bullish for the market, just that volatility can move independently of stocks in wider time frames.
On a day-over-day basis, VIX has a large negative correlation to the market, something like about -0.75 to -0.8, depending on how far back you go. But the absolute level of VIX does not have much correlation to SPDR S&P 500 ETF Trust (SPY) … in fact, it's almost zero. That's because the overall value of stocks tends to rise over time. SPY was in the 40s back in 1993, for example. VIX isn't a stock, it's a statistic which mean reverts over time, and thus never "grows."
Ergo, in the very short term we can see those huge correlations, but over the long term … not so much. Obviously, VIX will spike in reaction to market accidents, but those spikes will simply ebb in long-term low-VIX regimes.
Disclaimer: Mr. Warner's opinions expressed above do not necessarily represent the views of Schaeffer's Investment Research.