A volatile look at volatility

Traders sifting the tea leaves for market direction inevitably look at some point to the sentiment indicators. In Guambat's experience, it's a last resort, desperate act, when all else fails to reconcile his market view with price reality.

Sentiment charts tend to look like inverse related mirror images of price action, but Guambat is not at all sure if this is more correlative than causative, when starting from the sentiment perspective. That is, does sentiment lead or follow price? Take a look at this S&P500 chart and the StockCharts.com VIX chart below it.





The VIX is the "classic" proxy for market sentiment, but what it really measures is a forward-looking indication of traders' pricing expectation of options, and the VIX charts these expectations.

In the S&P price chart above, there are Bollinger Bands, and these, too, are something like sentiment indicators in that they measure volatility, that is variations in price ranges over similar periods of time, but they are more backward-looking, calculated by standard deviations from a moving average, which is historical data.

Mark Hulbert, who is a divine diviner of market behavior, but no one gets it right all the time, looks at the current VIX, compares it to the relative price swings of recent times, and concludes the VIX is not expressing the truth about state of current perceived volatility.

That is, he is expecting a retreat from current price levels even though the VIX is "low" by recent standards (but certainly high by the standards of recent years). He suggests, to use the term generally assigned to low volatility numbers, the market is "complacent", which is a contrarian set up for a fall.

Thus, he reckons prices will react to sentiment, rather than the other way around.

But he explains it all so much better in his MarketWatch blog:

[The VIX] is not, strictly speaking, a measure of the stock market's past volatility. It instead reflects option traders' expectations of future volatility.

The two are related, to be sure, but nevertheless not the same.

Consider the number of sessions in which the stock market rises or falls by at least 1%, as it did on Thursday, when the Dow Jones Industrial Average fell by 102 points, or 1.2%. Over the last 63 trading days (a calendar quarter, in other words), there have been no fewer than 43 such sessions -- or more than two of every three days in which the stock market was open. The comparable number as of last Oct. 24, when the VIX hit its all-time high, was essentially the same -- 44 days.

So, at least from this perspective, recent volatility is just as high today as it was last October.

The data paint an even starker picture when we focus on daily percentage changes of at least 2%. There have been 21 such days over the last quarter, or one out of three days in which the market was open. The comparable number last October was 27, not that much higher than the current reading.

Looking at actual volatility, therefore, it is difficult to justify the VIX being barely a third as high today as it was last October.

Why is it nevertheless so much lower? Because options traders are less concerned about the stock market today than then -- far less.

And that, as I have argued in recent columns, is a bad sign from a contrarian point of view. It adds yet more evidence that the market's impressive rise since March 9 may nevertheless be a bear-market rally.

Guambat defers to Hulbert's more experienced and successful viewpoint, and dearly hopes he's right, as it would confirm Guambat's disposition du jour.

But Guambat is squeamish about putting his money on these particular tea leaves. He turns, nevertheless, to them in these desperate times.