In my post last week I noted a number of positive divergences in the FX market (see Waiting for the bounce) and thought that equities were poised for a bounce. Now I am seeing bullish divergences in the VIX Index as well, which suggest that the markets are poised for a more sustainable rally into December much like the traditional Santa Claus rally.

First, consider this chart of the VIX Index (top) and the SPX (bottom). The normal inverse relationship of these two indices appears to be broken. Stocks have been falling all of last week, so why is the VIX down on the week as well? Even if you were to ignore the lockstep downward movement of the VIX and SPX on Monday and Tuesday, the VIX was flat to slightly down from Tuesday's close to Friday (green line) while the stock market continued to fall for that period.

(click to enlarge)

Is fear falling? If so, then is this a bullish divergence?

Volatility term structure forecasting a bullish reversal
In addition, I read an interesting post by Vance Harwood entitled Protecting High Yield Bond Investments with VIX/VXV Based Timing where he used the term structure of volatility for market timing (mostly in high yield bonds):

Recently I've been looking at volatility metrics for predicting market action. The CBOE's VIX index gets a lot of attention, but using absolute values of the VIX to trigger investments is almost certainly useless. On the other hand, volatility prices over different time frames, often called the term structure, does show significant predictive value.

In a truly bearish market the short term expected volatility, typically cheaper than longer term volatility, climbs higher than the longer term value. This behavior is shared between flavors of VIX (e.g., the one month VIX and its three month version VXV), VIX futures, and the implied volatility of same strike options of different months...

A simple metric that captures this behavior divides the short term volatility number by a longer term number. If the ratio is below one the market is relatively calm, if above one the market is especially nervous. I've been using the CBOE's VIX and VXV indexes as a convenient way to implement this volatility metric.

The same principles he uses can be applied to the stock market. In "normal" markets, short-term volatility (VIX) is slightly lower than longer dated volatility (e.g. three-month VXV). Harwood uses the VIX to VXV ratio as a way of measuring how much angst the market has about volatility. If one-month volatility (VIX) significantly rises above three-month volatility (VXV), then watch out below.

He uses a threshold of 0.917 for the VIX to VXV as his trigger point. If the ratio is below that number, then he will hold high yield bonds, but he will exit his position if the ratio rises above the threshold.

While the 0.917 threshold is a trigger that could be the result of data mining, I examined the history of ratio based on data I downloaded from CBOE and found the median ratio was 0.928. The difference between Harwood's 0.917 and the median is, as they say, close enough for government work. As a test, I added a further refinement. I created a second "crash warning" trigger, based on the median plus two standard deviations, which came to about 1.10.

The chart below shows the results of my backtest of a modified version of Harwood's model. First consider the lower panel showing the VIX to VXV ratio. I have defined a high-risk zone as ratio readings that are two standard deviations above the median and a caution zone as readings above the median of 0.928.

(click to enlarge)

Buy signals are generated when the ratio descends from the caution zone below the median ratio reading, which are shown as blue arrows in the upper panel of the S&P 500. Sell signals occur when the ratio rises above the median (red arrows) and "crash warnings" are flashed when the ratio rises above two standard deviations above the median.

As you can see, the model seems to have backtested reasonably well and will bear watching to see how it performs in real-time. Buy signals have been relatively good entry points and sell signals have been good exit points to go to cash. Moreover, the two "crash warnings" have been excellent signals to enter into short positions.

With the caveat that this is only a backtest, I would highlight the back that the VIX to VXV indicator has fallen and generated a buy signal in the last week. Based on past experience, this suggests a sustained advance of several weeks for the stock market.

NYSE Summation Index turning up?
In addition, I have been watching the slow stochastic of the NYSE Summation Index, which is shown below in the top panel, with the SPX shown in the bottom panel. In the past, cross-overs in the stochastic have marked good entry points to get long the stock market and this indicator has only failed once out of six in the past two years.

(click to enlarge)

In summary, the stock market has fallen hard in the last few weeks and is oversold. While oversold markets can get more oversold, this market is poised for a rally. Intermediate term indicators, such as the VIX Index, the term structure of volatility (which measure risk appetite) and breadth indicators (NYSE Summation Index) are pointing to a multi-week advance once a bottom is established.

Get ready for the Santa Claus rally.