Back in March, we spent some time exploring how investors could use the VIX in order to help them determine safe entry points in the market after corrections. We looked at a couple of different buy signals that were tied to the VIX. As we discussed a really simple approach to identify turning points in the market after a correction is to wait for the VIX to first spike and then reverse lower.

After looking at an admittedly simplistic approach, we then did some math and took the idea a bit further. We looked for a way to define extreme moves in the market and reviewed what happened when the VIX first moved more than 2 standard deviations above its 20-day moving average and then reversed. In the article, we wrote that the move above the 2 standard deviation line told us that the market had experienced a meaningful correction. And then, when the VIX moved back below the two standard deviation line, it told us that volatility had begun to retreat. This meant that the correction in stock prices had likely run its course.

History shows this to be a pretty effective buy signal and the statistics on the buy signal were impressive. However, we noted at the time that while this was a pretty darned good buy signal, it did not work quite as well on the sell side.

In Search of a VIX Sell Signal

Given that everybody and their grandmother has been yammering on lately about the low readings seen on the VIX, we thought it might be a good idea to look a little harder at our research to see if we couldn’t come up with a sell signal that might be useful.

To be sure, the VIX has been moving lower in recent months. In fact, the weekly chart of the VIX shown below illustrates that the current reading is the lowest since early 2007.

VIX – Weekly

In looking at the data, it is worth noting that turning the buy signal discussed on its head in order to produce a sell signal isn’t too bad.

Since 1995, the S&P; 500 moves down -0.09 percent on average 5 days after a sell signal is given. This represents a pretty good departure from the average return for all 5-day periods in the sample, which is +0.18 percent.

Ten days after a sell signal, the market is -0.03 percent lower, which, again, compares pretty well to the average return of +0.35 percent for all 10-day periods.

However, the problem is that when the bulls get on a roll to the upside, this type of system gives signals that are oftentimes VERY early. As such, one winds up missing some of the biggest moves to the upside. And this is simply unacceptable.

Looking For Clumps of Signals

While there isn’t that much data to review, it does appear that there is an answer. You see, when stocks start to enjoy a serous joyride to the upside – a move that tends to defy logic – this indicator will occasionally fire off several sell signals in close proximity to each other. This tells us that the market is overbought and that the uber-low reading of the VIX suggests that things have gotten a little frothy.

So, while a single sell signal from this indicator during a serious uptrend should be ignored, if three or four signals come bunched together – well, now we’re talkin’.

The chart below shows the S&P; 500 plotted weekly with arrows depicting the times when “bunches” of our VIX sell signals occurred.

S&P; 500 – Weekly

Again, there isn’t that much data to review on this chart. However, you have got to admit that the arrows in 2010, 2011, and 2012 were, at the very least, pretty good warning signs, right?

In 2010 there were 5 sell signals clumped together about two weeks before the ultimate top. In 2011, there was a clump of 3 signals that occurred just as the market peaked out. And then in 2012, there were actually two clumps of 3 signals before the decline began in earnest.

So, what’s happening now, you ask? While this is admittedly not a terribly scientific system and my analysis isn’t exactly rigorous, it is worth noting that our little VIX indicator has fired off 4 sell signals in the last two weeks, which in my book is the very definition of a clump. And the most recent signal occurred… wait for it… yesterday.

So, will stocks begin a decline tomorrow or the next day? Frankly, there is NO way to tell. However, given what has transpired after clumps of sell signals since 2010, this is an indicator that might just bear watching in the near-term – especially if the bears can come up with some sort of bad news with which to work.

Leave a reply