The first deduction we can make from the chart is that absolute volatility readings have no bearing on market trend. Markets have risen during times of low volatility (e.g. 1995), as they have with high VIX readings (e.g. 1999), and the same can be said of declining markets. As one would expect, the higher the volatility, the larger the distance between price highs and lows, which is what volatility is all about. For example, the lowest VIX reading of the fifteen year period happened during 1993 (9.31) which saw an 18% price swing between the high and the low. On the other extreme, the VIX high occurred during 1998 (45.74) when prices fluctuated by 35%. Yet, both 1993 and 1999 turned out to be up years.

So instead of absolute VIX readings it must be relative VIX values, or changes in volatility which show the way, right? Wrong. What about the ratio of the index and the VIX (e.g. S&P 500 divided by VIX) which some advisors claim to be the magic formula? Nope. In fact, Mark Hulbert, the well respected publisher of the Hulbert Financial Digest, has done extensive analysis on this subject and in recent articles he concludes that there simply is no historical support for the notion that a low VIX indicates that a market top is near, and that while high VIX readings often have come before market rallies, the same also appears to be true for low VIX readings.

What the chart above reveals however, is that large upward VIX spikes consistently coincide with market lows as can be seen repeatedly, including the one that resulted from the September 11, 2001 terrorist attacks. A well known behavior of markets is that declines are generally much more explosive than increases. Markets can drop further over a shorter period of time than they rise, as exemplified by the "Black Monday" on October 19, 1987 when the Dow Jones Industrials dropped by a whopping 22.6%. This primarily has to do with human self-preservation psychology in which fear and panic are much stronger emotions than enthusiasm and excitement. Even during manias driving market bubbles skyward, volatility does not come close to the spikes seen during bear market declines and crashes. As a matter of fact, the chart reveals that there are no downward volatility spikes.

As Trend Timers we have profited in times of both high and low volatility. Significant bear markets and crashes coinciding with volatility spikes are good for market timers because, by definition, it is during Sell signals that we have an opportunity to substantially gain on Buy and Hold approaches. Just because we enjoy such periods of extreme volatility does not mean we can use the VIX index as a reliable market trend predictor. For that, we will continue using the market itself and follow the trends it exhibits.

Warm wishes and until next week.

The TimingCube Staff