Board-Ansicht
Mix-Ansicht- Groupthink on VIX - Amanito, 05.10.2004, 08:53
Groupthink on VIX
--> VIX readings -- such as those we've seen in recent weeks -- are bearish, right?
"Yes," is the resounding answer from virtually every investment adviser that I monitor who regularly refers to the VIX -- which is a measure of the implied market volatility based on a basket of widely traded options on the S&P 500 (SPX: news, chart, profile).
Just as high readings supposedly reflect the extreme fear typically seen at market bottoms, low VIX levels indicate the widespread complacency and lack of fear and usually seen at market tops.
These contrarian notions are so taken for granted in the investment advisory community, in fact, that few actually take the trouble to check whether they are true. So right now it is simply assumed to be self-evidently true that because the VIX (VIX: news, chart, profile) is so low, the market must be on shaky ground.
The market very well may be on shaky ground, of course. I actually am sympathetic to that conclusion. But, try as I might, I cannot find historical support for the notion that a low VIX is the cause of any such shakiness.
And believe me, I've been trying.
When I had the audacity in mid-September to suggest that low VIX readings might not be particularly bearish, I received many e-mails protesting that I must be wrong. (Read archived column)
So I returned to the drawing board (or my PC's statistical software, to be more precise) to see if I had missed any pattern in the data that shows low VIX readings to be bearish.
Not only did I not find any such pattern, I became even more confident than before that my original conclusion is right.
One of the many suggestions that readers provided in their e-mails was that I should look at relative VIX levels rather than absolute ones. So I focused on where the VIX stood each day relative to its moving average over the previous 20 trading days (a period covering the equivalent of about a calendar month).
I found the same pattern that I did when focusing on the VIX's absolute levels. I won't repeat each of my findings from before, since you can retrieve them easily from my earlier column. But, in general, I found that, as expected, high VIX readings often have come before market rallies. However, the same also appears to be true for low VIX readings.
And that's directly contrary to what contrarians would expect.
Take a look at the accompanying table, which breaks down all trading days since 1990 into five equal size groups according to how far above or below the VIX stood relative to its 20-day moving average.
20% of trading days since 1990 on which: Average Wilshire return over next 1 week Average Wilshire return over next 4 weeks Average Wilshire return over next 13 weeks
VIX was highest relative to 20-day MA 0.27% 0.86% 3.25%
Next to highest 20% 0.39% 1.12% 3.27%
Middle 20% 0.10% 0.85% 2.89%
Next to lowest 20% 0.08% 0.73% 2.49%
VIX was lowest relative to 20-day MA 0.23% 0.93% 3.03%
As you can see, the highest VIX relative readings are followed by above-average returns. But the same is true for the lowest readings.
There is irony in these findings: They show that many contrarians are guilty of the very"groupthink" of which they accuse others.
It isn't supposed to be that way. Humphrey Neill, the father of contrarian analysis, recommended that we be suspicious of beliefs that are widely held or taken on blind faith to be true, since those are the very ones least likely to have been subjected to a reality check.
So the next time you read or hear that today's low VIX readings is a bearish omen, demand the historical evidence that supports such a claim. I doubt you'll ever see it.
<ul> ~ http://cbs.marketwatch.com/news/story.asp?guid={3B2B8E0A-FD62-4F23-9B9C-162DA5CC</ul>
Thread gesperrtgesamter Thread:
