The Psychology of the Herd

by: Barry Ritholtz

I got an interesting email last night about the Everyone is Bullish! Kudlow commentary. Mike writes:

"With all due respect, you are in no position to be ridiculing anyone's wild prognostications."

Why not? I am not suggesting my prognostications are any better than anyone else's. That misses the ENTIRE point. The issue at hand isn't if a single forecaster is right or wrong; we are focusing here on group sentiment, the psychology of the herd. What should matters to investors and traders is the psychology of what happens when EVERYONE flips bullish.

That is a very significant occurrence, as it is both a rare event and one that tends to occur at rather ill timed places in the market cycle. The last time such an alignment of sentiment took place was inauspicious to say the least. [see this Bloomberg article: Stock Strategists Raise Alarms With Call for Rally; I assume this article is based on the recent Barron's roundtable].

Speaking of forecasting: I must admit to being totally surprised by how people responded to last year's Dow 6,800 discussion. Perhaps this was naive of me, but there you have it. Why surprised? First, because I put so little weight on forecasts, I was truly surprised just how seriously so many people take them. Where the markets will be precisely 12 months from now is a guess, pure and simple. I have always put precisely zero credence in year ahead forecasts, so I was quite stunned that others took it as gospel.

Indeed, long before that forecast, I explicitly discussed why these are so silly in The Folly of Forecasting:

• No one truly knows what tomorrow will bring. Nobody. Any and all forecasts are, at best, educated guesses.

• All prognostications are instantly stale, subject to further revision. Conditions change, new data are released, events unfold. Yesterday's prediction can be undone by tomorrow's press release.

• In order to "become right," some investors will stand by their predictions despite a stock or the market going the opposite way, hoping to be proven correct. Ned Davis called this the curse of "being right rather than making money."

And yet, people get genuinely worked up over these. Go figure.

Before the year is over, I plan on revisiting the Cult of the Bear series, offer up my mea culpa, and discuss what was wrong -- and right -- about the forecast. Keynes was certainly correct about failing conventionally versus succeeding unconventionally [he never said much about failing unconventionally!]

The second surprise was how others read just what they wanted into the forecast and the Cult Series: Call it selective perception: I got emails from Perma bears who were upset anytime I was long/discussed/recommended/owned anything -- except gold; that was okay. [What's that about?] And the Perma bulls utterly refused to acknowledge my published expectations for the "Rally Before The Fall:"

"Why the bull call before the fall? Because that's how market tops get made: In the 12 months leading up to the October 1987 highs, the Dow ran from 1800 to 2700 [a 50% gain], while the S&P 500 sprinted from under 240 to about 340 [about 42%]. From October 1999 to March 2000, the Nasdaq nearly doubled. Although I don't expect anywhere near those gains in the first half of 2006, the pattern could be quite similar: A leap to new highs on some widely held assumption, which subsequently turns out to be false." [emphasis added]
-Cult of the Bear III (1/20/2006 5:28 PM)

Hmmm, a widely held assumption which subsequently turns out to be false . . . soft landing anyone?

When the Businessweek piece was published, the bears ridiculed the 11,800/1350/2620 expectation prior to a correction [which I still expect]. At the same exact time, the bulls trashed the 6800/880/1100 low forecast.

A brief explanation of how those levels came about: those high numbers were my expectation of a double bubble -- the re-inflation of a bubble that occurs post crash [there's lots of academic papers on this], leading to an eventual parabolic move. That's why my Nasdaq top numbers was so aggressive [2620]. The low numbers were derived from an eventual retest of the March 2003 lows -- something that has happened in prior post crash periods -- Japan (1989-98), and the US Dow [1929-33]. Like I said, these were edumacated guesses.

I'll have more on this before the year's over, as well as that mea culpa.