Confidence Springs Eternal

In this morning's Wall Street Journal, Jenny Strasburg has a short piece on the "cash on the sidelines" currently held by hedge fund investors (as distinct from hedge fund managers). We're not particularly impressed with "cash on the sidelines" per se as a contrary indicator, as only increased risk appetites, which leave buyers more enthusiastic than sellers, can translate cash into higher asset prices.

But for our purposes the most interesting nugget, a truly classic behavioral finding, is buried in Strasburg's last two paragraphs:

The moneymaking outlook is so-so. Investors surveyed by Deutsche Bank said they expected hedge funds to earn a median 7.5% return this year.

The same investors apparently believe they possess superior skills in choosing managers: For their own portfolios, they are projecting a 10% return.

This disconnect between one's projected average for the masses and expectations for oneself--also known as overconfidence--is a cornerstone of the behaviorist understanding of market participants.

One minor statistical quibble while we're at it. Investopedia's short description of investor overconfidence opens with this (emphasis added in bold):

In a 2006 study entitled "Behaving Badly", researcher James Montier found that 74% of the 300 professional fund managers surveyed believed that they had delivered above-average job performance. Of the remaining 26% surveyed, the majority viewed themselves as average. Incredibly, almost 100% of the survey group believed that their job performance was average or better. Clearly, only 50% of the sample can be above average, suggesting the irrationally high level of overconfidence these fund managers exhibited.

When making arguments like this one, we think it's smart to be as precise as possible. And as it turns out, more than 50% of the sample can be above average--as long as the median outcome is higher than the mean outcome. Think of it this way: Just a few terrible outcomes can drag the average (i.e., the mean outcome) lower, but because they're already below the median (i.e., they're already in the bottom 50% of performers), their truly awful underperformance can produce a population in which more than 50% enjoyed above-average outcomes.

Using the numbers in Strasburg's story, think of an average outcome of 7.5% per year, where the bottom 10% of participants produced returns of -10% per year (in a kind of "fat tail" phenomenon). One can easily imagine a world in which more than half the participants did better than 7.5% per year. A general assumption here would be that, as a group, any outcomes in the above-average "fat tail" outperformed by less than the worst laggards underperformed.

Now, our point about overconfidence remains. It's plenty real (and may even be a necessary condition for risk-taking). But it is mathematically possible for more than 50% of participants to do better than average.

Comments

[billddrummer]

A very good point, and you bring up an issue that seems more prevalent now than before WSJ sold to NewsCorp: People are more willing to challenge assertions made in the WSJ now than in the past. An opiinon piece that declared "The Housing Crisis is Over" generated lively discussion today, virtually all of it contrary to that opinion.

Could the "independence&quo... and "editorial integrity" promised to the Bancroft family prior to the sale be a specious proposition, considering who now holds the reins?

[fatcat]

you should have seen the glee on cnbc with nat gass going to 16.0 0..If our economy is truly 75% consumer,that one item will kill consumer spending. Its like they are in in some kind of bubble or suspended animation!!Maria Stupidloma was all giddy over record oil..How does this help equity markets?