Wanted: Bubble Detector

Imagine if somehow we could rig up a bubble detector: something to tell investors when stock prices had reached unsustainable levels. Would stock market crashes disappear? Or could you at least learn to avoid them?

See Also: Hot Air Behind the Stock Bubble Talk

A professor at the California Institute of Technology thinks he might be onto such a detector. Colin Camerer is a genius, which isn't so unusual on the geek-packed campus of Caltech. But Camerer is certified, having been awarded a so-called genius grant from the MacArthur Foundation last September. The no-strings-attached fellowships bestow a total of $625,000 on each recipient over five years. Camerer will use the windfall to further his studies in the emerging field of neuroeconomics, which uses magnetic resonance imaging to analyze brain activity to better understand human decision-making.

The idea of a bubble detector comes from a recent study coauthored by Camerer in which Caltech students watched replays of trading sessions rigged in a lab. Half of the sessions resulted in bubble markets, in which prices significantly exceeded the value of the asset. The Caltech students, given a stake of $60 at the outset, were periodically asked whether they would buy, sell or hold shares of the asset at the going price. Despite the fact that the fundamental value of the lab-created security was easy to calculate, based on dividend payments that decreased at regular intervals over the sessions, some of the students got swept up in the bubble anyway. "These kids are extremely analytical," says Camerer. "The median math SAT score here is 800 [out of 800]. It's hard to believe they can't count and keep track of a dividend value."

Scans of the students' brains found that susceptibility to bubble markets is associated with two types of activity. One is in the area where we make social calculations—figuring out what other people are going to do, what they want, what they think of us. "That supports the greater fool theory" of bubbles, says Camerer. "It's not that these investors don't realize that in 15 minutes the asset will be worthless. Instead, they think, Maybe I can sell [to a 'greater fool'] and get out."

Camerer also found that the human brain seems to recognize (probably not consciously) that a bubble market has a restless pattern that doesn't conform to the rhythm of normal markets. "People recognize that there's something unusual, some social contagion that's moving prices around," says Camerer. Ultimately, any early-warning system would include information from brain activity, prices and trading volume, as well as chatter in social media and the news. "I'm very optimistic about bubble detection," says Camerer.

Ka-ching Factor

Camerer also looks to the brain to discover why investors tend to sell winning stocks too soon and procras­tinate about dumping losers—behavior documented years ago by finance pro­fessors Terrance Odean and Brad Barber. Camerer found that when people sell winners, even those destined to rise further, activity flares in a part of the brain that assigns value and recognizes rewards. "The ka-ching feeling of locking in that profit, even though it's a mistake, feels good," says Camerer. "The lesson is that you somehow have to override that signal."

If you're reluctant to sell a poorly performing stock because you think it will come back, try this test: Ask yourself if you're willing to double down and buy more. "If your heart and brain tell you that's dumb, that's your way of knowing you don't really think it's going to bounce back," says Camerer.

Sounds a lot like the old Wall Street adage: Cut your losses short, but let your profits run. Camerer says he would not be at all embarrassed if "all we discovered is that Warren Buffett is right."