From the price of tulips to housing prices to Japanese stocks in the 1990s—financial “bubbles”, where the value of a product rises and then falls dramatically, have happened over and over again. But economists still don’t fully understand why bubbles happen in the first place. Why would people suddenly begin to trade an asset at a price much higher than its fundamental value?
Benedetto De Martino, a social scientist at California Institute of Technology, and colleagues recently published a paper in the journal Neuron that sheds light on what goes on in traders’ minds when they act on a bubble. In economics, price and fundamental value are supposed to be the only two factors needed for making trading decisions. While behavior should be predictable based on those values, this isn’t true in a bubble market.
De Martino and colleagues suspected that there is some other type of information that traders use. Based on research in the past decade, the authors believed that an area of the brain located just between the eyebrows, the ventromedial prefrontal cortex, or vmPFC, might be involved. “This area is known to code for any type of subjective value,” explained De Martino.
The researchers set up an experimental market and recruited 21 subjects, running half the sessions as a non-bubble market control and half as bubble markets, in which market prices skyrocketed past intrinsic values. They then used fMRI to measure brain activity. As expected, the vmPFC showed an inflated response to value in bubble markets. To figure out what was going on, De Martino created a bubble susceptibility index, a measurement of “how keen you are to jump onto the bubble.” Scores on this index varied between subjects but correlated with activity in the vmPFC.
When the researchers next looked at whole brains, another region, the dorsomedial prefrontal cortex (dmPFC), emerged as a potential contributor to bubble behavior. The dmPFC is known to be associated with theory of mind (ToM)—the ability to make inferences about the mental states and intentions of other people. Activity in this region had already been correlated with the ability to predict price changes in markets when insider traders were involved. “But we were very careful to avoid reverse inference,” said De Martino. “We retested the subjects to assess their ToM skills.” The authors confirmed that the activity they were seeing in the dmPFC was related to ToM.
Connectivity between the dmPFC and the vmPFC increased during a bubble market. “What we suspect is that while you’re constructing the value on the fly, some information about the intention of the others gets percolated and computed somewhere. So people are not only considering price and fundamental value, they’re also taking social signals into account,” said De Martino.
De Martino explained that people have the tendency to see intentional patterns everywhere, even when there is no intention behind the pattern. And this region of the brain responds dramatically to intentionality. “But in this case it’s just an illusion, almost like a group illusion,” said De Martino. ToM in this particular situation might be more of a handicap than an asset, because the better ToM a person has, the more likely they are to ride a bubble.
De Martino cautions that generalizing ToM as disadvantageous in all financial situations would be premature. “It can be good in other things. When people really do know better than you, you can actually predict some relevant information,” he said. “But in the bubble, people don’t have better information; it’s just an illusion. But the brain still uses the same, quite well-developed machinery to check for intentionality.”
De Martino M, O’Doherty JP, Ray D, Bossaerts P, Camerer C. In the Mind of the Market: Theory of Mind Biases Value Computation during Financial Bubbles. Neuron, Volume 79, Issue 6, 1222-1231, 18 September 2013