Superfreakonomics_ global cooling, patri - Steven D. Levitt [52]
After a few years, List was invited to join Vernon Smith, the godfather of economic lab experiments, at the University of Arizona. The job would pay $63,000, considerably more than his UCF salary. Out of loyalty, List presented the offer to his dean, expecting UCF to at least match the offer.
“For $63,000,” he was told, “we think we can replace you.”
His stay at Arizona was brief, for he was soon recruited by the University of Maryland. While teaching there, he also served on the President’s Council of Economic Advisors; List was the lone economist on a forty-two-person U.S. delegation to India to help negotiate the Kyoto Protocol.
He was by now firmly at the center of experimental economics, a field that had never been hotter. In 2002, the Nobel Prize for economics was shared by Vernon Smith and Daniel Kahneman, a psychologist whose research on decision-making laid the groundwork for behavioral economics. These men and others of their generation had built a canon of research that fundamentally challenged the status quo of classical economics, and List was following firmly in their footsteps, running variants of Dictator and other behavioralist lab games.
But since his days at Stevens Point, he had also been conducting quirky field experiments—studies where the participants didn’t know an experiment was going on—and found that the lab findings didn’t always hold up in the real world. (Economists are known to admire theoretical proofs; thus the old quip: Sure, it works in practice, but does it work in theory?)
Some of his most interesting experiments took place at a baseball-card show in Virginia. List had been attending such shows for years. As an undergrad, he sold sports cards to earn cash, driving as far as Des Moines, Chicago, or Minneapolis, wherever there was a good market.
In Virginia, List cruised the trading floor and randomly recruited customers and dealers, asking them to step into a back room for an economics experiment. It went like this. A customer would state how much he was willing to pay for a single baseball card, choosing from one of five prices that List established. These offers ranged from lowball ($4) to premium ($50). Then the dealer would give the customer a card that was supposed to correspond to the offered price. Every customer and dealer did five such transactions, though with a different partner for each round.
When the customer has to name his price first—like the white men who visit Chicago street prostitutes—the dealer is plainly in a position to cheat, by giving a card that’s worth less than the offer. The dealer is also in a better position to know each card’s true worth. But the buyers had some leverage, too: if they thought the sellers would cheat, they could simply make a lowball offer each round.
So what happened? On average, the customers made fairly high offers and the dealers offered cards of commensurate value. This suggests that the buyers trusted the sellers and the buyers’ trust was rewarded fairly.
This didn’t surprise List. He had simply demonstrated that the results you get in a lab with college students could be replicated outside the lab with sport-card traders, at least when the participants know a researcher is carefully recording their actions.
Then he ran a different experiment, out on the real trading floor. Once again, he recruited random customers. But this time he had them approach dealers at their booths, and the dealers didn’t know they were being watched.
The protocol was simple. A customer would make a dealer one of two offers: “Give me the best Frank Thomas card you can for $20” or “Give me the best Frank Thomas card you can for $65.”
What happened?
Unlike their scrupulous behavior in the back room, the dealers consistently ripped off the customers, giving them lower-quality cards than the offer warranted. This was true for both the $20 offer and the $65 offer. In the data, List