Colin F. Camerer
1200 E California Blvd
MC 228-77
Pasadena, CA 91125
(626) 395-4054

Email me

I am lucky to be an "experimental coach" and collaborator in a remarkable project on fieldworkers doing experimental economics "in the wild"— at a dozen simple small-scale societies in remote parts of Africa, Asia, and South America. Venturing outside our usual university labs—far outside!—inspired others to do similar experiments in interesting field settings like everyday workplaces (see Jeff Carpenter's work with freight handlers ), Israeli kibbutzim, and Siberia (ask Rick Wilson).

The anthropology project began when an adventurous UCLA graduate student, Joe Henrich (now at Emory) did ultimatum game experiments with the Machiguenga in Peru. In ultimatum games, one player—a Proposer—has a fixed sum of money (typically $10) and offers some of it to a Responder. The Responder can accept or reject the offer. If she accepts it they both earn the money they agreed on; if she rejects it they get nothing and the game is over. (The ultimatum game is like a monopolist fishmonger offering fish which is about to spoil at a fixed price, or like the last-minute offer management or a union make just before a labor strike deadline.) In about a hundred studies with people in advanced societies, most Proposers offer about 40% of the "pie" because they anticipate that Responders might reject the offer if it is too low (about half of 20% offers are rejected, even for high stakes up to $400); see my Behavioral Game Theory book, chapter 3, for details. This is surprising to economists because if people want to have the most money, they should take even a low offer, and Proposers should anticipate this and offer little. In striking contrast to all the previous studies, Joe found that the Machiguenga offered very little (offers of 15% were common) and Responders accepted almost every offer. (Even the offer which was accepted is a little suspect because Joe's interpreter expressed surprise that the Machiguenga Responder would take it!) Joe wasn't sure whether he had made a mistake in the design or execution of the experiment, or discovered the first group of people who behave according to game theory! (Since then, Elisabeth Hill and my former student Dave Sally have found that autistic adults—who tend to have difficulty in guessing what other adults will feel and do—also make very low offers. Neuroscientist Jonathan Cohen and his collaborators have also found, using fMRI brain imaging, that people who show high activation in insular cortex—an area associated with emotions like disgust—when they contemplate unfair offers are also more likely to reject them, which shows a clear neural underpinning for the interpretation that rejections are due to a distaste for being treated unfairly.)

Joe's remarkable result in Peru led to a project, funded by the interdisciplinary MacArthur Foundation Preferences Network, in which 12 fieldworkers went to their remote sites and played ultimatum games so the results could be compared across cultures that vary widely in important cultural variables. Their findings are reported in an article by the entire team published in Behavioral and Brain Sciences, and in a bookpublished by Oxford University Press. The main findings are that some societies do make low offers and accept anything, and that the degree to which the societies have gains from cooperation (e.g., where public goods are valuable) and use markets for exchange (e.g., selling cows or food they grow at a periodic market) are highly correlated with fairness. The fact that market integration is correlated with fairness turns Adam Smith's famous dictum ("it is not for the benevolence of the butcher and baker that you get your daily meal") upside down: In theory markets may tolerate self-interest, or thrive because buyers and sellers are trying to get the most for themselves and their families. But in practice, in societies with a lot of market activity people bargain more fairly than in societies with little market trade. Of course, we don't know the direction of causality, and the empirical findings need to be replicated more widely (a follow-up project is going on now), but the correlation between market integration and fairness should push economists to think very differently than we have in the two centuries since Smith's famous words were written.

The anthropology project is important for three other scientific reasons. It showed that there is more to study in economics than college students in America. The anthropologists also face more dramatic challenges in running experiments in the wild than we usually do in university labs—equalizing stakes across cultures (many of which use little money), getting illiterate subjects to comprehend the idea of an experiment, limiting post-experimental interactions among subjects, etc. Their work will provoke lively discussion about experimental methods in our own labs and other domains. And finally, the project showed that truly interdisciplinary collaborations can succeed—provided scientists from different intellectual communities begin with deep curiosity and mutual respect, and are prepared to spend a lot of time becoming "multilingual" in the language, methods, and intellectual culture of other fields.

A May 2004 story from the Dallas Morning News.

Division of the Humanities and Social Sciences
Mail Code 228-77
California Institute of Technology
Pasadena, California 91125

Office: Room 101
Phone: (626) 395-4054
Fax: (626) 432-1726
email icon  E-mail: camerer@hss.caltech.edu
Caltech Home Page