A Fair Share of the Pie

People everywhere put a social spin on economic exchanges

Hadza hunter-gatherers in Tanzania chow down on gazelle meat, fruit, honey, and other mealtime staples without spending a dime. The Au and Gnau foragers of Papua New Guinea generously give food and other gifts to their neighbors without bothering to take out bank loans. Kazakh herders in western Mongolia monitor the weather and their animals’ health but ignore the ups and downs of international stock markets.

SPLIT DECISION. In Kenya, two Orma players sit in a grass hut and learn to play an economic game. Ensminger

GAMES PEOPLE PLAY. During fieldwork in Peru, Joe Henrich, at right, demonstrates an economic game among the Machiguenga. Henrich

IS THAT YOUR FINAL ANSWER? Machiguenga players go over the rules of an economic game with Joe Henrich. Henrich

These groups have little need for the colored paper and bits of metal known lovingly as money in industrialized nations. Nonetheless, small societies lying outside the Western world’s corporate bustle are hotbeds of economic activity. In fact, more than a dozen such groups, including those just mentioned, have yielded insights into social forces that may shape economic behavior from Tennessee to Timbuktu.

There’s a simple explanation for this unlikely development: Experimental economics has gone native. Until now, this academic exercise in controlled financial exchanges has rarely strayed from the laboratory. Researchers typically recruit college students to play games in which two or more strangers divvy up a sum of money using a set of ground rules.

In a break from this tradition, 11 anthropologists and an economist spent the past couple of years probing how people in diverse cultures play economic games. Initial results come from 15 small-scale societies located in 12 countries that span the globe. Participating groups consist of three hunting-and-foraging societies, six communities that rely primarily on slash-and-burn agriculture, four nomadic-herding groups, and two farm villages.

This extension of experimental economics far beyond the realm of indentured undergraduates has proven enlightening, says project director Joseph Henrich, an anthropologist at the Institute for Advanced Study in Berlin. Henrich and several other project scientists described themes emerging from their data in a session at last November’s annual meeting of the American Anthropological

Association, held in Washington, D.C.

Traditional economic theory assumes that basic human self-interest lies at the heart of commerce. As a result, regardless of his or her cultural background, a volunteer who receives a wad of money in an economics experiment should offer as little as possible to a partner, even if both players will be left penniless if that partner rejects the offer. In theory, a self-interested partner should accept even a stingy offer since it’s better than nothing.

However, a chief discovery of the project, Henrich says, is that nowhere do individuals behave out of pure self-interest, whether they live in college dorms or thatched huts. In one group after another, a person given a chunk of money or other valuable stuff tends to offer a substantial, although highly variable, share to an anonymous partner. Moreover, the partner often rejects any offer perceived as too low and contentedly departs empty-handed.

Cross-cultural results indicate that economic games tap into collective notions of what makes for a fair transaction. These flexible rules of thumb for sharing resources run circles around any brute instinct for self-interest.

In economic games, members of societies that feature lots of bargaining and bartering gravitate toward dividing available goods equally. Communities in which families are isolated come closest to exhibiting the traditional economic model.

As bastions of market economies and cooperative business ventures, industrialized nations promote a stronger ethic of fairness than do many of the traditional societies studied so far, Henrich notes.

These findings raise profound issues that have been little explored, says economist Colin Camerer of the California Institute of Technology in Pasadena, a participant in the cross-cultural project.

“The opportunity to trade in economic markets may create social expectations about sharing and trust that exist over and above individual decisions and motivations,” Camerer remarks. “Such findings can show us how to look at economic-game data with a fresh eye.”

Take it or leave it

The idea of going global with experimental economics grew out of Henrich’s initial findings among a community in Peru. Each forest-dwelling Machiguenga family lives in near-isolation and subsists on slash-and-burn farming, hunting, foraging, and fishing.

Henrich administered to pairs of unrelated Machiguenga volunteers a version of the so-called ultimatum game. Each duo had at its disposal a relatively large sum of money, equal to about 2 day’s worth of labor by a Machiguenga adult.

The first player, the proposer, was free to offer any part of the total to the other player, the responder. If the responder accepted the offer, each player received those amounts; if the responder turned it down, neither player got anything.

Machiguenga proposers displayed a greater streak of self-interest than any college student had in previous laboratory studies (SN: 3/28/98, p. 205). The Peruvian forest dwellers usually offered 15 percent to 25 percent of the pot. Responders agreed to nearly all offers, including those below 15 percent. In contrast, undergraduate proposers in the United States and elsewhere usually tender 30 percent to 40 percent of the total, and most of their responders reject anything below 20 percent.

Rules for fair behavior with nonfamily members have little chance to flourish among the Machiguenga, according to Henrich. A nomadic existence in which goods are exchanged only within families results in a relatively self-interested approach to the ultimatum game played by unrelated people. Even the Machiguenga, though, don’t display purely selfish behavior in this situation.

Research elsewhere underscores the impact of everyday opportunities for exchanging goods and participating in cooperative activities, such as hunting and food growing, on how people play economic games.

Consider the Hadza. These hunter-gatherers treat meat and other food as public property if it’s brought back to camp and others see it. The Hadza enforce extensive sharing through gossip and outright punishment of cheaters. Occasionally hunters beat the system by eating part of their kills in the field and sneaking some of the rest into their families’ shelters at night.

In ultimatum games conducted by Harvard University’s Frank Marlowe, Hadza players displayed their tension between resenting forced sharing and wanting to punish those who defied it. When dealing with an anonymous partner, Hadza proposers made offers almost as low as those of the Machiguenga. In turn,

Hadza responders usually turned up their noses at offers.

Individual traits, such as age, sex, and number of children, did not affect how the Hadza played the ultimatum game, Marlowe says. However, members of a particularly large camp made much higher offers as proposers than did members of several smaller camps. The Harvard researcher suspects that the large camp had developed the harshest sanctions against those who tried to flout food sharing.

“The Hadza have donor fatigue,” Marlowe says. “When they have a chance to escape from forced sharing, they do.”

Yet donor fatigue doesn’t create all-out selfishness. In dictator games, where proposers simply give what they want to another player who has to accept the offer, Hadza players forked over around 10 percent of the pot rather than keeping it all.

Paraguay’s Aché foragers provide an interesting contrast to the Hadza. Aché hunters often leave their killed prey outside of camp to be discovered by others, so as to avoid looking boastful. Game then gets distributed equally among all households.

In ultimatum games conducted by Kim Hill of the University of New Mexico in Albuquerque, Aché proposers usually offered either 40 percent or 50 percent of a sizable sum. Many others offered as much as 70 percent. There were no rejected offers.

When carving up the experimental pie, Aché proposers perceived themselves as parceling out meat that they or a male member of their family had hunted, Hill proposes.

Whatever the case, the contrasting ways in which the Aché and the Hadza play the ultimatum game “seem to reflect their differing patterns of everyday life, not any underlying logic of hunter-gather social organization,” Henrich contends.

Whale of a deal

The whale-hunting Lamalera of Indonesia made the most generous offers of all in the ultimatum game, according to Michael Alvard of Texas A&M University in College Station. A majority of Lamalera proposers, who exchanged packs of cigarettes, offered half or more of their booty. Offers lower than 50 percent were frequently rejected.

Because the cigarette account in Lamalera experiments represented 10 days’ wages, making an offer was like dividing up a whale, Alvard says. Lamalera players refused to use money in order to avoid the appearance of gambling, an act frowned upon in their community.

Again, volunteers played the game with their daily interactions in mind. When a Lamalera whaling team returns with a whale or other big catch, a specially designated crewmember separates it into preassigned parts for the harpooner, other hunt participants, the sail maker, and nonhunting community members. In other words, Lamalera hunters are accustomed to getting a modest portion of the booty they secure.

Ultimatum exchanges take a radically different turn when players are accustomed to giving and getting gifts that come with strings attached. Unlike any other groups, the Au and Gnau speakers of Papua New Guinea’s northern coast avidly rejected both stingy and generous offers, says David Tracer of the University of Colorado at Denver. Like the Hadza, these foraging villages relentlessly enforce food sharing. Selfish offenders face physical attacks that can result in severe injury or death.

Moreover, Au and Gnau frequently give gifts of food and other items to neighbors in order to cement local alliances and to create social debts that they can later collect on when, say, food runs low or fighters are needed to wage warfare.

Au and Gnau responders feared incurring such debts if they accepted large offers from proposers, Tracer holds. “They either didn’t believe that an anonymous windfall was real or feared the consequences of taking it,” he says.

Unusual findings also occurred in Mongolia, where Francisco Gil-White of the University of Pennsylvania in Philadelphia studied two neighboring groups of nomadic herders, Mongols and Kazakhs. Proposers in both populations offered relatively high amounts—on average, 40 percent—although responders demonstrated a willingness to accept even extremely low offers. This pattern makes sense in light of the herders’ great concern for nurturing a good personal reputation with others and smoothing over interpersonal conflicts, Gil-White says.

In violation of numerous Western social psychology experiments documenting in-group favoritism, Mongols made higher offers to Kazakhs than to players from their own ethnic group. Kazakh proposers also upped their offers when they knew they were dealing with a Mongol. Concern for avoiding conflict between closely interacting groups may have sparked this surprising generosity across populations, Gil-White suggests.

Market-based trust

A few researchers in the project have extended their work beyond the ultimatum and dictator games. For instance, Jean Ensminger of the California Institute of Technology administers the “trust game” to pairs of anonymous players. Each player receives $40 or an equivalent sum. Player 1 gives any part of his or her pot to Player 2. The experimenter then triples Player 2’s total. Player 2 can then give any portion of that much larger sum back to Player 1. In this game, the more money Player 2 returns to Player 1, the more trust exists between them.

So far, the most trusting players have been from a rural Missouri town, Ensminger says. Among these folk, on average, Player 2 gave back enough to Player 1 to create roughly equal final shares.

Returns to Player 1 were somewhat less among the Orma, a group of African livestock herders studied by Ensminger. In other experiments, college students have displayed relatively low trust, with Player 2 usually keeping the entire bonus.

“There seems to be more trust and greater concern about fairness and going for 50-50 splits in places where people are somehow involved in a market economy,” Ensminger contends.

Her argument that a society’s structure molds its members’ selfish and altruistic behavior clashes with evolutionary as well as economic theory. Many evolutionary biologists hold that natural selection has favored individuals who are genetically inclined to act out of self-interest in order to propagate their own genes.

Henrich theorizes that, throughout humanity’s evolution, groups that devised the most successful social guidelines for pursuing fair interactions left competing groups in the dust. This process advanced genetic traits in the surviving groups that proved conducive to hashing out equitable deals.

“This new cross-cultural research is cutting-edge stuff,” comments anthropologist Stuart Plattner of the National Science Foundation in Arlington, Va. “It will advance economic theory.” NSF has funded a second phase of the project. A larger contingent of researchers will study additional societies and probe economic behavior using food and other nonmonetary items.

It’s encouraging, Henrich notes, that all people studied so far treat the economic games as if they’re genuine encounters. Once they grasp the point, nomads, foragers, and farmers draw on their own group-specific assumptions about fairness and sharing.

Henrich and his colleagues have yet to formulate an airtight case for group differences in concepts of fair exchange, remarks economist Kevin McCabe of George Mason University in Fairfax, Va. The cross-cultural findings may instead reflect disparities in researchers’ ability to articulate the point of economic games to players and to administer the games consistently, he contends.

McCabe suspects that much cooperative behavior represents a form of delayed gratification, in which individuals forgo immediate self-interest to work with others for a greater personal payoff in the future. “Still, the cross-cultural work is provocative and interesting,” he says. “It’s definitely worth pursuing.”


Bruce Bower has written about the behavioral sciences for Science News since 1984. He writes about psychology, anthropology, archaeology and mental health issues.

More Stories from Science News on Anthropology

From the Nature Index

Paid Content