Coalitional Effects on Reciprocal Fairness in the Ultimatum Game: A Case from the Ecuadorian Amazon

Is there a universal human sense of fairness? Initial experiments with the Ultimatum Game seemed to lean that way: Two participants who do not know each other's identity and are not allowed to communicate are held in separate rooms. The first participant is given an amount of money to divide -- say $100. The participant can keep 90% and leave 10% for the other, unknown participant, or can choose to divide it more fairly. The second participant is then told of the proposed division and can accept the proposal, in which case both players are paid and the game is over; or the second participant can reject the proposal, in which case neither is paid and the game is over. Classic economic theory proposes that it is only rational self-interest to accept any amount of money. Who cares that the other player gets $90 if you get $10? But thousands of trials -- entirely with American, European, Japanese participants (usually college students) -- showed a perplexing result: A significant percentage of second participants choose to reject divisions that are unfair. When the game was tried with people in non-industrial societies, however -- hunter-gatherers, slash-and-burn agriculturists, nomadic herders -- the results varied widely. Participants in some societies are more or less generous, accept more or less generous offers -- which suggests that a sense of fairness is culturally influenced, depending on the circumstances of the society. These summaries were created by me in collaboration with the Institute for the Future.

Source: Foundations of Human Sociality
Pub: Oxford University Press
Publication Date: 2004

Authors:

Patton, John Q.

Findings:

Market integration in nonwesternized cultures might have had a negative impact on cooperation by introducing unequal access to goods and conflict between western institutions and local nonhierarchical structures. The more market-integrated group (and the one described as more cooperative), the Quichua, proposed significantly smaller divisions compared to the Achuar. This poses a counterexample to a hypothesis presented in the Overview to this book, that market integration means frequent interactions and a familiarization with the process of cooperating with strangers, which translates into cooperation in the Ultimatum Game.
The most important factor in cooperation during the game was group coalition strength, over the much less significant or near significant factors of sex, age, and social status. Alarmingly high homicide rates among males in the area magnify the importance of coalition strength. "Coalitional instability undermines one's ability to condition future play, and discounts the benefits derived from investing in the creation and maintenance of a reputation as a fair player."
Longstanding habits of reciprocal fairness that flourish in politically stable environments overrode factors that would otherwise tempt self-interested behavior, such as the anonymous and one-shot properties of this Ultimatum Game.

One Sentence:

Patton attributes differences between two Ecuadorian ethnic/political groups in their willingness to cooperate in the Ultimatum Game to the groups' "differences in coalitional stability, perceptions of trust, and needs to maintain reputation," and emphasizes properties of the groups' political environment over individual differences.

One Paragraph:

This study examined patterns of cooperative behavior of two ethnic/political groups in Conambo of the Ecuadorian Amazon, the Achuar and the Quichua, with the Ultimatum Game. The participants were randomly divided into proposers and responders. Proposers were told to divide 20 coins worth a total of a days labor (approximately $3.85) into two piles, one for them and one for the responders. The proposer then left the room and a responder was brought in, not knowing the identity of their proposer, and asked to accept or reject the division (rejection of the division entailed no money for either participant, aside from the 5 coins given to all at the start for their time). A successive pile technique was used to determine the alliance strength of all participants. Informants were asked to divide photographs of the participants according to who would be most reliable in maintaining a coalition during a conflict. The researchers found that proposers with higher average alliance strength gave more generous offers and that the Achuar, with higher average alliance strength, had an average proposal of 42.9 percent, while the Quichua, with lower average alliance strength, had an average proposal of 24.6 percent. "The relationship between average alliance strength and amounts offered appears to be a group effect rather than an individual effect."