Bestiary of Behavioral Economics/Ultimatum Game

From Wikibooks, open books for an open world
Jump to navigation Jump to search

The Ultimatum Game is an economic game that is designed to question the standard economic assumption of universally self-interested action.

Description[edit | edit source]

Created by the Israeli game theorist Ariel Rubinstein, the ultimatum game, like the dictator game, usually involves two people.[1] In the ultimatum game, after the first player is given some quantity of money, said first player must make an offer to the second player of how much of the money he is willing to share. If the offer of the first player is rejected by the second player, neither player gets to keep any money.[2]

Predicted Results[edit | edit source]

Under the standard assumption of self-interested rationality, the first player is predicted to offer the smallest non-zero amount he can to the second player, in order to maximize his own share of the money. Because the second player would be better off with even “a vanishingly small offer," as opposed to zero should he reject any kind of offer, he ought to accept even the lowest offer of money, so long as it is not zero.[2]

Actual Results[edit | edit source]

According to experiments involving the ultimatum game done by Kahneman, Knetsch, and Thaler (1986), the majority of first players in basic ultimatum games offer to split the money equally with the second player.[2] Further, “offers of less than $5 [representing 50% of the total allocation in the Khaneman et al. experiments] are sometimes rejected” by the second players.[2] These results contradict both the predicted results of the first player and those of the second player.

As with the dictator game, the results of the ultimatum game can be modified by changing more subtle parameters of the game (while leaving the overall premise intact). For example, there is a version of the game where both players are given a quiz before allocation and the first player is told that he “earned the right” to distribute money as he had the higher score. Offers by the first player in this version of the game are typically lower than in the standard experiment.[2] Rather than offering an equal split, the first player offers the second player a lower amount, typically a 70-30 distribution.[2] Despite this difference in the amount of money offered, rejection rates by the second player do not change significantly between permutations of the ultimatum game.[2]

Explanations[edit | edit source]

The explanations as to the anomalous behavior (according to standard economic theory) observed in the ultimatum game are varied.

The most obvious explanation to these results is “a fear of rejection.”[1] As the second player may decide to punish the first player for proposing too low an offer, causing neither to receive anything, the proposer may feel the need to offer a more equitable distribution as a means of self-protection.[1]

A second explanation is that human beings are inherently driven to cooperate and find a more equitable solution to such situations. The origin of this drive to the equitable is usually taken to be cultural ideas of “fairness” or “manners.”[2] As evidence to this, Hoffman states, “although the forms and norms of reciprocity vary endlessly across cultures, they are functionally equivalent in promoting cooperation.”[2] Indeed, the near universal presence of such norms of reciprocity can be taken to be evidence of a larger, evolutionary, predisposition to reciprocity among Homo sapiens.[2]

The problem with differentiating between these two (and other) competing explanations of the results of the ultimatum game is one of the appeals of the dictator game, which controls for the "fear of rejection" explanation.[2]

References[edit | edit source]

  1. a b c Clark, Josh. "What\u0027s the ultimatum game?" 26 February 2008. <> 06 May 2012.
  2. a b c d e f g h i j k Hoffman, Elizabeth, Kevin McCabe, Vernon Smith. “RECIPROCITY IN ULTIMATUM AND DICTATOR GAMES: AN INTRODUCTION.” Handbook of Experimental Economics Results 1, (2008): 412, 411-416. DOI: 10.1016/S1574-0722(07)00046-7