Research

Abstract

The Abilene paradox (AP) is a situation in which a group of people implement an outcome which no one prefers. We show that in a sequential voting game, a combination of social preferences and incomplete information can lead to the AP. This model is tested in an online experiment. We find evidence suggesting that on-line subjects vote according to the group’s preference rather than their selfish preference, potentially leading to the AP. Moreover, we find a position effect: when subjects are the first to vote in the group, they are more likely to vote according to their selfish preference than when they are the second, third, and so on.

Abstract

Recent evidence suggests that a person's reciprocity toward others depends on the amount of sacrifice made by others. We develop a reciprocity model that incorporates the concept of sacrifice, defined as the payoff forgone when choosing an action. Our model retains many desirable properties of the existing reciprocity models, while capturing the nuanced psychology of reciprocity. In our model, one's motivation to reciprocate increases in the size of others' sacrifice. In two-person games, it can be shown that our model strengthens reciprocity motivation in reciprocity equilibria that are not Nash equilibria under selfish preferences.

Abstract

Despite the growing interest in leadership in the experimental lit- erature, theoretical analyses of leadership have received relatively lim- ited attention. This study investigates the effectiveness of leading by example in public goods games among reciprocal agents. The “leading- by-example” game (LBE) is a two-stage game with the same payoff structure as the classic voluntary contribution mechanism (VCM), in which a leader first makes a contribution decision, followed by the fol- lowers who make a contribution decision upon observing the leader’s move. If the marginal per capita return (MPCR) is sufficiently high, LBE is more effective than VCM at inducing contributions. However, if MPCR is low, VCM is more effective than LBE. An experimental test of the comparative statics is conducted.

Abstract

Psychological motivation is generally private information which generates fundamental uncertainty about others' psychological sensitivity. We incorporate uncertainty about others' reciprocal motivation into existing models of reciprocity (Dufwenberg and Kirchsteiger, 2004; Rabin, 1993) and propose an equilibrium concept, called Bayesian reciprocity equilibrium. We prove that an equilibrium exists. Then, we apply the theory to solve the prisoners' dilemma. In this example, we show that a small amount of uncertainty in reciprocal motivation can eliminate cooperation via iterated elimination of dominated strategies, unlike the complete information case in which cooperation is supported as an equilibrium outcome.