Three Essays on Other Regarding Behavior in One-Shot Anonymous Games

Robert Mayo

Advisor: Kevin A McCabe, PhD, Department of Economics

Committee Members: Thomas Stratmann, David M. Levy

Buchanan Hall, #D180
April 26, 2017, 01:00 PM to 10:00 AM


Chapter 1: I use subjects recruited from an online employment exchange to study the robustness of the triadic trust design with a different subject pool. In running my experiments, I take advantage of the cost reducing features of the micro-employment culture found on Amazon’s Mechanical Turk. I find that first mover trust is robust to the change in subject pool, but second mover reciprocity is not. In exploring this failure, I examine some of the procedural requirements necessitated by Mechanical Turk. I conclude that procedures play a significant role in the robustness triadic trust design and are likely to play a significant role in comparing the results from other experiments across experimental modalities.

Chapter 2: I conduct an on-line experiment to decompose giving in a dictator game into amounts motivated by pure altruism and amounts motivated by warm glow. By manipulating the price of benefit to the recipient while holding the price of the act of giving constant, I estimate values for a coefficient of altruism α, as defined in Andreoni (1989). I find significant evidence of both pure altruism and warm glow as motivations for giving. The fraction of amounts sent to an anonymous recipient that are motivated by warm glow is increasing in age of the dictator and decreasing in dictator’s income. Female subjects are not sensitive to price suggesting they are mostly motivated by warm glow rather than pure altruism, while male subjects demonstrate the opposite behavior.

Chapter 3: I conduct a series of experiments to determine if the difference in behavior between men and women found in chapter two is caused by subjects following gender specific economic norms. I use adopt the definition of norms from Bicchieri (2006) that a norm is composed of empirical expectations, normative beliefs, and conditional preferences relative to a reference group. Using an incentivized questionnaire, I show the existence of gender specific empirical expectations and normative beliefs. To establish conditional preferences, I adapt the procedure of Bicchieri and Xiao (2009) and show causality.