• An Investigation of Behavioral Influences in Strategic Decision Making

      Dufwenberg, Martin; Cardella, Eric; Kugler, Tamar; Wooders, John; Dufwenberg, Martin (The University of Arizona., 2012)
      In this dissertation, I study the impact of behavioral influences on strategic economic decision making in three essays.The first essay explores the interpersonal implications of guilt aversion in strategic settings. In doing so, I first introduce a stylized 2-player game where one players has an opportunity to induce guilt upon the other player in a manner derived from findings in the psychology literature. I then develop an experimental design, centered around this game, that allows me to test (i) whether agents attempt to induce guilt upon others in self-serving ways, (ii) whether agents are susceptible to the guilt induction of others, and (iii) whether agents are more trusting when they have an opportunity to induce guilt upon others. Furthermore, I theoretically show, via an application of the Battigalli and Dufwenberg (2007) model of simple guilt, that effective guilt induction can be supported as an equilibrium of the game considered.In the second essay, I explore the influence of posted price fairness concerns in bilateral negotiation settings. In doing so, I propose a price fairness model where, in addition to their material payoff, buyers receive disutility from engaging in negotiations, and aggressively negotiating, when the price is fair. As a result, the model predicts that buyers will negotiate less aggressively and possibly even forgo profitable negotiations when the posted price is fair, which is consistent with prior survey evidence on negotiation behavior. I also include a thorough discussion of the differences between the price fairness model and main alternative approaches to modeling fairness that exists in the literature.In the third essay, I experimentally investigate how the decision making quality of an agent's opponent influences learning in strategic games. In particular, I test whether learning-by-doing and learning-by-observing become more effective in games when agents face an optimal decision making opponent. To test these hypotheses, I propose a novel experimental design that enables me to measure strategic decision making quality and control the decision making quality of the opponent.