Essays on Image Concerns and Norm-Enforcing Behavior

Abstract

This thesis consists of four essays which contribute to a better understanding of the behavioral impact of image concerns as well as norm-enforcing behavior. All of them employ experiments, either in a lab setting or a lab-in-the-field setting, and together focus on the behavior of adults as well as the behavior of children and adolescents.<br /> In Chapter 2 (joint work with Armin Falk and Simone Quercia), I study how self and social image concerns influence prosocial behavior. A stream of evidence from economics and psychology indicates the importance of the two notions of image concerns as drivers of prosocial behavior; however, no study compares them. We, therefore, develop a symmetric design that allows us to contrast the influence of the two notions. In a large scale dictator game, we exogenously increase self-awareness and observability in order to direct subjects' focus on their private and public self, respectively. We show that both self and social image concerns are drivers of prosocial behavior. We observe, however, that manipulating observability causes a stronger increase in prosocial behavior compared to self-awareness. We also document a marked gender difference in self-image concerns, shedding light on the different effect sizes across our treatments. While both genders react similarly to the observability manipulation, only men react to the self-awareness manipulation, closing the initial gender gap in prosocial behavior. We report evidence indicating that men tend to be generally less self-aware than women, i.e., they are less focused on their prosocial standards, but can be "nudged" towards them.<br /> In Chapter 3 (joint work with Armin Falk and Simone Quercia), I study how self and social image concerns influence lying behavior. Image concerns are often hypothesized as an important component of lying costs. In a standard die-rolling paradigm, we expose subjects to the same manipulations as in Chapter 2, and find that an increase in self-awareness has no effect on their reports. In contrast, we show that an increase in subjects' observability, while still maintaining their private information, significantly decreases their reports (and with it, their profits). We finally show in a survey experiment that respondents believe that the likelihood of a lie increases with the reported outcome, and attribute negative traits to people who make high reports. This further supports reputation concerns as the explanation behind the results of our social image treatment.<br /> In Chapter 4 (joint work with Armin Falk and Simone Quercia), I study how self and social image concerns influence prosocial behavior in childhood and adolescence. While lots of studies focus on the effect of image concerns in adults, very little is known about their roots in young age. We show that in a dictator game with 7-14 year-old subjects, both self and social image concerns matter, however, only for boys: i) boys give more when social image concerns are increased, significantly more so than girls, who do not react, ii) boys give more when self-image concerns are increased, while girls do not, and iii) supporting evidence indicates that the observed gender asymmetry is not due to differences in understanding of normative behavior, suggesting an actual difference in concerns for appearing prosocial. Our results support evolutionary theories suggesting that men should be more inclined to signal prosociality as well as recent predictions about the age when reputation should start playing a role. <br /> In Chapter 5 (joint work with Armin Falk and Fabian Kosse), I study the development of egalitarian norm enforcement in childhood and adolescence. Egalitarian norm is a long-existing organizing principle, commonly enforced by adults across societies. We investigate the development of this enforcing-behavior with 9-18 year-old subjects, by taking the most commonly-used third-party punishment game where a third party is added to a dictator game, and adapting it for children. We show that children start enforcing the egalitarian norm at the age of 11-12. In addition, we show that: i) as the egalitarian norm enforcement emerges, a non-negligible proportion of punishers also disapprove of overly-generous transfers that exceed the norm, ii) the punishers' behavior only changes until 13-14 years of age, indicating that egalitarian norm enforcement is mainly developed by that period, and iii) the dictators increase their transfer in the direction of the egalitarian norm primarily in the period when norm enforcement develops.