Three essays on experimental economics: Social recognition via charity, gender difference in risk preference and property rights through effort investment

Sukanya Chaudhuri, Purdue University

Abstract

In this dissertation we add two new experimental studies to the growing catalog of non-economic factors that influence one's economic decisions. Specifically, we examine the effect of social recognition on charitable donation and the effect of context on gender difference in risk preference. In addition to these studies, this thesis examines how the origin of wealth affects decisions to give in laboratory experiments involving charitable donation. The theories of public good and warm glow have shaped much of our understanding about why people charitable contributions. Despite the tremendous empirical success of these theories, some commonly observed aspects of charitable giving remain unexplained. One such aspect is social recognition. It has been shown repeatedly that one's decision to donate and how much to donate may depend on whether others get to know about this act of charity. In Chapter 2, we explore how this effect can be partially explained by the theory of conspicuous consumption due to Glazer and Konrad (1996). According to this theory, when social rank is determined by factors that are unobservable to others, people may use charitable donation as an instrument to signal their social status, by ensuring that their observed donation rank exactly matches their unobserved social rank. We present an experimental study demonstrating this phenomenon where subjects suitably change their initial allocation to charitable donation to reduce the mismatch between their donation rank and the rank determined by their unobserved endowment. With the help of varying treatment conditions and a careful hierarchical Bayesian data analysis we establish that the effect of rank mismatch on the change in donation is more than what can be explained by the intrinsic utility subjects derive by donating to charity. It is commonly held that women are more risk averse than men. Some empirical and experimental studies provide evidence toward this stereotype while others report no significant difference. Experimental studies usually employ an abstract setting to elicit risk propensity. It is often argued that abstract settings, with the lack of any familiar cues, do not well represent the natural world. Thus it would be important to build a catalog of gender difference studies that make use of a diverse set of familiar contexts. In Chapter 3 we present one such study where gender difference in risk aversion is examined in the context of emission trading and its mathematically equivalent abstract gamble counterpart. Our experimental study indicates a greater risk propensity by women in the context of emission trading, though the stereotype of their greater risk aversion than men prevails in our abstract game treatment. Our detection of a greater risk propensity by women is a first of its kind, as the best documented evidence against the stereotype has so far been in the form of “no significant differences between the sexes”. Emission trading is fast becoming an important social issue, involving intricate sub-contexts of self-restraint and law abidance under imperfect enforcement. Why women tend to be more risk seeking in such a context poses an interesting question for future research. It is a common practice in economics experiments to gift subjects with an initial endowment. It is debatable whether subjects gifted with an endowment have the same kind of attachment to it as people do in the natural world where endowment is hard earned. To answer this, we study in Chapter 4 whether greater effort investment leads to bigger property rights in a modified dictator game where recipients are well known charities. Our choice of charities as recipients eliminates a peer-to-peer comparison and the associated relative property rights that are commonly observed in dictator, ultimatum and trust games. We find that whether endowment is earned or received as a gift has no effect on what subjects to donate to charities. We also observe that whether endowment is earned by defeating a competitor or cooperating with a partner has a minimal effect on the donation amount. The effect can be explained by relative property rights that two cooperating subjects may develop toward the common sum they earn as a team. Therefore, we conclude that effort investment alone does not lead to stronger property rights – property rights develop only in a context where two peers compare their relative worths.

Degree

Ph.D.

Advisors

Cason, Purdue University.

Subject Area

Economics

COinS