Punishment, Cooperation and Emotion Expression in Economic Exchange

Erte Xiao

Advisor: Daniel E Houser, PhD, Department of Economics

Truland Building, 400-R
May 17, 2006, 08:00 PM to 07:00 PM


Punishment is widely used to maintain social norms and enforce cooperation. Yet many aspects of punishment remain poorly understood. This dissertation is one step towards a deeper understanding of (i)why people punish; (ii) how and why punishment can sometimes fail to promote cooperative economic exchange; and (iii) what factors influence the efficacy of punishment in dynamic social environments. My findings have useful implications for economic problems including (i) how to design and implement sanctions to promote cooperation in commons environments, and (ii) the role of sanctions in efficient labor contracts, and (iii) the design of efficient online trading institutions. Chapter 1: Emotion Expression, Punishment and Fairness (Job market paper) Although it is accepted that emotions are connected to punishment decisions, there remains substantial debate over why humans use costly punishment. This chapter reports data from laboratory experiments with human subjects on this topic. The data suggest (i) constraints on negative emotion expression can increase the use of costly punishment, and (ii) emotion expression plays an important role in helping to promote fairness and maintain social norms in economic exchange environments. The results highlight the important role of communication systems, as channels for emotion expression, in individual decision-making. I argue that human demand for emotion expression can have significant behavioral consequences in social environments including families, courts, companies and markets. Chapter 2: When Punishment Fails Cooperation can be negatively affected by threats of sanctions, and researchers have pointed to both "intentions" and incentives as sources of this counterproductive effect. This chapter presents a novel experimental design aimed at determining the relative importance of negative intentions and incentives in producing non-cooperative behavior. The data indicate that incentives have a much more substantial influence on behavior than negative intentions, and that the nature of non-cooperative behavior is significantly affected by threats of sanctions. When not threatened with sanctions the most common behavior is to ?partially? but not fully cooperate, while this behavior is least common when punishees are threatened with sanctions. In particular, weak punishment significantly decreases trustworthiness and leads to completely selfish behavior. The results suggest that credible threats of sanctions can crowd-out norm-based motivations and increase the likelihood of income-maximizing decisions. Chapter 3: Publicly& Privately Implemented Punishment in a Public Goods Game People are affected both by events they experience and events they observe. This leaves open the possibility that, within a social context, privately implemented punishment might have different effects than publicly observable but otherwise identical punishment events. I report data from public goods game experiments with public or private punishment to investigate this issue. The data indicates that public punishment has a significantly different affect than private punishment, and that these differences vary with punishment?s severity. In particular, private severe punishment is most effective at promoting cooperation, while, consistent with the findings of weak punishment in Chapter 2, private weak punishment leads to less cooperation than found in a baseline public goods game without punishment. However, weak punishment implemented publicly increases group cooperation significantly. An important implication is that social effects of incentives are important to consider when designing mechanisms to deter misconduct.