Essays in Discrimination, Informational Preferences, and Social Preferences Under Uncertainty

This thesis uses online experimentation to explore discrimination, informational preferences, and social preferences under uncertainty using methodology commonly found in experimental economics. The abstracts for each piece of work are as follows: Chapter 2: The Effect of George Floyd Protests on Experimental Evaluations We utilize a unique panel of experimental data to investigate whether the George Floyd protests and Kenosha Unrest affected race in worker evaluations. To manipulate race, we randomly assigned white and black names to worker-generated output. We establish that white individuals discriminate solely within in-group domains. Black names received lower ratings during the most active months of the George Floyd protests and showed no racial differences around Kenosha Unrest. Outside of active months of social unrest, black names received higher evaluations. The data suggest that our findings are likely driven by taste rather than belief-based partiality. Chapter 3: The Sunk Cost in Performance Beliefs We test whether real-effort costs influence how people update beliefs about their performance. We find that paying a cost to observe feedback alters belief adjustments, but is quite sensitive to informational quality. When feedback is highly accurate, under-inference is moderate and does not differ between cost conditions. However, when feedback is less accurate, those who received free feedback showed extreme under-inference, while those who paid a cost were approximately Bayesian. We argue that behavioral costs can stimulate responsiveness to feedback in environments that would otherwise motivate significant conservatism. Chapter 4: Talent, Sorting, and Redistributive Uncertainty It is well understood that many laborers choose occupations to maximize earnings with the intent to donate large sums to charitable causes. However, how others judge this strategy is less understood, as forgone opportunities are left unknown. Using a third-party dictator game, we evaluate such judgments. We find that redistribution is less common and more sensitive to efficiency considerations for equitable allocations. Furthermore, third-parties punish (reward) inequitable (equitable) allocations, a pattern amplified by deviations from efficiency. Some evidence suggests that the extent of punishment (reward) is more (less) weighted in equity than in the efficiency considerations that determined such outcomes.<p></p>

Author dc:creator

• Nicholas Lacourse (21038123)

• economics
• behavioral economics
• gender and race economics
• beliefs
• informationa preferences
• Bayesian updating
• belief updating
• social preferences
• fairness norms
• dictator game
• third party punishment
• payoff ambiguity
• racial discrimination
• social unrest
• stereotypes

Statement dc:rights

• All rights reserved
• Open Access after 2027-07-13

Identifier

10779/exe.31054657.v1

OAI identifier oai:identifier

oai:figshare.com:article/31054657