Aurélien Salas

Research

Job Market Paper

• Price Discrimination with Endogenous Participation

  Abstract: This paper studies a platform that informs a monopolist seller about buyer valuations to enable price discrimination. The platform commits to a segmentation rule, and buyers decide whether to participate by comparing their expected surplus under this rule to an outside option. The platform thus internalizes how its information affects both participation decisions and seller pricing responses. I vary buyers’ information about their valuations and analyse how endogenous participation reshapes optimal segmentation. Endogenous participation creates a fundamental trade-off between extraction and participation, shifting optimal segmentation toward more buyer-friendly allocations. When participation is increasing in valuations, the most buyer-friendly designs are not feasible and the welfare frontier shrinks. I characterize this constrained frontier and show that the platform’s optimal surplus choice is determined by a reverse-hazard condition balancing participation incentives against profit extraction. Overall, endogenous participation can align seller-aligned platforms with buyer welfare, but differential participation across buyer valuations simultaneously limits the platform's ability to implement buyer-friendly allocations.

Publications

• Strategic Information Disclosure to Classification Algorithms: An Experiment Forthcoming Experimental Economics
  with Jeanne Hagenbach
  Paper Online Appendix

  Abstract: We experimentally study how individuals strategically disclose multidimensional information to a Naive Bayes algorithm trained to guess their characteristics. Subjects' objective is to minimize the algorithm's accuracy in guessing a target characteristic. We vary what participants know about the algorithm's functioning and how obvious are the correlations between the target and other characteristics. Optimal disclosure strategies rely on subjects identifying whether the combination of their characteristics is common or not. Information about the algorithm functioning makes subjects identify correlations they otherwise do not see but also overthink. Overall, this information decreases the frequency of optimal disclosure strategies.

Work in Progress

Disclosing Proxies
with Paula Onuchic