Incentive Design for Equitable Resource Allocation: Artificial Currencies and Allocation Constraints

Devansh Jalota (Stanford University)

Abstract

The proliferation of algorithmic decision-making systems in resource allocation applications has enabled the efficient allocation of scarce resources. However, in the pursuit of an efficient outcome, such systems often discriminate against some users who may be made disproportionately worse off. To address this concern, in this work, we aim to develop equitable allocation mechanisms that respect the often complex preferences of users and associated allocation constraints while catering to the needs of all groups of society. In particular, we will develop conceptual frameworks to embed fairness and equity constraints in designing resource allocation mechanisms for emerging transportation and labor market applications. Furthermore, we will devise artificial currency market mechanisms that ensure users have an equal opportunity to avail resources. The results of this research will help (i) catalyze support from multidisciplinary stakeholders at the intersection of economics and optimization, (ii) inform policymakers about how to improve regulation in transportation, labor, and artificial currency markets, and (iii) lay the groundwork to pilot the mechanisms to tackle important resource inequity challenges in real systems.