Empirical essays on dynamic allocation mechanisms
Author(s)Waldinger, Daniel Cane
Massachusetts Institute of Technology. Department of Economics.
Nikhil Agarwal and Parag Pathak.
MetadataShow full item record
This thesis contains three chapters which empirically study how dynamic decision making affects the allocation of public resources. In the first chapter, I study the problem of allocating public housing. In the U.S., public housing authorities (PHAs) allocate apartments using a wide range of choice and priority rules. I evaluate how these allocation mechanisms affect the efficiency and redistribution achieved through assignments. Using waiting list data from Cambridge, MA, I estimate a structural model of public housing preferences, finding substantial heterogeneity in applicant outside options and preferred apartment types. Counterfactual simulations suggest that the range of mechanisms used by PHAs involves a significant trade-off between efficiency and redistribution. However, some commonly used mechanisms are never optimal. In the second chapter, joint with Nikhil Agarwal, Itai Ashlagi, Michael Rees, and Paulo Somaini, I study the allocation of deceased donor kidneys. In the U.S., patients on the kidney waiting list are offered organs in order of priority, and may decline an offer without penalty. This paper establishes an empirical framework for analyzing the design of these waiting lists. We model the decision to accept an organ as an optimal stopping problem and use waiting list data to estimate the value of accepting various kidneys. We then show how to solve for counterfactual equilibria under different priority rules, and search for mechanisms that improve the match quality of transplants and reduce organ waste. In the third paper, joint with Sydnee Caldwell and Scott Nelson, I investigate how beliefs about risky future income influence households' financial decisions. We quantify one contributor to income uncertainty by surveying low-income tax filers' expectations of and uncertainty about their tax refunds, and link the survey with administrative tax and credit report data. Households face substantial refund uncertainty, and both refund expectations and surprises influence financial behavior. Households borrow in anticipation of their tax refunds, and this pattern is less pronounced for more uncertain households, consistent with precautionary behavior. Surprisingly, positive refund surprises induce higher debt levels by relaxing down-payment collateral constraints.
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Economics, 2018.Cataloged from PDF version of thesis.Includes bibliographical references (pages 305-314).
DepartmentMassachusetts Institute of Technology. Department of Economics.
Massachusetts Institute of Technology