Date of Award
Doctor of Philosophy (PhD)
This dissertation proposes new estimators of program treatment effects in the presence of spillovers, a situation where one person's treatment can affect another's outcome.The first chapter focuses on a setting where treatment decisions depend on observables and there are spillovers from friends in a social network known to the researcher. I show that primitive conditions on the treatment, a dyadic model of network formation, and a flexible random coefficient model on the outcome, can motivate high-level unconfoundedness and rank conditions to identify average treatment effects and spillovers effects. I show that a novel finite-dimensional statistic (the network propensity score) can summarize the relevant observables under key conditions. I propose feasible estimators of the average effects that are consistent and asymptotically normal in settings with multiple networks. Finally, I study the effects of an intervention on political participation in Uganda where I find evidence of spillovers on non-participants. I also evaluate an information intervention about microfinance adoption in India where I find large treatment effects but limited spillovers effects. The second chapter (coauthored with Francis J. DiTraglia, Camilo García-Jimeno and Rossa O'Keeffe-O'Donovan) shows how to use a randomized saturation experimental design to identify and estimate causal effects in the presence of spillovers and one-sided non-compliance--subjects can only be offered treatment, not compelled to take it up. We consider the case in which spillovers occur only within known groups, and take-up decisions do not depend on peers' offers. In this setting we point identify local average treatment effects, both direct and indirect, in a flexible random coefficients model. We go on to propose a feasible estimator that is consistent and asymptotically normal as the number and size of groups increases. We apply our estimator to data from a large-scale job placement services experiment, and find negative indirect treatment effects on the likelihood of employment for those willing to take up the program. These negative spillovers are offset by positive direct treatment effects from own take-up.
Sánchez Becerra, Alejandro, "Essays On Peer Effects And Network Econometrics" (2021). Publicly Accessible Penn Dissertations. 3718.