My job market paper explores the optimal distribution of a limited number of preventative treatments (eg: vaccines) for a deadly communicable disease such as malaria or Ebola among individuals on a network. While the literature has considered the optimal distributions of treatments in the presence of heterogeneous treatment effects, the main contribution of this paper lies in accommodating for spillovers in treatments. This paper explicitly models disease propagation on the contact network for diseases such as Ebola or malaria. I extend the empirical welfare maximization (EWM) procedure considered in Kitagawa, Tetenov (2018) to estimate an optimal treatment assignment rule using data from a randomized control trial (RCT). Like Kitagawa, Tetenov (2018), I provide a finite sample bound for the effectiveness (measured in terms of uniform regret) of the proposed procedure. This is the first paper to establish theoretical guarantees for the treatment assignments under restriction on the spillovers and the network structure. I demonstrate that the welfare attained by EWM converges to the maximum attainable welfare as the size of the RCT grows. I also show that there can exist no other statistical procedure with a faster rate of convergence.
This paper presents a game theoretic model of Joint Liability Lending (JLL) microfinance programs with endogenous peer pressure to repay. In addition, this paper describes a role for institutional pressure applied by microfinance institutions. We find that this model helps better explain two important empirical findings in the literature. Firstly, observed repayment rates in not for profit microfinance programs are very high. Secondly, the welfare implications of these programs (as evidenced by RCTs) are small. A sequential game is analyzed where the MFI’s interest rate, the projects selected by the group members and the subsequent peer pressure and repayment decision are endogenized. I also characterize the solutions and analyze the outcomes computed over a large range of parameter values. The most striking intuition generated by the model is that when (risk-averse) participants can choose between low risk-low reward and high risk-high reward investments, and the MFI prefers to set low interest rates, the resulting equilibrium boasts inefficiently high repayment rates. This leads to an inefficient transfer of the burden of risk bearing onto the participants who respond by inefficiently choosing low risk-low reward investments. Thus, counter to the main purpose of these programs of poverty alleviation, this implies that growth generating investments (high risk-high reward) are left under funded in equilibrium. Thus, the model provides a more satisfactory explanation of some of the empirical findings in this literature.
Works in Progress
Sharp Identification Region from Pairwise Stable Networks with Francesca Molinari
Optimal Seeding of Office Lite on Collaborator Networks with Francesca Molinari and Sida Peng
The Effect of Evictions in the Azure Spot Market with Francesca Molinari, Sida Peng and Will Wang
Awards and Honors
Tapan Mitra Economics Prize, 2018
L.R. “Red” Wilson MA ‘67 Excellence in Economics Medal, 2018
Michael Brunn Family Goldman Sachs Scholarship, 2020