Research

Research Statement

Research Statement [Download]

My interests in economics are primarily in the areas of Macroeconomics, Macro-Labor, and Search Theory. I am interested in exploring the incentives created by search frictions and their macroeconomic implications. I believe that developing an understanding these facets of economic behavior greatly benefit society. I use both theoretical and empirical methods in my work. In my current project, I study how agents' job search behavior changes over their life cycle. The paper contributes to the literature because the model allows for structural estimation, and the structural model admits new insights on how education and gender affect labor search. I also analyze the welfare implications of retirement policies. In previous work, I have also studied labor search in academia, and sovereign default with optimal bailout. I have one paper already published and another under consideration. I look forward to producing more high quality research in the future.

I was motivated to pursue the topic in my job market paper by an important and not well explained stylized fact in the labor market. Over the life cycle, average wages and new job matches initially increase, then decrease towards retirement. To understand this, I study an overlapping generations model in which workers choose their search intensity. While a few papers have studied search in an overlapping generations context, I am able to estimate my model due to its clean theoretical results. I conduct comparative statics and estimate the model using Maximum Likelihood Estimation with Current Population Survey (CPS) data. Leveraging the structural nature of the model, I find that educated people have higher search costs, but also a higher arrival rate of jobs, and draw from a better wage distribution. I also show that women have higher opportunity costs to searching for work than men, but also see a lower arrival rate of jobs for the same effort. I find that self selection due to differences in preferences explain about one third of the employment and earnings gaps between men and women. Finally, raising the retirement age from 65 to 67 causes the employment to population ratio for workers under the age of 65 to increase by up to 2.5%, and wages for these younger workers to increase by up to 4%.

In a previous paper, I study a dynamic game in which members of a group vote on new inductions and prefer to induct applicants of their own type. These members know that newly inducted members will have a say in future elections so their own-type preferences are attenuated. I call this incentive which arises "Dynamic Homophily", and I characterize it theoretically and quantify it empirically. Empirically, I apply this model to the decisions of tenured faculty in an economics department who vote to hire a new candidate with tenure. To do so, I use a novel dataset on 40 top US economics departments over 24 years. This dataset includes the positions, citation counts, and JEL codes of the department members. I find significant evidence that own-type preference is strong in this environment. In addition, the role of the new hire in future elections factors strongly in decision making.

I look forward to continuing my work studying search. I am currently working on a project examining how frictions in the labor market affect individuals choices to run their own firm rather than supplying labor. This question is interesting because minorities face greater search frictions, but also tend to engage in less entrepreneurship. To answer this apparent contradiction, I introduce capital market frictions as well as labor market frictions to the model. In this line, studying entrepreneurship with search frictions, I also hope to explore the pro- or counter-cyclicality of entrepreneurship.

In my paper studying hiring in academia, I hypothesize that department members care about a researcher's quality, and the research synergies that can be realized by having him aboard. I hope that the my research shows through its rigor, creativity, and quality, that I would be a valuable and productive colleague.

Job Market Paper

Employment and Wages over the Lifecycle. [Download]

I study an overlapping generations model in which workers choose their search intensity. Over the lifecycle, wages and new job matches initially increase, then decrease towards retirement. Comparative statics are conducted on search intensity over the lifecycle. The model is estimated using Maximum Likelihood. I find that educated people have higher search costs, but also a higher arrival rate of jobs, and draw from a better wage distribution. I also show that women have higher opportunity costs to searching for work than men, but also see a lower arrival rate of jobs for the same effort. I find that self selection due to differences in preferences explain about one third of the employment and earnings gaps between men and women. Finally, raising the retirement age from 65 to 67 causes the employment to population ratio for workers under the age of 65 to increase by up to 2.5%, and wages for these younger workers to increase by up to 4%.

Other Research Papers

Search By Committee for new Voting Members: A Study of Tenured Hiring in Economics. (Under Consideration: Journal of Economic Behavior and Organization) [Download]

I examine a dynamic model of search by committee in which homophilic committee members vote to induct a candidate after observing the candidate's type and quality. A newly hired candidate joins the committee and votes in future elections. I study a new form of homophily that arises in this environment due to committee members' considerations of future elections, and call this ``Dynamic Homophily". Dynamic Homophily is characterized across variables such as committee size and composition, and the separation rate. I then estimate this model using a novel data set on the faculty of top public Economics departments in the US from 1994 to 2017. Researchers are clustered into types using JEL codes found in their publication histories, and their quality is measured using their citation counts. Estimation results suggest that to set a committee member indifferent between a same-type and a cross-type applicant, the cross-type applicant must have 80% more citations than the same-type applicant. Dynamic Homophily is responsible for up to an additional 20\% of the difference between same- and cross-type reservation values. In counterfactual experiments, I find that for high values of homophily, agents prefer a dictator, even of the other type, rather than voting, as the dictator solves the externality caused by Dynamic Homophily.

Birds of a Feather: Disentangling Spillovers and Shocks in Cooperative Convergence Groups. (Published: Annals of Finance)

In this paper, I propose a tractable model of sovereign default and the inter-state spillovers emanating from default. A coalition of nations may choose to insure against default, and the behavior of the coalition is used to examine the magnitude of the international spillovers. A voting structure for the coalition is proposed to examine idiosyncratic spillovers. Finally, the model is calibrated to the recent Greek Debt crisis to understand the spillovers from a default, and the moral hazard effect of the Troika.

Ongoing Work

Entrepreneurship and Search Frictions.

The literature on occupational choice studies frictions in the factors markets and the incentives they have on firm creation, but search frictions have yet to be discussed. Consider a model in which agents may choose to supply labor, or run a firm and demand labor. I introduce search frictions to this model by allowing firms post vacancies, and workers send an application to one vacancy per period. Some portion of workers are "stained", and when matching occurs, unstained workers will always receive a job over a stained applicant. I study the effect of search frictions on workers and on firm creation, both in and out of steady state. Stained workers experience greater search frictions and therefore have the incentive to run firms as an outside option. I study how these firms differ from those run by unstained agents.

Contact Info