Evgeni Drynkin

Evgeni Drynkin
PhD Program, Economics
PhD Program Office
Graduate School of Business
Stanford University
655 Knight Way
Stanford, CA 94305

Evgeni Drynkin

I’m a Ph.D. candidate in Economics Analysis and Policy. My research spans different areas of empirical IO with both applications to actual industries and developing techniques for structural estimation. I am available for interviews during the ASSA Annual Meeting in San Diego (Jan 3-5).
Research Interests
Empirical Industrial Organization
Structural estimation
Econometrics

Job Market Paper

Mergers and Investments in the Wireless Industry
This paper studies the outcomes of hypothetical T-Mobile/Sprint and AT&T/T-Mobile mergers in the U.S. wireless telecommunications industry. I propose a model in which consumers trade off price and network coverage, so firms have to compete on both price and investment. The key finding is that had T-Mobile and Sprint merged in 2009, consumers would have benefited from expanded network coverage. The two firms would have increased profits due to less duplication on the investment side. An acquisition of T-Mobile by AT&T, on the other hand, would have harmed consumers because it would not have resulted in better coverage. Additionally, the outcomes of the T-Mobile/Sprint merger vary across geographic areas. Markets with high population density or flat terrain typically have a strong initial Sprint or T-Mobile presence, and would therefore experience lower, often negative, changes in consumer surplus as a result of the merger. Conversely, markets where the merging parties struggle to enter separately, mainly those with lower population density and harder to cover terrain, benefit more because the merger would diversify carrier choices.
Working Papers
Random Coefficients Model: A Machine Learning Approach (with Nick Doudchenko)
In this paper we propose a new method to estimate a discrete choice demand model when individual level data are available. The method employs a two-step procedure. Step 1 predicts the choice probabilities as functions of the observed individual level characteristic. Step 2 estimates the structural parameters of the model using the estimated choice probabilities at a fixed point. We use simulations to compare the performance of the proposed procedure with the standard methodology. We find that our method delivers an improved precision as well as a substantially faster convergence time. We supplement the analysis by providing the large sample properties of the proposed estimator.
Work in Progress
Market Structure and Growth in Television (with Ali Yurukoglu and Chuan Yu)
Last Updated 14 Jul 2020