Research

Market Power in Input Markets: Theory and Evidence from French Manufacturing [Job Market Paper]
This paper documents the market power of large importing firms in foreign input markets, and evaluates its effects on the aggregate economy. I develop an empirical methodology to consistently estimate buyer power at the firm level. I apply the methodology to study imperfect competition in the market for foreign intermediate inputs, using longitudinal data on trade and production of French manufacturing firms from 1996-2007. My results show that buyer power is substantial, concentrated in key sectors, and it significantly correlates with the size and productivity of the firm. I then show that, in a simple general equilibrium model of production, buyer power has large distortionary effects, both at the firm and the economy level. This type of distortion could cost around 0.2% of total GDP in France. My analysis suggests that policies that spur import market integration can reduce the scope of buyer power and thus play a key role in stimulating aggregate production.
International Prices and Cost Shocks: The Shock Matters (draft coming soon)
This paper investigates how export prices in different destination countries react to cost shocks that originate from two different sources: fluctuations in the exchange rate, and in tariffs. Using longitudinal export and production data for the universe of French manufacturing exporters, I provide new evidence that, conditional on the size of the shock, the year-to-year elasticity of free-on-board export prices to cost shocks is higher when the shock generates from exchange rate than tariff movements. In other words, the exchange-rate pass-through is larger than the tariff pass-through. This fact stands in contrast to traditional, static models of international pricing behavior, which predict symmetric pass-through of all cost shocks. I argue that a menu cost model of price adjustment, where firms have to pay a cost to change their prices, can rationalize the empirical evidence.

Work in Progress

A Sufficient Statistic Result for Estimating Firm-Level Pass-Through (with C. Arkolakis)