Essays in Macroeconomics and International Trade

Abstract

Economic activity frequently takes place in markets that are subject to search frictions. To reduce search costs, agents often interact repeatedly with the same partner, for example through long-term employment contracts or in long-term business relationships. The main purpose of my dissertation is to study long-term relationships between firms. Chapters 1-2 use confidential, transaction-level import data from the U.S. Census to comprehensively analyze long-term business relationships, and use theory to illuminate their potential macroeconomic effects. Chapter 3 studies how labor market institutions affect worker-firm relationships, specifically the allocation of workers to firms.

In the first chapter of my dissertation, I investigate how long-term buyer-seller relationships affect price rigidity. Economists have long suspected that firm-to-firm relationships might increase price rigidity due to the use of explicit or implicit fixed-price contracts. I study the responsiveness of import prices to exchange rate changes in the Census data and show that prices are in fact substantially more responsive to these cost shocks in older versus newly formed relationships. Based on additional stylized facts about the life cycle of relationships and interviews I conducted with purchasing managers, I develop a model in which a buyer-seller pair subject to persistent, stochastic shocks to production costs shares profit risk under limited commitment. Relationships that experience good shocks have lower costs, trade more, and survive longer, which generates the life cycle. Furthermore, since partners in older relationships on average enjoy a greater relationship surplus, alternative matches are less attractive to them, which enables the firms to share profit risk more completely by setting prices that are more responsive to shocks. Once structurally estimated, the model replicates the empirical correlation between relationship age and price flexibility. My results suggest that changes to the average length of relationships in the economy – e.g., in a recession, when the share of young relationships declines – can influence price flexibility and thus the effectiveness of monetary policy.

The second chapter (co-authored with Justin Pierce, Georg Schaur, and Peter Schott) examines how trade barriers affect firms' relationships with suppliers. In our theoretical framework, firms can conduct purchases under two opposing systems. Under the "Japanese" system, buyers motivate sellers to maintain product quality via small, frequent orders at a price above seller's cost and by promising a continued relationship if the quality is good. Under the "American" system, buyers place large orders with the lowest cost bidder and quality is ensured via costly inspection. A lower probability of a trade war increases firms' incentives to switch to Japanese-style procurement since it makes long-term relationships more sustainable. We test the model's predictions using the Census data. We use a differences-in-differences strategy to exploit a shift in U.S. trade policy which eliminated the threat of a rise in U.S. tariffs for Chinese imports to potentially prohibitive levels. Consistent with our predictions, we find a shift of procurement practices towards the Japanese system after the policy change. The results suggest that trade agreements which allow firms to develop long-term relationships may give rise to a new source of welfare gains from trade associated with lower inventory and monitoring costs.

The third chapter of my dissertation (co-authored with Tommaso Porzio) focuses on worker-firm relationships. We investigate whether a change in labor market regulations can improve the allocative efficiency of worker-firm matches. We study this question using the German reunification as a natural experiment that exposed East Germany to Western-style institutions. Interpreting a firm's median wage as a measure of its inherent productivity, we use matched employer-employee data to examine the evolution of allocative efficiency, defined as the correlation between a firm's median wage and its number of workers. We find that East German allocative efficiency is significantly below West German efficiency levels even 20 years after the reunification, for two reasons. First, East German workers face a flatter job ladder: when moving job-to-job, the difference between their previous firm's median wage and their new firm's median wage is smaller than in the West. Second, East German workers more frequently become unemployed. We rationalize our findings in a job ladder model with low and high productivity firms in which East Germany has a higher risk of job termination than the West. This higher risk of employment relationship separation lowers the incentive for high productivity firms to post vacancies, which flattens the job ladder. Our work highlights that policies shifting a country's labor market institutions towards Western policies may fail to generate large efficiency gains when the shift is accompanied by a rise in unemployment.