Date of Award

Spring 5-22-2021

Degree Type


Degree Name

Doctor of Philosophy (PhD)




Devashish Mitra


Import Growth, International Trade, Labor Mobility, Preferential Trade Agreements, Terms-of-Trade, Trade Policy

Subject Categories

Economics | Social and Behavioral Sciences


This dissertation is comprised of three chapters on preferential trade agreements (PTAs). The first two chapters examine why countries sign PTAs with other countries, while the final chapter studies the effect of PTAs on imports.

The first chapter provides direct empirical evidence that PTAs are consistent with the terms-of-trade theory. In other words, PTAs internalize the temptation to increase tariffs when countries possess importer market power. Using PTAs that occur between 2001-2015, I study the formation of PTAs and the structure of first year PTA tariff cuts. I first show the likelihood of a PTA increases when both countries have more importer market power. This suggests countries form PTAs in a way that internalizes the terms-of-trade externality by agreeing to mutual tariff decreases. Second, using recently available tariff data for 39 bilateral PTAs, I show high importer market power leads to tariff cuts that are 103 percent larger in magnitude when PTAs enter into force. Thus, countries target tariffs they have a strong temptation to raise in the absence of an agreement, as predicted by the terms-of-trade theory. These results provide a rationale for why PTAs are allowed under the World Trade Organization’s rules.

The second chapter focuses on an alternative rationale for PTAs by examining the empirical relationship between trade agreements and domestic labor mobility. The domestic-commitment motive from Maggi and Rodriguez-Clare (2007) and standard trade models with labor frictions predict that trade liberalization should occur when labor is more mobile. I find support for this prediction. Using PTAs covering 56 countries in 2015, I show labor mobility is a strong predictor of trade liberalization. The probability of a PTA increases when the country pair’s average domestic labor market is less rigid. When the average labor mobility increases by 1 standard deviation from the mean, the probability of a PTA increases by 14-26 percent. These results are also consistent with results on the bound tariffs negotiated under the World Trade Organization, where less rigid labor markets are associated with lower bound tariffs.

The final chapter, co-authored with Xiuming Dong, estimates the causal effect of phase-in tariffs on import growth for 11 PTAs signed by the United States between 2000-2007. The phase-in hypothesis from Baier and Bergstrand (2007) implies that products with longer periods of tariff cuts should experience slower and longer periods of import growth relative to unchanged tariffs. Using recently available PTA tariff data, we extend the work of Besedes et al. (2020) and utilize a triple-difference strategy to provide additional evidence that import growth is not consistent with the phase-in hypothesis. We show that phase-in tariffs do not necessarily yield additional import growth relative to already duty-free products. Our analysis documents the average effect of phase-in tariffs and PTA-specific trends.


Open Access

Included in

Economics Commons