This dissertation explores the effects of international trade policy cooperation on firm export behaviour and competition in international markets. It is motivated by three facts. First, firms are increasingly exposed to international trade. World trade has grown by an average of 5% over the last 30 years, and the world trade to GDP ratio has increased by more than a third over this period to around 52% in 2020. Second, firms are subject to a large and growing number of trade policy interventions. The number of trade agreements notified to the WTO, for example, has increased more than tenfold over the past 30 years, to well over 300. And third, international trade agreements do more than just influence the behaviour of the firms they benefit. They shape the competitive environment of the markets they affect.

The first chapter argues that mutual recognition agreements have both direct and indirect effects on firms’ export behaviour. These agreements make it easier for international firms to demonstrate that their products meet the minimum requirements of a market, and so streamline market access. I build a multi-country model of trade with oligopolistic competition and variable markups that explicitly models the conformity assessment sector and allows firm decisions to be non-separable across markets. The model highlights three separate direct effects of mutual recognition agreements on firms, an effect on firms’ fixed costs, an effect on firms’ marginal costs, and an effect on the price of conformity assessment, and shows that each of these direct effects also has an indirect effect on the intensity of competition in final goods markets. I simulate my model with standard parameters from the literature and show that mutual recognition agreements benefit firms which are affected by their direct effects, but hurt firms which are exposed only to their indirect effects. I also show that both of these effects are particularly relevant for the most productive firms, and that the effects of mutual recognition agreements without rules of origin are essentially the same for firms based in signatory countries and third-country firms.

The second chapter estimates the effects of mutual recognition agreements on disaggregated firm exports. I build on the model presented in the first chapter and develop a new empirical approach which focuses on the experience of third-country firms to circumvent reverse causality concerns and distinguish between the direct and indirect effects of mutual recognition agreements. I then compile a new dataset on the product coverage and implementation timeline of thirteen mutual recognition agreements and combine it with the universe of firm exports for thirteen emerging and developing economies. My results show that firms which benefit from cost reductions as a result of a mutual recognition agreement export around 15% more, while firms which are primarily exposed to the agreement’s broader effects on a market’s competitive environment export up to 15% less. Both the direct and indirect effects of mutual recognition agreements also matter for firms’ extensive margin decisions, export volumes, export prices, the frequency and variety of firm exports, and firms’ import behaviour, as well as for their export performance in unaffected markets.

The third chapter explores firms’ markup decisions in international markets and highlights the implications of differences in the intensity of competition between different sets of firms. This chapter is joint work with Meredith Crowley and Lu Han and develops a new multi-country model of trade with variable markups in which firms from the same origin compete more fiercely with each other than with firms from other origins. This gives rise to a rich oligopolistic structure in which an exporter’s markup adjustment after a trade policy shock depends on two market share reallocation effects: (1) an across-origin reallocation effect which captures changes in overall competitive pressures for all firms from a given origin in the destination market and (2) a within-origin reallocation effect which captures changes in the competitive pressure an exporter faces from its compatriot firms from the same origin. To explore the implications of this model empirically, we combine data on trade agreements and tariffs with detailed administrative customs datasets for eleven emerging and developing economies. We find that the two reallocation effects move in opposite directions after a bilateral tariff liberalisation: while firms face less competition from other origins and the preferred origin as a whole gains market share, each exporting firm faces more competitive pressure from its compatriots due to additional entry and loses market share within-origin. Overall, our results suggest that the within-origin reallocation effect dominates and exporters reduce their markups in response to a bilateral trade liberalisation.