Essays on Firms in International Trade

My thesis is composed by three chapters that study the important role that firms play in mediating imports and exports between countries, and how their decisions and heterogeneity can influence the behavior of macroeconomic outcomes. The first chapter, explores the role of demand uncertainty and learning in explaining the low survival rates and sales of new exporters using a manufacturing survey for Colombian firms. In the second chapter, using a transaction-level dataset for Colombian exporters, I document novel facts about the post-entry dynamics of firm performance and prices in international markets, and study their connection with the transmission of exchange rate shocks into firm-level prices. To rationalize these facts, I propose a theory of firm dynamics emphasizing the interaction between demand accumulation and endogenous export participation. In the third chapter, using a novel administrative dataset for Canada, I explore the role of trade intermediaries in bringing foreign products into the domestic market and empirically estimate how domestic manufacturing firms are affected by being exposed to the imports of final and intermediate goods carried by wholesalers. In Chapter 2, I develop a model featuring Bayesian demand learning and endogenous export participation in which the learning mechanism is modelled as a signal-extraction problem. The underlying assumption is that firms are uncertain about their demand in the foreign market and, for each period they export, they observe a noisy realization of it. These observed signals determine firms\u27 posterior beliefs about their demand, on which they base their quantity and participation decisions. The parameters of the resulting dynamic discrete choice model are calibrated to match several key moments observed in Colombian firm-level data. This model is able to reproduce the dynamics of new exporters observed in the data, namely that new exporters sell only a small fraction in the foreign markets, and their survival rates are smaller than the ones of more experienced firms. Finally, the calibrated model is used to evaluate the efficiency of policies subsidizing a fraction of the fixed costs associated with exporting. The results suggest that, in terms of benefit-cost ratios, policies subsidizing exporters with less than 3 years of experience are more effective than the ones targeted to all the exporting firms. In Chapter 3, I further explore novel features about the dynamics of new exporters, and study how these dynamics are related to the firm-level response to exchange rate shocks. Using a transaction-level data for Colombia over the period 2008 to 2018, I document that: 1) export sales and survival rates are initially small, but then consistently grow in the years following entry to a new market; 2) prices of exported goods increase with the number of years a firm has exported to the same destination; 3) the exchange rate pass-through into international prices is incomplete and new exporters exhibit a pass-through rate that is, on average, 1.8 times higher than the one of incumbent exporters. To rationalize these empirical facts, I develop a dynamic discrete choice model of exporting introducing customer base accumulation due to deep habits. A key feature of this model is that pricing decisions become dynamic, and exporters have incentives to charge low prices upon entry to foster their future demand at the expense of their current profits. Moreover, this model has important predictions for the variation of the price response to exchange rate movements across firms. According to the model, young firms that are closer to the exit margin transmit a higher fraction of the exchange rate shocks into their prices, while experienced exporters partially absorb these shocks in order to keep their prices and customer base stable. I calibrate this model to match salient features of Colombian exporters, and in particular to generate new exporter dynamics that are consistent with the data. I then use the calibrated model to examine, through impulse-response analysis, the aggregate response of trade volumes to two types of trade shocks: a persistent depreciation of the Colombian peso, and a trade liberalization episode. The inclusion of deep habits and endogenous entry and exit has important aggregate implications. First, aggregate volumes are more responsive to tariffs than to exchange rates. On impact, I estimate an elasticity of trade volumes to tariffs of 1.5 and to exchange rates of 0.5. Second, for both shocks, the costumer base accumulation mechanism generates a sluggish response of trade volumes, which produces discrepancies between the short-run and long-run elasticities. Particularly, in response to a permanent tariff reduction, I estimate a long-run trade elasticity that is 2.6 times higher than the short-run elasticity. In Chapter 4, by using a unique administrative dataset containing detailed information on firm performance and import transactions for all Canadian firms, I study the role of wholesalers as international trade facilitators, and provide evidence on how their import decisions affect the performance of domestic producers through different margins. Initially, with this dataset, I document some features regarding the import behavior of wholesalers. First, I find that the share of total import value accounted by wholesalers increased from from 26.5% in 2002, to 34.3% in 2012. Second, I document that wholesalers were the dominant players on the final goods import markets, accounting for almost 50% of the total import value of final goods in 2012, up from 43% in 2002. Similarly, they earned notoriety in the import markets of intermediate inputs by increasing their import share from 16% to almost 22% ten years later. Additionally, I find substantial differences on how wholesalers and manufacturing firms engage in international trade. Specifically, I document that when compared to manufacturing firms, wholesalers imported more goods and from more countries, and over the observation period the number of importing firms grew more in the wholesale trade sector relative to the manufacturing sector. Finally, I explore how domestic producers respond to the increased import competition in the form of final and intermediate goods carried by wholesalers. To do that, I construct measures of exposure to these indirect imports at the local market level, and exploit the geographic variation in this exposure to empirically estimate the effect on sales, employment, productivity and exit rates of the domestic manufacturing firms located there. The empirical analysis reveals that a higher exposure to indirect imports of final goods has a negative effect on domestic firms\u27 employment and sales and a positive effect on exit probabilities. While a higher exposure to the imports of intermediate goods made by wholesalers has a positive effect on employment and sales, and a negative effect on exit rates. Recognizing the potential endogeneity of local market-level indirect import penetration, I also report the results obtained by specifications that instrument the indirect import penetration. Instrumental variable estimates confirm most of the previous results