The welfare effects of trade policy are shaped by the outcomes of imports reallocation and price changes. In this paper, I show that these outcomes crucially depend on whether importing firms are matched with multinational suppliers or with single-country producers. I study an antidumping duty imposed by Colombia on the imports of Chinese truck tires. In the data I observe the full network of Colombian importers and their foreign suppliers. For the latter, I use data on tire plant's location to distinguish between multinational (manufacturing in many countries) and single-country manufacturers. Due to the policy, approximately 75% of imports of Chinese tires where replaced with imports from other origins, and the bulk of this geographical substitution involved multinational suppliers. I estimate a quantitative trade framework to match the reallocation and price changes in the data. I analyse the policy under a counterfactual network without multinational suppliers, and find that pass-through increases. The analysis suggests that ignoring this type of network structure could lead to biases for the prediction of the welfare effects of tariff -and similar- shocks.
An importer, "Company X" (left), sources tires manufactured in various countries (middle), establishing commercial links with different suppliers (right). Khumo Tire Co (pink/right) manufactures tires in China and Korea and sells both products to Company X. Other suppliers (green/right) manufacture tires in a single origin and, while many of them sell relatively little amounts, altogether they constitute an important share of Company X's purchase. This paper analyses a policy that makes Chinese tires prohibitively expensive. The findings remark on the importance of these buyer-seller links in substituting Chinese tires with Korean and Mexican, and other origins in general.
Commodity price shocks have negative consequences for developed economies that rely heavily on imported materials. Consequently, firms employ risk-management instruments to reduce their exposure. In this paper we study how the use of supply contracts by firms can shape the transmission of commodity price shocks to aggregate variables. We focus on purchase obligations, which are supply contracts with fixed prices for the delivery of goods in future periods. We rely on a novel dataset to document two empirical findings. First, we find a large exposure reduction to commodity price risk for firms using these contracts; our estimates suggest a reduction of about 27% compared with non-users. Second, sector output and labor compensation have a smaller negative correlation with commodity prices when firms trade larger contracts. We assess the aggregate quantitative role of these contracts by introducing and calibrating a tractable general equilibrium model. We measure the contribution of purchase obligations to dampening the aggregate transmission of commodity price shocks by constructing a counterfactual in which firms are not allowed to trade these contracts. Our results show that when firms engage in purchase obligations, real consumption has a relative response of 4% less to a 10% commodity price shock.
This paper explores the dynamics of Mexican exporters upon entry to multiple foreign markets. I document that exporters tend to make greater sales adjustments in their first export destination compared to subsequent destinations upon entry. I develop a model of demand learning in which knowledge can be carried over across destinations according to their market similarity. The proposed model provides an explanation for the variation in the timing of expansion to new export destinations that is observed in the data. When markets are similar, exporting to a large number of destinations can enable firms to gain a quicker understanding of foreign market conditions. However, some exporters may find such strategies to be excessively costly. Thus, they may choose to begin with a few initial destinations as a way to "test the waters" before expanding to other similar markets.