Working Papers

“Two-Sided Market Power In Firm-to-Firm Trade” (with Michele Fioretti, Ayumu Ken Kikkawa, and Monica Morlacco). July 2025. R&R (2nd round) at American Economic Review. [NBER WP 31253] [ProMarket]

This paper develops a theory of bargaining in firm-to-firm trade with two-sided market power. The framework accommodates flexible market structures, yielding analytical expressions for pair-specific markups and pass-through elasticities. In U.S. import data, we estimate strong importer bargaining power and an upward-sloping export supply curve, consistent with oligopsony power. Pass-through of the 2018 tariffs in firm-to-firm relationships is incomplete, in contrast to product-level studies, primarily due to exporter cost reductions driven by falling demand from dominant buyers. Our study highlights how bargaining and network rigidities shape price outcomes, with implications for markup dispersion and shock propagation in global value chains.


“Multinationals and Structural Transformation” (with Cheng Chen, Kei-Mu Yi, Nitya Pandalai-Nayar, Liliana Varela, and Hongyong Zhang). February 2026. R&R at Journal of Monetary Economics. [NBER WP 30494] [IDB Research for Development]

We study how multinational corporations (MNCs) shape firm-level and aggregate structural transformation. Using confidential microdata from Japan and exploiting a quasi exogenous reform that expanded foreign investment opportunities in China, we assess empirically how this reform affected employment at firms in both the host country (China) and the home country (Japan). In liberalized industries, Japanese manufacturing affiliates in China expanded employment, while parent firms in Japan shifted out of manufacturing and into higher-value service activities, including R&D. To assess the broader relevance of this mechanism, we use microdata from several advanced and middle-income economies, and show that MNCs account for the majority of the middle-income countries' reallocation to manufacturing.


“Concentration and Markups in International Trade” (with Michele Fioretti, Ayumu Ken Kikkawa, and Monica Morlacco). July 2025. [SUERF Policy Brief]

This paper derives a closed-form expression linking aggregate markups on imported inputs to concentration in a model of firm-to-firm trade with two-sided market power. Our theory extends standard oligopoly insights in two dimensions. First, it reveals that markups increase with exporter concentration and decrease with importer concentration, reflecting the balance of oligopoly and oligopsony forces. Second, it adapts  conventional market definitions to reflect rigid trading relationships, yielding new concentration measures that capture competition in firm-to-firm trade. Analysis of Colombian transaction-level import data shows these differences are key to understanding markup dynamics in international trade.


“Whose Deficit? How Ownership Shapes U.S. External Imbalances” (with Laura Alfaro and Diego Santana). November 2026.

The consequences of using tariffs to reduce the U.S. trade deficit depend on the ownership of the firms behind the trade flows. A persistent deficit is financed by what the country earns abroad, and for the United States a large part of that income is what its multinationals earn on the production they own in other countries; that income depends on where they produce, and policy can move it. We make this point in three steps. First, using four decades of harmonized BEA data, we assign U.S. goods trade to its owner. Over forty years in which U.S. multinationals expanded abroad on an unprecedented scale, their trade stayed close to balance: they carry close to 40 percent of U.S. trade but run almost none of the deficit, and their trade with their own affiliates is close to balanced with almost every partner. The deficit is concentrated in foreign multinationals' U.S. affiliates and domestic firms, while the income that offsets part of it is what U.S. multinationals earn on their affiliates abroad. Second, a model of trade and multinational production makes these trade balances equilibrium outcomes, together with the income each owner earns. Given the current account, that income is the only channel through which a policy moves the aggregate deficit: tariffs, which move production across source countries without changing the return its owners earn, barely move it, while taxes that change what owners keep, and the cost of producing abroad, do. The model also delivers two sufficient statistics: the trade a tariff destroys per dollar of bilateral deficit it closes, and the welfare cost of a policy, the standard loss from trade plus the net taxes that cross the border. Third, ownership divides policies in two. Tariffs change who runs the deficit: the 2018--19 tariffs closed the bilateral deficit with China and left the aggregate where it was, as imports reallocated to third countries; and a tariff on Mexico, which falls on U.S. firms' own supply chains, destroys 2.0 dollars of trade per dollar of bilateral deficit closed against 1.2 with China. Taxes and the cost of producing abroad change its size: the 2017 rate cut narrowed the deficit by raising what foreign owners keep of their U.S. profits, a result a model without owners of capital gets with the opposite sign, while its exemption of foreign capital widened it; and the globalization of U.S. production since 1999 widened it by four tenths of a percent of GDP, through income rather than trade. Ignoring ownership therefore mismeasures how much a policy moves the deficit, what it costs, and where the income accrues.


“Tariff Day: Impact of Recent U.S. Tariff Increases on Latin America and the Caribbean” (with Angel Espinoza). June 2025. [IDB Tariff Tracker]

This paper evaluates the global economic impact of the 2025 U.S. tariff increases under the Trump administration, focusing on Latin America and the Caribbean. Using a quantitative general equilibrium trade model based on Eaton and Kortum (2002) and Caliendo and Parro (2015), we simulate six policy scenarios. The first three capture escalating trade tensions—from unilateral U.S. tariff hikes to retaliatory measures by China and Canada. The final three reflect partial de-escalation paths, including reciprocal tariff reductions, expanded exemptions, and tariff harmonization across North America. We find that LAC was the least affected region. Mexico, exempted from the April 2 measures but subject to Section 232 tariffs, benefits from trade diversion.


“Firm-to-Firm Trade Networks: A Focus on Latin America and the Caribbean” (with Juan S. Blyde). April 2025. Submitted. [IDB-WP-1679]

Trade flows are the result of the interaction between buyers and sellers. Understanding how these firm-to-firm relationships form, survive, and evolve over time helps identify how countries can improve their export outcomes. Using a very granular dataset of United States import transactions from all countries around the world, we present a battery of static and dynamic metrics of firm-to-firm relationships. We show how Latin American exporters compare with exporters from other regions across all the metrics. We show that trade costs negatively affect the formation and duration of firm-to-firm networks. Regional differences in these costs partially explain Latin Americas performance. The study discusses a series of measures to reduce trade-related cots in the region that, in light of the results, are likely to improve the trade networks of Latin American countries.