加成、生产重新配置与贸易收益

Journal of the European Economic Association · 2025 · [{"name": "Hamid Firooz", "affiliation": ["North Carolina State University", "Federal Reserve Bank of San Francisco"]}, {"name": "Gunnar Heins", "affiliation": ["University of Florida"]}]

中文摘要

摘要本文构建了一个关于国际贸易与利润转移的多部门、多国模型，将不完全竞争嵌入 Eaton 和 Kortum（2002）的李嘉图贸易模型，并允许进口与出口的加成率分布在部门和国家之间存在差异。我们首先从理论上说明，贸易自由化的收益如何取决于进口商品相对于出口商品的加成率分布。为将模型对接数据，我们估计了贸易弹性，以及针对超过 36,000 个不同部门—国家配对的一整套国家和行业特定的进口需求弹性。我们发现，出口加成率相对于进口加成率的跨国异质性，是贸易收益、尤其是关税所致福利损失的一阶决定因素。在灵活纳入加成率异质性后，对于高加成率产品的净出口国而言，这类损失最多可高达三倍。我们将模型应用于 2018—19 年中美贸易战，结果表明，一旦将加成率与利润转移纳入考虑，美国因关税战遭受的福利损失将是原来的两倍多，而中国总体上略有获益。

Abstract

Abstract This paper develops a multisector, multicountry model of international trade and profit shifting that embeds imperfect competition into Eaton and Kortum (2002)’s Ricardian trade model and allows markup distributions for both imports and exports to vary across sectors and countries. We first show theoretically how the gains from trade liberalization depend on the markup distribution for imported relative to exported goods. To bring the model to the data, we estimate both trade elasticities and a rich set of country- and industry-specific import demand elasticities for over 36,000 distinct sector-country pairs. We find that cross-country heterogeneities in export markups relative to import markups are a first-order determinant of the gains from trade and especially the welfare losses from tariffs. By flexibly taking heterogeneity in markups into account, these losses are up to three times larger for net exporters of high-markup products. We apply our model to the 2018–19 U.S.–China trade war and show that U.S. welfare losses from the tariff war are more than twice as high once markups and profit shifting are taken into account, whereas China benefited slightly overall.