Working Papers



[1] Markups, International Specialization an the Gains from Trade (2013, last updated August 2014)

This paper emphasizes the role of markups in international specialization. By accounting for across product/industry differences in markups, I develop a unified theory that tractably combines modern features of North-South and North-North trade into one model. Additionally, the unified theory delivers an alternative explanation for the "Washington Apples" effect -- a well-established fact regarding intra-industry specialization. In the unified framework, North-North trade is governed by national product differentiation. North-South trade is governed by a new channel of across-product specialization that has been overlooked in the literature. In equilibrium, rich countries specialize in highly differentiated, high markup products, while poor countries specialize in less differentiated, low markup products. The predicted patterns are supported by detailed trade data. I estimate the structural parameters of the model to quantify the gains from trade across rich and poor countries. Embedding the new channel of across-product specialization into a standard trade model magnifies the gains from trade by around 200% for the average country. Despite trading less, poor countries experience the largest gains from global trade.


[2] Weight as an Iceberg Trade Cost (2013, last updated November 2014)

Trade theory relies heavily on the iceberg (ad-valorem) trade cost assumption. Empirical studies, however, have generally rejected the iceberg specification. These studies assume independence between quality (unit value) and weight. By exploiting detailed import/export data, I directly test this restriction. My results reveal a positive and significant relationship between weight and quality within narrowly defined product categories. Incorporating this relationship, I find strong support for iceberg shipping costs in the US import data. Specifically, for every 10% increase in unit-value, shipping cost per item increases by 9.5% -- more than 80% of the effect is driven by the positive relationship between quality and weight. Remarkably, high-price items are exported to more remote locations, despite being heavier and paying proportionally higher shipping costs. An across-industry analysis reveals that the majority of this effect is due to markups.


Work in Progress



[1] Additive Trade Costs in a Ricardian Model of Trade: a Quantitative Analysis (with Konstantin Kucheryavyy)

[2] Learning and Firm Size Distribution: "The lost cause of Industrial subsidies in Iran" (With Farid Farrokhi)

[3] Hit and Run Exporters (with Gustavo GudiƱo)