Essays on International Trade

I combine micro-level data and structural models to study the interaction between trade, innovation, and human capital. In the first chapter, I examine how China's expansion of college education since 1999 affects innovation and exports' skill content. This policy change is interesting because of its sizable scale: the annual quota on the number of newly admitted college students increased from 1 million in 1998 to 7 million in the 2010s; as a result, the number of college-educated workers more than tripled between 2000-2015. I develop a two-country spatial equilibrium model, featuring skill intensity differences across industries and heterogeneous firms' innovation and exporting choices. I highlight three main channels at work, including: (1) with an increasing number of college-educated workers, China shifts production to more skill-intensive industries and converts the excess supply of skill-intensive goods into exports; (2) the growing supply of college-educated workers promotes innovation; and (3) as skill-intensive industries tend to be more innovative, trade-induced industry reallocation reinforces the innovation surge. I empirically validate my model mechanisms about how the college expansion affects innovation and exports, exploiting differential supply shocks of college-educated workers across regions due to historical college endowments. Using the calibrated model, I find that China's college expansion explained 40-70% of increases in China's manufacturing R&D intensity between 2003-2018 and triggered export skill upgrading. I also find that trade openness amplified the impact of this education policy change on China's innovation and production.In Chapter 2, co-authored with Chen Liu, we build a multi-sector spatial general equilibrium model to account for China's export surge between 1990 and 2005. We focus on the role of the reductions in tariffs and internal migration costs during that period. Our model generates a closed-form aggregate trade elasticity that can be decomposed into four margins of adjustments. Two are the commonly studied intensive and extensive margins of exports. The remaining two margins are the new-firm margin and the export-regime margin, for which we have found empirical support and used our reduced-form evidence to discipline the structural parameters. We find that reductions in China's tariffs accounted for 21% of China's export growth in the 1990-2005 period, whereas reductions in migration barriers accounted for another 8%. We also find firms' location switches are important: in the absence of firms' relocation, the portion of China's export surge explained by the three policies combined would drop from 29% to 16%.In Chapter 3, co-authored with Alejandro Nakab, we study how exporting shapes experience-wage profiles. Using detailed Brazilian employer-employee and customs data, we document that workers' experience-wage profiles are steeper in exporters than in non-exporters. Aside from the self-selection of firms with higher returns to experience into exporting, we show that workers' experience-wage profiles are steeper when firms export to high-income destinations. We propose that this destination-specific effect is likely driven by faster human capital accumulation with exposure to high-income destinations