| dc.description.abstract | This dissertation investigates the relationship between firm-to-firm interactions, firm performance and trade flows. In the first chapter, I examine the diffusion of productivity gains among publicly-listed companies in the United States through buyer-supplier relationships. To do so, I develop an approach to estimating total factor productivity that accounts for firm interdependence through spillovers. My procedure flexibly incorporates models of network effects into standard production function estimation approaches. I find substantial variation in intra- and inter-sectoral spillovers, and my results highlight the importance of large, central firms to US productivity growth. The next chapter investigates the impact of US multinational enterprises (MNEs) on competitors, buyers and suppliers in their home country. In contrast to existing research that relies on industry-level measures of exposure to MNEs, I exploit variation in firm-specific interactions with MNEs through horizontal and vertical linkages. I find positive spillovers from MNEs on their competitors and buyers, and negative spillovers on their suppliers. For the average domestic firm, however, the combined effect of all three kinds of interactions with MNEs is modest and negative. In the final chapter, I investigate how import tariffs imposed in 2018 during the trade war between the US and China affected their trade relationships with importers and exporters in Mexico and Colombia. Using transactional customs data, I find no evidence of trade diversion or deflection to Mexico or Colombia due to the trade war. This dissertation demonstrates how the examination of firm networks can enrich our understanding of firm performance and the impact of international trade policy. | |
