UC San Diego

This dissertation contains four chapters of my research. In chapter 1, I show that higher military spending leads to higher exports of weapons. This is a manifestation of the home market effect - the prediction that countries with higher demand for a differentiated good will be net exporters of that good. The home market effect is specific to monopolistic competition models of international trade, and serves to distinguish empirically between these and comparative advantage-based models. I construct a monopolistic competition model with a military sector and a continuum of civilian industries, and derive empirical implications of the home market effect for arms and ammunition. I then use military expenditure as a measure of demand for military weapons to show that, indeed, countries with higher military spending as a share of GDP export more arms and ammunition relative to homogeneous, cheap-to-ship civilian goods. In my setup, military spending serves to introduce variation in demand across countries. This is an innovation over the typical approach in the literature, whereby consumers in different countries are assumed to have identical preferences, and home market effects stem from differences in goods' characteristics and country size alone. Chapter 2 tests the existence of the home market effect for construction materials, by using public infrastructure spending as a measure of demand. I construct a theoretical model that suggests goods with high transport costs and high differentiation are most likely to display home market effects. I test this prediction empirically for a handful of construction materials that meet the necessary criteria. As expected, I find that the home market effect holds for alloy steel and construction machinery. However, cement and glass display the opposite trade pattern, whereby increased domestic demand leads to reduced exports. I discuss potential explanations for this result. In the next two chapters, I examine the determinants of firm success. In chapter 3 (co-authored with Sarada) we find that new firms with higher network concentration, i.e. wherein initial employees have worked together previously, are on average larger, have higher wages and survive longer. This association increases with the initial size of the newly founded firm. However, we find a negative relationship between network concentration and initial firm growth. Finally, chapter 4 (co-authored with Marc Muendler and James Rauch) gauges the prevalence and performance of firms founded as employee spinoffs relative to other entrants. We find that size at entry is larger for employee spinoffs than for new firms without parents but smaller than for diversification ventures of existing firms. Similarly, survival rates for employee spinoffs are higher than for new firms without parents and comparable to those for diversification ventures of existing firms