This thesis is divided into three separate chapters. The first chapter, joint work with Alessandro Dovis, analyzes the effect of financial frictions on the gains from undergoing trade reform. Financial frictions may limit the ability of exporting firms to expand, or keep new firms from becoming exporters in response to a trade liberalization. The way that financial frictions are modeled crucially determines whether or not this happens. Empirical evidence from a trade liberalization shows that capital flows to new exporters efficiently in response to the liberalization, even though those firms are still financially constrained. This suggests that financial market inefficiency is not a barrier to realizing gains from international trade. The second chapter, joint with Pau Pujolas, examines contemporary and historical international trade data, and establishes new patterns of trade between countries of differing income levels. Existing trade models cannot simultaneously rationalize high intra-industry trade between countries of similar income levels, and low intra-industry trade between countries of different income levels. A model of non-homothetic preferences is consistent with these facts, and implies higher gains from trade than standard models. The third chapter considers the problem of a government that wants to use immigration policy as a source of fiscal revenue. Governments can allow highly productive immigrants to enter and then tax them to finance spending on their native population. Optimal policy is considered in the context of a fixed native population and pool of potential immigrants. Optimal immigration policy follows a cutoff rule in the productivity of the entering immigrant. When agents are privately informed about their own productivity, immigrants are taxed at a higher rate than natives and the effect on the native population of increased fiscal transfers may be very large.