Customs union formation: Theory and evidence from a public choice perspective

This dissertation consists of three essays that examine the political economy of customs unions. If the selection of trading regime is endogenous, then what would cause a government to abandon a restricted trade, status quo for a bilateral or multilateral trading agreement? The answer lies not in the movement toward free trade, but in the relaxation of a political resource constraint that faces policy makers.;It is the purpose of the first essay to design a model that describes the choice problem facing a government that is considering forming a multilateral trading agreement. The formation of a customs union is politically feasible when the member countries together have market power in an export or import good, but individually do not have that power. The formation of the customs union makes a policy feasible that was once infeasible--an optimal tariff.;The second essay addresses lobbying and the associated rent-seeking behavior and their role in regime selection. When the ability of the government to supply protectionist policies is limited constitutionally, as is the case in a customs union, the resources that were once devoted to persuasion are now devoted to production. This essay will show that the reallocation of these resources to productive activities is Pareto improving for both interests groups and policy makers. In this case, customs unions provide a constitutional limitation to inefficient rent seeking that is beneficial to both voters and policy makers.;The third essay, "Self Interest, Constituent Interest, and the Customs Union Issue: Evidence from the 1787 Constitutional Convention," continues this theme of policy responses to exogenous changes in economic conditions. Using a combination of the voting records from Constitutional Convention in Philadelphia and biographical data on the delegates in attendance, I find that voting over the selection of trading regime, a customs union, is consistent with the predictions of standard Vinerian customs union theory. Using a simultaneous mixed-probit estimation I find that the addition of personal economic characteristics as explanatory variables fails to explain as much about voting behavior over trade policy as do constituent characteristics.