2 Doctoral Research
Eventually, Paul returned to Oxford for postgraduate study at Nuffield College. The subject of his doctoral thesis (the economics of customs unions, supervised by Max Corden) was motivated not by a passion for the theory of international trade, but by a perception that at that time, only a “mainstream” economics postgraduate education would give him the intellectual tools needed for a career in development economics.
The thesis led to two publications highlighting the importance of the way in which commodities are aggregated in theoretical models of customs union formation (see Collier 1979a, 1985). These papers are purely theoretical and strictly neoclassical: they introduce extra layers of complexity into the standard analysis of customs unions originally developed by Viner (1950) and Lipsey (1957), but do not stray beyond the boundaries of standard trade theory with perfectly competitive markets. In the original Vinerian theory, which was based on a model with two commodities (“imports” and “exports”), calculation of the welfare effects of customs union formation was based on the concepts of “trade creation” and “trade diversion”. Begin with a world in which each country is imposing tariffs on imports from the others. If two of the countries form a customs union, then there are aggregate welfare gains from the increased trade between them that follows from the abolition of intra-union tariffs. However, the customs union also creates a wedge between the consumer price of imports from inside the union relative to those from outside and the relative producer price. As a consequence, importers may switch from a low-cost producer outside the union to a high-cost producer inside the union, and this entails a welfare loss.Paul's main contribution was to extend the analysis to a case of more than two commodities, making it possible to analyse the reallocation of resources (and of consumption) not just between imports and exports, but also between different types of import, and between different types of export. In this case, the traditional categories of trade creation and trade diversion become redundant, and the aggregate welfare effects of customs union formation depend on the elasticities of production and consumption between different types of import (and export), not just imports as a whole and exports as a whole. The analysis of customs union formation demonstrates Paul's ability as an economic theorist, but it does not reflect the interests apparent in his later work. For example, there is no analysis of the effect of customs union formation on income distribution or poverty through Stolper-Samuelson effects.
3