The traditional theory of international trade, the Heckscher-Ohlin-Samuelson (HOS) theorem, suggests that a country exports those goods whose production involves relatively intensive use of factors of production with which the country is relatively well endowed. This approach does not, however, account for several aspects of the observed pattern of international trade.
In Discussion Paper No. 237, Research Fellow Patrick Minford outlines an alternative explanation of trade patterns that focuses on differences in relative labour costs. In its simplest form, the HOS approach assumes that all factors are immobile between countries. Minford assumes instead only labour immobility: all other factors are mobile. He retains the HOS assumption of constant returns to scale in production as well as the assumption that technology is identical across frontiers. Thus, with mobility of capital, resources and technology and no barriers to trade, production costs differ across frontiers only because of differing labour costs.
Minford's approach departs from the HOS approach by treating `commodities' as heterogeneous. They consist, however, of subcommodities, organized in a hierarchy from low to high `value added', which are homogeneous and whose price and cost therefore must be equated across frontiers. For simplicity he divides `subcommodities' into three types: `services', which require labour of the highest skill; `innovations', which require high technology and skilled labour; and `mass manufactures', which are standardized products requiring unskilled labour implementing well known practices.
Minford also treats labour as heterogeneous in `quality' or productivity. Each country has a labour endowment whose quality or productivity differs both on average and by industry from that of other countries. Differences in labour productivity are the fundamental explanation of observed trade patterns in Minford's approach. Differences in relative labour costs result from differences in endowments of skilled and unskilled labour. Labour supply is elastic because social support systems create a reservation wage. If the wage set by international competition is less than the reservation wage for a particular group of workers, they will all be unemployed and the industry in which they would normally work will not be viable.
The index of real wages per head (deflated by the country's consumer price index) plays an important role in Minford's model and is a measure of `competitiveness', or `terms of trade', which in turn is a determinant of net exports at constant demand levels. Countries face downward-sloping demand curves with respect to competitiveness. Because activity levels, real wages and employment depend upon the position of the demand curve for exports, the country will be tempted to shift the curve through `mercantilist' policies. Minford argues that aggregate wages will not be equalized in this model because there are different types of labour and because the benefit system creates unemployment and non-specialization. Wages of workers of the same type, if employed, will, however, tend to be equalized.

Minford tests the model by examining whether its predictions are consistent with cross-country data. The theory suggests that high-wage countries are net exporters of goods using relatively expensive labour and net importers of goods using relatively cheap labour. Minford therefore regresses net exports by industry on the average wage level in those industries, for high- and low-wage countries separately. The theory suggests a positive relationship between exports and wages for the high-wage countries and a negative relationship for the low-wage countries: these predictions are confirmed in the regression.

A Labour-Based Theory of International Trade
Patrick Minford

Discussion Paper No. 237, April 1988 (IT)