Generalizations of HOV Theory

Abstract

The model derived in this section is based on the one developed previously. It allows for non-neutral technological differences in the parameters G, the direct factor cost shares, across countries. The technological differences are non-neutral, as the adjustment of the direct factor cost shares within each country is allowed to be different across factors. The model hinges on the observation that, even after considering the differences in factor prices, w, and unit factor requirements, R, the available data show, for some countries, important departures from Ɵ computed for a reference country, typically the US. This would be, for example, the consequence of different access of countries to technologies. The previous model can be quite easily transformed if we assume that the matrix Ɵ is modified in a non-neutral way and differently across countries. In Section 2.2 we had equation (2.13) for a particular country:

$$WR = \theta P$$

where Ɵ was invariant across countries. Now, we assume that Ɵ is a country-specific matrix of effective direct factor cost shares, following the adjustment for non-neutral technological differences in the parameters. Remember that in Section 2.2 we defined the factor price equalization in the presence of different factor productivities and different factor prices in such a way that the latter exactly reflected the differences in the former across countries.