Development-related biases in factor productivities and the HOV model of trade

Past empirical failures of the basic Heckscher-Ohlin-Vanek (HOV) model related to the inability of data to meet its restrictive assumptions, particularly identical international technologies and factor price equalization. Trefler (1993) tried to resuscitate HOV by introducing a simple Hicks-neutral (HN) factor-productivity adjustment, an approach that was heavily criticized. In this paper, we re-examine the productivity question by estimating factor-specific productivities from the individual technology data of multiple countries. Using a dataset of 29 countries, both developed and developing, we find evidence of factor-augmenting technological differences. In particular, the factor-productivity adjustment works well for developed members of the OECD. Further, we find that the ratios of factor productivities are strongly correlated with corresponding factor endowments. This systematic bias implies that the ability of HOV to explain North-South factor trade depends both on relative factor abundance and factor-augmenting productivity gaps.