This paper explores the link between trade structure, trade specialization and per capita incomegrowth. It is argued that industrial upgrading in export specialization patterns has a positive long-rungrowth effect, while the effect of structural change in industrial import patterns is in principleambiguous. A standard empirical growth model is augmented by various measures of structuralchange. The hypothesis that not trade per se matters, but that various types of trading activitiesimpact differently on economic growth is tested on a sample of 45 countries (OECD members andselected Asian and Latin American countries) over the period 1981-1997. The data set comprisesexports and imports for 35 manufacturing industries at the 3-digit level of the ISIC classificationwhich are grouped according to skill intensity. The results of the dynamic panel estimation pointtowards a positive long-run growth effect arising from trade specialization in medium-high-skill-intensive industries. Further, important distinctions between the skill intensity of export and importpatterns and their respective influence on economic development, as well as between the group ofdeveloping countries and OECD members are observed in this relationship.
trade structure Balassa specialization index economic growth spillovers