Skill Upgrading and Imports in US Manufacturing

The majority of skill-upgrading in the United States in the 1980s and early 1990s was due to within industry shifts in skill-intensity. Traditional trade theory, represented by the HeckscherOhlin model, predicts that trade affects relative employment primarily by shifting resources across industries. For this reason, many economists have concluded that international trade has played a minor role in skill-upgrading in the US. However, more recent theoretical models show that international trade can induce within-industry skill-upgrading by raising R&D intensity and creating skill-biased technologies. This paper tests this assertion by estimating the effect of rising import shares on the change in wage shares for 4-digit SIC US manufacturing industries. Taking long-differences on the data, I find that changes in the import share are positively correlated with changes in the non-production wage share between 1980 and 1988. Rising imports can explain roughly half of the average increase in skill-intensity for this time period. However, imports do not appear to have a significant impact on skill-intensity between 1989 and 1996. The results hold up to several robustness checks. JEL Codes: F14, F16, J21, J30