Content area
Full text
Since the late 1970's, the wages of lessskilled U.S. workers have fallen dramatically, both in real terms and relative to the wages of more-skilled U.S. workers. There is considerable debate over whether international trade has contributed to the declining economic fortunes of the less skilled (Richard Freeman, 1995; J. David Richardson, 1995). Skeptics about the role of globalization in U.S. wage changes claim that technological innovation is a more plausible explanation for the shift toward skilled labor. While technology is surely an important contributing factor to rising wage inequality in the United States, the current trade-versus-technology debate obscures a more fundamental question about how firms respond to import competition and how these responses, in turn, are transmitted to the labor market.
In previous work, we have argued that outsourcing, by which we mean the import of intermediate inputs by domestic firms, has contributed to an increase in the relative demand for skilled labor in the United States (Feenstra and Hanson, 1996). This feature of globalization-the fragmentation of production into discrete activities which are then allocated across countries-has received little attention in the literature. If firms respond to import competition from low-wage countries by moving non-skill-intensive activities abroad, then trade will shift employment toward skilled workers within industries. Most previous studies of the labor-market effects of trade presume that import competition shifts resources across industries, without changing their internal composition. Ignoring outsourcing misses an important channel through which trade affects the demand for labor of different skill types.
In this paper, we extend our previous work by incorporating new data on manufactured imports. The revised National Bureau of Economic Research (NBER) trade data base (Feenstra, 1996) contains import data for U.S. manufacturing industries over the period 1972-1994. We combine the import data with disaggregated data on input purchases from the Census of Manufactures to construct industry-by-industry estimates of outsourcing for the period 1972-1992. We then reexamine whether outsourcing has contributed to an increase in relative demand for skilled labor. I. Data
As described in Feenstra (1996), the import data were computed by starting with the most disaggregated import figures collected at the border by the U.S. Census for the years 19721994. These data were summed according to import-based SIC (MSIC) at the four-digit level. This classification...