The service sector has accounted for virtually all U.S. job growth since 1972, employing 70 percent of the labor force in 1986 and 77 percent in 1990 (Carey and Franklin, 1991). However, international trade's demographic effects on service sector employment remain uncertain.
This paper examines the demographic and industrial characteristics of trade-sensitive service industries.(1) Trade sensitivity is the degree to which the service industries experience net trade-related employment gains or losses. Estimating net trade-related service employment involves input-output simulations of the total (direct and indirect) labor inputs required to satisfy a hypothetical change in the level of net exports. The analysis here assumes that this change equals the ratio of 1985 to 1977 export and import values. It uses nominal values in order to reveal domestic employment effects of export and import price changes over time. However, one should interpret the findings within the context of input-output analysis limitations and assumptions. For instance, the assumption of fixed technological coefficients requires one to interpret long-term policy implications with care.
BEA: Bureau of Economic Analysis
BLS: Bureau of Labor Statistics
CPS: Current Population Survey
MOFA: Majority-owned Foreign Affiliates
PPI: Producer Price Index
II. TRADE IN SERVICES AND DOMESTIC EMPLOYMENT
Nusbaumer (1987) and Herman and van Holst (1981) express skepticism about traditional trade theory's applicability to services. However, Deardorff (1985), Hindley and Smith (1984), Oulton (1984), and Sapir and Lutz (1980) suggest that normative conclusions from international trade theory are equally valid for both goods and services. However, researchers have not fully identified the determinants of comparative advantage (Richardson, 1987). Quantitative studies suggest that while differences in physical and human capital endowments explain trade patterns between developed and developing countries (Sapir and Lutz, 1980), comparative advantage among developed countries appears to be industry-specific (Oulton, 1984), ranging from differences in skill-intensity to disparities in natural resource endowments.
Skill-intensity or knowledge is a crucial component of most service activities (Nusbaumer, 1987; Gibbs, 1986). Nevertheless, since the bulk of U.S. trade in services is with similarly endowed developed countries (Whichard, 1988), asymmetries other than skill-intensity and relative factor endowments apparently influence comparative advantage in services.(2) Such asymmetries may reflect imperfections in the labor, product, or capital markets and differences in technology (Goto, 1990).
Making a priori predictions about the characteristics of service industries that would benefit from trade is difficult. Predicting the incidence of service trade on the labor market status of women and minorities is equally difficult in the absence of explicit information on the determinants of service sector comparative advantage.
Input-output analysis provides an appropriate framework for identifying the service industries that trade affects based on trade-related employment performance. Applying this framework to Current Population Survey (CPS) data reveals these industries' demographic and industrial characteristics.
One can disaggregate service trade's employment effects into direct and indirect components. Direct effects arise from initial changes in industry output and employment levels required to satisfy a unit change in foreign procurement of domestic services--e.g., a U.S. firm's providing international marketing services to a Bolivian leather manufacturer. The indirect effects, on the other hand, arise from the interdependent nature of the economy's productive systems. A change in output level in one service producing activity in turn can induce corresponding changes in other sectors of the economy, including the non-service producing sector. …