- 18 - 3 Empirical literature on the US and Europe This section is organized as follows: First the empirical findings for US and Europe about the effects of trade on relative employment and wages are discussed based on both trade theoretical and labor economics approaches (Sections 3.1.1and 3.1.2). Then the empirical findings regarding the outsourcing and FDI effects are discussed with a particular emphasis on the debate on the effects of eastern enlargement for the EU (Section 3.2). Finally the empirical political economy literature on the changes in the labor’s share in income is reviewed. 3.1 Effects of trade on sectoral or relative employment and wages 3.1.1 US: evidence and methodological debates The discussion on trade effects in the last one and a half decade is dominated on whether rising trade with low wage countries has been an important factor in the job losses in certain industries as well as rising inequality in the wage share of skilled vs. unskilled labor in the US. Landesmann et al. (2001) report that the increase in import penetration took off much earlier in the US than in Europe; therefore it is no surprise that much of the initial research focused on the US. In the US the wages of the low skilled workers have not only declined dramatically during the 1980s and 1990s, but also decreased in absolute terms (Feenstra and Hanson, 1996, 1999). The interaction between the effects of skill-biased technical change and trade with low wage countries, and more recently international outsourcing have been explored to account for this change. Most researchers have concluded some negative effect of trade on both aggregate employment and wages, and inequality between skilled vs. unskilled labor, but the magnitudes change and differ across sectors.8 Based on factor content analysis for the US, Sachs and Shatz (1994) find that trade had a significant but small effect on the skill premium and job losses for the unskilled9. Borjas et al (1992) find some negative trade effect for high school dropouts. Revenga (1992) finds that the decline in import prices causes a decline in employment and wages in the trade impacted industries.10 On the other hand, the trade-theoretical critique of these studies, which focus on trade as an explanation for skill premium, have argued that one has to build the link through relative price changes, however the observed changes in import prices and volumes, or relative changes in the price of skill-intensive vs. unskilled-intensive products have not been sufficient to explain the large changes in relative wages, or they are inconsistent with the expectations of Stolper-Samuelson theorem (e.g. Krugman, 1995; Lawrence and Slaughter, 1993). Lawrence and Slaughter (1993) argue that technological change is the main reason for decline in the relative wage of the 8 See Greenaway and Nelson, 2001 for a review. 9 They show that increasing import shares resulted in 7% job-loss among manufacturing workers and 2% among non-manufacturing workers. They estimate that 40% of the difference in employment growth during 1950-78 and employment decline during 1978-90 can be attributed to trade. 10 Based on reduced form regression method, the effect of import prices (interacted with import share) on employment, average weekly hours and wages of production workers in the US over the period of 1977-1987 for a panel of manufacturing industries is estimated. It is found that a 10% reduction in import prices is associated with 2.5% to 4% decline in employment and 0.5% to 1% in wages in the trade impacted industries.