Krueger’s results from equation (K-1) are reported in Figures 1 and 2, Table 4 (columns 1 and 2), and Table 5 (column 2). His main finding is that b is estimated to be statistically significantly less than zero with virtually the same point estimate either with or without the computer dummy variable. That is, the data indicate a positive correlation between product-price increases and skill intensity. Measuring skill intensity by educational attainment yields the same result.
The result from equation (K-2) is reported in Table 5 (column 1). The main finding here is an estimated mandated rise in wage inequality: the difference between the estimated bi’s for skilled and unskilled labor is 0.52 and is statistically significant at the .0001 level. In an unreported robustness check Krueger obtains qualitatively similar results from OLS and median regressions, although the estimated rise in warranted inequality is smaller (p. 17). Thus Krueger draws the same conclusion from the results of both equation (K-1) and (K-2): “fairly robust evidence that price growth was relatively lower in less-skill intensive industries between 1989 and 1995 … The magnitude of the price changes is roughly compatible with observed wage changes for skilled and unskilled workers” (pp. 19-20). Electronic Payday Loans Online
Feenstra and Hanson (FH) (1997) also use the mandated-wage framework used by Leamer, Baldwin and Cain, and Krueger. However, FH differ from these earlier studies in an important way. They do not consider product-price changes to contain any direct evidence of the role of international trade on factor prices. Rather, building on their earlier model of outsourcing (FH (1995)) they argue that outsourcing manifests itself in industry-level data as tfp growth. Because outsourcing changes the mix of activities done within industries, “this will shift the entire production function for activities done at home, and therefore show up as a change in total factor productivity” (p. 13).