Offshoring causes economic inequality

We bring our theory to data to test the model’s key predictions. First, our novel prediction from the model is that offshoring has two countervailing effects, complementary effect and substitution effect, on labor markets. We use data on U.S. BEA multinationals and U.S. BLS employment, and link them 45 to U.S. ACS individual-level information, and distinguish individuals by occupation, workers, and managers. Then, exploiting industry-year variation of offshoring measure, we have found compelling evidence that offshoring has differential impacts on workers and managers. We have also found that offshoring generates within task inequality using conditional quantile regression technique. Furthermore, we quantify the relative contribution of offshoring to the widening inequality and find that offshoring explains about 12 percent to 21 percent of the rising inequality during the period 2002 and 2011 in the U.S. Lastly, we use data on U.S. BEA multinationals and Barro and Lee (2013) educational attainment, and set up an offshoring gravity equation to test an inverted U-shape relation between the volume of offshoring and skill overlap between two countries. Using Poisson pseudo-maximumlikelihood (PPML) estimation technique, we have confirmed the inverted U-shape relation, and this explains why the U.S. is shifting most of its production activities to countries such as Brazil, China, Mexico and India whose skill distributions have intermediate skill overlaps with the U.S.