Can't SBTC explain the U.S. wage inequality dynamics?

Based on the effect of skill-biased technology change (SBTC), this paper builds a search model with heterogeneous firms and workers to explain the dynamics of the wage inequality in the U.S. from 1963-2005. Firms differ in capital intensity (technology content) of the job created and workers differ in their education level. As the match-specific productivity is stochastic, the productivity threshold of employment of each education-job pair matched is endogenously determined. The advance in skill-biased technology increases the productivity of the highly educated workers as well as the capital intensity facing firms when creating high-tech jobs. We argue that in response to the rise in the capital intensity, high-tech firms increase the productivity thresholds of hiring, which leads to wage increment in the high-tech sector, and thus widening the residual wage inequality. Meanwhile, the increase in the productivity of the highly educated worker in the high-tech sector results in higher education premium in both the high-tech and low-tech sectors. Using the historical U.S. data, calibration shows that SBTC can explain the general trends in the education premium and the residual wage inequality from 1963-2005. In particular, it solves the puzzle why the education premium fell but the residual wage inequality grew in the early 1970s.