Technical change, wage inequality, and optimal taxes in an assignment model

This paper studies income inequality and optimal taxation policies in a talent‐to‐task assignment model of self‐selection. Our model considers relative capital‐skill complementarities across tasks, leading to the polarization of capital and technology by task complexity, which in turn drives the polarization of job and wage growth by talent levels. Regarding optimal tax policy, the wage compression channel remains effective through the trickle‐down effect of subsidizing high‐wage earners and taxing low‐wage earners. Yet, the wage compression channel via capital, corporate, and R&D taxes, aimed at reducing wage inequality, does not operate via a trickle‐down effect. Instead, it works by taxing capital income and R&D investments in high‐task‐complexity sectors while subsidizing those in low‐task‐complex sectors. Moreover, we identify a Pigouvian effect that arises to address spillovers, which modifies the marginal tax rates on labor income, capital income, firm profits, and R&D investments.