Automation and labor market polarization in an evolutionary model with heterogeneous workers

The purpose of this paper is to investigate the mechanisms underlying the relationship between automation and labor market polarization. To do so, we build an agent-based model (ABM) in which workers, heterogeneous in nature and level of skills, interact endogenously on a decentralized labor market with firms producing goods requiring specific set of skills to realize the tasks necessary for the production process. The two scenarios considered, with and without automation, confirm that automation is indeed a key factor in polarizing the structure of skill demand and increasing wage inequality. This result emerges even without reverting to the routine-based technical change (RBTC) hypothesis usually found in the literature, giving some support to the complexity-based technical change (CBTC) hypothesis. Finally, we also highlight that the impact of automation on the distribution of skill demand and wage inequality is correlated with the velocity of technical change.