When Robots Do (Not) Enhance Job Quality: The Role of Innovation Regimes

Abstract

Whether robots have a positive or negative impact on job quality and wages depends on the dominant innovation regime in an industry. In an innovation regime with a high cumulativeness of knowledge, i.e. if accumulation of (tacit) knowledge from experience (embodied by workers) is important for innovation, robots enhance the probability that workers will get permanent (other than temporary) contracts and they earn higher wages. The opposite holds for industries with a low-cumulativeness regime when innovation depends mainly on general (and generally available) knowledge. Our results emerge from multi-level estimates of two countries (Italy and Germany), combining sectoral data on robot use with person-level data on properties of workers. Our results imply that previous studies tended to find weak effects of robotization as they did not control for innovation regimes. An implication for European industrial policy is that the hiring of more flexible personnel (and shorter job tenures) that has become popular in the period of supply-side economics is likely to have a negative impact on the productive use of robot technology in industries with a high cumulativeness of knowledge, and less so in low-cumulativeness industries. Unqualified pleas for labour market deregulation can have a problematic impact on technology and should be reconsidered.

Keywords: robots, quality of work, innovation regimes, knowledge cumulativeness

JEL Classification: J3, J5, M5, O3