Skills gap: The timing of technical change

•This article generalizes the business cycle model in Jovanovic (2006).•I generalize the utility function and the distribution of the productivity shocks.•These assumptions are consistent with the skill-biased technical change.•Optimal technical change is time-varying.•In Jovanovic (2006) optimal technical change is constant.

This article generalizes the business cycle model in Jovanovic (2006) along two important and meaningful dimensions: (i) more general utility function; (ii) more realistic distribution properties of the productivity shocks. Unlike the original model, I assume the power utility function of the representative agent, and a non-zero expected value of the distribution of the shocks. I include the non-zero expected value of the productivity shocks to account for the skill-biased nature of the technical change in the post-war period. The model implies an endogenous time-varying technical change as an optimal investment policy, consistent with the data.