Technical Progress and Early Retirement

This paper claims that technical progress induces early retirement of older workers. It presents a model where human capital is technology specific, so that technical progress erodes some existing human capital. This affects mostly older workers, who have a smaller incentive to learn the new technology, since their career horizon is shorter. Hence, their labor supply declines. We find strong support to this erosion effect in HRS data, which shows that labor of older workers is negatively related to technical progress in their sectors. Unlike the effect across sectors, the model is ambiguous about the aggregate effect of technical progress on labor supply of older workers. While in sectors with many innovations it falls due to the erosion effect, in other sectors it increases due to higher wages. To examine which effect dominates we run a time series test using US data and find that the effect of average technical progress on aggregate labor force participation by the old is negative. Namely, the erosion effect dominates. JEL Specification: J24, J26, O15, O33.