No abstract

This research explores theoretically, empirically and quantitatively the role of age diversity in determining aggregate productivity and output. Age diversity has two conflicting effects on output. On the one hand, due to skill complementarity across different cohorts, age diversity may be beneficial. On the other hand, rapid skill-biased technological change makes age diversity costly as up-to-date education tends to be concentrated among younger cohorts. To study this trade-off, I first build an overlapping-generations (OLG) model which, in view of these two opposing forces, predicts a hump-shaped relationship between age diversity and GDP per capita. This prediction is established analytically, and also quantitatively using real-world population data in an extended computational version of the model. The prediction is then tested using country-level panel data with a novel instrument, and regional data from Europe. Moving one standard deviation closer to the optimal level of age diversity is associated with a 1.5% increase in GDP per capita. In addition, consistent with the predictions of the model, the optimal level of age diversity is lower in economies where skill-biased technological change is more prevalent.