Using different data sources from local labour markets (LLMs) in Italy between 1971 and 2011, we document a number of stylized facts: a) local differences in the ratios of private employment to population are highly persistent; b) the population has a limited reaction to labour demand shocks, consistent with the high rigidity of nominal wages and pro-cyclical variations in rents, which absorb the gains (losses) from higher (lower) employment rates; c) labour demand shocks are fairly persistent and unevenly distributed, to the detriment of those areas that were already lagging behind and boosting the more advanced ones; d) shocks are amplified by the non-linear employment adjustment, which reacts more to negative shocks than to positive ones. The estimated reactions to shocks are then used to perform policy-motivated simulations. We find that allowing greater population reactions is a superior policy option. Had Italy experienced the population reactivity of the US, local disparities would have been significantly less, to the same extent as with a sizeable public intervention in areas lagging behind.
Published in 2019 in: Regional Science and Urban Economics, v. 75, pp. 1-21
