There is widespread concern about the toll of the pandemic on local economies, but little causal evidence to assess its real costs. This column presents an impact evaluation of the local economic effects of the COVID-19 crisis in Italy, based on a counterfactual application of machine learning algorithms. It documents that, to date, impacts on employment and firms have been dramatically uneven across the Italian territory and spatially uncorrelated with the epidemiological pattern of the first wave. It shows that this heterogeneity is associated with sectoral specialisation, exposure to social aggregation risks, and pre-existing labour market fragilities. Finally, it argues that such diverging local trajectories call for a place-based approach in the policy response to the crisis.