Marco Letta (Sapienza University of Rome)
Augusto Cerqua, Costanza Giannantoni and Gabriele Pinto
Despite increasingly stringent regulations, there has been a concerning stagnation in reducing workplace fatalities recently. Can place-based targeting help? By coupling machine learning techniques with comprehensive data from Italy, we develop a placebased approach to workplace fatalities. Harnessing accurate forecasts, we construct a granular risk map and compare it to the allocation of on-site inspections and public subsidies for occupational safety, uncovering limited overlap. Counterfactual estimates reveal that current measures are effective only in areas flagged as high-risk by ex-ante machine predictions. AI-powered territorial targeting can reduce the incidence of this chronic issue while lowering the costs of policy implementation.