April 2025
JEL Classification: C53, C63, H79, K42
Keywords: Organized crime, Municipalities, Mafia infiltration, Financial statements, Machine learning.
This study investigates the risk of criminal infiltration in Italian local administrations by analysing municipalities' financial data from 2016 to 2021 using statistical and machine learning techniques. If compared with a control group, the budgets of city councils being dissolved for mafia infiltration exhibit higher operating costs, lower spending on education and social services, increased budget rigidity, and misallocation of funds toward sectors like construction and waste management. These municipalities also display lower revenue collection efficiency, likely due to decreased administrative capability. The study also develops a machine learning algorithm to assess the risk of infiltration in Italian municipal administrations. The risk indicator developed in this paper can contribute to the design of more effective anti-mafia policies.
