Measuring spatial patterns with the procurement index of corruption shape

This study proposes an approach for measuring the spatial configuration of corruption risk in public procurement across regions, moving beyond traditional incidence-based indicators to capture how corruption risk is spatially distributed and clustered. Using a novel municipal-level public procurement dataset for Italy, we apply Local Indicators of Spatial Association (LISA) and develop the procurement index of corruption shape (PICS). The empirical application to the Italian case illustrates the analytical properties and interpretability of the index. Results reveal heterogeneous spatial configurations across Italian regions. Northern regions display concentrated high-risk clusters, consistent with localised and more visible rent-seeking behaviour. In contrast, southern regions exhibit widespread low-risk clusters, indicative of more systemic and less detectable forms of corruption, often linked to pre-bidding collusion and cartelisation. The paper’s main contribution is methodological as it introduces a spatial dimension into corruption risk measurement that can be adapted to different institutional contexts. Empirically, it provides granular sub-national evidence and, theoretically, it suggests that corruption adapts to institutional contexts and may coexist with economic performance. Our findings highlight the importance of spatially and context-specific anti-corruption policies and have broader relevance for countries characterised by regional disparities.