Corruption Red Flags in Public Procurement: New Evidence from Italian Calls for Tenders

Abstract

This paper contributes to the analysis of quantitative indicators (i.e., red flags or screens) to detect corruption in public procurement. Expanding the set of commonly discussed indicators in the literature to new ones derived from the operating practices of police forces and the judiciary, this paper verifies the presence of these red flags in a sample of Italian awarding procedures for roadwork contracts in the period 2009-2015. Then, it validates the efficacy of the indicators through measures of direct corruption risks (judiciary cases and police investigations for corruption-related crimes) and indirect corruption risks (delays and cost overruns). From a policy perspective, our analysis shows that the most effective red flags in detecting corruption risks are those related to discretionary mechanisms for selecting private contractors (such as the most economically advantageous offer or negotiated procedures), compliance with the minimum time limit for the submission of tenders and subcontracting. Moreover, our analysis suggests that greater standardization in the call for tender documents can contribute to reducing corruption risks. From a methodological point of view, the paper highlights the relevance of prediction approaches based on machine learning methods (especially the random forests algorithm) for validating a large set of indicators.

Gotham city. Predicting ‘corrupted’ municipalities with machine learning

“…Gallego et al (2021) use a large micro dataset with more than 2 million public contracts to investigate the potential of ML to track and prevent corruption episodes in public procurement in Colombia and understand its main drivers. Finally, Decarolis and Giorgiantonio (2020) use three machine learning routines, namely LASSO, ridge regression and random forest, on data concerning the procurement of public works to predict indicators of corruption risk, showing the potential of such algorithms in detecting corruption in public procurement.…”

“…Apart from their procurement focus,Gallego et al (2021) andDecarolis and Giorgiantonio (2020) are micro-level analyses not aimed at forecasting the spatial distribution of white-collar crime occurrences. In contrast, our explicit aim is to provide an ML model able to map corruption episodes and thus help police investigations on the ground.3 In particular, Durham Constabulary has employed a risk assessment tool, constructed using random forests, to predict the risk of reoffending and used to decide whether some individuals should be prosecuted or not(Oswald et al, 2018).4 Toward an AI strategy in Mexico.…”

The extra‐legal governance of corruption: Tracing the organization of corruption in public procurement

“…Empirically, our article adds to the fast‐growing literature on measuring corruption through risk indicators (Fazekas et al, 2018), showing how “red flags” can detect institutionalized forms of corruption, such as the ones involving mafia‐like organizations on the local level. Methodologically, our article adds to the small methodological literature which uses supervised machine learning methods for identifying and validating proxy indicators for corruption (Decarolis & Giorgiantonio, 2020). Finally, from a policy perspective, while our results are specific to mafia‐like infiltration in Italy, we offer pointers at how they can be applied more broadly, assessing mafia‐like EGO presence in local public procurement across Europe.…”

“…To assess the impact of mafia-like groups' governance on public procurement, the empirical analysis relies on data from public procurement over the period 2008-2014 compiled by the Italian Anticorruption Authority (ANAC), and it exploits exogenous variation in law enforcement: the dissolution of local city councils under suspicion of being infiltrated by mafia-like groups (Minister of Interior data). Using a host of indicators such as the number of bidders, we compare traditional regression methods with tree-based machine learning algorithms (for a similar methodological approach see: Decarolis and Giorgiantonio (2020)).…”

Rules, Discretion, and Corruption in Procurement: Evidence from Italian Government Contracting

“…Decarolis and Giorgiantonio [2019] analyze the universe of court sentences for corruption in public auctions finding that only 2% of the firms awarded public contracts were thus implicated. In the same set of auctions, our measure flags 17% of contract winners as potentially criminal (note thatDecarolis and Giorgiantonio [2019] use a smaller and different set of auctions than the one used in our paper).…”