Perceptions of Local Organized Criminality

In an online survey of 4000 Italian residents, we study how perceptions of the presence of organized crime at the local level influence citizens’ preferences for being tough on crime and their propensity to run for local office. We start by measuring respondents’ beliefs about the presence of organized crime in the town where they live. We then provide half of respondents – randomly selected – with accurate information on the presence of organized criminality in the province where they live, relative to other Italian provinces. More precisely, we rank provinces using an index capturing the presence of organized criminality developed by De Luca (2025) et al. and based on crime data reported by the Italian Statistical Agency (ISTAT) and the Italian Ministry of Interior, and we group them into different quintiles of the distribution. We inform respondents whether the presence of organized criminality in their province is very low, low, average, high or very high relative to other Italian provinces based on this index – for provinces in the 1st, 2nd, 3rd, 4th, and 5th quintile respectively. We measure respondents’ perceptions of the presence of organized criminality in their province, to check for the effect of the treatment on beliefs. We then elicit respondents’ preferences for making punishment for organized crime harsher and for investing more private and public resources to fight organized crime. We also elicit their propensity to run for mayor, should their preferred political party ask them to do so.

2025-08-05

2025-09-30

Respondents’ beliefs about the presence of organized criminality in their town and province. Respondents’ beliefs about whether punishments for organized crime currently in place are adequate to act as a deterrent for criminals and to provide justice to the victims of organized crime. Respondents’ preferences for making punishment for organized crime harsher. Respondents’ preferences for spending more public money on fighting organized crime. Respondents’ willingness to pay in terms of higher income taxes for fighting organized crime. Respondents’ preferences for increasing oversight on public works to avoid infiltration by organized criminality. Respondents’ willingness to accept delays in public works to eliminate the risk of involving companies that have ties with organized crime. Respondents’ willingness to run for mayor of their town or of another town, should their preferred party ask them to do so.

Respondents’ beliefs and preferences are measured using 7-points likert scales ranging from “absolutely not/absolutely disagree” to “absolutely yes/absolutely agree”. We may transform these answers into dummy variables when we do the treatment effect analysis to improve power and interpretability.

Respondents’ wiliness to pay for fighting organized crime is measured using a slider going from 0% to 20% that they can use to select the higher income tax rate that they are willing to accept to increase public spending on fighting organized criminality. Respondents’ willingness to accept delays in public works to eliminate the risk of involving companies that have ties with organized crime is measured using a slider going from 0% to 20% more time to complete public works on average.

In addition to the standard treatment effect analysis on all the variables listed above, we also plan to explore the heterogeneity in the treatment effects by respondents’ socio-economic characteristics (such as gender, education, age), and crucially location – in particular, comparing location at lower and higher risk of organized criminality infiltration.

We provide half of respondents in our sample – randomly selected – with accurate information on the presence of organized criminality in the province where they live, relative to other Italian provinces. More precisely, we rank provinces using an index capturing the presence of organized criminality developed by De Luca (2025) et al. and based on crime data reported by the Italian Statistical Agency (ISTAT) and the Italian Ministry of Interior, and we group them into different quintiles of the distribution. We inform respondents whether the presence of organized criminality in their province is very low, low, average, high or very high relative to other Italian provinces based on this index – for provinces in the 1st, 2nd, 3rd, 4th, and 5th quintile respectively.

Respondents are randomly assigned to the control or treatment group as they take the survey.

Individual respondent.

No

4,000 respondents

4,000 respondents, age between 18 and 75

We are going to split respondents equally among the control and the treatment group. Hence, we expect to have about 2,000 respondents in each treatment arm.