All FSI Projects

An Impact Evaluation of the Pacifying Police Units (UPPs)

photo 10 police2
photo 10 police2
Pacifying Police Unit officers at favela da Rocinha, Rio de Janeiro, Brazil.
Photo credit: Vanessa Melo

Researchers

Beatriz Magaloni
Co-principle Investigator
Senior Fellow, Associate Professor
Edgar Franco Vivanco
Co-principle Investigator
Vanessa Melo
Project Manager
Project Manager and Research Associate
Veriene Melo
Investigator

BACKGROUND

Since 2008, Rio de Janeiro has implemented a new public security policy called the “Pacification”, a police strategy with full support from the Federal government that aims to improve the overall levels of security in the city and retake areas previously dominated by criminal organizations. Based on this new model of policing - that takes an approach on community policing initiatives – “Pacifying Police Units (UPPs)” are implemented in different poor communities in the city (shanty towns). In efforts to improve the entire police force based on this new approach, through the results of our research project, Rio’s Secretary of Security hopes to further understand the causes of police violence and abuse in the city. The goal is to assess the role of the Pacification in reducing criminality and violence levels in the city of Rio de Janeiro, shedding light to the importance of investing on efforts for police reform. Finally, through this research, we hope to add to the existent literature on criminal violence, police violence, and policereform initiatives in violent and impoverished communities and urban areas in international contexts.

PARTNERS

In collaboration with the Military Police of Rio (PMERJ) and the Secretariat of Public Security (SESEG), Stanford’s International Crime and Violence Lab is working to evaluate how the UPPs impact violence and police use of deathly force.

METHODS 

Our team has collected and geo-referenced all the homicides as well as civilian deaths by police action in the metropolitan area of Rio de Janeiro from 2005 to 2014. We identified the location of more than the 30,000 violent deaths taking place during this period.[1] This data is unique in the context of developing countries because it allows to identify with high accuracy where and when each homicide took place, thus reducing problems of spatial aggregation normally present in other measures of homicide counts. Geo-referenced murders will allow us to better understand how geographic, socioeconomic, and institutional factors have shaped homicides and deaths by police action in the recent history of Rio. Further, another advantage of geo-referenced murders is that it will allow us to better isolate the effect of the introduction of the UPP on violence.

STATUS

One of the challenges of any study seeking to evaluate the UPPs is that these have been assigned to favelas where violence and the use of police lethal force have traditionally been very high. Geo-referenced deaths will allow us to better address problems of identification than other studies by comparing similarly violent favelas, some with UPPs and without UPPs. The study also allows us to measure whether the UPPs have helped contain violence in the intervened favelas, the extent to which violence and criminal activity has spread to other areas of the city, and when and how that contamination has taken place. Moreover, using data at this level of precision is useful to identify heterogeneous responses, that is, if some favelas have been more successful than others in reducing violence.

In addition, our team has conducted a series of extensive and detailed qualitative interviews (including interviews with police commanders) in different UPP units as well as regular battalions within the metropolitan region of Rio - some of them located in areas with the highest levels of homicides in the city. Based on the results from these initiatives, the Stanford International Crime and Violence Lab will provide critical feedback to the Military Police of Rio de Janeiro and the Secretariat of Security to adjust their interventions to better protect the liberties and rights of citizens the favelas and to reduce violence in the city.

[1] Around 16,000 taking place in the municipality of Rio de Janeiro and around 17,000 taking place in the adjacent area of Baixada Fluminense.

Contact

Vanessa Melo