DENUNCIATION, CRIME AND PUNISHMENT THE CYCLE OF VIOLENCE IN THE CITY OF

SECURITY AND CITIZENSHIP DENUNCIATION, CRIME AND PUNISHMENT THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO

Rio de Janeiro, 2016 FGV/DAPP

Director Marco Aurelio Ruediger

DAPP +55 (21) 37994300 | www.dapp.fgv.br | [email protected]

TEAM

IMPLEMENTATION TEAM Coordination Marco Aurelio Ruediger

Research Coordination Roberta Novis Maria Isabel Couto

Researchers Rachel Bastos Lucas Silva Janaina Fernandes Miguel Orrillo Danielle Sanches

Graphic project Rebeca Liberatori Braga

FGV is an institution with a philanthropic, educational, technical and scientific nature, created on 20 December 1944 as a private entity, whose aim is to act in a broad manner in all questions with a scientific nature, with an emphasis on the social sciences, administration, law, and economy, contributing to the socio- economic development of the country.

Head Office Praia de 190, Rio de Janeiro | RJ, CEP 22250-900 ou Caixa Postal 62.591 CEP 22257-970 | Tel: (21) 3799-5498 | www.fgv.br

Founding President Luiz Simões Lopes

President Carlos Ivan Simonsen Leal

Vice-Presidents Sergio Franklin Quintella, Francisco Oswaldo Neves Dornelles e Marcos Cintra Cavalcante de Albuquerque SECURITY AND CITIZENSHIP

DENUNCIATION, CRIME AND PUNISHMENT THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO OVERVIEW

This new study reveals a geography of urban criminality in the city of Rio, showing the spatial reality of the cycle of violence. In this sense, the cycle of violence is understood as the different moments analyzed between denunciation, crime, prison and conditional release. Therefore is possible to point to clear and strategic public policies for the different moments in this cycle of violence.

Specifically, for the municipality of Rio de Janeiro, this study reveals areas in which the city government should implement support policies for public security in addition to state level decisions. The debate and the establishment of a municipal public security agenda demand actions by the municipal secretariats in priority areas for the prevention of violence through social actions, urban maintenance and planning, and especially the use of technology and intelligence.

In this sense, based on a territorial analysis of this cycle of violence, it is clear that the City Government must have an active role in a connected public security policy, which is transversal to other secretariats and which is equipped with result monitoring tools. Therefore, public choices in the municipality can be supported by an alignment between public resources and efficiency in strategic and intelligent public security.

Municipalities are important agents in the public security agenda to the extent that they are called to support efforts to implement public security with funding, public services and, increasingly, by providing municipal guards. In this scenario, FGV/DAPP believes that municipalities should assume an active role in the definition of directives for public security policies in their territories jointly with the state administration.

5 | DEPARTMENT OF PUBLIC POLICY ANALYSIS DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 6 HOW CAN THE MUNICIPALITY PLAY A ROLE IN PUBLIC SECURITY? Principal considerations of this analysis

Elaboration of local plans to combat violence: Bangu Policies for preventing crime and the social (in absolute terms) and (in relative terms) reinsertion of those leaving the penitentiary stand out in all stages of the cycle of violence. As a system: Cidade de Deus stands out, both in absolute result, it is therefore justified to elaborate local terms and in rates per thousand inhabitants, in the plans to confront violence, prepared with the aim of concentration of the original self-declared residence preventing the development of criminality, to repress of prisoners and their destination when given it when necessary, and mitigate its effects. These plans parole. Although it is not the only neighborhood could involve the strengthening of their respective where these two stages are important, it is the administrative regions and the coordination of efforts only one where only these two stages stand out. in other areas, such as: education, health, transport, Furthermore, analyses of the 2010 Census indicate urbanism, housing, public order, social work, and a low intermediate level of social development in employability and income. the neighborhood, as well as an average household income which does not reach a minimum monthly wage. For this reason, it is a neighborhood where Deployment of the Municipal Guard in areas where qualification and employability policies could have a there is a significant perception of insecurity and significant impact on security in the city. a high concentration of crimes, especially against property: The Municipal Guard (GM) can complement the conspicuous policing work of the Polícia Militar, Use of data for planning and identification of even without firearms. Centro and Madureira, successful practices: Planning based on daily neighborhoods which stood out in the spatial analysis information from the city can help to replicate of the distribution of crimes against property, in successful practices to contain criminality. For absolute terms and weighted by population, could example, the neighborhood of Grajaú is the territory benefit significantly from preventive action by with the highest incidence of crimes which led to the Municipal Guard, in the sense of occupying imprisonment in the period studied, but it does not the streets to reduce the opportunities to become stand out as a neighborhood where reported crimes involved in small thefts or burglaries. Anchieta and are concentrated. This can point to the greater Guadalupe also stand out as neighborhoods where effectiveness of police investigation in the locality, this type of action by the GM could significantly resulting in a high rate of completed investigations impact on the feeling of security. and judgements.

Creation of a unified public security database: Standardization and obligatory fields are recommendable in crime reports and records, in criminal records and forms, as well as in any information system in the institutions which are part of the public security system.

7 | DEPARTMENT OF PUBLIC POLICY ANALYSIS DENUNCIATION, CRIME, AND PUNISHMENT (IN ABSOLUTE NUMBERS) Data obtained between 1st January and 31th July 2015

In absolute numbers, Bangu had the highest incidence of crime in the cycle of violence analyzed.

The following maps present the numbers, in absolute terms, of what we call the cycle of violence in the municipality of Rio de Janeiro between 1st January and 31th July 2015. Based on this we can build an interpretation of perceptions of insecurity which lead the population to take part in the fight against violence (whose proxy in this study is Disque-Denúncia); about the spatial distribution of reported crimes; about the neighborhoods of origin of those imprisoned after having engaged in illegal activities; and about the neighborhoods to which those who leave the prison system on parole return.

It is important to make a remark here about the data used. The crime category does not incorporate all crimes committed in Rio de Janeiro, rather specific types of crimes reported in police stations. In the methodological notes of this research, we explain in a detailed manner the choice of crimes. However, in general terms we selected types of crimes what had an acknowledged impact on the perception of security and the quality of life of the population. These are: murder, robbery followed by death, assault followed by death, rape, apprehensions of drugs and weapons, and various types of robbery (of passersby, mobile Dephones,núnci anda in groups).

Denunciation

No value 1 - 150 151 - 500 501 - 1000 1001 - 2000 Above 2000 sem valor 1-150 151-500 501-1000 1001-2000 acima de 2000 Source: Disque-Denúncia RJ between 1st January and 31th July 2015 Crime Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 8

1-50 51-200 201-500 501-1500 acima de 1500 Castigo

sem valor 1-50 51-100 101-200 201-600 acima de 600 Condicional

sem valor 1-50 51-100 101-200 201-600 acima de 600 Denúncia

Denúncia

sem valor 1-150 151-500 501-1000 1001-2000 acima de 2000 Crime sem valor 1-150 151-500 501-1000 1001-2000 acima de 2000 Crime Crime

1 - 51 51 - 200 201 - 500 501 - 1500 Above 1500 1-50 51-200 201-500 501-1500 acima de 1500

Source: ISP between 1st January and 31th July 2015 Castigo Prepared by: FGV/DAPP 1-50 51-200 201-500 501-1500 acima de 1500 Castigo

Punishment

sem valor 1-50 51-100 101-200 201-600 acima de 600

CondNoi valueciona1l - 50 51 - 100 101 - 200 201 - 600 Above 600 sem valor 1-50 51-100 101-200 201-600 acima de 600 Source: SEAP between 1st January and 31th July 2015 Condicional Prepared by: FGV/DAPP

9 | DEPARTMENT OF PUBLIC POLICY ANALYSIS

sem valor 1-50 51-100 101-200 201-600 acima de 600

sem valor 1-50 51-100 101-200 201-600 acima de 600 Denúncia

sem valor 1-150 151-500 501-1000 1001-2000 acima de 2000 Crime

1-50 51-200 201-500 501-1500 acima de 1500 Castigo

sem valor 1-50 51-100 101-200 201-600 acima de 600 Condicional

Parole

No value 1 - 50 51 - 100 101 - 200 201 - 600 Above 600 sem valor 1-50 51-100 101-200 201-600 acima de 600 Source: SEAP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

The previous maps call attention for the concentration in Bangu of all the stages of this cycle of violence, considering absolute terms. This indicates a strong perception of insecurity which is compatible with rates of crime, as well as that this neighborhood produces — and there return to it — a significant quantity of individuals institutionalized in the prison system. Bangu, therefore, in relation to the diversity of issues linked to security, is presented as a priority neighborhood for the intervention of public authorities in all stages: preventive policies, conspicuous policing policies, and social reinsertion policies. Although not standing out as much as Bangu, other neighborhoods are also of importance in this sense, namely: Campo Grande, Santa Cruz, , Centro, Bonsucesso, , , and Copacabana.

On the other hand, the map highlights the neighborhoods which stand out from the others in relation to the universe of denunciations and crimes (Anchieta and Guadalupe)— or just crimes (Irajá, Madureira, , , Taquara, Cascadura, Botafogo, Bento Ribeiro, , Coelho Neto, and Brás de Pina). In these cases, investment in conspicuous policing policies and urbanistic crime prevention policies should be a priority, and probably more efficient in relation to the reduction of violence.

There is also the case of the neighborhood of Jacarepaguá, which has a high incidence of denunciations, of the original residence of the imprisoned populations, and of those leaving the penitentiary system on parole, but which does not stand out in relation to crimes reported in police stations. In addition, there exist cases of neighborhoods which stand out in the concentration of the origin of the imprisoned population and/or people on parole, but in none of the other stages of the cycle of violence — Senador Camará, , Padre Miguel, , Inhaúma, Jacaré, , and Cidade de Deus. In these cases, given the lack of resources, the characteristics of the cycle of criminality of the neighborhoods in question points to the need for the prioritization of social insertion and reinsertion policies— such as education, training, and employability — instead of just conspicuous policing policies.

Finally, there are also the cases of São Cristóvão and Penha, which have significantly high rates of crime, original residence of prisoners, and those leaving the penitentiary system on parole.

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 10 DENUNCIATION, CRIME AND PUNISHMENT (PER THOUSAND INHABITANTS)

Data obtained between 1st January and 31th July 2015

Bonsucesso is the neighborhood in relative numbers, per thousand inhabitants, with the highest incidence in the cycle of violence;

Centro and Cidade Nova stand out in relative terms in the stages of denunciation and crime.

As important as spatial perspectives of the cycle of violence in absolute terms is the analysis weighted by population. While absolute numbers help to identify the most affected neighborhoods, the incidence per thousand inhabitants helps to identify tendencies which lie outside the normality of determined phenomena. In this sense, absolute numbers are a good measure for understanding a problem which affects the population of that neighborhood as a whole, while the incidence per thousand inhabitants shows the chances which each individual from that neighborhood would have in general of encountering some of the phenomena studied, Densimplyúnc byia being (po inr thatmi lspace. hab Theita followingntes) maps show the same incidence as the previous maps, weighted by the population of each neighborhood.

Denunciation (per thousand inhabitants)

No value Até 10 10,01 a 15 15,01 a 25 25,01 a 40 Above 40 *** Nota: Cálculo não realizado para o bairro da Lapa por ausência de dados sobre população. Ver detalhes nas notas metodológicas. sem valor até 10 10,01 a 15 15,01 a 25 25,01 a 40 acima de 40 ***

Source: Disque-Denúncia RJ between 1st January and 31th July 2015 Crime (por mil habitantes) Prepared by: FGV/DAPP

11 | DEPARTMENT OF PUBLIC POLICY ANALYSIS

até 5 5,01 a 15 15,01 a 30 30,01 a 60 acima de 60 *** Castigo (por mil habitantes)

sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20 *** Condicional (por mil habitantes)

*** sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20 Denúncia (por mil habitantes)

Denúncia (por mil habitantes)

sem valor até 10 10,01 a 15 15,01 a 25 25,01 a 40 acima de 40 *** Crime (por mil habitantes)

sem valor até 10 10,01 a 15 15,01 a 25 25,01 a 40 acima de 40 *** Crime (per thousand inhabitants) Crime (por mil habitantes)

Until 5 5,01 a 15 15,01 a 30 30,01 a 60 Above 60 *** Note: Calculation did not include Lapa neighborhoodaté 5 due5 to,01 absence a 15 of population15,01 a 30 data. See30 ,details01 a 60 in methodologicalacima de 6 0notes ***

Castigo (por mil habitantes) Source: ISP between 1st January and 31th July 2015 Prepared by: FGV/DAPP até 5 5,01 a 15 15,01 a 30 30,01 a 60 acima de 60 *** Castigo (por mil habitantes)

Punishment (per thousand inhabitants)

sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20 *** Condicional (por mil habitantes)

No value Until 1 1,01 a 4 4,01 a 10 10,01 a 20 Above 20 *** sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20 *** Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes

Condicional (por mil habitantes) Source: Disque-Denúncia RJ between 1st January and 31th July 2015 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 12

*** sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20

*** sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20 Denúncia (por mil habitantes)

sem valor até 10 10,01 a 15 15,01 a 25 25,01 a 40 acima de 40 *** Crime (por mil habitantes)

até 5 5,01 a 15 15,01 a 30 30,01 a 60 acima de 60 *** Castigo (por mil habitantes)

sem valor até 1 1,01 a 4 4,01 a 10 10,01 a 20 acima de 20 *** Condicional (por mil habitantes)

Parole (per thousand inhabitants)

No value Until 1 1,01 a 4 4,01 a 10 10,01 a 20 Above 20 *** Note: Calculation did not include Lapa** neighborhood* sem va duelor to absenceaté 1 of population1,01 a 4 data. See4,0 details1 a 10 in methodological10,01 a 20 notesacima de 20

Source: SEAP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

With this weighting significant differences can be noted. Bangu, which previously had been outstanding in all areas, is now outstanding in none. Only Bonsucesso and Cidade Nova have outstanding rates in all stages of the cycle of violence. Centro stands out among denunciations and crimes, but not among the residence of convicts nor among the residence of those on parole. Jacaré and Saúde stand out both among denunciations and the residence of prisoners and those on parole, but not among reported crimes. Cidade de Deus and Mangueira appear among the neighborhoods with the highest rates of self-declared residence of prisoners and those on parole.

On the other hand, some neighborhoods have significant rates of crimes: , Campos dos Afonsos, Campinho, and Cidade Universitária. Finally, there are also neighborhoods which have significant rates of denunciations, but not of the other stages of the cycle of violence: Anchieta, , Guadalupe, Costa Barros, and Cocotá.

NEIGHBORHOOD ATTRIBUTES

A wide-ranging and integrated public security policy requires not only a profound comprehension of the territorial dispersal of the cycle of violence, but also the understanding of a set of socio-economic characteristics of neighborhoods where the preventative and social reinsertion policies should be implemented. Below, we present some examples of these characteristics.

13 | DEPARTMENT OF PUBLIC POLICY ANALYSIS Índice de desenvolvimento social (2010)

Social development index (2010)

0 a 0,400 0,401 a 0,500 0,501 a 0,600 0,601 a 0,700 0,701 a 1 *** *** 0 a 0,400 0,401 a 0,500 0,501 a 0,600 0,601 a 0,700 0,701 a 1 Note: Calculation did not include Lapa* neighborhood** 0 a 0, 4due00 to absence0,40 1of a population0,500 data.0,5 0See1 a details0,600 in methodological0,601 a 0,70 notes0 0,701 a 1 Rendimento domiciliar per capita em salários mínimos (2010)

Household income per capita in minimum monthly wages (2010)

0 a 0,400 0,401 a 0,500 0,501 a 0,600 *** 0,6010 a-1 0,700 1-20,701 2a -15 ***5-10 Mais que 10 Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes Postos formais de trabalho (2014) Source: IPP/IBGE 2010 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 14

sem informação 1-1.000 1.001-10.000 10.001 a 50.000 50.001 a 100.000 acima de 100.000 Índice de desenvolvimento social (2010)

*** 0 a 0,400 0,401 a 0,500 0,501 a 0,600 0,601 a 0,700 0,701 a 1 Rendimento domiciliar per capita em salários mínimos (2010)

Formal Labor Positions (2014)

0 - 1 1 - 2 2 - 5 5 - 10 Above 10 *** Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes *** 0-1 1-2 2-5 5-10 Mais que 10

Source: MTE (RAIS) 2014 Postos formais de trabalho (2014) Prepared by: FGV/DAPP

A rapid comparison between this map of socio-economic characteristics of neighborhoods and the map of the cycle of violence in absolute terms shows a significant concentration of all stages of this cycle in the areas which have lower levels of social development and lower per capita household income. This concentration is even stronger when we talk about the self-declared residence of people who are imprisoned or on parole.

In this sense, the comparison between the cycle of violence and the attributes of the neighborhoods cited here points to the central role which the Carioca municipal executive can assume in the public security agenda. The Social Development Index constructed by the Pereira Passos Institute (IPP), for example, aggregates information about urban conditions in each neighborhood, literacy, and the income of residents. Municipalities are responsible for acting in the following three areas: guaranteeing the expansion of urban infrastructure to guarantee access to water, sewage, and garbage collection for all residents; universalizing access to basic education; and making the economy dynamic throughout the city in order to include the economically active population. Here, what calls attention is the case of Bangu – first in absolute terms in the stages of denunciation, crime, residence of prisoners, and residence of those on parole -, which has low/intermediate levels of social development and formal labor positions, and low levels of average household income. The neighborhood could therefore be made a privileged space for urbanization policies, policies aimed at stimulating economic dynamization, and income complementation policies.

sem informação 1-1.000 1.001-10.000 10.001 a 50.000 50.001 a 100.000 acima de 100.000

15 | DEPARTMENT OF PUBLIC POLICY ANALYSIS WHAT THE DENUNCIATIONS OF CRIMES SAY IN THE CITY OF RIO DE JANEIRO? Data obtained from anonymous denunciations received by Disque Denúncia RJ between 1st January and 31th July 2015

After analyzing the general spatial cycle of violence, we decided to present some other characteristics of each stage, which can help understand them.

Below we show the themes of the universe of denunciations dealt with here — which also follow the more general tendencies from January 2015 to March 2016 as demonstrated in a previous FGV/DAPP study. In the first figure there are the more general categories which encompass the denunciations, while the second figure refers to the activities directly denounced — which, as can be seen, can be either a crime or an infraction of municipal bye-laws.

Observed jointly, these highlight for us the most important concerns which mobilize the population to become engaged in the denunciation of crimes are related to the drug trade, armed violence, and crimes against property.

ClaTypessse of denunciation (general category in which it fits)

NARCOTICS 13006

CRIMES AGAINST PROPERTY 9278

FIRE ARMS AND WEAPONS 8540

CRIMES AGAINST DE ADMINISTRATION 5454

DISTURBING THE PEACE 4003

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000

Type of denunciation (specific subject of denunciation)

DRUG TRAFFICKING 11096

ILLEGAL POSSESSION OF FIREARMS 7642

NOISE 3517

OBSTRUCTION OF PUBLIC ROADS 1800

ROBBERY / BURGLARY OF PASSERSBY 1663

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000

Source: Disque-Denúncia RJ between 1st January and 31th July 2015 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 16 AN ATTENTIVE LOOK AT CRIMES IN THE CITY OF RIO DE JANEIRO Data obtained from the Institute of Public Security between 1st January and 31th July 2015

In the maps of the cycle of violence, we present the data about the reporting of crimes by the ISP as a whole. However, we know that there exist specifities in 35.851 the public security planning. Crimes against life possess different characteristics and tendencies from crimes Crimes against property against property. For this we will focus on the universe of crimes analyzed3 here.5.851 63.50.38451 Crimes against property ACprpimreehse angsaiionnsst opfr odpruegrtsy and firearms 63.50.38541 16..602314 CrimesACpripmresh eaagainstngsaiionnsts popropertyrfo dpreurgtsy and firearms ApprehensionsRAappperse rheepnosriotends ooff ddrugsrugs aandnd ffirearmsirearms 61.063241 11..761211 RapesARappperes hrreportedepnosirotnesd of drugs and firearms CrimesCRraipmeess r aeagaistgpaoirntse tdlife life Crimes reported 1.67211 1.711 Crimes against life 2321 Crimes against life Rapes reported 5918

116 8455 1.711 1621 Crimes against life 1603 94 14 Crimes reported 3602

21473

Apprehension of drugs Apprehension of firearms Rape Murder Robbery followed by death Assault followed by death Robbery of passerby Robbery of mobile phones Robbery of vehicles Robbery in public transportation Apprehension of Drugs Apprehension of firearms Rape Murder Robbery followed by death Assault followed by death Source: ISP between 1st January and 31th July 2015 Robbery of Passersby Robbery of Mobile Phones Robbery of Vehicles Robbery in Public Transportation Prepared by: FGV/DAPP

17 | DEPARTMENT OF PUBLIC POLICY ANALYSIS Apreensões de armas Followinge d thero tendencyga ofs denunciations (ab sin theo samelu period,to the) registrations of crimes which most stand out are those related to crimes against property and drug use and dealing. In relation to these, the spatial analysis of crime registrations identifies the prominence of Centro, both in absolute terms and in rates weighted by neighborhood population. Bangu is also of importance, but only in the absolute accumulation of occurrences.

In addition to Centro, Jacaré, Bonsucesso, Cidade de Deus and São Cristóvão also stand out in terms of the registration of apprehension of drugs and weapons, both in absolute terms and in the incidence per thousand inhabitants. In relation to the registration of crimes against property, only Centro and Madureira lead the ranking of neighborhoods in absolute terms and in rates per thousand inhabitants.

Finally, the map of crimes against life registered in the same period follows tendencies relatively different from the rest. In absolute terms, Bangu and Centro are still prominent, together with Santa Cruz, Campo Grande, and Realengo — thereby placing an accumulation of crimes against life in the Zona Oeste of the city. Nevertheless, when the incidence per thousand inhabitants is counted, the axis of crimes against life is partially altered with Grumari and Gardênia Azul, in Zona Oeste, and Zumbi and in Zona Norte being of most importance

Apprehensions of firearms and drugs (absolute numbers)

No value 1 - 10 11 - 50 51 - 100 101 - 500 Above 500

Source: ISP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 18 sem informação 1-10 11-50 51-100 101-500 acima de 500 Crimes contra o patrimônio (absoluto)

sem informação 1-150 151-500 501-1000 1001-2000 acima de 2000 Crimes contra a vida (absoluto)

sem informação 1-10 11-20 21-40 41-100 acima de 100 Apreensões de armas e drogas (absoluto)

Apreensão de armas e drogas (por mil habitantes)

Apprehensions of firearms and drugs (rate per thousand inhabitants)

sem informação 1-10 11-50 51-100 101-500 acima de 500 Crimes contra o patrimônio (absoluto)

No value Until 1 1,01 a 2 2,01 a 4 4,01 a 10 Above 10 *** Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes

Source: ISP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

Crimes against property (absolute numbers)

sem informação até 1 1,01 a 2 2,01 a 4 4,01 a 10 acima de 10 *** Crimes contra o patrimônio (por mil habitantes)

No Value 1 - 150 150 - 500 501 - 1000 1001 - 2000 Above 2000

Source: ISP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

19 | DEPARTMENT OF PUBLIC POLICY ANALYSIS sem informação 1-150 151-500 501-1000 1001-2000 acima de 2000 Crimes contra a vida (absoluto)

sem informação até 5 5,01 a 10 10,01 a 20 20,01 a 50 acima de 50 *** Crimes contra a vida (por mil habitantes)

sem informação 1-10 11-20 21-40 41-100 acima de 100

sem informação até 0,50 0,51 a 1 1,01 a 1,40 1,41 a 10 acima de 10 *** Apreensões de armas e drogas (absoluto)

Apreensão de armas e drogas (por mil habitantes)

sem informação 1-10 11-50 51-100 101-500 acima de 500 Crimes contra o patrimônio (absoluto)

sem informação até 1 1,01 a 2 2,01 a 4 4,01 a 10 acima de 10 *** Crimes contra o patrimônio (por mil habitantes)

Crimes against property (rate per thousand inhabitants)

sem informação 1-150 151-500 501-1000 1001-2000 acima de 2000 Crimes contra a vida (absoluto)

No value Until 5 5,01 a 10 10,01 a 20 20,01 a 50 Above 50 *** Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes

Source: ISP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

Crimes against life (absolute numbers)

sem informação até 5 5,01 a 10 10,01 a 20 20,01 a 50 acima de 50 *** Crimes contra a vida (por mil habitantes)

No value 1 - 10 11 - 50 51 - 100 101 - 500 Above 500

Source: ISP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 20 sem informação 1-10 11-20 21-40 41-100 acima de 100

sem informação até 0,50 0,51 a 1 1,01 a 1,40 1,41 a 10 acima de 10 *** Apreensão de armas e drogas (por mil habitantes)

sem informação até 1 1,01 a 2 2,01 a 4 4,01 a 10 acima de 10 *** Crimes contra o patrimônio (por mil habitantes)

sem informação até 5 5,01 a 10 10,01 a 20 20,01 a 50 acima de 50 *** Crimes contra a vida (por mil habitantes)

Crimes against life (rate per thousand inhabitants)

No value Until 0,50 0,51 a 1 1,01 a 1,40 1,41 a 10 Above 510 *** Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes

Source: ISP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

There is also another level of information about the occurrence of crimes which has to be highlighted. Using the sources and data available, we can compare the places where crimes were reported and where crimes were committed by the imprisoned population. It is immediately noticeable that the neighborhood which was most identified as the territory of the crimes of those imprisoned, Grajaú, does not stand out at all in relation to the quantity of records of criminal occurrences, while Realengo, Anchieta, Coelho Neto, Irajá, Cascadura, and Brás de Pina, which are in the second category of neighborhoods which most report criminal occurrences do not appear among the neighborhoods of the crimes of those imprisoned.

In the weighted analysis per population, once again Grajaú has a leading place in the category ofthe sem informaneighborhoodção whereaté crimes0,50 which resulted0,51 in aimprisonment 1 1 were,01 committed, a 1,40 but which1, is4 1not a of1 importance0 acima de 10 *** in the weighted incidence of crime reported by police stations in the city. On the other hand, Grumari, Campos dos Afonsos, and Cidade Universitária, which have significantly high rates of reported crimes, do not have crime reports leading to imprisonment in the same period.

The number of crimes which led to arrest in Grajaú can be related to a Polícia Militar project in that neighborhood, launched in February 2015, called Integrated Neighborhood Police Company, which may have significantly increased arrests of people caught in the act in the region.

21 | DEPARTMENT OF PUBLIC POLICY ANALYSIS DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 22 crimes_ISP_absoluto

CRIMES REPORTED IN POLICE STATIONS X CRIMES WHICH LED TO IMPRISONMENT

Crimes ISP (absolute numbers) crimes_SEAP_absoluto

1 - 50 51 - 200 201 - 500 501 - 1500 Above 1500

Source: ISP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

Crimes SEAP (absolute numbers)

1-50 51-200 201-500 501-1500 acima de 1500 crimes_ISP_taxa (mil habitantes)

No value 1 - 300 301 - 1000 1001 - 2000 2001 - 5000 Above 5000

Source: SEAP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

23 | DEPARTMENT OF PUBLIC POLICY ANALYSIS

sem valor 1-300 301-1000 1001-2000 2001-5000 acima de 5000 crimes_SEAP_taxa (mil habitantes)

até 5 5,01 a 15 15,01 a 30 30,01 a 60 acima de 60 ***

sem valor 0,01 a 10 10,01 a 20 20,01 a 50 50,01 a 100 acima de 100 crimes_ISP_absoluto

crimes_SEAP_absoluto

1-50 51-200 201-500 501-1500 acima de 1500 crimes_ISP_taxa (mil habitantes)

Crimes ISP (rate per thousand inhabitants)

sem valor 1-300 301-1000 1001-2000 2001-5000 acima de 5000 crimes_SEAP_taxa (mil habitantes)

Until 5 5,01 a 15 15,01 a 30 30 a 60 Above 60 *** Note: Calculation did not include Lapa neighborhood due to absence of population data. See details in methodological notes

Source: ISP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

Crimes SEAP (rate per thousand inhabitants)

até 5 5,01 a 15 15,01 a 30 30,01 a 60 acima de 60 ***

No value 0,01 a 10 10,01 a 20 20,01 a 50 50,01 a 100 Above 100

Source: SEAP between 1st January and 31th July 2015; IBGE (Censo) 2010 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 24

sem valor 0,01 a 10 10,01 a 20 20,01 a 50 50,01 a 100 acima de 100 THE UNIVERSE OF THOSE DEPRIVED OF THEIR LIBERTY IN RIO DE JANEIRO CITY Prison population data for Rio de Janeiro city between 1st January and 31th July 2015

The following figures show the crimes which led to arrests of people in Rio de Janeiro city between 1st January and 31th July 2015. Presented in the first figure was the population deprived of their liberty during that period and in the second the population who obtained parole. As we can note, in both cases what stand out are the so-called crimes against property and crimes associated with the drug trade. The only difference is in the presence of murder as the fifth most committed crime by the imprisoned population, which does not appear among the five most committed crimes by those on parole — it is replaced by burglary.

Most practiced crimes which led to detention in the municipality

14.243

14.243 10.401

10.401 19.992

19.992 4.474

4.474

EImprisonedncarcerados

4.195 4.195

Robbery TrafficRoubo sim ofple snarcoticsTráfico deAggraveted entorpecentes robberyRoub o qualIlligalificado possessionPosse ou pofor tae ilfirearmegal de arm awith de fo grestrictedo de uso res truseito HMurderomicídio simples

Source: SEAP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

25 | DEPARTMENT OF PUBLIC POLICY ANALYSIS 2335

2335 2023

5222 10.401

1269 5222 1269

ConParoledicional 1190 1190

Roubo simples Tráfico de entorpecentes Roubo qualificado Posse ou porte ilegal de arma de fogo de uso restrito Furto simples Robbery Traffic of narcotics Aggraveted robbery Illigal possession of a firearm with restricted use Burglary

Source: SEAP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

Finally, we present below some characteristics of the population on parole in the period studied, since we believe that it can help to construct specific policies aimed at preventing violence and social reinsertion for this group.

Average profile of convicts on parole

2222 a yearsnos old 5,6% dos apenados em liberdade condicional no período estudado tinham 22 anos - compondo a idade de maior representatividade naquele m5.6%omento . A of m é convictsdia de toda a p onopul aç paroleão, no en ta innto , theera d e period30 anos. studied were 22 years of age – 2composing2 anos the most representative age at that moment. However, the average 5,6% dos apenados em liberdade condicional no período estudado tinham 22 anos - compondo a idade de maior representatividade naquele m1ofo8m ethe naton. Apopulationo msédia de toda a p opwasulaç ão30, n o yearsentanto , eold.ra de 30 anos. 124,25% adans poessoas em situação de liberdade condicional no período estudado haviam entrado no sistema com 18 anos. A média de entrada no sistema, por sua vez, era de 25 anos. 152,682% d aaonsn aopoesnsados em liberdade condicional no período estudado tinham 22 anos - compondo a idade de maior representatividade naquele m18ome nyearsto. A média oldde toda a população, no entanto, era de 30 anos. 14,5% das pessoas em situação de liberdade condicional no período estudado haviam entrado no sistema com 18 anos. A média de entrada no sP5is,6tae%mr da,o psoo arp seunaa vdeozs, eerma dlieb e2r5d andoes c.ondicional no período estudado tinham 22 anos - compondo a idade de maior representatividade naquele m214.5%o2me naton. A oof msé dthoseia de toda on a po pparoleulação, no ehadntant oentered, era de 30 an theos. prison system at 18 years of 416,83 % a auntodoecslararam-se pardos. 25age.,62% dao Thesn apoens aaveragedos em liberd entranceade condiciona lto no ptheeríod osystem, estudado ti ninha mturn, 22 ano swas - com p25ond oyears. a idade de maior representatividade naquele P1m4oa,5mr%edn dtaos. Ape msséodaias deem t soidtuaa aç pãop duel aliçbãeor,d naod en ctoanndtoic, ieornaa dl en o3 p0e arníoodso. estudado haviam entrado no sistema com 18 anos. A média de entrada no 51,68% daosn apoensados em liberdade condicional no período estudado tinham 22 anos - compondo a idade de maior representatividade naquele F4si6su,t3e %nm aad,u ptaodrm esculaae rvaenrza,mt ea-rsael d peia n2rd5co sao.noms. pleto m124o,2m5%e na dtaons. Apo emsésodaias deem t osidtau aaç pãop duela lçibãeor, dnaod en ctoandtoic, ieornaa dl en o3 0p earníods.o estudado haviam entrado no sistema com 18 anos. A média de entrada no 4s17is,2t8e%m pao, pnssouroí asmusa o v feuzn, dearam deen t2a5l i nacnooms.pleto. P5,6a%r ddoso apenados em liberdade condicional no período estudado tinham 22 anos - compondo a idade de maior representatividade naquele Fm1Mixed4ou,8m5n%e na dtaonas. Apom em sésRaceeodainas detema t oslidt aui anaç pãcopo duelma lçibãepor, dlnaeod etn cotoandtoic, ieornaa dl en o3 0p earníods.o estudado haviam entrado no sistema com 18 anos. A média de entrada no E4sP46.3%i6ss,ta3et m%rua da, dupotooa rddeclared sneucalta vereazra, emr-as dee pthemselves 2ar5d aons.os. as being of mixed race. 4174,2,85%% pa doansss upoíeasms oa fsu nedma msiteunatçaãl ion cdoem lipbleertdoa. de condicional no período estudado haviam entrado no sistema com 18 anos. A média de entrada no 4sP49i6s,7ta,3e % mr% ad a,e upcotoloarrd saeuvcalm avre aczor,a memroa- sp dereo p f2ais5rsdã aoons: o.ess.tudante. EF14su,5t%nu ddasa apemnsstoeaesn emt asitlu aiçnãoc doe limberpdalde ctoondicional no período estudado haviam entrado no sistema com 18 anos. A média de entrada no sP4i6sta,3e mr%ad a, upotoord seucal avreazr,a emra-s dee p 2a5rd aonso. s. R44FIncomplete97,ou,72 %un dpbdeocsolasaur maísavmaime o cfnoupmntdolae pmBasicrlsoe fininstasãlc ion:o ceos mtEducationupdlaepnttole.e. to 46,3 % autodeclararam-se pardos. 3P487,8a,2 %r fdpooroasmsu cíaomnd oe nfuanddoas mà epnritvaalç inãoc odme plibleetrod.ade por roubo simples. EF47.2%ustnudda ahadmnete nincompletetal incom basicplet oeducation. R46o,3 u% bauotod secilmararpaml-eses pardos. 3R4EF897,eu,8,s72 % it%nn uf pdodcoerdascaimsladu mrcíaoenvmnadne tmoe enft cuaoendtmdo aosn m àpl oer poinr niftvsiasalscçi iãnãsooct: o demes tplmiublpedetaroalnd.etaed.te opor roubo simples. 44% eram reincidentes no sistema. F497u,72 %n pdoescaslaumríamvae om fn cuontmdaom lpe rionnftiaslcs iãnoc:o emsptulpedtaoln.etet. o REStudentesoitunubcdioda senintmetep lneos sistema 47,2% possuíam o fundamental incompleto. 43RE449.7%489%,,so87 e t%urau fdmobedr rcaeolm aideclarednr cansoivdanietmdnet cenosapm dnols e sp ài rstheirs otpefrimisvsaã.çoã: o eprofession dsteu ldibaenrtdea. de por asrou bstudents.o simples. 49,7 % declaravam como profissão: estudante. E38,s8 t%u fodrama cnontdeenados à privação de liberdade por roubo simples. Reoiunbcoid seinmtpel enso sistema 49,7 % declaravam como profissão: estudante. 4R3Robbery48%,eo8 e%iurn afbmocr aroiemdin cscoeinidmdneenntepaesd nol eosn àss oipstre ivsmaçiaãs. ot de lmiberadade por roubo simples. R4338.8%48%,eo8 %eiurn afbomcr aro imeweredi ncscoeinidmdneen ntcondemnedepaesd oln eson àss oipsrt ievsmaçiaãs. ot de lmitober adimprisonmentade por roubo simples .for robbery. R4348%,e8 %eirn afomcr arimedi nccoeinddneennteaesd on son às oipsrt ievsmaçiaãs. ot de lmiberadade por roubo simples. 44% eram reincidentes no sistema. RRepeateincide Offendersnte no sistema 444%4% era mwere reincid erepeatntes no sis toffenders.ema.

Source: SEAP between 1st January and 31th July 2015 Prepared by: FGV/DAPP

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 26 FINAL CONSIDERATIONS

The municipal government of Rio de Janeiro needs to be an important actor in the design and planning of integrated public security policies. The need for strengthened municipal participation, through its strategic knowledge, in decision making in relation to public security policy is obvious. With the monitoring and assessment of local policies, programs, and projects, city government should intervene focused at the local level on social prevention, the modernization of monitoring systems and intelligence, and social reinsertion.

The participation of the municipality in the prevention of violence is essential for public security. The implementation of a municipal citizen security agenda goes beyond the Secretariat of Public Order and the Municipal Guard, which, in turn, should assist the actions of the police, thereby increasing the capillarity of the on-site and preventative scope of security agents. City governments can and should support the elaboration of public security through localized prevention actions, community oriented activities, and projects aimed at youth.

Municipal programs such as Agents of Liberty, which support those leaving the penitentiary system, should be expanded and publicized, though reallocated to strategic regions of the city. Social policies and actions to support and insert those leaving the prison system in the community and in the labor market are fundamental and have to be incorporated at the municipal level.

A critical alignment is necessary between public resources, efficiency, and public security through the intensive use of data, technology, intelligence, and methodological rigor.

PRÓXIMOS PASSOS

Cleaning and standardizing the data used in this research for the state of Rio de Janeiro, adopting it to the level of specific addresses.

Publication of the study Denunciation, Crime, and Punishment for the state of Rio de Janeiro.

Studies about the Municipal Guard using the raw material of this research.

27 | DEPARTMENT OF PUBLIC POLICY ANALYSIS METHODOLOGY

This new research from FGV/DAPP drew on data from three different sources: Disque-Denúncia Rio de Janeiro (DD-RJ), the Institute of Public Security (ISP), and the State Secretariat for Penitentiary Administration (SEAP). Through the spatial analysis of data, maps were constructed of the concentration of denunciations, crimes, and origins of convicts in Rio de Janeiro city per neighborhood. The temporal scope adopted was 1 January - 31 July 2015.

As different databases are combined, it is important to explain some characteristics of each one. First, the data coming from Disque- Denúncia is not about the occurrence of crimes in themselves, but the suspicions of part of the population that some illicit activity could be occurring. However, this does not signify that they are not important. They constitute interesting indicators of feelings of insecurity that are often diffuse, and which are capable of mobilizing people to denounce suspicions of crimes or problematic situations. In addition, they can be very useful in re-planning public security policies and various types of urban planning, as well as, in the case of denunciations of known criminals, are an important instrument of intelligence, capable of preserving the lives of public security agents and the population in general.

Another important specificity about the DD-RJ database is that it is composed of anonymous calls which often contain more than one denunciation. The attendants of Disque-Denúncia who receive the calls hear the anonymous report and record in the agency’s database as many denunciations and themes that are mentioned. In other words, each denunciation can be related to others. In this sense, a call can contain multiple denunciations. For this research, we use as a reference the lowest unit possible, in other words, the denunciations, which allow us produce a richer analysis.

The data obtained from the Institute of Public Security (ISP) is related to criminal occurrences reported to the Civil Police. Similar to what occurs with DD, although less commonly, an event reported in a police station can contain more than one crime. Therefore, we once again privilege the lowest analytical unit possible, that of crimes and not events. Even more specifically, the counting of the crimes adopted takes into account the quantity of victims involved. Therefore, a crime with multiple victims is counted multiple times. For this research the following crimes were counted: murder, robbery resulting in death, assault followed by death, rape, robbery of vehicles, robbery of mobile phones, robbery of passersby, robbery in groups, apprehensions of drugs, and apprehensions of arms. The choice of these types of crimes, specifically followed as a criterion the analysis of crimes which formed the Integrated Target System of the State Secretary of Public Security, complementing them with some crimes which had a recognizably high impact on the quality of life of the Carioca population.

Finally, in relation to the SEAP database, we used two analytical categories: the declared neighborhood of residence of origin of convicts and the neighborhood of residence of those on parole.

DENUNCIATION, CRIME AND PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 28 Taking into account the differences between each database, it is important to make some comments about their quality. For various reasons, the construction of public security information presents many problems. In analyzing the "neighborhood’ field, in all the databases, we can find errors which run from the mistyping of the name of the neighborhood to the non-existence of the name itself, also covering the identification of sub-neighborhoods, administrative regions, or favelas. Furthermore, when we compare the information of the address with the neighborhood, cases in which the data conflicts are not rare.

For this research we observed only the "neighborhood" field — admitting probable errors of addresses identified in wrong neighborhoods in a later stage of research. For each of the databases, we carried out a standardization process of the names of neighborhoods, adopting as a model the neighborhoods recognized by the Pereira Passos Institute (IPP). Typing errors were corrected, in order not to lose neighborhood information. It was not possible to identify the neighborhood in approximately 0.97% of records inthe Disque-Denúncia database, 7.87% in the case of ISP, and 8.38% in that of SEAP.

To estimate the rates per thousand inhabitants, we used the neighborhood population according to the last IBGE Census (2010). An observation should be made here: in 2010, the neighborhood of Lapa was still a sub- neighborhood of Centro, and for this reason we do not have the estimated population of Lapa, preventing the calculation of rates for this neighborhood.

TEAM

FGV/DAPP Department of Public Policy Analysis Getulio Vargas Foundation

DIRECTOR Marco Aurelio Ruediger

IMPLEMENTATION TEAM Coordination Marco Aurelio Ruediger

Research Coordination Roberta Novis Maria Isabel Couto

Researchers Rachel Bastos Lucas Silva Janaina Fernandes Miguel Orrillo Danielle Sanches

29 | DEPARTMENT OF PUBLIC POLICY ANALYSIS INNOVATION FOR PUBLIC POLICIES

dapp.fgv.br [email protected]

DENUNCIATION,+55 CRIME (21) AND 3799.4300PUNISHMENT: THE CYCLE OF VIOLENCE IN THE CITY OF RIO DE JANEIRO | 30