Violent Crime and Homicides in São Paulo: underlying factors and recent trends / Renato Sérgio de Lima

The State of São Paulo and its capital have registered successively lower rates of violence, particularly homicides. However, the reasons behind this phenomenon remain the subject of fierce debate. Elucidating these underlying factors is important not only in academic terms but particularly in the field of public policy, since identifying the factors leading to such a positive impact on rates of violence in São Paulo can help refine public programs currently underway within the State and possibly allow broader implementation throughout the country. Drawing on the current literature in the field, this study briefly reports and systematically organizes several theories for this fall in violent crime so as to foster further debate on the theme.

Brazil and the State of São Paulo have undergone – and continue to undergo – radical demographic and social changes which, in conjunction with new approaches to devising, managing and implementing public policies in public security and violence prevention explain the trends in these figures to a lesser or greater extent. When presenting the issue in these generic terms, no great divide is evident among scholars in the field. However, upon delving deeper into the issues, differences in opinion come to the fore.

Thus, although there is consensus on the presence of manifold determinants for the fall in rates of violence in São Paulo, there is broad divergence in the weight given to each of these underlying factors. These differences partially stem from the fact that this theme is often addressed by scholars drawing on a body of specific knowledge and practices limited to ad hoc experiences or international cases.

Example Cases and São Paulo State

Recent success stories concerning policies to combat urban violent include the experiences of New York and Bogota cities: homicide rates were reduced from 30.8 to 9.4 crimes per 100,000 inhabitants between 1990 and 2000 in New York, and cut by 63% in Bogota between 1993 and 2003. Although the falls seen cannot be definitively attributed to generalized decrease or higher levels of security in all its spheres and across different regions of these cities, the results achieved were nonetheless remarkable. It is natural therefore that proposals on public security policy for implementation in other cities attempt to interpret these cases and identify successful initiatives which may be replicated or promoted elsewhere.

Since 2000, the State of São Paulo has enjoyed similar results to those cited. Based on health records, a steep rise in violent death rates was followed by a subsequent fall from 43.2 cases per 100,000 inhabitants in 1999 to 22.0 cases per 100,000 in 2005. This decrease brought São Paulo rates to below the national average (26.2), reversing the historic trends seen in annual statistics since 1980 (Graph 1).


graf 11 copy

Violent Death Rates per 100,000 inhabitants by region

Brazil and São Paulo



São Paulo

Brazil (excluding São Paulo)

Source: MS/SVS/DASIS – Death Records System – SIM

However, comparisons made with international cases are limited in that, hand in hand with decreasing indices evidenced by falling murder rates is a structural change in the dynamics of the violence in the State. This is not consistent with the concepts put forward to explain the phenomenon in the 1990s, strongly influenced by the inciting of conflict in the metropolitan areas and the surge in violent deaths among youths.

The pattern for São Paulo State, namely, a swift fall in homicide rates in the city and metropolitan region of São Paulo, tend to lead to misinterpretation of violence as being a strictly metropolitan malaise. Given this fall has not been accompanied by increasing levels of violence in the State interior, where regional rates tend to converge toward the State rate, this points to a new “São Paulo pattern”. Moreover, if the rise in homicide rates over the past 20 years has indeed resulted from the incitement of violence among young males, then recent trends in these figures reveal a radical reversal among this population: between 1994 and 2006, the violent death rate among males in the 15 to 24 age bracket fell by half, to levels closer to those observed among the 25 to 34 year-old age group.

The data thus reveals that the factors leading to the fall in homicide rates in São Paulo were geographically concentrated and affected specific social groups.  However, the significance of this process and its root causes differs among researchers, policy-makers and the police, therefore the aim of the present study was to discuss these underlying divergences.

Multiple factors underlying reduced violence in São Paulo

A recent study by the Seade Foundation conducted for the Secretariat of Justice and Protection of Citizens’ Rights of São Paulo[3] highlighted several factors related to the social, political and institutional processes which should be taken into account in studies and in State interventions in the field of public security and violence prevention. According to the cited article, a brief examination of the institutions and the policies of justice and public security implemented in Brazil, particularly São Paulo, after the transition to democracy, reveals a course which, although well known, is worth revisiting.

Initially in the 1980s and 1990s, there was a steady rise in rates of crime and violence and a growing feeling of lack of security among the population, a situation which prevails in many States today. The specialized literature[4] links this increase to the continuation of organizational structures and legal characteristics inherited from the dictatorship era, which led to the persistence of obsolete operational practices and resistance to social control and participation on the part of police and judicial institutions. Furthermore, because these institutions comprise a system organized into different levels and spheres of authority and governance, the process of integrating and adopting work paradigms which are grounded in the principles of human rights and transparency, proved extremely complex.

However, the rising levels of violent incidents in previous decades drove the Federal Units, particularly the State of São Paulo, to bringing in a raft of managerial reforms in police forces and to seeking ways of pooling the efforts of several institutions comprising this system, ultimately tasked with handling conflicts in our society.  In this context, the view that the phenomena of crime, violence, unrest and disrespect of human rights result have multiple determinants and root causes took shape.

Thus, the implementation of changes to the system of justice and public security, together with a broader understanding of the causes of violence, led to recognition by the public authorities and scholars on the theme that violence and crime rates are influenced by a combination of social and demographic factors, the presence of risk factors, the implementation of social and urban policies and prevention strategies. Under these conditions, even if the police and criminal justice systems were pivotal in the efforts to combat violence and maintain the peace, this occurred in conjunction with new institutions and social actors, along with new practices adopted which undoubtedly contributed to the reversal in crime levels observed in recent years.

Outlining the role of each of these factors, with their multiple forms and complex interrelationships, is a challenging undertaking. However, identifying them is the first step toward assessing their influence and understanding their connections. This may give rise to new approaches and methods of combating urban violence for use by public authorities.

The ensuing sections of this study compare the key factors thought to underlie the reduction in rates of violence in São Paulo and which have been proposed in the literature. These contributions have been systematically summarized in a bid to contribute to the debate on the theme.

Refinement of the Mechanisms of Planning, Management and Control

One of the most common arguments put forward to explain the recent marked fall in homicides in São Paulo concerns the management of public security policies, particularly the specific management reforms and strategic planning adopted in the past ten years by the Secretariat for Public Security.  Numbering among these initiatives is the expansion of the DHPP, the creation the Superintendency of Technical and Scientific (Forensic) Police, the raising of priority to incarcerate serial killers, the appropriate division of function and spheres of activity of two territorial police forces and community policing.

Advocates of this notion point to the pooled efforts by Federal and São Paulo authorities to improve the technical aspects of police activity and the management of justice as well as the public security institutions on the following fronts:

• policy planning and appraisal;

• modernizing and expanding of communication systems and employment of latest technologies;

• information management (Infocrim, Police Service Levels System);

• practices and procedures of action in line with those intended to safeguard rights (Witness Protection, the ‘Bem Me Quer’ Program, creation of POPs – Standard Operational Protocols by the Military Police, saturation and social turnaround operations etc.);

• local prevention policies (such as community policing);

• training and valuing of professionals, including the inclusion of subjects on human rights in curricula of police academies and training schools for field practitioners;

• increased internal and external control mechanisms (Ombudsman and Investigators;Commission for the Reduction of Lethal Force by Police);

• creation of institutional entities which widen access to justice and safeguard rights (Public Defense Attorney, Special Courts, Scientific (Forensic) Police, new units of the Housing Foundation (Fundação Casa), new vision of land policy, etc.).

In addition, there has been an increase in the proportion of public funds allocated to the area of public security. In São Paulo for instance, funds were increased from 6.6% of total government expenditure in 1999 to almost 10% in 2006 (Graph 2).

graf 12 copy

Graph 2: Percentage expenditure on Public Security as a proportion of Public Spending

São Paulo


Source: Ministry of Finance, Secretariat of the National Treasury, General-Coordination of Relations and Financial Analysis of States and Municipalities – Corem
This school includes those who claim that the Disarmament Statute (arms amnesty) enabled arms to be withdrawn from circulation, which subsequently had a positive effect on the homicide rate. São Paulo was a pioneer in mounting spot checks to control firearms and ammunition. Studies carried out using information from the Datasus have corroborated this notion by assessing the impact of the disarmament campaign on firearm-related deaths. These studies found a 19.4% reduction in the number of such deaths which fell from 10,097 to 8,137 between 2003 and 2004, and coincided with the bringing in of the new legislation. Increased incarcerations (higher than the national average) are also thought to be partially responsible for the drop in homicides.

Role of Municipal Authorities

Another question which has been raised in the security policy ambit is the inclusion of new institutions to the public security agenda, as is the case for the municipal councils. According to the Constitution, public security (police management) comes under the remit of the State Governments, while the municipal authorities are only responsible for protection of their assets.

However, following the waves of violence in the 1980s and 1990s, the role of the municipal authorities in preventing crime and violence was redefined at a local level. This has apparently led to a change in approach from repressive to preventive, and has shifted focus from crime to violence, in which measures and interventions implemented in urban settings are key to remedying the state of insecurity.

Briefly, there are five key municipal-level policies: creation of the Municipal Guards; institutionalizing of councils, secretariats and municipal security plans; access by municipal authorities to funds held by the National Fund for Public Security; and, in some cities, creation of laws restricting the operation of bars and sales of alcohol.

According to the IBGE (Brazilian Institute of Geography and Statistics), 28.7% of municipalities within São Paulo had Municipal Guards in 2006, versus a national average of 14.1%. Notably, some States present higher homicide rates than São Paulo, even though a larger proportion of their municipalities have Guards. This is the case for Rio de Janeiro, where 71.7% of municipalities boast Municipal Guards. With regard to councils and similar institutions, the IBGE has no data available on how many municipalities per State have created these bodies, only information that 445 municipalities have set these up nationwide.

Public funds dedicated to municipal authorities through the National Fund for Public Security amounted to R$ 111.283.104,34, between 2003 and 2007. In 2007 alone, a total of R$ 42.363.550,76 went to 95 municipal authorities in the form of covenants, with 37 of these municipalities located (39%) in São Paulo State. These resources were largely to fund prevention actions and training (Graph 3).

In 2000, a paper by the author highlighted the role played by urban cities in homicides. Citing from the study, “crime in São Paulo became so central to everyday life that it reshaped landscapes and behaviors. The majority of violent crimes take place in the outlying suburban districts, generally affecting the poorer regions with the highest social-economic distress. Crime represents yet another element amid a setting of abject structural conditions and illegality. The State is not so much absent, but has an ambiguous and arbitrary presence:  repressive, paternalist, or clientelist. In this sense, the population of São Paulo is experiencing what could be called managerial breakdown of the city, where it faces an absence of public policies and mechanisms from Municipal, State and Federal governments, with a lack of global incentive and citizenship actions and also of social pacification policies”.

Following this line of reasoning, the decrease in homicides is directly linked to the incorporation of public security agendas by local authorities and consequently the reorganizing of their services and urban space under a preventive framework. Limiting the activities of bars and the consumption of alcohol is one such example. A number of important studies involving emblematic cases have been conducted on this theme, such as the Diadema study. The study found a statistically significant correlation between However, the theme had not been investigated exhaustively since ideally studies could be conducted comparing empirical evidence from several cities that had adopted this policy, against other cities which had not, with this latter group serving as controls.

Social Participation

By the same token, and irrespective of the role of the State, other findings have acknowledged that the shifts occurring in civilian society constitute a crucial element in understanding this situation. Experts draw attention to the role of social and community organizations which, in collaboration with the public authorities, contribute toward bridging a major institutional gap in public projects, by recruiting the vulnerable population as an active agent in policies for combating violence and in boosting the effectiveness of public actions.

Nevertheless, akin to the studies on the role of municipal policy, those aimed at increasing the level of community participation in high risk areas are localized in nature, with major inherent difficulties in extrapolating outcomes to larger geographic domains, notwithstanding the potential for intervention that these local schemes offer.

Curbing Turf Wars among Criminal Factions

Another line of argument, also centered on the forms of sociability of the population in the poor outlying districts and high-risk areas, focuses on a code of order-maintenance devised by organized crime. According to this theory, defended by police segments and ethnographic studies, there is a widely held belief in these areas that the hegemony of a criminal faction (the PCC – First Command of the Capital – an inmate gang operating from within penitentiaries) has contributed to lower homicide rates by mediating conflicts and maintaining order through “pacifying” territories hitherto dominated by several different gangs linked to drugs trafficking.

According to police reports, this hegemony results from the PCC having taken control of the wholesale trade of illegal drugs in São Paulo, obliging local groups to buy fixed volumes of drugs, which in turn has curbed disputes over territory. The question of generalizing of a localized action to wider geographic domains applies in this instance. Also, the perception of the social segments regarding lethal violence encompasses a broad notion of “homicide” as understood by these segments, which does not correspond to the legal and social categories formally accepted in Brazil.

Demographics and Socio-economic Aspects

Another line of argument holds that, although the behavior of homicide rates can be discussed within the specific context of public security policy or trends in society, the issue of crime must be addressed in a broader sense. Factors of a structural nature and social transformations over the short and long terms are also deemed to have contributed toward the lowering of murder rates in São Paulo in as far as they enhance the effectiveness of police actions. In this instance, the fall in crime results from a combination of demographic, social and economic factors, which have demonstrated to be highly conducive to reducing the murder rate.

Firstly, the demographic transformations have narrowed the young population strata. This group is more prone to becoming involved in violent acts. Predictably, it was among these very strata that homicide rates were seen to decrease. Graph 4 illustrates the proportion of 15 to 24 year-olds from the entire São Paulo population. The proportion fell from 19.4% in 2000 to 17.6% in 2006 (Graph 4).  Thus, it has been proposed that the reduced number of youths – a repercussion of the relentless drop in fertility which began in the 1980s – has caused the population segment most exposed to risks of violence to shrink. This in turn has increased the effectiveness of new standards of police work, leading to positive outcomes in combating violence in São Paulo.

graf 13 copy

Graph 4: Proportion of 15 to 24 year-old inhabitants of

São Paulo


Source: IBGE – Brazilian Institute of Geography and Statistics- National Household Survey

This shift in the demographic picture combines with another important aspect: the increased level of schooling of the population in general, particularly among younger age groups. More highly educated individuals are expected to resolve conflicts in a non-violent manner. Considering this hypothesis as a premise, it can be concluded that the rapidly broadened access to school, particularly to the younger generations, should reduce the rate of violent acts during conflict resolution. Moreover, besides the role in educating and providing socialization of its pupils, the school may also take on a protective role, especially for those students residing in high-risk areas. Thus, for several reasons, the provision of near universal access to schools may have contributed to the marked reduction in fatal violence in São Paulo.

Finally, another noteworthy point is that the demographic and social transformations coincide with an economic environment conducive to the entry of various social segments to the job market. Although the performance of the economy has proved erratic since the end of the 1990s, the São Paulo Real Index of Added Value presented an overall rise (Graph 5) which it turn was reflected in a steady increase in income (Graph 6). This scenario created favorable conditions for the job market, evidenced by a fall in jobless figures and increased rate of participation: between 1999 and 2006, whereas the unemployment rate fell from 12.6% to 10.0%, the rate of activity  – the proportion of individuals who have belonged to the job market either as occupied or unemployed for more than 10 years – rose from 56.8% to 62.6% in the State of São Paulo.  In addition to the recovery of self-esteem of those rejoining the job market with its associated pecuniary gains, the increased amount of time spent on productive activities tends to reduce exposure to high-risk situations among individuals rejoining the work universe.


One of the theories which has gained credence in the debate over the reduced homicide rate is undoubtedly the association of the fall in this crime to the imprisonment policy adopted by the State of São Paulo. According to some experts, the increase in the number of imprisonments and the building of several penitentiaries has affected the crime dynamic by getting delinquents with criminal tendencies “off the streets”. However, with regard to increases in the prison population of São Paulo State, the rate of incarceration has in fact been rising linearly since 1994 as depicted in Graph 5, yet the violent death rate grew until peaking in 1999, and subsequently fell more sharply than prison populations rose, particularly from 2003 onwards.

In other words, the fall in homicides occurred after six years of simultaneous growth in the number of prisons. This indicates that, in contrast to what the mass incarceration argument maintains, the success in reversing the trend in this type of crime must be attributed to factors other than prison. A plausible hypothesis is that from 1998 the São Paulo Police, more specifically the Homicides and Personal Protection Department – the DHPP, linked to the Civil Police and which operates in the Capital, prioritized imprisonment of “contumacious murderers” in a strategy to tackle these crimes by identifying and imprisoning individuals responsible for multiple deaths. In this case, what appears to have had a greater bearing than the number of imprisonments per se (shown in Graph 5), is prioritizing the imprisonment of persons who represent a high danger to society.

Graph 5: Incarceration rate and violent death rate State of São Paulo

graf 14 copy

Final considerations

The discussions at the Workshop which inspired the writing of the present article confirm the importance of researchers and managers from the area to consider multi-factorial approaches in relation to homicides, and from this, establish programs for research and monitoring of lethal violence which address both more immediate aspects such those associated to managing public security policy and police institutions, as well as socio-economic and demographic elements.   Certainly, in the agenda of public politics, this task may seem too great a challenge, but in fact, the challenge will be to structure information systems which are able to handle multiple factors and, within the bounds of the technical resources available, to identify preponderances and key characteristics. Rather than exhaust all the variables in this explanatory territory, the present article sought to provide a brief overview of the main determinants of lethal violence in Sao Paulo and Brazil.

In theoretical terms, the same argument put forward during the 1990s in a bid to explain the rise in homicides also applies to the debate on reducing these crimes in this first decade of the twenty-first century. Hence, within a context which associates spatial, urban, social, institutional and cultural issues, the conflicts which result in death in São Paulo are part of a logic (informed in a multifaceted way) which symbolically fires the social imagination of how these should be resolved, and therefore, future research needs to address the moral value of life and the roles of the State, and the violent death in the everyday language which pervades social relations in the metropolis.


[1] Fuller version of the article written by Renato Sérgio de Lima, Sinésio Pires Ferreira, Eliana Bordini and Vagner de Carvalho Bessa for the workshop “Violent Crime and Murders in São Paulo: Underlying Factors and Recent Trends (Criminalidade Violenta e Homicídios em São Paulo:Fatores Explicativos e Movimentos Recentes), organized and promoted by: São Paulo Law School – Edesp/FGV; the Brazilian Forum on Public Security; the Friedrich Ebert Foundation – FES; the State Data Analysis System Foundation – SEADE; the United Nations Institute for Crime Prevention and Offender Rehabilitation – ILANUD; the Sou da Paz (I’m for peace) Institute; the Center for Studies on Violence – NEV-USP; the Secretariat of State for Public Security Affairs for São Paulo; National Secretariat for Public Security – SENASP. Event held in August 2008.

[2] Renato Sérgio de Lima PhD in Sociology from the University of São Paulo. General Secretary of the Brazilian Forum of Public Security.

[3] Text produced to support the revising of the 1998 State Plan of Human Rights, and the State Conference on Human Rights to be held in August and September of this year. Full text available at

[4] “Violência, Criminalidade, Segurança Pública e Justiça Social” (“Violence, Crime, Public Security and Social Justice”), in Revista Brasileira de Informaçaão Bibliográfica em Ciências Sociais (the Brazilian Journal of Bibliographic Information on Social Sciences), Number 50. São Paulo, Relume Dumará,  2000.

About these ads

4 comentarios

  1. I think this article made some interesting points, I read a textbook directly related to this topic, its called Comparing Economic Systems in the Twenty-First Century by Gregory , I found my used copy for less than the bookstores at

  2. Your site was extremely interesting, especially since I was searching for thoughts on this subject last Thursday. :)

  3. Hey, I found your blog in a new directory of blogs. I dont know how your blog came up, must have been a typo, anyway cool blog, I bookmarked you. :) :)

  4. I don’t know If I said it already but …Hey good stuff…keep up the good work! :) I read a lot of blogs on a daily basis and for the most part, people lack substance but, I just wanted to make a quick comment to say I’m glad I found your blog. Thanks,)

    A definite great read….

Deja un comentario

Introduce tus datos o haz clic en un icono para iniciar sesión:

Logo de

Estás comentando usando tu cuenta de Cerrar sesión / Cambiar )

Imagen de Twitter

Estás comentando usando tu cuenta de Twitter. Cerrar sesión / Cambiar )

Foto de Facebook

Estás comentando usando tu cuenta de Facebook. Cerrar sesión / Cambiar )

Google+ photo

Estás comentando usando tu cuenta de Google+. Cerrar sesión / Cambiar )

Conectando a %s


Recibe cada nueva publicación en tu buzón de correo electrónico.

Únete a otros 74 seguidores

%d personas les gusta esto: