Imprimir Republish


How many former prisoners reoffend after release?

Researchers calculate and attempt to reduce the country's criminal recidivism rate

Police recidivism and criminal recidivism are concepts used to measure the phenomenon of imprisonment in Brazil

Léo Ramos Chaves/ Revista Pesquisa FAPESP

A recurring statistic has puzzled Brazilian researchers in public safety for many decades. Leaders such as Chief Justice of the Federal Supreme Court (STF) Gilmar Mendes or former Chief Justice Cezar Peluso, in addition to documents such as the final report of the Congressional Investigative Commission (CPI) on the prison system issued in 2008, repeatedly indicate that 70% of individuals serving prison sentences in Brazil commit the same crimes shortly after their release. This rate perplexes the experts: no one can tell where it came from nor how it was calculated. Recent studies carried out in different states around the country came up with recidivism rates between 24% and 51%, all quite far from the 70% used as a benchmark.

This discrepancy served as a catalyst for increased attention by academia in recent years. “This recurring, unsubstantiated statistic has helped to build consensus that the prison system has failed, not rehabilitating nor saving anyone. But is this what these data actually mean?” questions sociologist Luís Flávio Sapori, coordinator at the Center for Public Safety Studies and Research at the Pontifical Catholic University of Minas Gerais (CEPESP-PUC Minas). “Perhaps this consensus is the reason why crime, violence, and imprisonment in Brazil are so thoroughly investigated, while recidivism much less so. Why study something we already know does not work?”

In Brazil, scientific interest in imprisonment increased significantly in the 1980’s, according to sociologist Maiara Corrêa, researcher at the Violence Studies Center of the University of São Paulo (NEV-USP). Two decisive moments were the enactment of the Criminal Enforcement Act (LEP) in 1984, and when the Legislative Assembly made human rights a top priority in 1987 and 1988. Studies of the prison situation were also motivated by the massacre at the Carandiru Prison in the city of São Paulo, where 111 inmates were killed by police during a riot in October 1992, as well as the growing attention of public opinion regarding organized crime factions.

Within this context, the first Brazilian studies that attempted to estimate recidivism rates, still at a local level, were led by the NEV-USP sociologists Sérgio Adorno and Eliana Bordini. In 1988, they observed that 46% of convicts released from prison in São Paulo were later reincarcerated. In 1991, Adorno and Bordini reviewed a slightly different situation: how many former inmates were convicted again in court. They came up with 29%. In 1999, in Rio de Janeiro, sociologist Julita Lemgruber, coordinator of the Center for the Study of Safety and Citizenship at Cândido Mendes University, calculated a 30% return rate to the prisons.

Pisco del Gaiso/FolhapresIn October 1992, during a riot, the Carandiru Prison in the capital city of São Paulo was the stage of a massacre where 111 inmates were killed by the policePisco del Gaiso/Folhapres

During the last decade, we have seen a broader effort to understand recidivism in the country, particularly following the publication of the 2015 report Reincidência criminal no Brasil (Criminal recidivism in Brazil), a partnership between the Institute for Applied Economic Research (IPEA) and the National Board of Justice (CNJ). This study described the recidivist as someone who, after having served their time, was convicted a second time. This statistic reached 24%.

Other studies have been carried out, focusing on the prison system for adults or on the social educational system, which addresses adolescents in conflict with the law. They highlight the paper “Aí eu voltei para o corre: Estudo da reincidência infracional do adolescente no estado de São Paulo” (So, I was back in trouble: A study of criminal recidivism among adolescents in the state of São Paulo), published by the Sou da Paz Institute in 2018, which developed a profile of young recidivists, but did not calculate the relevant statistic, and the report “Reentradas e reiterações infracionais: Um olhar sobre os sistemas socioeducativo e prisional brasileiros” (Criminal reentries and reiterations: A view of the Brazilian social education and prison systems), also published by the CNJ in 2019, which calculated 23.9% of reentries into the social educational system and a 42.5% recidivism rate in prisons.

A more recent study with a broader perspective on recidivism was disclosed to the public late last year, resulting from a partnership between the National Department of Prisons of the Ministry of Justice (DEPEN) and the Public and Economic Policy Evaluation Group of the Federal University of Pernambuco (GAPPE-UFPE). The paper is a compilation of information from 12 states and the Federal District, including data of approximately 979,000 people who were incarcerated between 2010 and 2021.

The conceptual and methodological choices of each of these studies show that recidivism is defined differently, depending on the investigation’s area of knowledge or on the research objectives. Article 63 of the Brazilian Criminal Code considers recidivism as crime committed after final sentencing for a previous crime. Article 64 limits recidivism to a period of five years between the release from the first sentence and the second crime committed. This five-year window complies with international standards, as mentioned by economist Camila Gomes of Georgetown University, in the United States, and one of the coordinators of the study requested by the DEPEN.

Studies focused on the administration of the prison system or on public policies for prisoners and former inmates tend to prefer the concept of “judicial recidivism,” which considers the conviction of an individual for a new crime, and “prison recidivism,” which considers the return to prison pursuant to the second offense. Apart from imprisonment, public safety research usually focuses on “police recidivism,” which refers to new police records on a crime committed by the same person, or “criminal recidivism,” which considers the recidivist as the individual who is the subject of a new lawsuit.

Offenses connected to valuable items represent a higher recidivism rate than that of homicides or aggressions

Such a considerable difference between the relevant definitions is part of the explanation for the disparity among the percentages identified in the various studies. Nevertheless, the published rates are always significantly lower than the 70% statistic that is common in public debates throughout the country. Using five definitions, the DEPEN and UFPE reports came up with rates between 36% and 42%. In Minas Gerais, using a new police index as a benchmark, the rate found by Sapori and sociologist Roberta Fernandes Santos was 51%, as quoted in the article “Fatores sociais determinantes da reincidência criminal no Brasil” (Social factors determining criminal recidivism in Brazil), published by both authors in 2017. “Adopting a judicial criterion does not seem to work in a country like Brazil, where the criminal justice system is so slow. For this reason, we considered the recidivist as the individual who is investigated and, by the end of the investigation, the police conclude they committed a new crime,” says the sociologist.

Studies on recidivism can also be found in databases that are incomplete, poorly assembled, and lacking standardization, housed by public safety departments or state penitentiary agencies. This is one of the reasons why most studies, until recently, limited their scope to the investigation of a single state. Even the work developed by the IPEA for the CNJ restricted its research scope. It concentrated on the states of Paraná, Minas Gerais, Rio de Janeiro, Alagoas, and Pernambuco.

“Currently, the DEPEN is trying to compile administrative data from the states, aggregating information from all prisoners in Brazil, with the purpose of feeding the studies and preparing public safety assessments,” adds Gomes. The purpose of the partnership with GAPPE-UFPE aims to establish a profile for the Brazilian inmate. The economist adds that the full report is expected to be published before the end of this quarter and will likely include information on race, education level, and income, in addition to inmates’ participation in the labor market, before and after incarceration.

This initiative is justified by the need to produce evidence that will uphold the policies designed to reduce crime. “Knowing how many commit crimes after incarceration is only the first step in the study of recidivism,” mentions Sapori. “It is important to investigate the social and psychosocial factors, and other aspects that affect the probability of an individual committing a new offense after they are released from prison. With this information, the technical management of the prison system could include more robust policies to monitor prisoners and former inmates, directing outreach and inclusion activities to those who are more prone to recidivism.”

According to Sapori, there is an example of a successful policy in Catalonia, Spain. “For many decades, the Catalans have studied this subject and use the information to create a recidivism risk protocol, which measures from the point of prison entry, the probability of a person committing a new crime, with the goal of directing care policies for the individual, not only within prison but also after release.” The so-called RisCanvi protocol was established in Catalonia in 2009.

Mario Tama/Getty ImagensPedrinhas Prison Complex, in São Luís, Maranhão. In the records filed in 2015, a recurring problem in the Brazilian prison system: overcrowdingMario Tama/Getty Imagens

“Several factors must be taken into account to understand the dynamics of recidivism, from the high crime rate in Brazil to the conditions under which the prisons operate,” says sociologist Almir de Oliveira Junior, an IPEA researcher and one of the coordinators of the study completed in 2015. Oliveira Junior describes crime as a market, where there is an interaction between individuals’ motives, opportunities, resources involved, and other incentives.

According to the sociologist, overcrowded prisons facilitate the entry of the convicted person into the so-called “criminal career.” “Crime, just like any other occupation, demands learning. How to use a weapon. When to walk away from an attempted crime. How to deal with a person who resists or runs. What is the ideal place for a heist? The best place to learn all of this is in prison, where those who already have the knowledge are found,” he notes.

In Brazilian studies, offenses associated with valuable items, such as thefts and robberies, and with drug trafficking, show a higher recidivism rate than homicides or assaults. According to Oliveira Junior, one of the reasons for this is that the two first categories are more commonly associated with the criminal career, more specifically, with the performance of organized crime. “When someone enters the world of crime, the doors for the formal labor market are immediately shut. The community distances themselves when they know someone has committed a crime, even if the person is not convicted. Sometimes, the family also turns their back on the person. The chance of being drafted by criminal groups only increases,” adds the IPEA researcher

The DEPEN report reveals information that may have a significant impact on the development of a public policy for those released from prison. The new crime committed by the recidivist most often occurs shortly after they are released. Approximately two-thirds of these offenses happen during the first year after release. Among these, almost 30% happen in the very first month and 50% before the third month. Gomes interprets this information as evidence that the policies designed for former inmates are critical for reducing recidivism.

However, these policies are scarce in Brazil, are not allocated sufficient funding, and represent a low priority for the government. The Former Prison System Inmate Care Network (RAESP), in Rio de Janeiro, is a pioneering initiative being undertaken by community organizations such as the Santa Cabrini Foundation, the Inmates’ Pastoral Ministry, the Providência Bank, and the Consuelo Pinheiro Institute. In 2005, Sapori launched the Social Inclusion Program for Former Prison System Inmates (PRESP), in Minas Gerais, when he was the state secretary for public safety. In 2009, the CNJ, in turn, implemented the project Começar de Novo (A New Start), which sponsored programs of reintegration under the responsibility of state courts of law. Later, in 2019, the board developed a Former Prison System Inmate Care Policy as an incentive for the creation of entities known as social offices in at least 18 states, with the purpose of following up and assisting former inmates.

Léo Ramos Chaves / Revista Pesquisa FAPESPDeprivation of liberty: as provided in article 64 of the Criminal Code, recidivism involves a five-year period between the conclusion of the first sentence and the second crimeLéo Ramos Chaves / Revista Pesquisa FAPESP

Equally important as the policies directed to those released from prison are those that seek to prepare each individual to return to freedom while still in prison. These initiatives focus on reintegration, resocialization, and reeducation through working, studying and other activities, whereby the inmates reduce their time in prison by one day for every three days they participate in such activities. These measures have been provided through the Criminal Enforcement Act since 1984 but are not given priority nor sufficient budget.

“In most cases, the resocialization initiatives are undertaken by community-based organizations. The Inmates’ Pastoral Ministry of the Catholic Church is a typical example. College professors have also submitted projects to the states’ Prison Administration Agency to transport volunteers to the prisons, as well as offer classes and other resocialization projects. Work opportunities for inmates are offered by a few private small businesses that benefit from the low cost of this type of labor. The public investment is low because the government’s involvement is quite rare at this point,” says Corrêa.

Two critical factors that influence recidivism are one’s youth and the stigma that follows a former inmate. “An adult offender usually begins his journey during adolescence. For this reason, the more we understand the reasons and methods by which a youth enters the world of crime, the better prepared we will be to think of prevention policies,” argues Sapori. The stigma, however, is harder to tackle: former inmates are rejected by potential employers and run the risk of only being able to make a living with the help of organized crime, which in turn requires their participation in new offenses.

“The stigma experienced by former inmates is equivalent to having one’s destiny engraved on their forehead. It is easy to identify someone who has been in jail, looking at body language, verbal language, and the clothes they wear,” explains Corrêa. “When the person is capable of breaking through the first barrier to be hired, complaints usually arise that the person does not know how to behave or deal with peers, nor adapt to the discipline required by the work environment. Is that not ironic? After all, at least theoretically, the criminal system exists exactly to discipline individuals.”

Transformações, continuidades e tensões: O universo res no sistema prisional brasileiro contemporâneo (nº 22/07866-2); Grant Mechanism Doctoral (PhD) Fellowship; Supervisor Marcos César Alvarez (USP); Beneficiary Maiara Corrêa (USP); Investment R$106,430.04.

Scientific articles
ABBADIE, C. E. S. et al. A reincidência criminal do sistema penitenciário brasileiro. Iberoamerican Humanities, Science and Education Magazine. 7 (4). 2021.
FARIA, A. H. P. Reincidência criminal e criminalidade em série. Revista do Instituto Brasileiro de Segurança Pública. vol. 3, nº 6. 2020.
IFANGER, F. C. A. e GRAVINA, NO. B. Um estudo da agravante da reincidência na cidade de Campinas-SP. Revista de Estudos Empíricos em Direito. vol. 7, nº 1. 2020.
SAPORI, L. F. et al. A reiteração de atos infracionais no Brasil: O caso de Minas Gerais. Revista Direito GV. vol. 16, nº 3. 2020.