Imprimir PDF Republish


Justice is delayed but (apparently) is not failing

A study of the fight against corruption from the perspective of police investigations and criminal prosecutions

Augusto Zambonato

Impunity in identified cases of corruption in Brazil may occur in as few as 5% of such cases. This is the percentage of cases that, between 2010 and 2016, in court districts in the states of Alagoas, São Paulo, Rio de Janeiro, and the Federal District, were dismissed simply due to the slowness of the justice system. “The statute of limitations is commonly viewed and portrayed in the media as a mechanism by which those accused of crimes are able to avoid being tried, benefiting offenders who run out the deadlines,” observes political scientist José Álvaro Moisés, a professor in the Department of Political Science of the University of São Paulo (USP) and coordinator of a study developed by the Center for Public Policy (NUPPs) at USP in partnership with the Brazilian Association of Jurimetrics (ABJ). “We start from this point of view in order to verify if impunity is actually happening and to what extent,” Moisés explains.

The study “Criminal justice, impunity, and the statute of limitations” brought together six researchers, four interns, and the research coordinator with the aim of seeking evidence of impunity in what is called the Brazilian Integrity System. The SI, as it has become known, is composed of judicial and law enforcement institutions targeting legal compliance in cases of corruption and money laundering. “The term is related, in political science, to the study of the quality of democracy and to the vision of preserving the integrity of public administration,” says Moisés, who points to the system as being responsible for unleashing Operation Lava Jato (“Car Wash”) in 2008. To understand how Brazilian institutions have been confronting impunity since that time, researchers’ efforts have been divided along two main axes: mapping the flow and duration of police investigations and criminal proceedings involving crimes of this type and identifying the profile of the actors operating within the SI.

With the aid of information technology, data were collected from more than 4,000 district court cases involving crimes such as active and passive corruption, influence peddling, and money laundering. All these cases were classified according to one of six possible outcomes: active (when the case is still in progress), charges dismissed, conviction, case dismissed without a decision on the merits, mixed results (conviction and acquittal), and statute of limitations expired. In the judicial courts of São Paulo and Rio de Janeiro, from which the largest number of cases in the sample came—1,625 and 1,010, respectively—3% of the cases had exceeded the statute of limitations. The highest proportion of these cases, 10%, occurred in the Judicial Courts of the Federal District. In 3 out of 31 cases, the state lost the opportunity to prosecute those being investigated for the aforementioned types of crimes due to the long duration of the cases.

“The study is a milestone in the debate over impunity in Brazil,” says political scientist Rogério Arantes, also from the USP Department of Political Science and a researcher at the Institute of Advanced Studies (IEA-USP). “Surely, in the opinion of the Brazilian public, which believes justice isn’t working because criminal proceedings run out the statute of limitations, a much higher percentage would be expected than that found by the study,” Arantes says. Sociologist Ludmila Ribeiro, a professor in the Department of Sociology and a researcher at the Center for the Study of Crime and Public Security (CRISP) at the Federal University of Minas Gerais (UFMG), believes that although the index is below what might be expected, it is not good news. “Exceeding the statute of limitations is something that shouldn’t happen. It’s the state demonstrating all its inefficiency: the agents open a case but lose the opportunity to punish the person responsible for a certain crime because they take too long,” Ribeiro observes.


Comprised of judicial and police institutions targeting compliance with the law in cases of corruption and money laundering

• Active and passive corruption
• Influence peddling
• Money laundering

• Active, when the case is in progress
• Charges dismissed
• Conviction
• Dismissed without a resolution of the merits
• Mixed results (conviction and acquittal)
• Statute of limitations expired

Analyzing the proportion of time taken by each stage of the proceedings in one court case—which takes 6.5 years, on average, to complete—the study concluded that the greatest amount of time is dedicated to the judicial fact-finding phase of the process, i.e., the evidentiary hearing by the judge. According to Ribeiro, who, like Arantes is not part of the study’s research team, the data show that cases expiring due to statutes of limitation are not caused by the introduction of legal remedies by the defense in order to intentionally produce delays in the process but by the inefficiency of the judiciary itself.

Considered high, the case dismissal rate caught the researchers’ attention. In the domain that was analyzed, one in every five cases was closed without a judicial decision. “Strictly speaking, it cannot be said that dismissal means impunity because the state did, in fact, act in that case: after the police investigation, a member of the Public Prosecutor’s Office or a judge decided to dismiss the case. The rate is quite high, but it may be related to merely technical issues not associated with impunity,” notes Fernando Corrêa, a data scientist and the ABJ’s technical director. In Arantes’s view, the question of why cases are dismissed should be answered by future investigations.

The survey also analyzed more than 3,000 judicial decisions in the superior courts in the same states of Alagoas, Rio de Janeiro, São Paulo, and in the Federal District. The objective was to understand whether cases with jurisdictional privilege—which allow cases involving public authorities to be tried directly in the appellate courts—lead to bottlenecks. The study’s conclusion was that approximately 45% of the requests for jurisdictional privilege are denied, which, according to the researchers, would be an indication of the system’s inefficiency. This conclusion is reached because the reasons for the denial of a request are already stipulated in the legislation—for example, when the defendant is removed from office and the case returns to the lower court or when it is a case that is under the Electoral Court’s jurisdiction. “If the rules were different, maybe the system would be more efficient. That is, there would be fewer interruptions in processing cases, which in turn could lead to faster investigations,” says Corrêa.

Statistics regarding the Federal Police also caught the attention of the researchers. Over an average investigation period of 936 days, approximately 2.5 years, 95% of the 3,885 investigations that were opened and closed between 2003 and 2018 were resolved. Guilty parties were identified in 38% of these cases, while, for 57%, it was concluded that despite the charges filed, no crime had been committed or the individual accused was not in fact responsible. Although the overall data suggest high efficiency, the numbers are viewed with some reservations. For UFMG sociologist Ribeiro, it would be necessary to investigate what actually happened in these cases. “Since these police investigations are classified—and not public like most judicial proceedings—we can’t know why there are so many allegations that don’t result in criminal charges or indications of guilty parties,” says Ribeiro.

Cases should not run out the statute of limitations. Such a situation demonstrates the state’s inefficiency, says Ludmila Ribeiro

Actors in the debate
The second axis of the study, aimed at identifying the profiles of actors who fight against corruption, was developed with the application of the Q methodology, which the research coordinator defines as a “tool for capturing subjectivity.” The methodology consists of an electronic form with statements about the area of activity researched, in this case the criminal justice system and corruption. When filling out the forms, each respondent rated the information in two ways: by organizing each statement according to the relevance they assigned to it and by indicating the degree to which they identified with it on a scale ranging from -5 (completely disagree) to +5 (completely agree).

Some statements were of a general nature, such as “the minimum wage in Brazil is fair,” “the law is equal for all,” and “poverty and inequality are at the root of corruption.” Other statements specifically aimed at legal mechanisms, e.g., “prison detention during appeals combats corruption,” “pretrial detention is unfair,” “’coercive conduction’ [forcibly bringing a subject of interest to formal police questioning] violates individual rights.” The objective was twofold: to understand the respondents’ views regarding their activities and the way those activities were introduced into society. According to political scientist José Veríssimo, a researcher at NUPPs, “by inviting participants to assemble a fairly complex map of the circumstances and opinions held regarding their work, the Q method highlights the way they understand their daily activity.” The set of statements was prepared collectively by the research team, which proposed two hypothetical types of actors: garantistas-contratualistas [those who believe in contractual guarantees regarding the constitution] and garantistas-igualitaristas [believers in egalitarian constitutional guarantees]. These two types are defined, respectively, as those who prioritize individual rights before the law and those who “recognize that structural inequalities challenge Justice in the sense that the law should be ‘equal for all’.”

Although it was sent to 1,842 recipients, including judges, local and state prosecutors, and Federal Police investigators, the questionnaire was answered by only 40 people. The data then went through two phases of analysis. First, the two ideal types, designed while drafting the questionnaire, were adjusted, allowing the team to establish some correlations: among the 27 who fit the “contractual guarantees” profile, it is believed that “prison detention during court appeals combats corruption, that procedural slowness leads to impunity, and that multiple appeals cause the statute of limitations to expire,” according to the research report. The 13 who fit the “egalitarian guarantees” profile do not believe that faster methods and procedures will bring about effectiveness and guarantee justice; these people “prove to be contrary, for example, to coercive conduction, are less critical of politics as a factor in controlling the judiciary, and highly favor a structural evaluation of the circumstances surrounding criminal contexts.”

In the second phase, the data obtained from the questionnaires were combined with the first part of the research; that is, the researchers observed the progress of court cases under the responsibility of the judges who had responded to the questionnaire. Among the “egalitarian” judges, 8% of cases were closed due to the expiration of the statute of limitations, while this rate was 3% among the “contractual” judges; however, the latter group dismissed more cases (19% as opposed to 8%) and convicted fewer people (40% vs. 46%). The group under study was small, and the method itself was a nonrepresentative, inductive tool that related beliefs and behaviors. For these reasons, the researchers warn that the conclusions cannot be extended to the entire Brazilian justice system. “The Q method indicates that the way the group expresses issues encountered in professional practice is a trend within the community being analyzed,” explains Moisés.

For Veríssimo, the results reflect current legal debates, such as detention after criminal conviction in the appellate court [some types of cases in Brazil allow those who are convicted to remain free until all appeals are exhausted], which is now under discussion in the Federal Supreme Court (STF). “The question that summarizes the content of the questionnaire could be: what kind of guarantees to the individual, at what costs to individuals, can be applied to the judicial process? This is the debate the judiciary is having today,” explains the NUPPs political scientist. For Arantes, too, the two profiles found coincide with the principal debates around how to combat corruption: “One group of actors believes that justice should function according to procedural rules and guarantee to all involved the full right to act that these rules confer. For the other group, the justice system must embark on something new, going beyond the rules in search of an ultimate objective.”

Developed over approximately one year, beginning in 2018, the study is innovative in the way it conducted an unprecedented survey to gather procedural information with the aid of computational tools. These tools are described in detail in the report available on the CNJ website as part of the Justice Research series. “Among the research developed by the Brazilian Association of Jurimetrics, this is an atypical study. There are several layers of interpretation and a lot of material for discussion here,” Corrêa concludes. In addition to the final report, the survey data can also be found on the ABJ website.

Published in June 2019

MOISÉS, J. A. (coord.) Criminal Justice, impunity and the statute of limitations. National Council of Justice. Online. 2019.