From the Banestado Case to Operation Car Wash: Building an Anti-Corruption Institutional Framework in Brazil ^*

Do Caso Banestado à Operação Lava-Jato: Construindo uma Estrutura Institucional Anti-Corrupção no Brasil

De l’Affaire Banestado à l’Opération Lava-Jato: Construire une Structure Institutionnelle Anticorruption au Brésil

Del Caso Banestado a la Operación Lava-Jato: Construyendo una Estructura Institucional Anticorrupción en Brasil

ABSTRACT

The article analyzes the development process of a new anti-corruption institutionality in Brazil, based mainly on judicial tools and on the legal system – police officers, prosecutors, and judges – that finds its highest expression during the Operation Car Wash. The hypothesis is that the new anti-corruption institutionality was built in the first term through justice policies coordinated by the Federal government, but was taken over by the Public Prosecution’s Office and the Judiciary. The article highlights the main characteristics of the new anti-corruption institutionality and introduces its historical construction, frame by frame, based on the evolution of task forces in Brazil from the Banestado Case to Operation Car Wash.

task-force; corruption; democracy; accountability; institutional change

RESUMO

O artigo analisa o processo de construção de uma nova institucionalidade de combate à corrupção no Brasil pela via judicial que encontra na Operação Lava Jato sua maior expressão. A hipótese é de que essa institucionalidade tenha sido construída, em um primeiro momento, por uma nova política de justiça coordenada pelo governo federal, mas que foi sendo assumida pelo Ministério Público e pelo Poder Judiciário. Os resultados apontam avanços analíticos, na medida em que revelam as principais características da nova institucionalidade de combate à corrupção e apontam para sua construção histórica, cenário a cenário, a partir da evolução das forças-tarefa no Brasil, desde o Caso Banestado até a Lava-Jato.

força-tarefa; corrupção; democracia; accountability; mudança institucional

RÉSUMÉ

L’article analyse le processus de construction d’un nouveau système institutionnel de lutte contre la corruption au Brésil à travers le système judiciaire qui trouve sa plus grande expression dans l’opération Lava-Jato. L’hypothèse est que cette institutionnalité s’est construite, dans un premier temps, par une nouvelle politique de justice coordonnée par le gouvernement fédéral, mais qui était assumée par le Ministère Public et pour le Pouvoir Judiciaire. Les résultats pointent vers des avancées analytiques, dans la mesure où ils révèlent les principales caractéristiques de la nouvelle institutionnalité pour lutter contre la corruption et pointent vers sa construction historique, scénario par scénario, de l’évolution des task forces au Brésil, de l’affaire Banestado à Lava-Jato.

task force; corruption; démocratie; responsabilité; changement institutionnel

RESUMEN

El artículo analiza el proceso de construcción de una nueva institucionalidad de combate a la corrupción en Brasil por la vía judicial que encuentra en la Operación Lava Jato su mayor expresión. La hipótesis es que esa institucionalidad fue construida, en un primer momento, por una nueva política de justicia coordinada por el gobierno federal, pero que fue siendo asumida por el Ministerio Público y el Poder Judicial. Los resultados señalan avances analíticos, en la medida en que revelan las principales características de la nueva institucionalidad de combate a la corrupción y apuntan para su construcción histórica, escenario por escenario, a partir de la evolución de las fuerzas tarea en Brasil, desde el Caso Banestado hasta el Lava Jato.

fuerza tarea; corrupción; democracia; accountability; cambio institucional

INTRODUCTION

Although Brazil has recently been facing a deep political crisis, the political science literature points out that, since the early 2000s, it improved its democratic regime, both in procedural and in substantial terms. Inequality rates improved; free and regular elections took place, and a participatory institutional framework was built, which reinforced democracy in the sense that it included several new players and themes in the political arena ( Avritzer and Marona, 2017 ). In addition, an accountability institutional network was built as a result of a set of institutional changes within the state’s bureaucracy ( Power and Taylor, 2011 ; Praça and Taylor, 2014 ).

In line with international parameters, different administrations had led an anti-corruption agenda since the re-democratization ( Taylor, 2019 ). Specifically, the Workers’ Party (Partido dos Trabalhadores – PT) administrations established the model of independent anti-corruption agencies ( Sousa, 2010 ). In this sense, we assume that the new bureaucratic structure, expressed by the task forces, reveals the (Federal) government’s ability to induce institutional changes (Cingolani, Thomsson and De Crombrugge, 2015). Nonetheless, the government’s potential to lead the anti-corruption agenda and institutionalize cooperation between the Judicial Branch and several other judicial and quasi-judicial agencies decreased gradually.

Generally speaking, institutional reforms are not straightforward. No institution can induce solely positive aspects from its players’ performance. Agencies will always bring the moral hazard to the table. Likewise, any institutional design may lead to some undesired or unpredictable effects. Our argument is that the government has reduced its capacity to coordinate and shape the anti-corruption agenda, which led to adverse consequences for democracy.

In this regard, this article aims to track the new anti-corruption bureaucratic structure, expressed by the task forces, pointing out that it resulted from a set of institutional changes specifically induced by (i) a “justice policy” developed during the Workers’ Party administrations (2003-2016) and (ii) several changes within the judicial system. We will also examine the disconnections between exogenous variables - such as government policies - and endogenous ones - institutional learning process in the Public Prosecutor’s Office (Ministério Público Federal – MPF), the Federal Police, and the Judiciary - at the root of the gradual replacement of political players by judicial ones in shaping the fight against corruption. These disconnections result in negative externalities to democracy.

As such, this article fits into the debate on institutional change. We assume that theoretical models are not alternatives, but a mosaic of options. We consider exogenous and endogenous sources of institutional change. We also take into account that institutional change can occur incrementally and may allow adjustments in the policy itself: there is an attempt to re-establish institutional balance in the face of external shocks ( Peters, 2012 ), which, however, depend on how institutions perceive and adapt internally to them ( Mahoney and Thelen, 2010 ; Peters, 2012 ).

We argue that the results that the creation of an accountability network produce are not always positive for democracies, a fact that, by and large, is not sufficiently explored in the literature. Power and Taylor (2011) and Taylor (2019) , for example, overlook the negative externalities to accountability. On the other hand, Avritzer and Filgueiras (2011) caution about the risks involved in approaching corruption from the perspective of criminal justice and of over-empowering a public, non-governmental authority. Thus, we raise the hypothesis that the levels of autonomy and discretion attained by the players of the judicial system in the structuring of an anti-corruption agenda in Brazil played a role to erode the political system.

The structure of this article is as follows. The next section seeks to identify and classify the set of conditions (or causes) of the institutional change process, highlighting all the elements that make up the sources of exogenous and endogenous changes. Subsequently, we review three key anti-corruption cases (the Banestado Case, Satyagraha Operation, and Operation Car Wash) to dynamically illustrate the institutional changes. In this respect, we propose different scenarios considering (1) the construction of a legal framework that considers corruption as organized crime and reinforces the state’s criminal punitive force and threatens civil rights; (2) the streamlining of the accusatory dimension of criminal justice; and (3) unaccountable anti-corruption agencies framework. The final section suggests how these cases may help us to understand the new anti-corruption institutional framework as a result of several institutional changes. We will present some concluding insights about the relation between the latter and the fading of Brazilian democracy.

BUILDING TASK FORCES: INSTITUTIONAL CHANGE AND A POST-88 ANTI-CORRUPTION AGENDA IN BRAZIL

A growing literature has pointed to the institutional evolution of Brazil’s accountability network, charting it in an incremental theoretical model of institutional change ( Sola and Whitehead, 2006 ; Armijo, Faucher and Dembinska, 2006; Taylor, 2009 ). Consequently, although it examines the creation or reform of key institutions and even take into account the exogenous efforts to enhance their performance, specialists usually point out other important features of the transformative processes, whereby gradual institutional changes build new resources and roles ( Galvin, 2012 ).

In this regard, the Brazilian case is taken as a “complex, interdependent evolution of institutions in an accountability web” ( Praça and Taylor, 2014: 27). Drawing on this assumption, we shall present the task forces as the institutional expression of the anti-corruption agenda, embedded, in turn, in a broader accountability web present in Brazil.

The very first federal task force occurred in 2003 with the Banestado Case. Since then, there were more than 3,000 task forces. They soared from 18 in 2003 to 3,512 in 2016 ( Marona and Barbosa, 2018 ). The most significant example is Operation Car Wash ( Operação Lava Jato ), and its most emblematic trophy is the arrest of former president Lula (2003-2010). However, the task forces express more than just a coordinated effort amongst the Federal Police, the Public Prosecutor’s Office, and the Judiciary.

Networks of institutions become a source of transformation that generates evolutionary pressures as one institution stimulates another (either through cooperative or competitive processes) to improve in areas that influence their conjoint effectiveness; supports and defends the other against external pressures; and generates new ideas for further change ( Praça and Taylor, 2014: 28).

In this sense, we argue that they reveal a new institutional set-up, which reached its height during Operation Car Wash. Our analysis focuses on the investigative and punitive scope of the Brazilian accountability web ( Praça and Taylor, 2014 ). To do this, we will map the various factors that mutually implicate themselves in the processes of institutionalization of the anti-corruption agenda. We shall begin by classifying them into two axes of analyses, according to their sources. As shown in the diagram below, the first axis gathers exogenous sources together under the label of Workers’ Party’s justice policy. It combines the National Strategy to Combat Corruption and Money Laundering ( Estratégia Nacional de Combate à Corrupção e Lavagem de Dinheiro - ENCCLA) initiative, a new way to appoint the head of the Public Prosecutors’ Office, the increase of public investment in the Federal Police, and the strengthening of the State criminal punitive force, represented by the streamlining of criminal proceedings (criminal legislative changes).

The second axis expresses a set of endogenous processes of institutional change in the Brazilian Judicial System: the specialization of the Federal Judiciary associated with the new jurisprudence of the Brazilian Supreme Court. The initiatives of the Public Prosecutors’ Office must also be outlined; namely, the (self-)authorization to conduct a criminal investigation, the attenuation of the principle of “natural judge”¹ , and the discretionary use of the plea bargain.

Diagram 1
Conditions for the Task Forces

In keeping with the exploratory character of this work, it is worth highlighting that the combination of factors presented constitutes a “necessary” cause for the result observed: the rise of Operation Car Wash. These factors favor the outcome, but do not guarantee it. Conditions of this kind produce indirect effects ( Mahoney, 2008 ), which means that the result would not have occurred in their absence, even if their presence does not ensure the same result per se (Mahoney, Kimball and Koivu, 2009). Let us explore each one of them carefully.

The Workers’ Party Justice Policy

The category “justice policy” rests on two pillars. On the one hand, it draws on studies in the field of public policies ( Kingdom and Stano, 2011 ) that allow the association of the primacy of actors and judicial institutions with the various stages of the “public policy cycle” of fighting corruption – from the conception of the agenda, formulation of the policy, to implementation ( Oliveira, 2019 ). On the other hand, it draws on the concept of “policy burts” – political interventions that have immediate effects, but which can fade away if abandoned or not exploited ( Taylor, 2019 ). To justice policy we ascribe the sense of public policy, understood here, in general terms, as a set of practices and laws, which provide a guideline for the efforts of the players and organizations of the judicial system. Our focus is on the justice policies of the Workers’ Party administrations, which are part of an attempt at building an accountability web, which started on president Lula’s first term of office. This web is, on the one hand, broad, since it involves all three Branches of government, and on the other, complex, given that the institutions that it connects are responsible for inspecting, restraining, controlling and punishing abuses.

What is not clear, however, is if the justice policy in Lula I was linked to that anti-corruption agenda which had its climax in Operation Car Wash. We understand that Operation Car Wash is the apex of the institutionalization of a certain anti-corruption agenda. This agenda is embedded in the process of building an accountability web in which judicial system institutions increasingly play a leading role not only on the implementation, but also on the design of the anti-corruption public policy. Government actions to reinforce the accountability network seem to favor preventive rather than punitive approaches² . While the former is ex ante , the latter is ex post and more directly linked to the actions of players and institutions of the judicial system. The analysis of the evolution of the treatment given to the anti-corruption agenda in the government programs of the Lula and Dilma Rousseff campaigns supports this statement. Not until the government plan for Lula’s re-election did the anti-corruption agenda merge with the processes of institutional change and highlight the punitive dimension of the accountability web, giving priority to the players and institutions of the justice system. Therefore, the construction of a justice policy that involved increasing the autonomy and expanding the investments on judicial institutions was built over the years in connection with the justice system itself.

(A) The National Strategy to Combat Corruption and Money Laundering (ENCCLA)

ENCCLA is a well-known case of formal institutional change that resulted from exogenous factors ( Praça and Taylor, 2014 ). The Ministry of Justice created ENCCLA in 2003, an indication of the government’s concern about effective networking. The strategy focused initially on fighting money laundering. As of 2006, by recommendation of the Court of Auditors ( Tribunal de Contas da União - TCU) and the consent of the Controller General ( Controladoria-Geral da União - CGU), the ENCCLA started to take part in fighting corruption. In other words, the issue of corruption, as a strategy, only started to figure prominently in Lula’s second term. Over the years, this initiative has constituted the main articulation network between the Executive, Legislative and Judicial Branches, as well as the Public Prosecutor’s Office, for the formulation of policies aimed at combating money laundering and corruption. ENCCLA’s structure, goals, actions, and results highlight the importance of building a new judicial bureaucratic network to fight corruption. The result was an increase in the volume of integrated operations. All the primary control bodies of the Brazilian State take part in the ENCCLA initiative.

The activities that ENCCLA develop involve the production of legal-normative diagnoses, the organization of databases, the formulation of legislative proposals, and the promotion of events. Among the main results of ENCCLA there are: (i) the National Training Program to Combat Corruption and Money Laundering (PNLD), accessed by more than 18,000 public agents from all regions of the country; (ii) the preparation of the Anti-Corruption Guidelines Plan; (iii) the creation of Specialized Offices in Financial Crimes, within the scope of the Federal Police Department; and, (iv) the establishment of the National Group for Fighting Criminal Organizations, within the range of the Public Prosecutor’s Offices in the states.

ENCCLA is one of the conditions for the rise of task forces because it induces cooperation among anti-corruption agencies. ENCCLA also became very important in the streamlining of the criminal law process: its members have presented several proposals on the financial system, criminal organizations, money laundering, loss of assets related to illegal acts, the statute of limitations, lobbying, banking secrecy, administrative improbity, among others. Many of those proposals became pieces of legislation.

It is worth highlighting the new legislation on money laundering (Law12.683/12) that ENCCLA supported. The new bill expanded the possibility of fighting money laundering because it made it non-dependent on a previous offense. Agents in the judicial system criticized the legal requirement, instituted in the late 1990s, for some other crime to pre-exist the actual money laundering. In their view, the law did not allow them to achieve high enough benchmarks in battling this crime. This objection gave rise to the two new bills that resulted in the new legislation. The first one, in 2003, was in the Senate, and the second in the House of Representatives, in 2008.

This new law, owing to the ongoing process of specialization of the Federal Police, which we address below, had procedural impacts. The centralization of the money laundering trials by specialized courts weakens the constitutional principle of “natural judge”. According to this principle, a judge is assigned before the trial, regardless of who is being judged. That way, the chances of the judging body being unbiased and independent should increase, thus assuring the rights of the defendants. This aspect is, at least partially, lost with the new Money Laundering Law.

The ENCCLA was a key factor to build the anti-corruption task forces. Its creation assured the prominence of the judicial elite not only in implementing the anti-corruption agenda, but also in its formulation. This established a new institutionality that gradually came to reflect the mindset of the judicial players (police deputies, prosecutors and judges). Diagram 2 below summarizes the main indicators of ENCCLA’s capability of inducing cooperation among judicial players, bodies and institutions of the judicial system, which creates a new form of institutionality and imparts a new mindset regarding the parameters of the anti-corruption agenda in Brazil.

Diagram 2
ENCCLA as a condition to Task-Force emergence

(B) The Federal Police

The institutionalization of the Federal Police expresses a revolutionary institutional change prompted by exogenous causes, which facilitated the structuring of the task forces in two ways. First, both the budget and the staff of the Federal Police increased significantly during the Workers’ Party administrations. In 2011, the Federal Police had just over 11,000 active members, of whom 4180 (38%) joined after 2002 ( Arantes, 2011 ). From 2003 to 2016, government investments in the Federal Police increased by 2,733% ( Marona and Barbosa, 2018 ).

In addition, there was also a restructuring of the Federal Police career and it achieved de facto autonomy, although not de jure ( Arantes, 2011 ). The 1988 Constitution adopts a model that combines investigative autonomy of the Federal Police with administrative subordination to the government³ . Several government initiatives increased the autonomy of the Federal Police in the constitution of their staff and their professionalization. This has affected the dynamics of the workings of the public administration’s accountability network, facilitating a new standard in the fight against corruption ( Avritzer and Marona, 2017 ; Kerche and Marona, 2018 ).

Combined, those initiatives have resulted in the improvement of the Federal Police’s institutional capacity. The rise of this “new Federal Police” contributed to a more collaborative anti-corruption institutional environment ( Arantes, 2015 ). However, it is interesting to note that the strengthening of the Federal Police was not necessarily linked to fighting corruption in the beginning of PT’s government.

A cross reading between PT’s government programs suggests that reinforcement of the Federal Police was justified in different ways over time. They were at first linked to the need for border control (Lula I). During Dilma Rousseff’s re-election campaign, they were linked to fighting organized crime and money laundering. On the other hand, in her re-election campaign, the Federal Police was explicitly associated with the anti-corruption agenda.

This is an important indicator to understand the expansion of the Federal Police’s operating limits during the PT administrations. Nevertheless, the construction of the anti-corruption agenda has to be looked at from an endogenous perspective. Besides the changes that the administration produced, one must consider the institutional learning that arises from the (often uneasy) interaction with other relevant players. This way, it is possible to point out some concurrency between the government aims (reinforce the institutional capacity of the Federal Police) with the institutional development carried out internally since the 1990s ( Fagundes e Madeira, 2021 ) by the organization itself. However, one should not overlook the importance of ENCCLA. The strategy is a turning point in the construction of the anti-corruption agenda because it encouraged the coordination of the Federal Police with the other institutions of the judicial system, allowing it to take part in the policy modeling.

(C) The Public Prosecutor’s Office

Until the 1988 democratic Constitution, the Public Prosecutor’s Office was an agency subordinate to the government. The framers withdrew the agency from direct control of the Minister of Justice and guaranteed the autonomy of the Public Prosecutor’s Office concerning the Legislative and Judicial Branches as well. Besides, the internal hierarchy of the Public Prosecutor’s Office is weak: the head of the Public Prosecutor’s Office ( Procurador-Geral da República ) does not control promotions, cannot dismiss prosecutors, and cannot choose a handpicked prosecutor for a particular case ( Kerche, 2009 ). Both external and internal autonomy have been crucial aspects to understand the critical role of the prosecutor in the Brazilian public scene since the beginning of the re-democratization.

This institutional design that the Constitution defined, alongside the institutional instruments that the prosecutor’s control, such as the civil action, enabled the Public Prosecutor’s Office to defend collective and social rights. Until the Workers’ Party administrations, prosecutors’ efforts in the fight against corruption could be classified as more parsimonious and not based on criminal action ( Arantes, 2015 ).

The judicial fight against corruption can be carried out through civil legal instruments, making primary use of the Administrative Improbity Act (LIA) or through criminal prosecution, based on the Brazilian Penal Code. During the 1990s, the context comprised by judicial delays, rigidity of the accusatory system, police’s low efficiency, and the Public Prosecutors’ Office interdiction for leading criminal investigations encouraged the civil path instead of the criminal one ( Arantes, 2015 , 2018 ). However, this scenario has been changing since 2003. During the Workers’ Party administrations, the criminal approach to corruption became more attractive due to the increase of the Public Prosecutors’ Office autonomy and discretion in the criminal field.

The Public Prosecutor’s Office is the gatekeeper of the judicial system, selecting and prioritizing issues that will be ultimately decided by judges. Therefore, the institutional position of the Public Prosecutor’s Office is central to build the anti-corruption strategy that has developed in Brazil in recent years.

The Public Prosecutor’s Office has experienced both exogenous and endogenous institutional changes. Although the Public Prosecutor’s Office has remained as an auxiliary agency of the judicial system, prosecutors have made the issue of corruption a top priority. In other words, the Public Prosecutor’s Office, which was created to be a social rights defender and an ordinary criminal prosecutor, became an anti-corruption agency – at least concerning its public image.

One of these changes concerns the appointment process of the head of the Public Prosecutor’s Office. President Lula, without amending the legislation, which envisaged the appointment by the president to be approved by the ever-friendly Senate, decided to accept the most voted name by the members of the agency itself. Thereby, the primary voter of the head of the Federal Prosecutor’s Office is no longer the president, but his or her peers.

The adoption of this new appointment process has had a twofold effect. First, it reversed the logic of the relationship between the Public Prosecutor’s Office and the Presidency. Instead of a chief prosecutor (who is the only one who can prosecute the president and his ministers) seeking to please the president in order to be reappointed, now the chief prosecutor is more focused on the demands of his constituents, the federal prosecutors themselves. This change, therefore, allows for the chief prosecutor to be more independent concerning the Executive Branch’s interests ( Kerche and Marona, 2018 ).

Another effect was the change in the internal balance of power that resulted in the organization of a new majority field around the anti-corruption agenda ( Marona and Barbosa, 2018 ). Although the framers of the democratic Constitution had decided that the Police Officers are in charge of conducting criminal inquiries ( Kerche, 2009 ), part of the members of the Public Prosecutor’s Office did not observe the rule at various moments. The National Council of the Public Prosecutor’s Office⁴ has even authorized, through a quasi-legislative decision, the prosecutors to conduct criminal inquiries, behind police’s backs. In 2015, the Brazilian Supreme Court finally authorized that “heterodox procedure”, and the prosecutors became criminal investigators on top of all other duties that the framers of the 1988 Constitution framers originally prescribed.

Thus, investigations that the Public Prosecutor’s Office members conduct increase prosecutors’ discretion in criminal matters. Ensuing the Brazilian Supreme Court authorization, prosecutors can select and prioritize criminal cases regardless of the Federal Police expertise and, to some extent, the Judiciary’s intervention. This new scenario breaks with the division of functions between the Police and the Public Prosecutor’s Office such as provided by 1988 Constitution, which favored a sort of checks and balances within the judicial system.

In addition, the investigation of criminal subjects by prosecutors undermine the “principle of legality” that the Brazilian law provides. Unlike the “principle of opportunity”, in which the prosecutor decides whether a case deserves to be tried by the Judiciary, Brazilian law requires that a judge must always be consulted in face of a criminal prosecution proposal.

We should point out another crucial change concerning the expansion of the autonomy and discretion of the Public Prosecutor’s Office that further threatens the “principle of natural judge” ( Arantes, 2002 ). That constitutional principle prevents the chief prosecutor from replacing a prosecutor during an investigation and prosecution process. Once again, the Public Prosecutor’s Office bypassed the Constitution and created groups of prosecutors specializing in different issues (corruption control, combating drug trafficking, and so forth), undermining the principle of “natural judge”. In short: the high-level bodies of the agency began to select prosecutors of their preference to act in particular cases. The non-observation of the principle of the natural prosecutor is adopted systematically in the anti-corruption task forces.

Lastly, we would like to address the particular use of the plea bargain by prosecutors. This relatively new legal instrument, in the prosecutors’ hands, also reinforced their discretion in criminal matters. Prosecutors could negotiate the reduction of sentences, which would encourage whistle-blowing. Several criticisms can be made to the use of the plea bargain when we consider the fundamental rights and guarantees provided by the 1988 Constitution. Nevertheless, we would like to highlight more concretely the lack of criteria and limits in its use by the Public Prosecutor’s Office.

The plea bargain is a persecutory tool originally linked to the fight against organized crime. The Public Prosecutor’s Office made use of it according to that spirit: prosecutors mobilized the plea bargain to build a narrative that brings corruption closer to organized crime ( Avritzer and Marona, 2017 ). Furthermore, in Brazil, the plea bargain does not allow as wide a deal with the criminal as it does under US law, for instance. However, the Public Prosecutor’s Office has actively handled the plea bargain in Brazil, sometimes even negotiating advantages for the whistle-blower not provided by law.

On account of all these changes, the Public Prosecutor’s Office has become a crucial agent to investigate, prosecute, negotiate convictions, and encourage whistle-blowing, without the counterpart of democratic accountability ( Kerche and Marona, 2018 ). In summary: prosecutors are more discretionary than provided by the 1988 Constitution, and the principle of legality is partially overwritten by the principle of opportunity because they 1) designate the head of the Public Prosecutor’s Office; 2) conduct separate criminal investigations; 3) bypass the principle of natural judge; and 4) may negotiate plea bargains and impose their corruption narrative.

(D) Criminal Legislative Changes: Plea Bargain

Among all of the designed instruments, the plea bargain has been most widely used, especially in Operation Car Wash. This new legal tool was “shaped under the discourse of state inefficiency” (owing to the increase of crime and its sophistication) and “coined to respond to the demands of society in the field of public security” ( Avritzer and Marona, 2017: 385).

The evolution of the plea bargain is part of a strategy to create “emergency criminal laws” ( Ferrajoli, 2014 ), which is a policy designed to meet immediate goals, offering political answers to social pressures. This kind of legislation tries to offer easy paths to majority portions of society, rather than privileging the rule of law and democratic principles.

These instruments began to be introduced in Brazil in the 1990s, but have become more systematic in the Criminal Organizations Act ( Lei das Organizações Criminosas ), in response to the protests that erupted in 2013, known as the June Journeys ( Jornadas de Junho ), addressing different demands, including the fight against corruption. Gradually, there was an extension of advantages and guarantees to whistle-blowers. The number of plea bargains has increased significantly over the last few years, especially in task forces. In 2003, during the Banestado Case, the first plea bargain was negotiated in Brazil. During Operation Car Wash, just over a decade later, more than 290 deals of this type were made, especially by prosecutors.

The law requires that, in a plea bargain, the collaboration must be spontaneous, the whistle-blower must have participated in the infraction, and that the information has to be relevant and useful. The benefits can be legal pardon, reduced sentence by up to two thirds, and alternative sentences instead of incarceration.

This instrument reinforces the new strategy to fight corruption: the role of the Public Prosecutor’s Office in the criminal field, the weakening of the division of competences among judicial players and handling corruption as organized crime. The plea bargain involves the informational disadvantage of the defense concerning the prosecution, rigidity, and severity of criminal laws and the prosecutors’ discretion without internal and external accountability.

This legal tool contributed to an increase in the number of temporary and preventive arrests by task forces. In only three years, this kind of arrest has quadrupled ( Marona and Barbosa, 2018 ). The plea bargain has impacted the performance of judicial agencies in the fight against corruption leading to the streamlining of criminal legislation and, consequently, reinforcing the new anti-corruption institutional framework expressed by the task forces.

The Brazilian Judicial Branch: Endogenous Process of Institutional Change

Owing to their constitutional design, the agencies and players that make up the Judiciary operate in a more independent environment. Their degree of freedom cannot be compared to the other judicial bodies reviewed here. Usually, in democracies, the Judiciary is an independent Branch of the government, while prosecutor’s offices and police forces are subordinated to it. Thus, we can expect that magistrates will have greater capacity to produce endogenous change than, for example, Federal Police Deputies. The specialization of the Federal Judiciary, for instance, was conceived by the Judicial Branch despite any government or legislative agenda (Madeira and Gelisky, 2019). In addition, the matter of judicial interpretation is among the endogenous changes brought about by the Judiciary (Elkins, Ginsburg and Melton, 2009). Thus, “judicial interpretations may merely provide legal cover and legitimacy for what is clearly a rule violation, or maybe manipulated to produce frequent changes in response to changing preferences” (Brinks, Levitsky and Murillo, 2019:25). Let us look at them carefully.

(A) The New Brazilian Supreme Court Jurisprudence in Criminal Issues

The performance of the Brazilian Supreme Court was decisive to strengthen the State criminal punitive force. Besides the streamlining of criminal proceedings, some very important changes in criminal jurisprudence have contributed to this. The legal choices that the Supreme Court justices took in the trial of Criminal Action 470, known as Mensalão ⁵ , reveal a significant change in the Court’s position concerning the guarantees contained in criminal legislation and criminal procedural law, especially in corruption related criminal trials ( Avritzer and Marona, 2017 ).

The stretching of the “privileged forum”⁶ in addition to the judicial strategy of classifying defendants as members of a criminal gang to increase the sentences, the application of Roxin’s theory ( Teoria do Domínio do Fato )⁷ was a hallmark of the Mensalão trial and other judicial decisions on corruption issues.

The Brazilian Supreme Court decision concerning the “privileged forum” proved to be crucial for retrieving its institutional leadership within the Judicial Branch ( Arantes, 2018 ). Furthermore, the reasons for the decision reinforced the prosecutors’ narrative that pointed to the existence of a ring supposedly organized around a political crime, which undermined democracy. Besides, the use of “Roxin’s theory” allowed for the creation of a counter-narrative: from the criminal fact was recreated a “web of actions, relationships and responsibilities between players” until reaching the alleged “head of the gang.” ( Arantes, 2018: 365).

Besides, the Supreme Court gave some new interpretations to criminal legislation during Operation Car Wash. Among others, the inclusion of the possibility of imprisonment before the decision on all instances of the Judiciary ( res judicata ) and the expansion of the concept of corruption. While all these innovations were important to the new anti-corruption strategy, none of them alone would have been enough to put the task forces on another level.

(B) The Specialization Of The Federal Judiciary

The creation of specialized courts in the fight against corruption is one of the strategies of the public policy agenda against corruption in many democracies ( Madeira and Geliski, 2019 ). The objective is to concentrate corruption related criminal actions in the hands of a few judges ( Sousa, 2010 ) and to attempt to build expertise in complex cases.

In Brazil, Madeira and Geliski (2019) examined the role of the Judiciary in the implementation of the anti-corruption courts, which helped the Federal government to boost ENCCLA. On account of the reduced level of implementation of money laundering legislation, the Federal Justice Counsel (Conselho Federal de Justiça – CJF) organized a study commission that proposed the specialization of the Federal Judiciary. The intent was to create specialized courts to “prosecute and judge crimes against the national financial system, money laundering or concealment of goods, entitlements and assets” (Resolution CJF 314/2003).

This specialization guaranteed a higher degree of autonomy from trial court judges in relation to the Higher Courts. As previously stated, in response, the Brazilian Supreme Court changed jurisprudence concerning “privileged forum” to assert its institutional leadership – the capacity to enact binding decisions in the face of institutional shifts regarding corruption control.

Nonetheless, the Supreme Court has gradually ceased to exercise any control over the abuses committed by the lower courts in the context of integrated operations. In 2003, in the Banestado case, Justice Celso de Mello expressed concern regarding the blurring between the jurisdictional and the persecutory activities. Similarly, Justice Gilmar Mendes classified some of Judge Moro’s decisions as harmful to the rule of law ( Marona and Barbosa, 2018: 145).

During the Satyagraha Operation (2004-2011), the Brazilian Supreme Court made interventions that are even more forceful. However, this pattern was not repeated in Operation Car Wash, where, until May 2016, only 3.9% of judicial sentences from lower Courts had been reviewed by Higher Courts. Particularly in the Brazilian Supreme Court, out of 52 judicial appeals analyzed, only 13% were in favor of the defense.

The task force’s head prosecutor, Deltan Dallagnol, and the judge, Sergio Moro, argue that this data indicates the strict legality of Operation Car Wash. We are not so convinced about that. The low review rate may also indicate the loosening of internal control of the judicial system ( Marona and Barbosa, 2018 ). Combined, all these institutional changes have resulted in a new anti-corruption institutional framework, which is most clearly expressed by Operation Car Wash. As we pointed out, this set of institutional changes has induced cooperation among anti-corruption agencies acting in the criminal field. This was also possible because of the institutional revolution that took place in the Federal Police, which contributed to reinforce the criminal prosecution strategy for dealing with corruption.

Diagram 3
Public Prosecutor’s Office, Federal Police and Judicial Branch’s Institutional Changes as a Condition for Task-Force Emergence

Operation Car Wash adopted the strategy of linking corruption to organized criminality as a consequence of legislative changes and the Brazilian Supreme Court’s criminal jurisprudence shift. In their handling of plea bargains, the Public Prosecutor’s Office has not only hurt the accusatory model of criminal justice, but it also violated the Fundamental Rights system, imparting an inquisitorial character to criminal proceedings. In addition, considering the huge institutional autonomy of the Public Prosecutor’s Office, there are few possibilities of political accountability for its actions. Finally, the specialization of the Federal Judiciary weakened the internal control within the Judicial Branch. Under this scenario, Operation Car Wash systematically undermined the rule of law and democratic principles.

The set of institutional changes that led the task forces to become a dominant strategy in dealing with corruption resulted in a frame that favors the criminal treatment of the issue. This, in turn, resulted in linking political activity to organized crime and the strategic use of the corruption agenda for electoral and political purposes (Veiga, Dutt-Ross and Martins, 2019). The consequences for Brazilian democracy have been dire. After 2013, the country began to deal with a level of instability that presently threatens the regime’s stability.

At this point, we shall trace back three key cases of anti-corruption efforts (the Banestado Case, Satyagraha Operation, and Operation Car Wash) to dynamically highlight the institutional shift. Next, we propose different scenarios, considering three variables that emerge from the consequences of the institutional changes we have pointed out: (1) a “corruption as organized crime” legal framework that reinforces the state criminal punitive force, threatening individual rights; (2) the streamlining of the accusatory dimension of criminal justice; and (3) an unaccountable anti-corruption agencies framework.

FROM THE BANESTADO CASE TO OPERATION CAR WASH: SCENARIOS OF AN ANTI-CORRUPTION INSTITUTIONALITY IN BRAZIL

Since 1999, the Brazilian Judicial System has conducted several large-scale integrated operations to fight organized crime. In recent years, this has also been extended to political parties and politicians. These operations (task forces) are “evolutionary”, but not necessarily in a linear way. Over the years, some institutional innovations created an environment that leads players of the judicial system, especially prosecutors, to act more freely.

The fight against organized crime and corruption also involves changes brought about by the players of the judicial system themselves. In this sense, one can observe several effects that both politicians and the government did not foresee. In this combination of external shocks and endogenous institutional change following in the wake of several task forces, Operation Car Wash, the expression of the new anti-corruption institutional framework, has been devastating to the Brazilian political system and economy ( Paula and Moura, 2019 ).

The first milestone in this institutional evolution was the Banestado Case. The FT-CC5 task force identified around $120 million in currency evasion and money laundering through “non-resident” accounts, called CC5. The fraud happened through different banks, among which the Bank of the State of Paraná (Banestado). Informed of the transactions, the Public Prosecutor’s Office requested the Central Bank (the Brazilian Federal Reserve) to provide details of these operations, which took place in 1998. The Central Bank refused to provide them. In response, prosecutors sought to unlock the bank details of all CC5 accounts in the country and not just suspicious transactions. The Judiciary authorized the request. The Public Prosecutor’s Office opened thousands of inquiries based on this data.

Deriving from the Banestado Case, a few years later, in 2001, the Macuco Operation⁸ took place under the command of the Federal Police. These were the first systematic investigative efforts to fight financial crimes: foreign exchange evasion, crimes against the financial system and money laundering, among others. The analysis of this operation, however, reveals the challenges in the collaboration between the Federal Police and the Federal Prosecutor’s Office, which was reduced later in the new institutional and legal scenario, particularly because of the ENCCLA initiative and the Federal Police institutional revolution itself.

In fact, federal prosecutors were very critical of the Federal Police strategy during the Macuco Operation. As a result, in 2002, the Chief Police Officer of the Macuco Operation was removed, and the investigation suspended. At this point, it is clear that the Federal Police did not enjoy de facto autonomy, which resulted in a set of institutional innovations during the Workers’ Party terms in office.

That same year, federal prosecutors obtained judicial authorization to transfer to Curitiba – the capital of the state of Paraná and the scenario for the Operation Car Wash a few years later – the majority of inquiries related to the CC-5 accounts. The following year, in the beginning of Lula’s administration, prosecutors set up a task force without other judicial agencies. Their first step was to repair the alleged failures of Police investigations, demonstrating that at that time, agency cooperation was not part of the menu. The focus of the new inquiries and the selection of the most relevant suspects indicated the adoption of the principle of opportunity by prosecutors, even without a decision by the Legislature to abandon the legality principle.

A new clash took place in 2004. This time the divergence was within the Judiciary. The judge in charge of the case, Sergio Moro, attempted “legal experiments” that later became strategic for the Operation Car Wash “success,” although they did not work very well in the absence of other institutional changes at that time. Mr. Moro authorized many temporary arrests, but the Higher Court refused some of them. The Court of Appeals alleged that these kinds of arrests decided by Mr. Moro were based on fragile legal grounds. A few years later, the same Court of Appeals ( Tribunal Regional Federal da 4ª Região – TRF4 ) accepted temporary arrests determined by the same judge and on a very similar basis, ignoring the defense’s claims on how it violated fundamental constitutional rights. It was also Mr. Moro who authorized several plea bargains and several convictions of Banestado’s staff, in a kind of laboratory for what happened years later. However, in this case, no politicians ended up behind bars.

It is important to stress that judicial tools for organized crime-fighting still were not available for dealing with corruption. The Federal Police also lacked expertise and institutional robustness. There was not, in fact, any joint anti-corruption plan, neither was there any willingness, on the part of the Supreme Court, to undermine individual rights in the name of fighting corruption. Nonetheless, derived from the Banestado Case, Operation Lighthouse on the Hill ( Operação Farol da Colina ), was launched in 2004 and fulfilled 215 warrants for search and seizure and 123 arrest warrants that Judge Moro issued.

The Higher Court overturned some of these arrests because the judge was not competent to judge crimes committed outside the state of Paraná, a practice that became commonplace during the Operation Car Wash following the Supreme Court’s jurisprudential shift. In fact, another divergence took place in 2006 among members of the Judiciary. The Brazilian Supreme Court’s Justice Sepúlveda Pertence accepted the request to release an executive accused by the Public Prosecutor’s Office, arguing that the arrest had violated individual rights due to alleged pressure to negotiate a plea bargain. In response, the judge in charge of the case ordered a new preventive detention. However, another Justice of the Supreme Court, Nelson Jobim, ordered the fulfillment of Justice Pertence’s previous decision.

Even after the investigation into Banestado’s suspicious accounts came to an end, some prosecutors remained in the case. The Prosecutor Deltan Dallagnol – who would gain significant prominence later, coordinating the Operation Car Wash – joined the team in 2005. Deltan Dallagnol could be described as a critical figure ( Baez and Abolafia, 2002 ) concerning the process of organizational changes in the Public Prosecutors Office. So, do Sergio Moro and Justice Joaquim Barbosa, rapouter of the Mensalão Scandal trial, regarding the Judicial Branch.

The task force in charge of the Banestado Case learned with the US experience of combating drug trafficking and terrorism by means of identifying the link between organized crime and economic crime. However, it was not as successful at moving into corruption fighting. Directors and former directors of financial institutions were charged, but not the politicians involved. At that point, unlike Operation Car Wash, which took place a few years later, the interface between prosecutors and Federal Police officers was ineffective. Neither of the two agencies had gone through all of the institutional innovations that were created during the Workers’ Party administrations. Even legal innovations like the plea bargain were not fully mature, which weakened the use of this kind of persecutory instrument. The Higher Courts also reviewed many initiatives of the judge in charge of the case, blocking several judicial and persecutory initiatives from the Banestado Case, the Lighthouse on the Hill Operation, the Sand Castle Task Force ( Castelo de Areia ), and the Satyagraha Operation ( Avritzer and Marona, 2017 ).

The Satyagraha case is an excellent example of the dispute between the different courts within the judicial body. This case involved the maintenance by the Opportunity Group of different sub-funds and companies to allow residents to invest as foreigners, evading the supervision of the Central Bank and the IRS ( Receita Federal ). Banker Daniel Dantas was one of the central players involved. His defense requested a preventive habeas corpus , which was denied. Thus, the Federal Police carried out search and seizure warrants and preventive detention against more than 20 suspects, among whom was Mr. Dantas. At that point, the Brazilian Supreme Court’s Justice Gilmar Mendes decided to release him.

In response, the judge in charge of the case issued another arrest warrant for Mr. Dantas, but Justice Mendes granted a new habeas corpus . In December 2008, the banker was convicted of active corruption. However, in June 2011, a Higher Court overturned all evidence obtained during the Satyagraha Operation and canceled the legal action as a whole. The issue at that point was about whether the evidence was legal and whether the concept of a criminal organization could be used as a strategy to fight corruption. To a certain extent, a few years later, the Supreme Court, in the case dubbed Mensalão , created jurisprudence that would solve the issue.

Likewise, in this regard, we should refer to the approval of the Criminal Organizations Act in 2013, by Congress, a response from Dilma Rousseff’s government to the “June Journeys”⁹ claims against politicians in general, which were regarded as corrupt. Taking advantage of this scenario, the Public Prosecutor’s Office advanced even more on its self-proclaimed investigative prerogative in the criminal field. A few years later, in 2015, the Brazilian Supreme Court would assure broad criminal discretion to the Public Prosecutor’s Office.

The Criminal Organizations Act recognized the plea bargain as a legitimate mechanism to fight organized crime, including corruption. The new Supreme Court jurisprudence and the new legislation were crucial to the success of Operation Car Wash. Previous operations did not achieve the same success because they happened in a different institutional environment.

The ease in replacing key actors involved in the Macuco and Satyagraha Operations highlights the fragile autonomy of the Federal Police vis-à-vis the Ministry of Justice. The two police officers in charge of both operations were removed from the cases. The dependence on the Executive is even more evident in comparison with the institutional position that the Public Prosecutor’s Office has occupied since 1988 and with the de facto autonomy that the Federal Police received during the Workers’ Party administrations.

Although judicial players in charge before Operation Car Wash attempted to use the plea bargain, it was just after the Criminal Organizations Act in 2013 that this legal instrument was systematically drawn upon and taken beyond financial crimes. The plea bargain has been successfully used by the judicial agencies (Police and Public Prosecutor’s Office) as part of a specific strategy of fighting corruption. In addition, new understandings from the Supreme Court were also crucial to strengthen the fight against corruption by mobilizing criminal, instead of civil, tools. The Mensalão trial, particularly, helped to consolidate an apparent “culture” of punishment through criminal instruments in dealing with corruption. It is essential to point out that the rapporteur of the case in the Supreme Court was Justice Joaquim Barbosa, another “critical figure” in the process of organizational changes, whose career began in the Prosecutor’s Office.

During the Mensalão trial, the Supreme Court innovated completely. The justices created a new and relevant jurisprudence for other anti-corruption actions, increasing the discretion of judges and prosecutors, and making the demand for evidence easier. An example was the use of the so-called Roxin’s theory through which Judiciary can condemn someone by the mere assumption of the illicit source of goods or values. Another example was the “guilty laundering”: only hiding the source of the money characterizes itself a crime of money laundering. Finally, the “ Ex Officio Act”: it is not necessary to point out the causal link between inappropriate advantage that someone receives and a decision that a politician takes.

The legacy of the Mensalão trial concerns a shift in the evidence evaluation method and the blurring of the limits between crimes of money laundering and corruption. This very new jurisprudence enabled the mobilization of investigative mechanisms typical of organized crime-fighting to be widely used against corruption.

In short, the fight against corruption through the judicial process took place in a new institutional setting: (i) more autonomy for the Federal Police; (ii) cooperation between the Federal Police and the Public Prosecutor’s Office; (iii) broad autonomy and discretion for prosecutors; (iv) specialization of the Federal Judiciary in dealing with financial crimes, corruption, and organized crime; (v) the Criminal Organizations Act, which provides a systematic treatment of the plea bargain; (vi) criminal jurisprudence based on increasing the state criminal punishing force.

It is in this new, incrementally designed institutional environment that Operation Car Wash took place in 2014. As far as institutions are concerned, there is an impact on the performance of police, prosecutors, and judges. It is no coincidence that Operation Car Wash is considered the most significant investigation of corruption and money laundering in Brazilian history. According to the prosecutors’ charges, thousands of dollars were diverted from Petrobras, Brazil’s largest state-owned company. Additionally, the players involved are high-profile individuals, both in political and in financial terms.

Operation Car Wash represents the high point of the judicial anti-corruption strategy that works at the limit (in some cases beyond) of the rule of law ( Kerche and Marona, 2018 ). The chart below summarizes institutional scenarios concerning the task forces analyzed: Banestado Case, Satyagraha Operation, and Operation Car Wash.

Chart 1
Task Forces’ Institutional Scenarios

Task Force/Category	Banestado Case (2003)	Satyagraha Operation (2004)	Operation Car Wash (2014)
“Corruption as organized crime” legal framework + reinforcement of state criminal punitive force	Absent	Partially Present	Fully Present
	The judicial tools for organized crime fighting were not yet available for corruption fighting strategies. Federal Police lacked expertise and institutional robustness. Higher Courts overturned several persecutory initiatives alleging lack of competency and individual rights violations.	The judicial tools for organized crime fighting were not fully available for corruption fighting strategies, but the Federal Police had already gained some expertise and institutional robustness. The Higher Courts, however, overturned all evidence obtained during the task force and canceled it as a whole in 2015.	Criminal Organization Law (2013) New Money Laundering Law (2012)
Inquisitorial Model Streamlining	Absent	Partially Present	Fully Present
	Severe coordination problems among judicial institutions and between them and other accountability agencies.	Some joint legal actions between the police and public prosecutors taking place in a coordinated manner.	Criminal prosecution discretion
Political Accountability	Fully Present	Partially Present	Absent
	The police deputy in charge of the case was removed by the Ministry of Justice. Appointment model for Chief Prosecutor had recently changed.	The police deputy in charge of the case was convicted and banned from the Federal Police	New chief prosecutor appointment model impacts Public Prosecutor’s office strategies

Source: Authors’ own elaboration.

If we take the main characteristics of the anti-corruption agenda in Brazil as a result of the various factors we have mentioned, we can see how it has evolved. In the Banestado case in 2003, the conditions for the establishment of a legal framework of “corruption as organized crime” were absent and there were serious legal and institutional barriers to the effectiveness of the criminal justice system.

Some of these gaps and barriers were overcome between 2004 and 2011, the period in which the Satyagraha Operation occurred. Finally, in 2014, when Operation Car Wash was unleashed, the conditions were entirely in place. This time around the task force relied on two new instruments. On the one hand, there was the Organized Crime Law, which structured the plea bargain. On the other hand, the new Money Laundering Law, which expanded the enforcement scenarios and brought procedural unity to the specialized ruling.

In this way, the accusatory model was gradually eroded. This came about pari passu with the amplification of the institutional capability of the Federal Police, especially with the swelling of the autonomy of the Public Prosecutor’s Office in coordination with the Judiciary. This process was fraught with tension. The 1988 Constitution set the criminal accusatory model determining the division of tasks, which comprises the criminal justice system: investigation, indictment and trial. On the one hand, the sub-constitutional legislation bears an array of inquisitorial elements, expressed in the broad discretion that the judges enjoyed during the investigative pre-trial phase. In addition, the Public Prosecutor’s Office overran the investigative competences of the Federal Police with the consent of the Supreme Court, which equated in 2015 the equivalence of the players in terms of the investigation. In addition, the cooperation between the different judicial spheres – the most emblematic case being the plea bargain – promoted the blurring between the investigation, accusation and trial roles.

Finally, it is worth noting that the structuring of the anti-corruption agenda around the unrestrained independence of the entities of the judicial system also advanced incrementally. The category of political accountability helps to understand the evolving scenarios. This evolution happened alongside the broadening of the capability of the police deputies, prosecutors, judges, and justices not only of applying the anti-corruption policy, but also of participating in its formulation based on a particular conception of the relationship between corruption and democracy.

CONCLUDING REMARKS

In 2003, the Workers’ Party administration embarked on an anti-corruption agenda. Throughout their mandates, the institutional framework was renewed. As a result, task forces have emerged as the prime anti-corruption strategy. Operation Car Wash is the capstone experience in this field and gathers some very particular features, which are a consequence of an incremental process of institutional changes.

The “justice policy” developed during Workers’ Party administrations (2003-2016) induced several changes within the Judicial system itself in a way that the government’s capacity to lead the anti-corruption agenda has become gradually limited. The Executive Branch’s initiatives were important to induce, through ENCCLA, cooperation among anti-corruption agencies and to bolster the streamlining of the criminal law process. In particular, because it raised the players of the Judicial system to the condition of architects of the public policies on corruption.

The Federal Police “institutional revolution” contributed to reinforce the criminal prosecution strategy on handling corruption and to induce a new dynamic of interaction between anti-corruption agencies. The new criminal legislation, particularly the legal treatment of the plea bargain, was also decisive: it allowed for a “corruption as organized crime” framework and served as an institutional opportunity to, via the lead of the Public Prosecutor’s Office, undermine the accusatory model of criminal justice and the fundamental rights system.

On the other hand, the Public Prosecutor’s Office and the Judiciary’s initiatives contributed to an anti-corruption institutional framework marked by the reinforcement of state criminal punitive force and a lack of political accountability. The increasing autonomy and discretion of the Public Prosecutor’s Office in criminal matters resulted, to some extent, in the agency’s leadership condition concerning that anti-corruption institutional framework.

The 1988 Constitution framers created the “new” Public Prosecutor’s Office in Brazil. Public servants with autonomy from politicians and society and with powerful legal instruments is something that is unusual in democracies. This institutional model allowed prosecutors during the 1990s and early 2000s to become relevant political players. They were able to interfere in public policies and fight against corruption in municipalities by means of civil actions.

Delegating to unelected players is inevitable in complex government ( Kiewiet and McCubbins, 1991 ). However, what happened in Brazil concerning the Public Prosecutor’s Office was not just a task delegation to unelected players, but also a “quasi-abdication”: a delegation that brought with it few and ineffective accountability tools ( Kerche, 2009 ). Previously, the limits for a complete abdication had been the influential role of the president in appointing the head of the Public Prosecutor’s Office, the division of prerogatives among Police, Public Prosecutor’s Office and Judiciary, the restriction for prosecutors investigating criminal matters, the “principle of legality” and the “principle of natural judge.” During the Workers’ Party governments, all of these institutional limits were relaxed.

Therefore, the agency in charge of criminal prosecution has experienced a process of institutional change, which came about as part of a new administrative and bureaucratic structure to control and fight corruption that arose in Brazil in the last decade, especially at the Federal level.

The Public Prosecutor’s Office has been using the plea bargain as an institutional opportunity to link corruption to organized crime. One of the most dangerous consequences in these cases is a diffuse perception of politics as corruption and politicians as criminals. The approximation of corruption to organized crime diffusely struck the political system and criminalized political activity itself. In addition, the internal control within the Judicial Branch became more relaxed due to the specialization of Federal Courts in fighting corruption.

A set of innovations that politicians promoted paved the way for the endogenous process of institutional change. The division of prerogatives of the penal system (criminal process’ accusatory model) was superseded by tacit agreements among the legal players of the Operation Car Wash, which brought an inquisitorial character to criminal proceedings and the weakening of the fundamental rights constitutional system. Thus, the Public Prosecutor’s Office secures a form of monopoly of the field of criminal proceedings. The agency begins to undertake selective persecution, supported by the discretionary use of the plea bargains, which also leads to the criminalization of politics.

The Federal Government lost the role of head of the anti-corruption agenda and was replaced by Judicial players, especially unaccountable prosecutors. The model has been “improved” over the years. However, it was only in the Operation Car Wash that the necessary, though insufficient, set of measures made it possible to include politicians among the accused and convicted. The analysis of the Brazilian political crisis must necessarily include these aspects.

In recent decades in Brazil, a set of institutional reforms were made seeking to strengthen the democratic nature of government through reinforcing an accountability web ( Power and Taylor, 2011 ), under the assumption that corruption should be dealt with by developing the bureaucracy ( Aranha and Filgueiras, 2016 ). Brazil’s case seems to demonstrate that this kind of approach is not enough to explain some contradictions between the war on corruption and civil rights. The strengthening of the administrative, bureaucratic and Judicial control mechanisms carried out by players who cannot be fully held accountable favored the generalized corrosion of the country’s political system, which jeopardized Brazilian democracy.

Acknowledgment

* We are grateful to Matthew Taylor and Daniel Brinks who have kindly offered their views on previous versions of this article. We appreciate, as well, the anonymous reviewers who presented critical considerations for preparing the published version. Finally, we extend our gratitude to Instituto Nacional de Ciência e Tecnologia (INCT)/ Instituto da Democracia e Democratização da Comunicação (IDDC) for their financial support.

REFERENCES

• ARANHA, Ana Luísa; FILGUEIRAS, Fernando. (2016), Instituições de accountability no Brasil: mudança institucional, incrementalismo e ecologia processual . Brasília, ENAP.
• ARANTES, Rogério Bastos. (2002), Ministério Público e Política no Brasil . São Paulo: Sumaré/Educ.
• ARANTES, Rogério Bastos. (2011), “Polícia Federal e construção institucional”. In: L. Avritzer; F. Filgueiras (orgs.), Corrupção e Sistema Político no Brasil . Rio de Janeiro: Civilização Brasileira, pp. 99-132.
• ARANTES, Rogério Bastos. (2015), “Rendición de cuentas y pluralismo estatal en Brasil: Ministerio Público y Policía Federal”. Desacatos (CIESAS) , vol. 49, pp. 28-47.
• ARANTES, Rogério Bastos. (2018), “Mensalão: um crime sem autor?”. In: M. Marona; A. Del Rio (orgs.), Justiça no Brasil: às margens da democracia . Belo Horizonte: Arraes Editores, pp. 338-388.
• ARMIJO, Leslie Elliott; FAUCHER, Philippe; DEMBINSKA, Magdalena. (2006), “Compared to what? assessing Brazil’s political institutions”. Comparative Political Studies , vol. 39, n^o6, pp. 759-786.
• AVRITZER, Leonardo; FILGUEIRAS, Fernando (orgs.). (2011), Corrupção e sistema político no Brasil . Rio de Janeiro: Civilização Brasileira.
• AVRITZER, Leonardo; MARONA, Marjorie. (2017), “A tensão entre soberania e instituições de controle na democracia brasileira”. Dados – Revista de Ciências Sociais , vol. 60, n^o2, pp. 359-393.
• BAEZ, Bien; ABOLAFIA, Mitchel Y. (2002), “Bureaucratic entrepreneurship and institutional change”. Journal of Public Administration Research and Theory , vol. 12, n^o4, pp. 525-552.
• BRINKS, Daniel; LEVITSKY, Steven; MURILLO, Maria Victoria. (2019), Understanding institutional weakness: power and design in Latin American Institutions . Cambridge:Cambridge University Press.
• CINGOLANI, Luciana; THOMSSON, Kaj; DE CROMBRUGGHE, Denis. (2015), “Minding Weber more than ever? The impacts of state capacity and bureaucratic autonomy on development goals”. World Development , vol. 72, pp. 191-207.
• DUTRA, Bruna Martins Amorim. (2012), “Criminal responsibility in the crimes committed by organized structures of power: Jurisprudence analysis in the light of International Criminal Law”. Revista da Faculdade de Direito da UERJ , vol. 2, n^o22.
• ELKINS, Zachary; GINSBURG, Tom; MELTON, James. (2009), The Endurance of National Constitutions . Cambridge: Cambridge University Press.
• FAGUNDES, Andrea Lucas; MADEIRA, Ligia Mori. (2021), “A Polícia Federal e o Governo Bolsonaro: duas décadas de desenvolvimento e dois anos de ataque e resistência?”. In: L. Avritzer; F. Kerche; M. Marona (orgs.), Governo Bolsonaro: retrocesso democrático e degradação política. Belo Horizonte: Autêntica, pp. 175-186.
• FERRAJOLI, Luigi. (2014), Direito e Razão: Teoria do Garantismo Penal . São Paulo: Revista dos Tribunais.
• GALVIN, Daniel J. (2012), “The transformation of political institutions: investments in institutional resources and gradual change in National Party Committees”. Studies in American Political Development , vol. 26, pp. 50-70.
• KERCHE, Fábio. (2009), Virtude e limites: autonomia e atribuições do Ministério Público no Brasil . São Paulo: Edusp.
• KERCHE, Fábio. (2018), “Ministério Público, Lava Jato e Mãos Limpas: uma abordagem institucional”. Revista Lua Nova , n^o105, pp. 255-286.
• KERCHE, Fábio; MARONA, Marjorie. (2018), “O Ministério Público na Operação Lava Jato: como eles chegaram até aqui?”. In: F. Kerche; J. Feres Jr. (orgs.), Operação Lava Jato e a Democracia Brasileira . São Paulo: Contracorrente, pp. 69-100.
• KERCHE, Fábio; OLIVEIRA, Vanessa Elias de; COUTO, Cláudio Gonçalves. (2020), “Os Conselhos Nacionais de Justiça e do Ministério Público no Brasil: instrumentos de accountability?”. Revista de Administração Pública , vol. 54, n^o5, pp. 1334-1360.
• KIEWIET, D. Roderick; McCUBBINS, Mathew D. (1991), The logic of delegation: Congressional Parties and the appropriations process . Chicago: University of Chicago Press.
• KINGDOM, John W.; STANO, Eric. (2011), Agendas, alternatives and public policies . Boston: Longman.
• MAHONEY, James; THELEN, Kathleen. (2010), “A theory of gradual institutional change”. In: J. Mahoney; K. Thelen (eds.), Explaining institutional change: ambiguity, agency, and power. Cambridge: Cambridge University Press, pp. 1-37.
• MAHONEY, James. (2008), “Toward a unified theory of causality”. Comparative Political Studies , vol. 41, pp. 412-436.
• MAHONEY, James; KIMBALL, Erin; KOIVU, Kendra L. (2009), “The logic of historical explanation in the Social Sciences”. Comparative Political Studies , vol. 42, pp. 114-146.
• MARONA, Marjorie; BARBOSA, Leon Queiroz. (2018), “Protagonismo judicial no Brasil: do que estamos falando?”. In: M. Marona; A. Del Rio (orgs.), Justiça no Brasil: às margens da democracia . Belo Horizonte: Arraes Editores, pp. 128-150.
• MADEIRA, Lígia; GELISKI, Leonardo. (2019), “O combate a crimes de corrupção pela Justiça Federal da Região Sul do Brasil”. Revista de Administração Pública , vol. 53, n^o6, pp. 987-1010.
• OLIVEIRA, Vanessa Elias de (org.). (2019), Judicialização de Políticas Públicas no Brasil. Rio de Janeiro: Editora Fiocruz.
• PAULA, Luiz Fernando de; MOURA, Rafael. (2019), “Consequências econômicas da Operação Lava-Jato”. Valor Econômico , August 28.
• PETERS, B. Guy. (2012), Institutional Theory in Political Science . New York: Continuum.
• PRAÇA, Sérgio; TAYLOR, Matthew. (2014), “Inching toward accountability: the evolution of Brazil’s anticorruption institutions, 1985-2010”. Latin American Politics and Society , vol. 56, n^o2, pp. 27-48.
• POWER, Timothy; TAYLOR, Matthew. (2011), Corruption and democracy in Brazil: the struggle for accountability . Notre Dame: Notre Dame University Press.
• SOLA, Lourdes; WHITEHEAD, Laurence. (2006), Statecrafting monetary authority: democracy and financial order in Brazil . Oxford: Centre for Brazilian Studies.
• SOUSA, Luís. (2010), “Anti-corruption agencies: between empowerment and irrelevance.” Crime, Law and Social Change , vol. 53, pp. 5-22.
• TAYLOR, Matthew. (2009), “Institutional development through policymaking: a case study of the Brazilian Central Bank”. World Politics , vol. 61, n^o3, pp. 487-515.
• TAYLOR, Matthew. (2019), “Alcançando a accountability: uma abordagem para o planejamento e implementação de estratégias anticorrupção”. Revista da CGU , vol.11, n^o20, pp.1311-1330.
• VEIGA, Luciana Fernandes; DUTT-ROSS; Steven; MARTINS, Flávia Bozza. (2019), “Os efeitos da economia e da Operação Lava-Jato na popularidade da presidente Dilma Rousseff no período pré-impedimento”. Revista de Sociologia e Política [online], vol. 27, n^o72.

NOTES

• 1
  . “The principle of the ‘natural judge’ ( juez natural ) is a fundamental guarantee of the right to a fair trial. This principle means that no one can be tried other than by an ordinary, pre-established, competent tribunal or judge. As a corollary of this principle, emergency, ad hoc , ‘extraordinary’, ex post facto and special courts are forbidden” (ICJ, 2007:7-8; translated by authors).
• 2
  . Brazil possesses a complex and interdependent network that includes bureaucracies that focuses on supervision (CGU and TCU), investigation (Federal Police), and sanction (Public Prosecutor’s Office and courts).
• 3
  . The Federal Police is subordinate to the Ministry of Justice and depends on the government in various aspects (promotions, removals and budget allocation), but the government cannot say which cases should be investigated and which should not.
• 4
  . The National Council of the Public Prosecutor’s Office is a body responsible for controlling and supervising the administrative and financial performance of the Public Prosecutor’s Office, as well as supervising the fulfillment of the functional duties of its members. The prosecutors are the majority of the Council (Kerche, Oliveira and Couto, 2020).
• 5
  . Criminal Action 470, known as the Mensalão , was a scandal that happened during Lula’s first administration. The charge was that the government paid congresspersons through advertising agencies to get support for bills in Congress. This case was judged by the Supreme Court and resulted in the arrest of dozens of offenders. Among them, José Dirceu, an important leader of the Workers’ Party and former Chief of Staff of President Lula.
• 6
  . The “privileged forum” is a judicial principle that determines that top politicians can only be tried for crimes in Higher Courts. The Brazilian Supreme Court extended the “privileged forum” to the 38 defendants during the Mensalão process, even though only three of them had this right previously. The prosecution’s strategy was to link all defendants to the same fact.
• 7
  . Roxin’s theory “seeks to clarify how it should be given criminal responsibility of superiors who, through the organized structures of power controlled by them, determine the criminal practices by subordinates who act freely and voluntarily” ( Dutra, 2012 ).
• 8
  . Macuco, in Portuguese refers, to the bird solitary tinamou. We have chosen to use the original in Portuguese.
• 9
  . The 2013 protests in Brazil, also known as June Journeys, were public demonstrations in many Brazilian cities, initially organized to protest against public transportation fees. The protests took nationwide proportions and incorporated other issues, such as corruption in the government. They became Brazil’s largest protest movement since the 1992 protests against former President Fernando Collor de Mello.

Publication Dates

• Publication in this collection
  17 May 2021
• Date of issue
  2021

History

• Received
  4 Oct 2019
• Reviewed
  25 June 2020
• Accepted
  8 Aug 2020