**Organization, Management and Crime – Organisation, Management und Kriminalität**

Maria Eugenia Trombini Elizangela Valarini Vanessa Elias de Oliveira Markus Pohlmann Editors

# The Fight against Systemic Corruption

Lessons from Brazil (2013–2022)

## **Organization, Management and Crime – Organisation, Management und Kriminalität**

#### **Series Editors**

Markus Pohlmann, Max-Weber-Institut für Soziologie, Universität Heidelberg, Heidelberg, Germany

Stefan Bär, Max-Weber-Institut für Soziologie, Universität Heidelberg, Heidelberg, Baden-Württemberg, Germany

Friederike Elias, Max-Weber-Institut für Soziologie, Universität Heidelberg, Heidelberg, Germany

Julian Klinkhammer, Max-Weber-Institut für Soziologie, Universität Heidelberg, Heidelberg, Germany

Die Reihe ist im Bereich der *Organization Studies* angesiedelt und hat ihren Schwerpunkt im Schnittfeld von Wirtschafts- und Organisationssoziologie. Sie widmet sich dem Zusammenhang von verschiedenen institutionellen Feldern (Wirtschaft, Medizin, Staat etc.) mit den Organisations- und Managementformen, die in diesen vorherrschen. Neben organisationssoziologischen Studien werden managementsoziologische Analysen sowie Studien zur organisationalen Kriminalität in die Reihe aufgenommen. Darüber hinaus beschäftigt sich die Reihe mit aktuellen Themen (z.B. Liberalisierung der Wirtschaft, Subjektivierung der Arbeitswelt, Ökonomisierung der Medizin oder der Politik) und sie hinterfragt gängige Erklärungen öffentlicher Skandale, die von Manipulation, Korruption oder Betrug in Organisationen handeln.

This book series establishes itself in the field of *organization studies* and focuses on the intersection between economic and organizational sociology. It is dedicated to the context of various institutional fields (economy, medicine, state, etc.) with the predominant organizational and management forms therein. In addition to studies on organizational sociology, management-related sociological analyses as well as studies on organizational crimes are included in the series. Also enclosed are current developments, such as Neoliberalism in the economy, the economization of medicine as well as the state of and the explanation for current scandals of manipulation, corruption and fraud.

Maria Eugenia Trombini · Elizangela Valarini · Vanessa Elias de Oliveira · Markus Pohlmann Editors

## The Fight against Systemic Corruption

## Lessons from Brazil (2013–2022)

*Editors*  Maria Eugenia Trombini Heidelberg Center for Ibero-American Studies University of Heidelberg Heidelberg, Baden-Württemberg, Germany

Vanessa Elias de Oliveira Sociedade e Políticas Públicas Universidade Federal do ABC Santo André, São Paulo, Brazil Elizangela Valarini Max-Weber-Institute of Sociology University of Heidelberg Heidelberg, Baden-Württemberg, Germany

Markus Pohlmann Max-Weber-Institute of Sociology University of Heidelberg Heidelberg, Baden-Württemberg, Germany

ISSN 2945-9842 ISSN 2945-9850 (electronic) Organization, Management and Crime – Organisation, Management und Kriminalität ISBN 978-3-658-43578-3 ISBN 978-3-658-43579-0 (eBook) https://doi.org/10.1007/978-3-658-43579-0

© The Editor(s) (if applicable) and The Author(s) 2024. This book is an open access publication.

**Open Access** This book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer VS imprint is published by the registered company Springer Fachmedien Wiesbaden GmbH, part of Springer Nature.

The registered company address is: Abraham-Lincoln-Str. 46, 65189 Wiesbaden, Germany

Paper in this product is recyclable.

### **Foreword**

The present volume originated in a project that began in 2018 and ended in 2023, in which a binational group of scholars engaged in the study of corruption in Brazil. As chair of IPSA's Research Council (RC 20), Political Finance and Political Corruption, I found the subject fascinating and the focus on Brazil of special significance. Consequently, the RC conducted its last 'meeting '(conference restricted to members of the RC alone) in Curitiba, where the famous 'car wash' investigations began. Because of the outbreak of the COVID 19 pandemic, further meetings became impossible. We did however maintain close contact and several of the authors in this book joined the RC and published chapters in its latest edited book. In turn, and despite the fact that I am no expert on Brazil, I was invited to offer a few comments in the concluding (zoom) meeting of the binational project.

A few words are needed to explain our shared interest and the special significance of Brazil. The country shares universal trends that are bound to change our understanding of what constitutes corruption and why it occurs. But over and above this, Brazil prominently manifests processes that were pointed out by students who sought to explain the erosion of political trust in European Western democracies. The theories they presented differ, but most indicate a link between the reaction to corruption scandals and the economic crisis of 2007–9 on the one hand, and the deepening authoritarian tendencies and threats of corruption on the other. The difference in time and location as well as intensity presented by what happened in Brazil suggests broader applicability of such factors and the possibility that in the not too far future they will help us shift the study of corruption to make it better suited to a world characterized more than anything else by rapid change and transition in all spheres.

A quick, undetailed account of the main historical shifts in the understanding of corruption and what brings it about will help clarify this statement. The concept originated in the thought of Athenian and Roman thinkers, for whom it carried a double meaning. One was dynamic and "naturalist." According to it, there always exists a gap between the ideals that underpin society and the realities of day-today life. The breach grows progressively, and with it a wholesale process of moral and spiritual decay that indiscriminately infects society at large. The other meaning was the narrower and more static. What could be termed "public office corruption" results from the natural temptation of all, including those in office, to employ the power to further their selfish interest.

In later ages the two meanings tended to separate. Philosophers such as Ibn Khaldun, and later Machiavelli, endorsed the naturalistic view. On the other hand, medieval thought tended to adopt a more static view, where corruption is the violation of a divinely ordained order. Both, however, lacked reference to particular behaviors and failed to discuss procedures that would prevent them. These questions became central for James Madison and his fellow American institution builders towards the end of the 18th century. The question for them was practical: what arrangements would prevent public officers from violating the boundaries of their office in pursuit of their own interests rather than the public good. Such a pragmatic view was, however, based on several tacit suppositions that in time posed obstacles to the consideration of corruption in times of rapid social and economic change. The constitution builders assumed that the limits of office could be clearly demarcated, that unambiguous rules would ensure that they would not be encroached upon, and that a proper constitution would thereby prevent large scale corruption. In a way, this also implied that properly run democracies, where laws were carefully explicated and followed and the people could "throw the rascals out", would be immune from such evils and that the real issue was how to adapt the best practices to circumstances. Such a view goes far to explains the postulation that came to govern the thought of most Western scholars and policy makers in the 1960's and 70's, namely, that corruption was a characteristic of the "third world". With the dissolution of the European empires, newly independent states found themselves in transition: the proper institutions and legal arrangements were not yet installed, or, where they were set up, did not fully correspond to actual behaviors. The result was the perpetuation of traditional patterns in which power- holders treated their authority as a personal property. Consequently, the dominant belief was that rapid democratization would be the assured way to overcome endemic corruption.

These assumptions came under serious doubt once major scandals rocked Western Europe in the early 1990's. The most serious of these was revealed by the 'Mani Pulite' (clean hands) investigations into political corruption that brought down the Italian "First Republic". The network of political corruption that was exposed involved politicians from all parties and levels of authority (at one point more than half the members of Parliament were under indictment) along with party officials, members of some four hundred town councils, bureaucrats, and businessmen. Scandals in other West European states were dwarfed in comparison, but together they amounted to the falsification of the postulation made about the immunity of Western democracies to corruption. We need mention only a few of the more prominent ones. The German long- term chancellor, Helmut Kohl, was found to have set up a secret slush fund to channel money to the ruling Christian Democratic Union, NATO secretary general Willy Claus was forced to resign after he was found guilty of corruption, and a series of high profile scandals resulted in ministerial resignations and the discrediting of the Gonzalez government in Spain. The prevailing definition of corruption remained the one formulated in the context of Development studies, but its mooting in third world studies was detached and most were abridged: as in the Word Bank's "abuse of public power for private benefit", or Transparency International's "misuse of entrusted power for private benefit". Nevertheless, even such definitions retained the core of the "public office" perspective and the essential view of corruption as a violation of democratic principles. Other critics adopted a different strategy by the argument that we need to identify a concept of corruption that would apply only to democracies. Still others sought to escape the issue altogether by the postulation that "we know corruption when we see it" and hence we need not enter problematic characterizations and causations. Whatever the solution, it is reasonable to assume that once the mesmerizing combined effects of the COVID pandemic, Ukraine crisis and threat of worldwide recession would relax, greater attention will have to be invested in the effort to produce a broader and more inclusive understanding.

A starting point for such an effort could be found in the growing awareness of the trends referred to above. One is that we are in the midst of an industrial revolution that changes our realities in an ever escalating manner. Unlike the first industrial revolution it does not affect our muscles but directly our minds. The term "globalization" is appropriate: unlike the first industrial revolution, the current one visits most societies contemporaneously, shaping not only the way we do things but first and foremost the way we understand them. The implications for in our field- as for most others - are profound. Whatever we mean by "corruption", the term cannot apply to a stable reality but be placed in the context of a rapidly shifting reality. The corollary is not that the "public office" notion should be abandoned, but that it should it be placed within the framework of a dynamic theory, where boundaries are flexible and subject matters shift. Although this is easier said than done, what we need is an understanding of corruption that would reunited the two original meanings of the term yet be significantly applied to ever shifting realities. The second trend we ought to bring into consideration is that regardless of the shifting realities, societies preserve a core of traditional behaviors. This core may well shift contours but in itself the preservation of traditional ways should not be thought of as a malevolent. Without them we live in perpetual chaos. The question however is whether we are aware of practices that were once legitimate but in a different reality facilitate or are synonymous with corruption.

Whereas all this holds true for most countries, this is not the case when the European developments to which we alluded earlier are considered. One of these was the rise of populist leaders, movements and parties in reaction to the corruption scandals of the early 1990's and, later, the "Great Recession" of 2007-9. Populism is a contested concept. Nevertheless, most critics agree on a minimal definition according to which it is a vision of the political world that is based on three essential beliefs. First, that society is divided between two homogeneous and antagonistic groups: the people and the corrupt elite. Then, that the will of the former is identifiable by the populist leader and his supporters, and, finally, that the mission of these is to express and carry out the will of the people. The second development is paradoxical: the belief in the heroic mission to clear house from corruption and return the nation to economic health is liable to lead not merely to the erosion of political trust but to actual corruption. Where the political order is presented as expressing the interest of the corrupt few the result could be corruption in the name of the people. This, in turn, raises the threat of a blurred division between the good of the people and the private good of the heroic figure and his supporters.

For a Brazilian reader, this could read as an interpretation of what actually happened in the country. Despite obvious differences, operation "car wash", the criminal investigation that began in early 2014, could be compared only to the "clean hands" of Italy. A vast, intricate web of corruption was gradually exposed to shake the fragile democracy to its foundations. This was complicated by an escalating depression that started at about the same time after years of unprecedented growth (2003-14) brought about a significant shrinking of the economy and rise of unemployment. As the corruption scandal unfolded, the crisis became fueled by the revelation of illegal large flows of money among large private and state owned companies and between them and politicians. In October 2018 Jair Bolsonaro was elected as President of Brazil, and a new vision of mass politics began to grow, in which the hero—leader represents the will of the people against the corrupt elite. Almost immediately ministers in the new government became involved in corruption investigations, and the president's party was accused of campaign finance fraud. After a year and four months in Bolsonaro's cabinet, Sérgio Moro, the Minister of Justice and Public Safety who was previously one of the lead judges in the 'car wash' investigation stepped down. It appears that a cycle was closed. If this broad outline seems to follow a well-known script, the main point should not be forgotten. Dramatic stories serve to excite the reader's imagination. For students of corruption it may encourage thinking about the contribution it could make towards the crystallization of a revised understanding of corruption in a world of rapid change.

Jonathan Mendilow

## **Contents**




## **Introduction**

Political Corruption, Corporate Crime and the Rule of Law in Brazil

#### Elizangela Valarini, Vanessa Elias De Oliveira, Maria Eugenia Trombini and Markus Pohlmann

The fight against corruption in Brazil has crossed national borders since 2014, gaining worldwide visibility with Lava Jato Operation, or Car Wash. This is due not only to the amount of money surrounding illegalities committed by political parties and representatives, elected or not, but also because large companies and businessmen were involved in a huge corruption scheme relating money and politics. The involvement of politicians and executives from several countries was also a hallmark of this corruption scandal. In addition, their confrontation involved an extensive effort by international anti-corruption agencies to track and recover the money of the Brazilian State. Like other Western democracies, Brazil

E. Valarini

M. E. Trombini (B) · M. Pohlmann Max Weber Institute of Sociology, Heidelberg, Germany e-mail: maria.trombini@mwi.uni-heidelberg.de

M. Pohlmann e-mail: markus.pohlmann@mwi.uni-heidelberg.de

M. E. Trombini Heidelberg Center for Ibero-American Studies, Heidelberg, Germany

Alumni Max Weber Institute of Sociology, Heidelberg, Germany e-mail: eliz.valarini@gmail.com

V. E. De Oliveira Federal University of ABC, Sao Paulo, Brazil e-mail: vanessa.oliveira@ufabc.edu.br

was neither immune to misconduct and nor were its institutions ineffective in detecting it and pursuing legal accountability.

In a methodological and empirical effort to understand the phenomenon's magnitude, contributors to this volume have focused on the causes and consequences of the anti-corruption fight in Brazil. The chapters range from presenting the interactional perspective, of the relationships between political, judicial, or business actors in this process, to the institutional perspective, by analyzing the institutional and legal advances associated with the proceedings of Lava Jato Operation in Brazil. We are not taking a moralist stance vis-à-vis corruption, given that anticorruption campaigns and scandals may be ambivalent as they delegitimize representative institutions and the political system. To sidestep a normative view of the object under study, scholars cover both the roots of and the reactions to wrongdoing in the political and economic systems, aiming to address the historical shift in corruption research.

This is the context in which the present work is inserted. The book explores, firstly, state actors and institutions directly involved in Car Wash, focusing on accountability mechanisms, government entities and public servants. Secondly, it investigates non-state actors that were engaging with the state apparatus and public officials in the corruption cases unveiled, with special attention to a pivotal construction company in this process: Odebrecht. While the first part casts light on the internal dynamics of the legal system and how members of legal careers think and act, the second is more concerned with the economic system and how for-profit organizations manage illegality. Boundaries between each 'system' are merely an epistemic heuristic, so the lines between the market and the State, public and private, legal and illegal, will be far more blurred throughout the book. We will move seamlessly back and forth in the corruption research landscape, even if the two parts in the volume should orient our readers.

The chapters that follow are the result of the binational research project "Organizational Crime and Systemic Corruption in Brazil" and of intensive debates to examine and explain the interplay between public and private sectors. Funded by the Deutsche Forschungsgemeinschaft (DFG) and the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), the research project began in 2018 and ended in 2023, carried out in cooperation between the University of Heidelberg, University of Sao Paulo (USP), Federal University of ABC (UFABC) and Federal University of Paraná (UFPR). Briefly, the specific contributions from each chapter are discussed in this introductory session.

Following the introduction, the book starts with a methodological discussion on how to study patterns of systemic corruption, which are widespread around the world and difficult to prevent. Contrary to the ritual of simply condemning systemic corruption and insisting on its negative consequences, the article shows what makes systemic corruption so endurable and why societies rely on systemic corruption when formal institutions fail. But it also introduces the theoretical tools to conceptualize systemic corruption, and the indicators and analytical tools to measure the degree of systemic corruption in a given society. Instead of complaining about it, the article by Markus Pohlmann "Systemic Corruption—How to analyze and measure it" suggests to move one step forward and analyze the phenomenon in a sound empirical way. In view of the low level of research interest in systemic corruption beyond the usual corruption expert ratings, on the one hand, and the importance of the topic for the development of suitable anticorruption measures in democracies, on the other, Pohlmann emphasizes the significance of improving the methodological tools for dealing with this subject.

The first aspect that has to be clarified is the conceptualization of systemic corruption and the bystanding behavior of many. Systemic corruption is related to legal organizations with legal objectives, providing services or selling goods within the framework of legally recognized organizations, but from time to time using illegal means to achieve legal objectives, and illegal personal gain. Petty corruption, in this sense, is carried out partly for the benefit of and based on the informal rules of a community, an organization or a society. As a hidden social order, it emerges when formal institutions fail or are not trusted. This definition is the basis for further analysis on systemic corruption. Some central methodological steps frame and understand the institutional order involved, analyze the collective action problems, and the costs and benefits involved for becoming part of it. Considering the complexity of this phenomenon, and the array of institutions and actors implicated on it, research on systemic corruption should follow a multimethod strategy that combines quantitative and qualitative techniques. This is the rationale behind the different chapters that constitute parts 1 and 2 of this book.

Part 1 is dedicated to the understanding of institutions and actors involved in the detection, investigation and sanctioning of wrongdoings. Each chapter tackles either accountability mechanisms, government entities or public servants, addressing how are they organized, what do they think and how do they behave.

Institutional reforms can deter or create disincentives to corrupt behaviors. According to different Brazilian scholars, the fight against corruption has been driven by incremental institutional changes that bolstered domestic accountability institutions. Providing a general map of the institutional improvement process that occurred in the country before Car Wash, Fabiana Rodrigues demonstrates in her chapter "Anticorruption institutional improvements and the Operation Lava Jato" how the country has become more efficient in identifying and punishing corruption. This process was highly influenced by international anti-corruption and money laundering reforms, as a result of international commitments undertaken by Brazil. A series of institutional changes over the years were essential to enable an investigation of the size and scale of Car Wash to take place and be successful in terms of legal outcome. Therefore, Rodrigues tracks where the investigation emerged from and presents it as the outcome of a long path of institutional incrementalism.

Maria Eugenia Trombini and Elizangela Valarini ask what guides the action orientation of people working in the anti-corruption complex when working with the people labeled white-collar criminals. The objective of their chapter entitled "Rational Executives and Selfish Politicians: the Cultural Repertoires of Legal Experts" is to advance the research agenda on Law and Politics by shedding light on the views held by members of the justice system. Their qualitative findings show the group locates the problem of corruption among politicians. The collective mindsets of legal experts reconstructed at the chapter criticize patrimonialism and a lack of separation between the public and private domains in the Brazilian context. The expert group with previous experience prosecuting, sentencing and defending people targeted for white-collar crime recommend ethical guidelines and knowledge-based management to remedy widespread corruption.

Powerful actors play a special role in any political process. That is not different for actors in the judicial system, although they insist on affirming that they are technical and not political actors. When former public officials in legal careers decide to follow a political career, like lawyers who worked in corruption scandals did, studying their preferences becomes an interesting endeavor. Noting the scarce literature on judicial decision-making in corruption cases, Luciano Da Ros, Luísa Zanini da Fontoura, Sérgio Simoni Junior and Matthew M. Taylor have attempted to bridge this gap by focusing on an important player from Car Wash: judge Sérgio Moro. In "Moro's opinions: a quantitative analysis of sentencing in Operation Car Wash", the authors analyze Moro's decisions in criminal cases as an exercise to understand the drivers of judicial decision-making in corruption trials. By quantitatively analyzing an extensive dataset of cases compiled from official sources, they demonstrate that politicians have higher chances of conviction and are awarded longer sentences when compared to all other types of defendants. This is an important finding for studies on corruption and the money-and-politics relations, not only for Brazil but for all democratic countries. Furthermore, the study emphasizes that decisions in cases of corruption can be analyzed considering both the decision to either convict or not convict the accused, and the decision on the length of the sentence imposed on convicted defendants.

Part Two, "Companies and the interplay with key sectors", is devoted to understanding the private actors and State institutions they interact with in corruption cases. These interactions are especially important in Car Wash Operation, considering the central role played by companies and their executives. Together with political actors, business elites challenged the country's accountability system, acting to promote individual and organizational gains for both the companies and the political parties engaged in illegal activities. The framework in which decision-making by top executives from a construction company takes place distinguishes itself from the framework regulating how judges sentence criminal cases about bribery. Also the worldviews about the State held by public officials working in the web of accountability stand in contrast to those held by businesspeople.

Corruption is often modeled in association with development, as Jonathan Mendilow stated in the Prologue. As such, to gain a revised understanding of the phenomenon using data from a site located in the Global South, factors pertaining to the market order need to be accounted for. Brazil is considered a developmental state, where the bureaucracy has leeway to "guide" business using its discretion. The contributions from Part Two refer to bargaining processes between state and non-state actors that may or may not be labeled illegal. Again, instead of deeming a country more corrupt because of the size and power of the public sector, we follow a non-normative approach in this collected volume.

At the business level, consolidated but informal relationships constitute a form of "durable capital" that rather than eroding, as with most tangible forms of productive capital, becomes more entrenched through repeated use. The first chapter of this part is "The public and the private in Odebrecht's vocabulary", where Trombini and Bischof dos Santos discuss the boundaries between the two by using narratives from executives of the company. In addition to historical and bibliographic sources, the authors collected evidence from a database consisting of 8 criminal cases sentenced in Car Wash, with testimonies from 48 Odebrecht members. Contrary to traditional approaches to the intertwining of businessmen and politicians, the chapter argues that corruption, understood as the interference of private interests in the public sphere, is not a maladaptive reaction to regulation, but an adaptive reaction due to the competitive advantages it provides to its agents, businessmen or rulers.

"A pragmatic giant: the logic behind Odebrecht's campaign donations in the 2014 elections", analyzes the enormous campaign contributions from the Odebrecht group (approximately 88.9 million reais, circa 37.8 million US\$) to 2014 election campaigns in Brazil. In order to understand the company's decision to donate or not and, when doing so, deciding on how much to contribute and how to distribute it among candidates and parties, Mancuso, Horochovski and Speck focused on candidate's political capital and region, and political party's ideology, its belonging to the government's ruling coalition in Congress, and its size. The quantitative analysis and its results demonstrated that Odebrecht concentrated its donations on the three largest parties and most important political parties in Brazil at that time (PT, PSDB and PMDB). It means that the company ignored the status of being a ruling or opposition party, as well as the ideological position of the party, concentrating its political investments based on the size (and chances of winning) of the Brazilian parties during the 2014 elections. These findings, therefore, demonstrate the company's extremely pragmatic approach, seeking to expand the company's political leverage in the upcoming political scenario.

Another chapter focused on corporate behavior is that of Costa, Ferreira and Campos "Businessmen, political financing, and corruption: Odebrecht in Operation Car Wash". The authors analyzed the managerial culture at Odebrecht concerning legal and illegal political financing aiming to grasp the "*insider perspective*". Their empirical material were testimonies of people in close proximity to the chairman of the holding company and descendent of the founders, Marcelo Odebrecht. Results illustrate how an important group of Brazilian economic elites dealt with the real working of Brazilian democracy through illegal methods. Using the category of "representation" the chapter studies a specific and underexplored dimension of the political culture and political action of entrepreneurs: the practice of corruption.

The chapter "Compliance Implementation Challenges in the Shadow of Corporate Crime: A Case Study of Odebrecht S.A.", by Mario H. Jorge Jr., presents a case study of Odebrecht in order to understand the features and implementation challenges of compliance systems. The study provides insights into how anticorruption practices are integrated into processes and structures, how they are handled and perceived by management, and highlights important factors for a successful system implementation. The study also examines the measures taken by Odebrecht, such as the termination of employment of individuals involved in misconduct, the development of its accountability system, and the increase in the Compliance departments' budget and resources. The paper suggests that compliance implementation could be further developed by considering elements such as the most vulnerable job positions, internal communication, power dynamics and culture among leadership, structural incentives (especially financial), sustainability and using technology to create an ongoing monitoring system.

The closing chapter by Vanessa Elias de Oliveira and Conrado Hübner Mendes delves deeper into the kind of corruption practiced within the public administration by centering on the organs of the justice system. Introducing a new research agenda, the authors present a typology of judicial corruption derived from Brazil and outline the importance of understanding how the office-holders who sought to combat political and corporate corruption instrumentalized their position to pursue their own personal interests or institional ends. The fact that anti-corruption efforts championed by the justice system coexist with a usage of office for nonrepublican purposes is a subject that is worthy of attention and further research by comparative politics.

Acknowledging that corruption is a complex and multicausal phenomenon, the contributions in this book are complementary inasmuch as they analyze the practices and institutions by using different types of data, methods and considering different analytical layers.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Systemic Corruption—How to Analyze and Measure It**

Markus Pohlmann

#### **1 Introduction**

Too often, the catchphrase "systemic corruption" leads to political and normative statements usually expressing its strict rejection and blaming the actors engaging in it. But despite decades of fighting systemic corruption, the world map of corruption is still red to dark red in many countries. We can keep complaining how bad this is. Scientists, and activists are telling us a lot about the negative consequences of systemic corruption. And we do not deny the negative consequences. But aren't there any positive consequences, and not only for those who enrich themselves personally in a criminal way? As sociologists, however, we are used to ask ourselves also the other way around. If systemic corruption is so long lasting, what is the productivity, indeed functionality, of systemic corruption for a given society?

Leys raised this question as early as 1965, pointing out that most authors—as is still the case today—deal only with the bad consequences of systemic corruption without having examined all the consequences—both the good and the bad—in greater detail (Leys, 1965; De Sardan, 1999). And it is precisely this question that will concern us in what follows: What causes systemic corruption to be so sustainable and reproduce itself despite all the means of combating it?

Before we are able to answer the question, we will first take a closer look at the conceptualization of the phenomenon of systemic corruption.

M. Pohlmann (B)

Max Weber Institute of Sociology, Heidelberg, Germany e-mail: markus.pohlmann@mwi.uni-heidelberg.de

M. E. Trombini et al. (eds.), *The Fight against Systemic Corruption*, Organization, Management and Crime – Organisation, Management und Kriminalität, https://doi.org/10.1007/978-3-658-43579-0\_2

Systemic corruption is first and foremost a political term that is often used to criticize or discriminate against certain deviations from the rules in a society. It often refers to the difficulties of states, their politics and their administration to achieve sufficient social acceptance for rules that are supposed to protect the public sector from the influence of private interests. In a first access, reference is made to the fact that corruption penetrates various aspects of a system or society and becomes firmly established in the structures, processes and institutions of a system. This access still remains wedded to an everyday theory approach to the phenomenon, but already introduces four important indicators and distinctions: (1) the regularity in the occurrence of corruption. "The relevant kind of corruption (1) is sustained and continued over time, (2) involves the participation, whether conscious or not, of multiple agents, and (3) takes the form of an identifiable institutional practice, (4) thus implying some degree of coordination among the participants" (Ceva & Lubomira, 2018: 3). In other words, systemic corruption occurs not only regularly, but also simultaneously or interconnectedly in different areas of society, involving many members of a given society.

In what follows, we want to look a more closely at these four aspects of our common understanding of systemic corruption and ask which theory we can best use to explain the emergence and reproduction of systemic corruption.

When systemic corruption is observed from an inside perspective by ethnographers and anthropologists, leaving aside the moral accounts of the observers, a complex understanding of systemic corruption emerges, taking up the productive side of systemic corruption as well (ibid. p. 45). For an anthropologist perspective on systemic corruption it is quite clear, that different societal contexts can have diverse and multiple moralities, and that "in some cultures, petty corruption may be accepted and also justified as a practice that smoothens social exchange" (Tornello & Venard, 2016: 35). "Considering the phenomenon from the perspective of the local people leads anthropologists to refuse to accept any a priori moral condemnation of corruption" (ibid. p. 48).

Systemic Corruption is also fundamentally understood as a social relationship and institutionalized practice in many sociological approaches (see Höffling, 2002; Pohlmann et al., 2016, 2019).

For a strong part of the recent social science literature, systemic corruption is related to the Rational Choice approach and described as a stable system, "where most individual actors expect others to act corruptly. The result is a collective action problem: the environment sets incentives for actors to act corruptly, even if it is not their personal preference to do so" (Mota & Pimenta, 2021: 74). An individually rational strategy produces an outcome that is collectively inferior."In such a context, the costs of refraining from corrupt practices may be high and the benefits are likely to be low, especially if those defectors lack sufficient numbers or influence to create meaningful, systemic change through their honest behavior" (ibid, p.77). But the framework of a rational choice explanation is still underdeveloped, when it comes to the operationalization of indicators that enable the empirical measurement of systemic corruption in a given society.

The perspective of the institutionalist approach is presented in recent social science literature as well. In an institutionalist perspective, systemic corruption is not merely a property of an official's individual behavior, but is to be understood by the properties of institutional practices (Ceva & Lubomira, 2018: 7). But till today, the theoretical frame of an institutionalist explanation is still underdeveloped as well as the operationalization of indicators, that beyond international statistics, allow field-based research (see e.g. Trombini et al., 2022; Valarini & Pohlmann, 2019).

This is, what our article is heading to: to provide explanatory frames and empirical indicators, that can help to answer this question. We introduce two analytical frames that supports us to get access to the phenomena behind and empirical measures to understand the mechanisms better, that lead to the sustainability of systemic corruption. We apply the institutional theory as well as the rational choice theory as explanatory frames (2), move on from there to the bystanding behavior of many actors, stabilizing systemic corruption (3), and explore the empirical indicators derived from the two explanatory frames and describe the methods we can use to verify their existence (4). Thus, we do not want to demonize systemic corruption from the outset, but rather we want to trace its productivity and answer the question of what the societal benefit, the societal function of systemic corruption is.

#### **2 Two Theories of Systemic Corruption**

Both theories, the **Institutional theory** and the **Rational Choice Theory** provide analytical frames to explain the "success" and the productivity of systemic corruption. In this chapter, we're going to explore them further concerning the mechanisms that contribute to the reproduction of systemic corruption.

#### **2.1 The Analytical Frame of Institutional Theory**

According to institutional theory, systemic corruption builds on centuries-old social institutions, social systems of exchange that pervade society on the backstage of the formal institutional order. If we rely on the concept of institutional multiplicity, which addresses the coexistence of independent normative systems, e.g. one that the formal rule of law implies, and the other informal that legitimates the use of petty corruption, we can also assume the coexistence of different institutional orders that both complement and oppose each other (Mota & Pimenta, 2021: 75 f.; Mota & Cornelius, 2020: 3). Systemic corruption in a society builds on long-established, social systems of exchange whose unwritten rules of give and take come to the fore whenever the established formal rules and institutions, and the personnel who represent them, receive little recognition and trust in a society. It seems to be functional, a form of useful illegality, because it keeps society functioning even when—in the worst case—many official roles in society are seen as to be filled by "rogues" and "greedy elites". Even in quantitative political and economic science studies it was observed "that corruption is less detrimental to efficiency in countries where institutions are less effective". It was even "positively associated with efficiency in countries where institutions are extremely ineffective" (Méon & Weill, 2010: 244; see also Méon & Sekkat, 2005). Dreher and Gassebner did show in their study also, "that corruption facilitates firm entry in highly regulated economies". (Dreher & Gassebner, 2013: p. 413).

The question of how to explain the persistence of systemic corruption despite permanent attempts to combat it can thus be answered by referring to the functionality of systemic corruption. Whenever the formal rules loose acceptance, the prevailing informal rules of the society take over.


enrichment is accepted to the extent that it serves to "make ends meet," i.e., to ensure economic subsistence. Subsistence-oriented acceptance of personal gain is an important indicator of systemic corruption. At the same time, it makes clear that pro-organizational crime, that is executed without illegal personal gain (see Pohlmann et al., 2020), is more closely interwoven with personal enrichment than it is the case in countries with casual corruption.


#### *Sticking to the Approach of the Institutional Theory, we have to answer these questions:*



**Table 1** Indicators of Systemic Corruption according to the Institutional Theory


Sometimes opposed, but partly connected to the institutional theory approach, **Rational Choice theory** is using another foundation and other assumptions to provide an analytical frame for the phenomena of systemic corruption.

#### **2.2 The Analytical Frame of Rational Choice Theory**

Persson et al. point out that systemic corruption solves not only problems of collective action, but can only be changed by the collective action of the majority of actors. They argue that the focus shall not be on principal-agent problems, but rather on the logic of collective action. Systemic corruption is prevailing and successful, because agents react to collectively anchored expectations to supplement society and its institutional order with socially embedded give-and-take structures where serious deficits in the societal order occur or trust in social institutions and elites is lacking. This also finds its expression in the fact that the principals do not adhere to the rules, laws and statutes they have enacted, but rather orient themselves to the ulterior, unwritten rules of the game of systemic corruption. The immunization of systemic corruption from the numerous attempts to combat it, is then achieved by "omission" of many to speak up. Those who speak up or take action against it have higher costs to take into account, and a greater likelihood that they will occur, compared to those who participate in or tolerate the systemic corruption game. When it comes to the combat against systemic corruption, the question is, whether and under which conditions the selective incentives are large enough to get enough players to stop playing the game (Persson et al., 2013: 450). "Quite contrary to what the principal-agent framework assumes, rather than reporting and punishing corrupt behavior, political leaders, as well as citizens, seem to at least passively maintain the corrupt system" (ibid p.454).

Although there have been some discussions concerning the two explanatory frames of rational-choice theory and their combination (Persson et al., 2013; Rothstein, 2018; Marquette & Pfeiffer, 2018; Persson et al., 2019), for the purpose of our article, we ourselves like to combine aspects of both frames: the collectiveaction and the principal-agent framework.

The question, then, of what makes systemic corruption so enduringly successful is answered in RC theory by reference to the following factors: 1. Corruption is perceived in a given society as an expected behavior; 2. The benefits for behaving corrupt are higher than the costs times the probability to be detected or that the crime is reported (Perrson et al., 2013, 2019; Marquette & Pfeiffer, 2018; Coleman, 1990); 3. The costs of speaking up are higher than the benefits, and the probabilities that the perpetrators will be punished are low. 4. On the backstage of legal organizations, the principals indirectly are rewarding the corrupt behavior, and not as much the whistleblowing; 5. The supposed "principled principal(s)" are also corrupt and not acting in the interest of the society;

Many do act in a corrupt fashion, and the expected payoffs of the corrupt behavior, as well as the odds of receiving the payoffs, are higher for most than the expected costs, as well as the probabilities of incurring any costs at all. Conversely, the benefits of formally doing the right thing and reporting wrongdoing seem negligible compared to the disadvantages of being socially discriminated as a traitor.

*Sticking to the RC-Theory, we have to answer these questions, if we want to measure the grade of systemic corruption, that accompanies a given society:*



**Table 2** Indicators of Systemic Corruption according to Institutional Theory, and Rational Choice Theory


#### **3 In Support of Systemic Corruption: Bystanding as Collective Action**

Systemic corruption could not be so stable and durable as a social and organizational form if it did not also have support through the collective omission to do anything about it. This collective neglect occurs even when most members of a society want to do something about it. For rational choice theory, it is a suboptimal outcome, for all concerned, of the cooperation game between rational egoists. For institutional theory, there is a difference between the public articulated will to do something and the actual willingness and opportunities behind the facade, behind the stage of public life. The habits of thought and action, as well as the established forms of justification for failing to intervene, create a habitual pattern that is difficult to break even if one wants to do something about it in any case. The cognitive and normative restructuring in the perception of the situation often makes people following the regular course of action even when the persistence leads to crisis. And that is not even the case with systemic corruption. On the basis of such support structures, we assume, systemic corruption manages to protect itself from the regular normative hostilities, even to immunize itself to a certain extent through "systemic bystanding".

By "bystanding as collective action" we mean the social establishment of mechanisms that lead to collective omission in a given society, to collective turning a blind eye to useful petty corruption and functional wrongdoings.

Thus, we ask: What are the mechanisms and structures that promote speakingup behavior in the context of systemic corruption, and how strong are the mechanisms that leads to covering up behavior and silence in organizations? When we look for answers to what makes systemic corruption so resilient, we must also identify at what their support structures are. And an important factor here is always that there are incentives or habits in a society or organization to look the other way when others break the rules and thus to tolerate and accept petty wrongdoings. Why is this so common in the context of systemic corruption?

In this chapter, we will again stick to our introduction of the two explanatory frames, and give two different answers to this question. On the one hand, one can answer with the help of the institutional theory that in societies and organizations "parallel worlds" emerge in which informal interpretative orders and unwritten rules come into effect, which on the basis of hierarchy, and socialization processes provide strong justifications for the collective silence (IT1, IT2, IT4). Here, the collective omission no longer appears as an unintended consequence of action in the tragic interaction of rational egoists, but as a "normal", norm-conforming action inside a given society, a legal organization and its personnel in a parallel interpretive order with the organization's own unwritten laws and rules. On the other hand, the rational choice theory proposes to understand the collective silence about the punitive acts as a rational action of rational egoists, who in their interaction realize a suboptimal outcome. It is suboptimal because it does not correspond to their primary preferences, because the perpetrators are not stopped. "In a context of systemic corruption, anyone seeking to report corruption, even the accused, may not resort to accountability institutions because there are no benefits (as punishment is unlikely) and there may be costs (in addition to resources—for example, time—there could be retaliation)" (Mota & Pimenta, 2021: 75).

#### **3.1 Bystanding, and Organizational Bystanding**

Especially in psychology and social psychology, there are numerous studies on so-called bystander effects (see only Latané & Darley, 1968; Latané & Nida, 1981; Fischer, 2011; Liebst & Philpot, 2018; Hussain et al., 2019 and many others). This refers to the phenomenon that bystanders do not intervene when they witness a crime and that the probability of intervention decreases the more observers there are at a crime scene. This is usually explained in the literature by three factors, in addition to personality traits such as "self efficacy" and gender, as well as effects of different situations (see, e.g., Krieger et al., 2017; Mabry & Turner, 2016; Leone et al., 2017).


In the psychological and social psychological literature, it is rarely known neither exactly which explanatory approach lies behind these factors, nor how they should be classified in a theoretical explanation. The following remarks attempt to remedy this deficit. The currently available literature is also devoted to bystanding effects at the workplace. In addition to characteristics of the workplace and the employees (see only Hellemans et al., 2017), three organizational effects are particularly emphasized in the various studies:


Once again, it is noticeable that the organization effect is not theoretically illuminated and not traced back to the basis of rational choice theory or the institutional theory. Although the effects are mentioned as results of the studies, they are not further substantiated theoretically.

#### **3.2 Bystanding as Legitimate Behavior: The Institutional Theory**

In this context, an institutional theory perspective shows that omission should not be understood as isolated opportunistic-criminal behavior of individual actors, but can be linked to institutionalized, i.e., taken-for-granted expectations and practices in their societal and organizational field. Within the particular societal or organizational field, it may be rational and legitimate to behave in silence because behind the formal facades collective bystanding fits the unwritten rules in a given society or organization.

Niklas Luhmann (1964, 304 ff.) refers to the organizationally useful deviation from formal rules as "useful illegality" in early work. In the case of organizationally useful illegality, there are, in turn, rule deviations that appear legitimate and rule deviations that appear illegitimate. The culture of a society or an organization then determines which rule deviations receive recognition and which do not. In our argumentation, we move to the next level of explanation, which is about how the organization and its employees deal with these rule deviations. We want to show that on this level, organizationally useful illegality or organizational deviance can occur, which does not report the serious appalling rule violations to the public prosecutor's office but endures them according to the informal rules, interpretative orders, and informal norm orientations of the society.

Thus, applying the frame of the institutional theory of bystanding, we have to answer four questions: 1. Are there informal norms broadly accepted in a given


**Table 3** Indicators of Bystanding according to Institutional Theory

society, that allow to accept, legitimate, or even justify petty corruption (I1, I4)? 2. Is it useful and legitimate to tolerate petty corruption and do we have access to more or better resources by bystanding (IT2)? 3. Do we have cognitive and normative institutions (IT4) and institutional complementarities (IT5) that foster bystanding by acting just formal, without detecting and punishing bystanding behavior in the case of petty wrongdoings. 4. Do we have low levels of trust in dealing with whistleblowing cases and in punishing the perpetrators (IT6)? (Table 3)

#### **3.3 Bystanding as Rational Action: The Rational Choice Theory**

If we want to explain how, in situations of systemic corruption, actors do not take action against corrupt acts but rather condone them, we need to draw on James Coleman's (1990) bathtub framework or "boat model" for a macro-micromacro-explanation in social science. In doing so, we have to relate the situations of systemic corruption, that the actors perceive on the macro level to their decisions on the microlevel. With reference to the perception of the situation, the alternatives to act for an actor are then provided with cost-benefit calculations and the likelihood that they will occur. Thereupon, the collective interaction of different actors is explored, explaining what happens back on the macro level (see Coleman, 1990; Esser, 1993). If we apply this "boat model" of Coleman, we can explain why actors cover for the misbehavior of others even when their preference is to do something about it. When acting together with other actors, they realize only a suboptimal result, which itself does not correspond to their preferences. This also helps to explain how this collective acquiescence stabilizes and secures patterns of systemic corruption in a given society.

**(1) The Systemic-Corruption Situation:** In the RC theory of systemic corruption, an important assumption is that systemic corruption develops along the logic of collective action. In countries with systemic corruption, it is part of the logic of the situation that expectations are socially institutionalized that do favor non-reporting of wrongdoings over reporting (RC1). When there is a perception that many will not speak up and get away with it, non-reporting behavior is further established. Thus, there will be no advantage of speaking up, quite the opposite. In criminology, this corresponds to the theory of differential contacts, which goes back to Edwin Sutherland (1973, 1983). A person becomes delinquent when there are primarily attitudes that favor deviations from rules over attitudes that negatively value wrongdoing. Low generalized and institutionalized trust in the regulative institutions is adding to the systemic corruption situation (RC6). There's uncertainty concerning the answer to the questions: Will I be protected against discrimination, and stigmatization? What are the benefits of speaking up? What will follow after blowing the whistle, will the regulative institutions follow up? *The hypothesis here is: The stronger the perceived expectations of the others towards omission (RC1) and the lower the trust in institutions and the supervisors are (RC6), the more likely wrongdoings are not reported.* 

**(2) The Bridging Hypothesis**: In the theory of cognitive framings, that we use as bridging hypothesis in the framework of Coleman's boat, corrupt action shall first be recognized as wrongdoing and not be booked under the usual routine framings of a given society or organization (see e.g. Ceva & Lubomira, 2018). *The hypothesis is, the more ambivalent the cognitive framing in a society is, the more wrongdoings are not acknowledged as such (RC2), the lower is the probability to report misconduct in an organization.* 

**(3) The Logic of Individual Action:** The benefits and costs of the individual neglect to speak up in the logic of selection are now determined according to the actor's payoff matrix and the probabilities of their occurrence according to the societal or organizational culture. *The hypothesis here is: The higher the costs of speaking up and their probabilities of occurrence are, compared to the benefits (RC3), the more likely it is that misconduct will not be reported.* 

**(4) The Rule of Transformation**: We need to also keep in mind the aspect of stigmatization of whistleblowers as traitors. The fear of stigmatization as a "denunciator" can be a significant deterrent to speaking-up. *The hypothesis is: In a context where reporting misconduct is associated with negative labels and high risks of social ostracization (RC3), there is a higher probability to stay silent to protect their social acceptance and reputation among colleagues.* Stigmatization does perpetuate a culture of silence and inhibits the development of a speak-up culture.

**(5) The Logic of Collective Action:** The game between actor A and actor B in the logic of collective action is not a game of prisoner's dilemma, but a game of cooperation. In this game it can be rational that—under the condition that A and B have to achieve a common result—actor A tries to contribute as little as possible to this result, especially if he sees that actor B does a lot. This problem is known as the free-rider problem. Again, the suboptimal outcome in dealing with this so-called free-rider problem may be that neither A nor B contributes much to the joint outcome of collectively speaking up—unless there's a strong pressure from outside ("principals"), from inside (supervisors) (RC5) or informal institutions rewarding the speaking up (RC4). It may follow from this form of cooperation that the logic of aggregation in systemic corruption situations does not lead to the joint outcome of reporting the crime, but to collective omission. *The hypothesis is therefore: The lower the formal and informal pressure on the joint outcome of reporting appears for A and B in a given society (RC4), the more principal and supervisors are also expected to not follow up the "whistle" (RC5), the more likely such a suboptimal form of cooperation sustains* (Table 4).

By applying this explanatory program, one can thus understand the interaction of actors in societies with systemic corruption that, under certain premises, contributes to the stabilization of systemic corruption even if the actors reject it individually.


**Table 4** Indicators of Bystanding according to Rational Choice Theory

#### **4 How to Analyze and Measure Systemic Corruption**

So how can we empirically determine whether there are more or less pronounced patterns of systemic corruption in a given society? How can we operationalize and measure the indicators presented in the previous chapter? The concluding chapter is devoted to this question. We will see that an understanding of systemic corruption and bystanding inspired by institutional theory often needs a different approach than that of rational choice theory. We also often need to combine methods, qualitative and quantitative, laboratory and field studies, experimental methods and group discussions. Very often, we cannot make use of the usual corruption indices, which are based on expert ratings, because they neither address the indicators nor allow for deeper assessments. From a methodological point of view, a test of the assumptions concerning systemic corruption has empirically high prerequisites that are not easy to meet methodologically. First, rule deviations are always embedded in a "dark field" that can usually only be approximated empirically by proxy variables (on dark field research, see e.g., Haverkamp, 2019). The "bright field" of detected cases tells us something about detected and possibly sanctioned rule violations, but always according to the attributions and investigative activities of social control agencies, e.g., according to the manner of prosecution. Second, the actual validity of the rules is therefore not easy to measure or assess. Although surveys and experiments provide access to rule addressees' knowledge of rules and how they deal with hypothetical case situations, the external validity of such results is controversial (see, e.g., Petzold & Wolbring, 2018). Conversely, the de facto validity is often quantitatively determined by the number of rule violations detected, but this in turn depends on the control and sanctioning activities of social control agencies (see Kersting & Erdmann, 2014, p. 17).

There is no silver bullet for these methodological challenges of the research field. We suggest to address them in four ways:


4. Experimental, as well as simulation and evaluation methods shall also be used in order to better deal with the problem of proving the RC-Theory assumptions.

The research on systemic corruption shall follow a multi-method strategy, combining quantitative and qualitative techniques, assessing the detected cases and exploring the dark field.

#### **4.1 Content Analysis of Detected Cases: Court Records**

Our first task to assess the cultural acceptance of useful petty corruption and find out to what extent we are dealing with an expected behavior (IT1, RC1) cannot be solved by a content analysis of court records. The number of anti-corruption cases and convictions in a given society cannot help us in this regard, because they only reflect the work of the judicial system or law enforcement agencies.

According to Kersting and Erdmann, the analysis of court records refers to the crime scene which is known and registered by law enforcement institutions. Unrecognized acts form the so-called "dark field" (Kersting & Erdmann, 2014, p. 11). Pending or non-judicialized cases are thus not available in the analysis of detected cases. While the detected cases can be measured well, the ratio between detected and undetected cases is usually indeterminate because the "dark field" can only be determined indirectly with the aid of self-reported delinquency or other tools (see below, see also Köllisch & Oberwittler, 2004; Naplava & Walter, 2006; for discussion: Oberwittler & Köllisch, 2004). For example, an increase in police crime statistics does not automatically imply an increase in criminal behavior (see also Köllisch & Oberwittler, 2004, p. 709; Oberwittler & Köllisch, 2004, p. 145). An intensification of prosecution by increased sensitivity of the investigating authorities, or the employment of new prosecutors, etc. may have led to this result. According to Naplava and Walter (2006), increased sensitivity to violence by the general public may be conducive to increased judicial activity through legislative measures, for instance. These can lead to the intensification of control measures, as a result of which, for example, due to increased readiness to report, an increase in detected cases occurs, which in turn can further an increase in the sensitivity to acts of violence (see Kersting & Erdmann, 2014, p. 17).

The method of court record analysis refers to data that are not scientifically produced but mediated by the practice of police, investigating authorities, courts etc. The information in the criminal records or the testing reports may differ from the actual events and provisional occurrences (see also Peter & Bogerts, 2010, p. 46; Kersting & Erdmann, 2014, p. 10). Therefore, it is important to know the conditions of data acquisition in order to assess the data of the court records given that the analysis of detected cases corresponds to a content analysis of documents that were not produced for the purpose of scientific analysis. The mechanisms of data production must be considered in the interpretation, but it still is a preferred choice of method to make sense of cases because of the lack of other methods and approaches.

The analysis of detected cases shall be supplemented by interviews with the involved defense attorneys, prosecutors, and judges.

*Quantitative Content Analysis:* However, since we assume that in societies with systemic corruption there are rather less proceedings with convictions, they can be an indicator of the functioning of the judicial system in a given society (RC4, RC5). Gaining Gaining access to a sample of detected cases, sentenced by the courts, we can idetify, count, and estimate" how high the benefits for perpetrators in detected cases have been, how high the potential costs for useful petty corruption, and the probabilities to be detected were (RC 2) as well, in the case of whistleblower cases, how whistleblowers are treated and what the benefits of speaking up have been (RC3). The ratio of tried cases and sentences per capita provides an indirect measurement for the costs of corrupt acts, the benefits of speaking up and the likelihood to be detected.

*Qualitative Content Analysis:* The qualitative content analysis of court records may add to this information how principals, high ranked executives in political or economic organizations did contribute to establish a corrupt legal organization (RC4, RC5), and if there have been departments, positions assigned to deal with the execution of useful corrupt actions (IT3). By analyzing detected cases, we can also find out, how much cross-over of societal sectors and actors in different societal sectors was revealed (IT4): Judges, Politicians, Third Parties, Companies etc.

The qualitative evaluation of court judgments, trials and court proceedings can be carried out through a classifying, summarizing content analysis like that of Mayring (Mayring, 2010). Based on the research question, the analyst charts categories that result from the material itself, so that argumentation structures become clear across all cases. This is a particularly suitable method for the evaluation of larger text corpora. The content analysis can be carried out with the help of MAXQDA, a software for qualitative data analysis and coding of relevant text passages—a method that has already been used several times in criminological studies (Table 5).



#### **4.2 Survey Research**

Systemic or institutional trust is seen to play a crucial part in the stability and maintenance of the social, political and economic system of a given society (IT6, RC6). "When trust breaks down, the social system is threatened with unrest, the democratic legitimacy of the political system is endangered and the legitimacy of the market-based economy is called into question" (Roth, 2009: 220). There are many surveys available that measure citizens' level of confidence in various institutions (IT6).

*Generalized Trust:* The World Values Survey provides e.g. information on generalized trust and the Latinobarometer shows, that generalized trust is under 20% in 14 of the 19 countries surveyed. "The Latin American Public Opinion Project (https://www.vanderbilt.edu/lapop/) shows in 2021: "More than three in five people in the average LAC country believe that most or all politicians are corrupt. Those who are more educated are more likely to believe that there is widespread corruption among politicians" (Lupu et al., 2021). Surveys such as Latinobarometer and World Values Survey show Latin Americans trusting families and friends (inner circle) rather than others not personally known. (Labarca & Mujika, 2022: 322). Most of the times, historical explanations of particularistic trust are available, highlighting the historical roots and the path dependency that led to different patterns of generalized and institutional trust (IT6, RC6).

*Institutional Trust:* In terms of institutional trust, e.g. "the Latinobarometer survey shows that, in the last decade, the percentage of people that trust their local Congress has surrounded 20%. This number has remained relatively stable during the last decade (2 points of difference between 2011 and 2020), and is a low figure if compared with Europe, for example. Political parties and governments have shown an even sharper decline. In 2011, people who trust political parties reached an average of 22%, while in 2020, it had declined to only 13% in the region. Government trust has had the sharpest decline: 40% in 2011 and only 27% in 2020." (Labarca & Mujika, 2022: 322). Thus, indicators on trust on societal level can be derived from public opinion polls (IT6, RC6).

#### **4.3 Exploring the Dark Field: Lab Experiments, Lab-In-The-Field Experiments, and Factorial Surveys**

When dealing with rule violations, there is always a confrontation with a field in which illicit behaviors are carried out in a concealed and hidden manner. The so called "dark field" can often only be measured indirectly, through anonymous statements or with the help of experiments. We suggest to map that side of the phenomenon through a mixed methods approach combining three different methods:

*a. Lab experiments with students in different countries:* For the paper & pencil research by means of hypothetical case descriptions, so-called vignettes, it is a good starting point to carry out laboratory experiments with students for internal validity testing. The paper & pencil experiments essentially consist of a vignette (hypothetical case description) in which the influence of informal norms in a society resp. in a company on the propensity to cheat, and the acceptance of cheating can be tested (IT1, RC1). The operationalization of these informal expectations and patterns of acceptance can be varied: 1) they can be presented only in a general way (control group), 2) under the condition of a generalized pattern of systemic corruption in a given society (RC1); 3) under the condition of high benefits and low costs concerning cheating (RC2); 4) under the condition of including illicit personal gain necessary for economic reproduction (IT2) and 5) on the condition of a strong relationship of trust with the supervisor within a given company, informally rewarding the useful corrupt behavior (RC4);. In these different settings defined in this way, the participants are able to decide whether they are not acting corruptly, whether they deviate from the rules with personal gain, or whether costs, and benefits or the behavior of the supervisors have an impact on the propensity to cheat.

*Labs-in-the-field:* To measure the acceptance of rule deviations (IT1, RC1) and the likelihood to report (RC3), role-play experiments can be conducted in the context of compliance, vocational training or further education events inside organizations. The participants of the workshops should be randomly recruited as test persons. Before the role-play experiments surveys are carried out with the test persons to identify the baseline concerning the propensity to cheat and to speak up in the case of wrongdoings. After the role-play experiments, group discussions (see Sim/Waterfield 2019; Hennink et al.: 2019 etc.) with all participants take place and are analyzed with the hermeneutical methods to identify the collective mindsets at work (Pohlmann et al., 2014; Pohlmann, 2022), and relating it to the cognitive and normative institutions (IT1, IT4, RC1).

For the implementation, we can draw on a "construction test" and modify it for the purposes of the analysis of systemic corruption and systemic bystanding. The task for the participants as a team is to build a bridge of a certain length and width out of paper. For this purpose, materials such as paper, scissors, glue, etc. are provided. At the end, the team must not only have agreed on the architecture and the manner of construction of the paper bridge, but also present a concrete result, e.g. a paper bridge of 1.5 m length and 1 m wide.

When performing the construction tests, different types of rule violations are built in at different positions. A whistleblower hotline is available in the setting of the game. It then will be recorded how strong the propensity to cheat and to report deviations from the rules of the game is (RC2, RC3). The role of the supervisor in the set up of role-play workshop will be varied in order to find out, if it has an impact on the propensity to cheat (RC4, RC5) (Table 6).

When carrying out the workshops, different types of rule violations can be built in at different positions. Since we are particularly interested in useful rule violations in favor of organizations (organizational crime or useful illegality), which often take place with the participation of managers, we can adopt the


**Table 6** Evidence-Based Research design of the role-play experiments

following experimental arrangement: at the level of middle management, a manager (initiated subject as part of the experimental arrangement) provides illicit resources to support the teams in building up the bridge construction in the specified time and quality. Information about this is communicated to all experimental participants in the form of deliberate hints with the help of an initiated observer. In the initial situation, a compliance department with a whistleblower hotline is available (baseline). We then record whether the willingness to report increases when two additional measures are introduced: (1) informal advice from the management board ("tone from the top"); (2) introduction of negative sanctions, related to the capacity of the teams to build the bridge. Thus, we could find out whether different measures had an impact on the likelihood to cheat and to report the wrongdoings.

*Factorial Survey Experiments:* Factorial survey experiments can be carried out also to clarify what is done by a member of a given society or an employee when faced with petty corruption. In addition, information is to be obtained on whether and which measures are chosen by "confidants" and which are not. For this purpose, a case of misconduct is presented to the test subjects in writing in the form of an online survey. The case is constructed in such a way that it takes place in a typical grey area of the society. The test subjects in the respective society or company under investigation are randomly divided into three groups. Depending on the group to which they belong, the setting in which the fictitious case of misconduct takes place changes:


Now all three randomly selected groups can be compared and, measured by willingness to engage in corruption or to keep silence about corrupt acts, the impact


**Table 7** Mapping the "dark field" of useful petty corruption

of each factor can be accurately determined. With this evidence-based approach, the individual factors that play a role in systemic corruption along the two theories presented can each be examined in their modes of action and thus the dark field can be measured (Table 7).

#### **4.4 Qualitative Interview Analysis: Mapping the Institutional Order**

For Scott (1995), cognitive elements form the framework through which reality is perceived. Scott understands institutions as consisting of cognitive, normative, and regulative structures. Cognitive institutions offer a taken-for-granted perspective, "natural" habits, and accepted rules for understanding the world (cf. Scott, 1995, p. 40). Normative institutions introduce collectively recognized rules about what is perceived as "right" and "wrong" and what "obligations" should guide action (cf. Scott, 1995, p. 37). Regulative institutions are understood as established rules to govern behavior and review others' conformity to them, and to reach it by sanctions, rewards, or punishments (Scott, 1995, p. 35).

Collective ways of thinking are therefore important elements of cognitive and normative institutions and provide the cultural repertoire of how problems can be perceived and solved. Any institutional analysis therefore relies on assumptions about the underlying knowledge bases in a given culture and the collectively accepted rules. However, these bodies of knowledge and underlying cognitive and normative rules are rarely systematically analyzed or mapped. While regulative institutions are easily traced, cognitive and normative institutions are often derived from a general understanding of the culture or formal institutions without empirically elaborating them. However, research on systemic corruption in our perspective shall also be based on the empirical reconstruction of cognitive and normative institutions.

Addressing systemic corruption, we are relying on cognitive and normative institutions to identify the set of underlying informal rules that lead to systemic corruption (IT1, IT2, IT4). They become natural patterns of how to perceive, how to evaluate social facts, and how to act in a specific context of rule settings. The more they become part of the natural life worlds of the people in the social setting, the more institutionalized and stable the informal rules grow.

Taking cognitive and normative institutions seriously helps to understand the fact that even when the organization has set up the necessary compliance rules, deviant self-regulation can be persistent. As pointed out before, compliance rules are often myths and ceremonies in organizations (Meyer & Rowan, 1977). Organizations like to dress their windows to receive legitimacy and subsequently try to acquire impunity, to generate individual liability, or to just reduce the risks of severe penalties. Their cognitive and normative institutions are often very different from these formal rules, even deviating from the formal legal side of an organization. In our perspective, one of the most important characteristics of organizational crime and systemic corruption is that the unwritten cognitive and normative rules of an organization lay the foundations for the use of illegal means, no matter how strict formal compliance rules are.

Especially unwritten rules can be reconstructed qualitatively by means of a Collective Mindset Analysis, as far as they come to bear in the collective thought patterns, given they are collectively valid in the respective field and in the respective organizations. Group discussions can then help to also reconstruct collective thinking patterns as well as norms and habits that answer the question of how to prevent organizational wrongdoings.

#### **4.5 Combining Methods**

Each set of theoretical assumptions, questions, and indicators concerning systemic corruption requires different methodological approaches which can control each other and or can simply complement each other.

Concerning the both approaches to analyze systemic corruption, we thus suggest to measure the following indicators by combining the analysis of detected


**Table 8** Indicators of the institutional theory, and Applicable Methods

cases with the investigation of the dark field of systemic corruption, using qualitative methods as well as standardized ones (Table 8).

#### **5 Conclusions**

Although the topic of systemic corruption has attracted a lot of attention and literature in recent years, the state of research in terms of empirical studies is still not very rich. A theory development on a larger scale has not taken place so far. It was important for us to briefly outline the state of research and the theoretical discussion here. More important, however, is to conceptualize systemic corruption and to operationalize the factors that come into play when conducting research on the topic. The usual recourse to international statistics on the subject is just as pointless as simply taking the data generated by the police. Ultimately, the number of rule deviations that are uncovered largely depends on the activities of the investigating authorities. So, we have to connect "dark-field analysis" with "bright-field analysis" in order to understand at least approximately how systemic corruption occurs, why it is sustainable and productive and how we can explain it.

We sought the starting point in the assumption that systemic corruption and modern formal rule of law embody two different social orders. Both orders provide sufficient guidance for the members of a given society. In societies where the rule of law is not fully established and enforced, we see that cutbacks in one cause the other to flourish (see also Méon & Weill, 2010; Dreher & Gassebner, 2013). Thus, whenever the institutional foundations of modern societies are challenged and lose their instructiveness and steering capacity, such societies can resort to systemic corruption.

Institutional theory explains this by the fact that the older patterns of systemic corruption have been preserved in the cultural repertoire and that it is easy to fall back on them in crisis situations of modern institutional orders. It is institutional path dependencies that take care of the cultural reproduction of this repertoire and thus make recourse to the parallel world of systemic corruption easy. The productivity of this older institutional order consists in the fact that it informally endows subsistence-oriented deviations from the rules with social recognition and legitimacy and thus, bypassing the formal rule of law, allows the supply of relevant goods and services to be maintained through reciprocal give and take. The fact that some people fare better than others is, after all, also true under the condition of a modern rule of law.

Rational choice theory explains this by saying that systemic corruption develops and stabilizes from its logic of collective action, which, assuming that we are dealing with rational egoists, continues to function even if they would prefer the modern rule of law. In the cooperation game of rational egoists, systemic corruption prevails even under this condition and reproduces itself as long as this cooperation game is not broken by the defection of many.

The validity of both explanations and also the degree to which systemic corruption can gain a foothold in a society are questions that can only be answered empirically. And it is precisely for this purpose that this article aimed to provide the tools. Applying these tools will hopefully enable us to gain a better, theoretically instructed and empirically controlled understanding of systemic corruption.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **The Field of Corruption and Anticorruption in Brazil**

## **Anticorruption Institutional Improvements and Operation Car Wash**

Fabiana Alves Rodrigues

#### **1 Introduction**

The task of understanding how corruption occurs within the political and economic systems involves an ambitious undertaking that cuts across different branches of knowledge and different approaches, ranging from the definition of the phenomenon itself and its causes to the proposition of institutional frames designed to curb or discourage corrupt behavior. The most recent researches on corruption highlight that, since the end of the 1970s, there has been a significant increase in academic production that has come to treat corruption as a harmful phenomenon that imposes bottlenecks on economic development. This literature points out that corruption is a disincentive to private investments (Habib & Zurawicki, 2002), shifts talent from creative activities to rentism (Acemoglu & Verdier, 2000), and distorts the allocation of public resources (Gupta et al., 2001; Mauro, 1998). Besides, corruption undermines trust in social relations and trust in government (Morris & Klesner, 2010; Seligson, 2002), undermining democracies' core values.

These analyzes start from a theoretical approach that postulates that institutional reforms can deter or create disincentives to corrupt behaviors (Klitgaard,

F. A. Rodrigues (B) University of Sao Paulo, Sao Paulo, Brazil e-mail: fabi.far@gmail.com

This text is part of the research for a master's degree in Political Science at the University of São Paulo (2019).

1998; Rose-Ackerman, 1999). This approach influenced the literature that investigates the functioning of institutions responsible for the control of corruption. At the same time, it also seeks to identify bottlenecks that prevent effective control of corruption in countries. The issue of corruption is present throughout Brazil's republican history. However, the re-democratization and promulgation of the 1988 Constitution paved the way for the country to undergo progressive institutional improvements that increase transparency and create and strengthen mechanisms of monitoring, inspection, and punishment of corruption. These incremental institutional changes promoted the development of the accountability institutions by expanding state capacities, optimizing specific interactions, and improving the fight against corruption (Power & Taylor, 2011; Carson & Prado, 2014; Machado & Paschoal, 2016).

This chapter seeks to provide a general map of the institutional improvement process that occurred in the country in the years that preceded Operation Lava Jato1 and that somehow outlined the field of action of members of the political class and economic agents in their relation to public assets. This includes creating and improving the country's accountability institutions and an entire repertoire of rules aimed at an increase of efficiency in identifying and punishing corruption. This creation occurred in a scenery of significant influence of international relations in the fight against corruption and money laundering since Brazil has made several international commitments on these topics.

While it is not the subject of this chapter, it should be noted that these changes were essential for the results achieved by Lava Jato, and part of them provided its actors with the capacity for discretionary action, which, in turn, produced politicized criminal control with selectivity traits, as shown in a previous work (Rodrigues, 2020). In this research, we mapped the institutional context in which Operation Lava Jato appeared and carried out a dense analysis of 144 criminal actions that took place in the nuclei of the operation in Curitiba, Rio de Janeiro, and Brasília, since the first ostensive police phase, in March 2014, until December 2018, focusing on the performance of the Federal Justice. This analysis made it

<sup>1</sup> Operation Lava Jato is the largest Brazilian initiative to fight corruption and money laundering. It encompasses dozens of investigations and criminal actions on embezzlement of public resources involving financial operators, large national companies, notably in the construction business, and middle and high-ranking politicians. The first and foremost unit of the operation focused on deviations linked to Petrobras, a state-owned oil company, but there were dozens of developments for other state agencies. The name Lava Jato was used from the identification of a gas station used for illicit financial transactions, which led to the "jet wash" pun as a reference to the high dimension of the volume of resources involved, something large enough to wash a jet. The English translation loses its original meaning because "lava jato" in Portuguese can mean "jet wash" or "car wash".

possible to formulate the argument that Lava Jato results from the combination of institutional improvement and learning, added to the strategic action of the criminal justice system actors, especially the Federal Justice. The notion of strategic action of the argument consists of the performance of actors in the judicial system who place ends above means and calculate their steps and decisions according to the results they intend to achieve, which somehow distorts the essence of Justice, which, unlike Politics, finds its legitimacy in the means and procedures that are followed in the process of decision-making.

This chapter summarizes the first part of this explanatory mechanism for Lava Jato, with some additions about other institutions that are part of the country's anti-corruption system and that participated in the operation, in addition to examples related to the Odebrecht Group. Nevertheless, it should be noted that this institutional dimension alone does not explain the results of Lava Jato, since the strategic action of the actors was essential for the operation to take place as it did and, more importantly, at the rate at which it occurred. Time was a key variable strategically calculated by the actors of the justice system that was essential to obtain the confession and cooperation of financial operators and business people who pointed out the responsibility of leading figures of the political class, as well as to ensure Lula's ineligibility in 2018 and maintain public opinion pressure on the Courts. The strategic action also made it difficult to have effective control over the conduct of the first instance of the Federal Justice, at least until the first half of 2019, when a movement of settlement of accounts started mainly coming from the Supreme Federal Court.

Four sections compose this chapter, in addition to this introduction. The second section presents the history of Brazil's international commitments in the fight against corruption and the crimes associated with it, particularly money laundering. The following section provides an overview of the Brazilian accountability institutions, focusing on those acting within Lava Jato, emphasizing the advancements in the years preceding the operation. Finally, the fourth section presents the main changes in the regulatory frameworks aimed at the severity of punishment and the improvement of the tools for monitoring and investigating corruption and money laundering in the country, followed by a brief section for final considerations.

#### **2 International Commitments Linked to the Fight Against Corruption**

The first aspect that stands out in the institutional changes in Brazil in the years that preceded Operation Lava Jato, notably from the 2000s, is the domestic repercussion of the internationalization movement to fight corruption, money laundering, and organized crime. Brazil internalized several international agreements through which its authorities committed to improve the fight against these crimes and to maintain comprehensive assistance with other countries to ensure the effectiveness of the investigations and punishment of the illegal acts and the repatriation of the agents' advantages. Three international conventions against corruption stand out, one against transnational organized crime, three multilateral instruments of mutual assistance, several bilateral cooperation agreements, in addition to adhering to two international control mechanisms over the country's involvement in fighting corruption and money laundering. Figure 1 shows the time series of international conventions and agreements described in this section, considering the dates on which the country internalized them.

The first international instrument on corruption introduced in the country reflects the global context of the expansion of US influence due to the alleged

**Fig. 1** International conventions and agreements aimed at fighting corruption, money laundering and organized crime (1999–2018). *Source:* Prepared by the author

competitive disadvantage of companies based in that country. In the USA, companies were subject to punishment for corrupt acts committed abroad since the Foreign Corrupt Practices Act (FCPA) enactment in 1977. The Convention on Combating Bribery of Foreign Public Officials in International Business Transactions, concluded in Paris in 1997, settled within the Organization for Economic Cooperation and Development (OECD) Council, started to integrate the Brazilian domestic legal order in 2000 (Decree 3678). Two years later, it had repercussions on creating two new crimes involving international transactions: active corruption and influence peddling (Law 10,467 / 2002). The introduction of these two new crimes did not produce relevant direct effects on the results of the national accountability institutions' repressive actions since none of them appears in the criminal charges filed by Operation Lava Jato until December 2018. According to the implementation report of the provisions of the OECD Convention of October 2014, there were only three cases of investigation of international corruption and one formal charge in the country (OECD, 2014).

On the other hand, besides the provision of transnational corruption, the convention imposed a commitment to assist other countries in supplying information and documents instructing criminal investigations and proceedings, and to cooperate in the execution of a systematic monitoring program to watch and promote the convention's full implementation. Furthermore, Brazil's adherence to the commitments to prevent and repress transnational corruption in some way legitimizes the actions of the United States authorities aimed at punishing Brazilian companies listed on stock exchanges in that country or that are required to file periodic reports with its Securities and Exchange Commission. Also, starting in 1998, cases in which the payment of bribes has somehow occurred in the U.S territory entered the jurisdiction of these authorities. In the case of Odebrecht, for example, cooperation between the authorities of the two countries and Switzerland led to the signing of the world's largest leniency agreement on anti-corruption matters, whereby Odebrecht and Braskem agreed to pay the penalty in the order of three billion and five hundred million dollars, subsequently reduced due to the company's financial incapacity (USA, 2016).

The annual goals proposed by the National Strategy to Combat Corruption and Money Laundering (Estratégia Nacional de Combate à Corrupção e à Lavagem de Dinheiro—Enccla) mentions the OECD convention, of which the impact on the accountability institutions in the country is especially noticeable in reaffirming institutional values to fight corruption. The Enccla is a collaborative network with approximately eighty bodies and agencies, from civil society, the public prosecutor and the executive, legislative and judiciary, the state and federal levels. Established in 2003, the Enccla articulates the arrangement and discussions that formulate public policies and solutions to combat corruption and money laundering. Its plenary meetings have the participation of the main agencies involved with Operation Lava Jato2 : Attorney General's Office (Advocacia-Geral da União— AGU), Administrative Council for Economic Defense (Conselho Administrativo de Defesa Econômica—CADE), Council for Financial Activities Control (Conselho de Controle de Atividades Financeiras—COAF), Office of the Comptroller General (Controladoria-Geral da União—CGU), Department of Asset Recovery and International Legal Cooperation of the Ministry of Justice (Departamento de Recuperação de Ativos e Cooperação Jurídica Internacional—DRCI), Federal Prosecution Service (Ministério Público Federal—MPF), Federal Police (Polícia Federal—PF), Federal Revenue of Brazil (Receita Federal do Brasil—RFB), Federal Court of Accounts (Tribunal de Contas da União—TCU) and Federal Justice (Justiça Federal—JF, through the National Association of Federal Judges—Ajufe, and the Federal Justice Council—CJF). Aside from Ajufe, the other agencies mentioned are part of the Integrated Action Cabinet of Enccla,3 which meets quarterly to follow up the implementation of actions and targets set in the plenary meetings.

Besides the recommendation for revision of the text translated into Portuguese of the OECD Convention in 2007, the Enccla included as a 2012 goal the improvement of the national fight against transnational bribery system to adapt to the convention commands. This set the strategic objective of strengthening governance, integrity, and control instruments in the federal public service, which involved ten participating institutions' engagement. This strategic objective was reaffirmed in 2013 when it hosted a conference to train investigators and prosecutors on international bribery.

<sup>2</sup> We chose to translate the names of institutions and public offices as indicated in the English version of the website of the respective institutions. At the first instance of each of these terms, we will present the original term, followed by the official acronym, if there is one.

<sup>3</sup> Here is the list of the other agencies that are part of the Enccla Integrated Management Office: Brazilian Intelligence Agency (Agência Brasileira de Inteligência—Abin), Central Bank of Brazil (Banco Central do Brasil—Bacen), Special Advisory on Legislative Matters of the Ministry of Justice, Securities and Exchange Commission (Comissão de Valores Mobiliários—CVM), Ministry of Justice, Ministry of Economy; Attorney General of the National Treasury (Procuradoria-Geral da Fazenda Nacional—PGFN), National Superintendence of Complementary Pensions—(Superintendência Nacional de Previdência Complementar—Previc), Special Secretariat of Social Security and Labor of the Ministry of Economy; Superintendence of Private Insurance (Superintendência de Seguros Privados— Susep), National Council of Justice (Conselho Nacional de Justiça—CNJ), National Council of the Public Prosecutors (Conselho Nacional do Ministério Público—CNMP).

The second relevant milestone in the country's adherence to international commitments to prevent and repress corruption, this time without being restricted to international bribery, was the introduction of the Convention against Corruption of the Organization of American States in 2002 (Decree 4.410), which materialized the country's commitment to the primary purposes of the convention: the promotion and strengthening of mechanisms necessary to prevent, detect, punish and eradicate corruption, as well as the incentive to acts of international legal assistance. This commitment was reaffirmed in the same year, with the adhesion to the Mechanism for Implementation of the OAS Convention (Mesicic), through which member states review their legal frameworks and institutions in the light of that convention. Mesicic reports bring recommendations that have been reflected in several institutional changes aimed at improving efficiency in the cycle of monitoring, investigation, and punishment of corruption in the country. One such change is adopting models of specialization of the justice system institutions and the intensification of cooperation between the national accountability agencies. These models' enactment led even to the Constitutional Court's jurisprudence changes, such as the one in May 2018 (STF, 2018). In this instance, the Federal Supreme Court started to adopt a more restrictive understanding of special jurisdiction for high-level authorities, which was also possible due to the strong public support for Operation Lava Jato.

The Convention of Mérida, held under the United Nations, completes the trio of international instruments against corruption internalized by Brazil (Decree 5687 / 2006). This convention reinforces the commitment to cooperate with other countries in investigations involving corruption, in addition to encouraging the adoption of mechanisms for granting benefits to defendants who cooperate with the authorities, which occurred later with an important advance in the legislation on cooperation agreement (plea bargaining), in the year 2013, as will be explained below.

In addition to the corruption itself, the significant investigations that have taken place in Brazil in recent decades point out that the large schemes of diversion of public resources involve several people's activities and the articulation of several companies, with a reasonably organized structure for the division of tasks. This organization was to ensure the success of deviant acts and guarantee that these acts were carried out employing sophisticated financial arrangements that would hide the illicit plot and its members' identity. For this reason, the crime of corruption is usually associated with organized crime, which also stands out in the context of institutional improvement aimed at curbing the practice of high-level corruption. Brazil's adherence to the United Nations Convention against Transnational Organized Crime, known as the Palermo Convention and introduced in the country in 2004 (Decree 5015), was another step toward conferring more efficiency against systemic corruption that is said to exist in the country. In addition to strengthening the legitimacy of the anti-corruption speeches, the convention encouraged the adoption of some changes that have occurred in the following years: the autonomous criminalization of participation in a criminal organization; the use of mechanisms of a premium nature (plea bargaining), and the suppression of the list that restricted the crimes that could be antecedents to money laundering.

When it comes to corruption involving prominent businesspeople and highranking politicians, the scenario of illegality likely accompanies the practice of concealment acts to hide the illicit origin of the resources and their real owners. That is why the investigations into grand corruption are expected to include investigating the crime of money laundering, based on a strategy known as "follow the money", emphasizing the institutional design developed to curb money laundering when seeking to map the anti-corruption system from any country. Operation Lava Jato offers an excellent example of the importance of anti-money laundering mechanisms in corruption investigations. Of the eighty-two indictments (charges) between April 2014 and December 2018, only two did not accompany money laundering charges, by far the most recurrent crime in the operation. In the nucleus with charges filed with the Federal Court of Curitiba, it was attributed to 337 people in 515 charges, while the crime of active corruption appears against 173 people in 165 charges.

The first international convention introduced in the country that dealt with money laundering, celebrated within the United Nations' scope and called the Vienna Convention (Decree 154/1991), aimed at combating the illicit traffic in narcotics. In this international scenario in which countries concentrated efforts in the fight against drugs, the law that criminalized money laundering (Law 9,613 / 1998) was introduced in the country, initially with a specific list of crimes that should precede the practice of money laundering to allow its autonomous criminalization.

One of the most relevant international landmarks for combating money laundering in Brazil occurred in 2000 when the country became a member of the Financial Action Task Force against Money Laundering and Terrorist Financing (FATF), an agency created in 1989 within the scope of the G7 (group of the seven wealthiest countries in the world). The FATF is an intergovernmental organization that develops and promotes national and international policies to combat money laundering and terrorist financing. Although the agency's work has no binding force (hard law), it has a strong influence and effectiveness (soft law) on member countries. This influence is especially notable through the forty recommendations that constitute the international anti-money laundering standard (ALD), recognized by the World Bank and by the International Monetary Fund (IMF). The organization also conducts periodic visits to member countries to evaluate the implementation of the measures of prevention and repression of money laundering and terrorist financing. A great indicator of the FATF's influence on institutional changes linked to the fight against money laundering (and corruption) appears in the Enccla's annual actions. Table 1 summarizes the actions from 2004 to 2019 that expressly refer to the FATF recommendations, indicating the eventual results.

The data show that compliance with the FATF recommendations has been the object of concern for the country's accountability institutions, suggesting that money laundering through international financial transactions now has major obstacles, considering the country's strong commitment to international agreements of legal assistance in criminal matters. Moreover, one can point to the strengthening of the agencies responsible for the actions of mutual legal assistance (Office of the Prosecutor General and Department of Asset Recovery and International Legal Cooperation).

In addition to adopting the Mercosur Protocol on Mutual Legal Assistance in Criminal Matters (Decree 3468 / 2000), the Inter-American Convention on Mutual Assistance in Criminal Matters (Decree 6340 / 2008), and the Convention on Judicial Assistance in Criminal Matters between the Member States of the Community of Portuguese Speaking Countries (Decree 8833 / 2016), Brazil signed several bilateral agreements that allow the streamlining of assistance in investigations and production of evidence. This process also entails seizure measures, retention, and transfer of goods, essential to prevent the fruition of concealed goods obtained illicitly. Between 2001 and 2013, sixteen bilateral assistance agreements were introduced, nine of which involve countries that collaborated with the Lava Jato investigation until March 2019: Canada, China, South Korea, Spain, United States, Ireland, Panama, United Kingdom, and Switzerland (Brazil, 2019).

The Federal Prosecutors Service's task force reports that more than 630 acts of international legal assistance were carried out from 2014 to December 2019, which involved relations maintained with almost a third of the countries in the world (MPF, 2020a). This number shows the relevance of international cooperation for the results of the Operation Lava Jato. The charges against Odebrecht executives, for example, were not only partially supported by documents obtained through assistance with other countries but led the company to start a legal dispute to prevent the use of documents sent by the Swiss authorities regarding


**Table 1** Description and results of Enccla's actions that mention the FATF (2004 to 2019)

(continued)


**Table 1** (continued)

(continued)


**Table 1** (continued)

Prepared by the author from Enccla (2020)

movements in offshore accounts that the Odebrecht Group allegedly controlled. Although the practice of "entraide sauvage" (wild legal aid) was recognized, as the documents that were of interest to Brazilian investigators were sent in a request for diligence made by Switzerland to instruct an investigation that was going on in that country, Odebrecht was not successful in excluding the evidence that instructed their executives' first conviction. That trial included Marcelo Odebrecht, the president of the holding company at that time, who was sentenced to nineteen years and four months in prison in a closed regime, in addition to the obligation to pay damages of approximately sixty-three million dollars.

The relevance of the international instruments concerning the country's anticorruption activities described in this section goes beyond their imposing force by internalizing the national judicial order. They also reverberate the discourse about the importance of combating corruption in the national institutions, contributing to the engagement of the control agencies, although each of them also seeks their own interests, which are not always declared, as they find support in the legitimacy that public opinion confers to the fight against corruption.

#### **3 Strengthening of the Accountability Institutions**

Harmful actions to public assets or the abuse of public power for private benefit may constitute illicit acts as provided for in a broad national legislation. This legal characterization can be arranged in several analytical frameworks, as there are relevant variations in several elements. These variations may refer to the quality of the agents that concur for the illicit (public or private agents; persons or companies); the nature of penalties (economic or monetary, restricting rights and custodial); the procedures required for sanctions impositions (administrative or judicial); the relationship between the nature of the infraction and the competent authority for sanctioning (political, administrative or judicial).

The detailing of these different frameworks escapes this chapter's objectives. Nevertheless, it is important to highlight that this complex legal structure of protection of fairness in dealing with public affairs helps to explain the existence in the country of a complex design of agencies with attributions that at times compartmentalizes and, other times, integrates continuous flows of information. These flows depend on more than one agency to incorporate the process that can go from identifying the illegal act to the application of the sanction (Power & Taylor, 2011; Machado & Paschoal, 2016; Carson & Prado, 2014). Criminal proceedings, for example, are restricted to charges against people (do not include companies, except in some environmental crimes), they involve the necessary participation of the Public Prosecution Service and the Judiciary, and are the only ones that can lead to imprisonment. The Prosecution Service is responsible for criminal charges, which are usually supported by investigations carried out by the police (Federal Police and civil police of the states), or can rely on evidence shared by other agencies. These agencies produce evidence in their regular exercise of inspection and punishment of conduct framed in different types of offenses, which is quite common when the same behavior is classified as a crime and as an administrative or civil offense. The Courts are responsible for conducting criminal actions in which evidence is produced under adversarial proceedings (possibility of the defense participation), in addition to the final judicial decision on the guilty of the defendants and the application or not of sanctions.

On the other hand, the administrative procedures of accountability of individuals and businesses for engaging in fraudulent acts and embezzlement of public funds can operate through several agencies with relative autonomy to investigate and apply penalties. Administrative contracts between private businesses and the government, for example, are subjected to the supervision of the contracting part and sanction power of the respective Minister of State or State Secretary, in addition to monitoring and sanctioning by the Federal and State Accounting Courts (Bidding Law). The Anticorruption Law (Law 12,846 / 2013), which provides for civil and administrative liability of companies for the practice of acts harmful to the national or foreign public administration, describes conduct similar to the illicit provided for in other legal norms. This provision leads to overlapping agencies' attributions concerning supervision and punishment related to the same facts.

One can say that the country has a tangle of rules on types of offenses and sanctions, different forms of accountability for its practice, and distribution of competencies among different accountability institutions. This leads the literature to focus not only on each agency's functioning, but also on the forms of interaction between them and the result of these interactions, which can lead to conflicts and inefficiency, but also competition, cooperation, collaboration, and compensation. The complexity of these interactions keeps the debate open whether the system should be better integrated and coordinated or more decentralized to optimize the performance of the anti-corruption system (Prado & Cornelius, 2020). It is important to highlight that the debate focused on the efficiency of accountability institutions in general gave little importance to equal and even more relevant issues, such as the deficiency in the control over some of these agencies and its reflexes on democracy.

This section does not intend to deepen this debate. It presents a general picture of the country's anti-corruption institutions, focusing on those involved with the results of Operation Lava Jato, assuming that the state capacity of these agencies is included in the strategic calculations of people and companies that carry out activities in Brazil, eventually on risks associated with the practice of illicit acts, but above all as a constraint on the choice of means to be employed to achieve their political and economic goals.

The literature dedicated to studying the Brazilian accountability institutions recognizes a progressive process of institutional improvement by creating and developing agencies devoted to monitoring, investigating, and punishing illegal acts involving corruption and misappropriation of public resources (Power & Taylor, 2011; Aranha, 2018).

The central agencies dedicated to anti-corruption and anti-money laundering actions are on the list of twenty-two agencies that make up the Enccla's Integrated Management Office. Still, we highlight those expressly mentioned by the task force Lava Jato as agencies that put forth the operation results (MPF, 2020a). These institutions operate in the federal level, but it should be mentioned that the Federal District and the twenty-six states have their accountability institutions, whose designs bear similarities with the federal model. However, they do not always possess equivalent state capabilities. We summarize these institutions' roles in the anti-money laundering and anti-corruption activities, indicating the most significant changes in the last decades and their role in Lava Jato. The institutions analyzed are listed in Table 2, with an indication of the main activities aimed at controlling corruption and money laundering.

The Federal Police and the Public Prosecution Service, next to the Federal Justice, may be considered the main protagonists of the Lato Jato Operation and of the most significant anti-corruption operations in the country, either by direct action in criminal investigations or by that part of its agents who engaged in a kind of advertising crusade seeking public support for Lava Jato, under the pretense of preventing their strategies of being hampered by other institutions and actors. (Dallagnol, 2017; Moro, 2018). The justification presented by these actors gave them legitimacy in society. However, that justification shroud the personal and institutional interests and advantages achieved through this media role in the criminal justice system. It also blocked debates about the limits that must be observed by these institutions and removed discussions about structural reforms that would best achieve the objective of discouraging the systemic corruption that is claimed to exist in the country.

The Public Prosecution Service conquered relevant spaces of power after the 1988 Constitution, which separated the institution from the Executive Power and gave it significant autonomy to carry out a wide range of functions. These functions were supported by the broad task of defending the legal order, the


**Table 2** Institutions involved with Lava Jato and its main anti-corruption and anti-money laundering activities

Source: Prepared by the author

democratic regime, and the unalienable social and individual interests (diffuse and collective rights). Concerning the control of corruption and the misuse of public resources, the Public Prosecution Service can use two judicial means to obtain the punishment of those responsible for the illicit acts: public civil action for acts of administrative improbity and criminal prosecution, which are not mutually exclusive and were adopted by Lava Jato.

The institution also has at its disposal an instrument of consensual nature to protect transindividual rights, the Term of Conduct Adjustment. This term allows the replacement of penalties provided by law for obligations assumed by the offending agent to adjust his conduct to the scope of lawfulness. There was controversy about the possibility of reaching this agreement in cases involving administrative improbity, prohibited by law but regulated in 2017 by the National Council of the Public Prosecutors. The controversy ended with Law 13,963, of December 24, 2019, which expressly authorizes the use of agreements in cases of administrative improbity. This change follows the tendency to appreciate consensual mechanisms in the face of the results obtained by Lava Jato from the collaboration and leniency agreements, introduced by US influence, which will be described in Sect. 4.

In the case of MPF, the process of institutional development was accompanied by investments in human and financial resources (Arantes, 2011b) and more pronounced independence between 2003 and 2017, when presidents Lula and Dilma Rousseff, both from the Workers' Party, adopted the practice of selecting the first name of a list of three drawn up by the members of the Federal Prosecution Service for the post of Attorney General (Viegas, 2019).

The MPF's great autonomy and discretion led some authors to claim that the institution has become a kind of fourth power (Arantes et al., 2010; Kerche, 2018b), devoid of accountability mechanisms over the performance of its members (Avritzer & Marona, 2017; Kerche, 2018a), emphasing its leading role in major operations to combat corruption.

Lava Jato illustrates the conjunction of the MPF's wide discretion devoid of accountability, since this allowed the creation of anti-corruption task forces with different action capacities, without legal justifications or indications of objective parameters for differentiated treatment in the three main cores of Lava Jato, which explains, at least partially, the production of disparate results, as can be seen in Table 3. The data on the respective task forces refer to their creation and the initial number of federal public prosecutors who made up the task force. For each of the operation's units, the results produced regarding the number of phases of police operations, the charges presented, and the trials at the first (judgments) and second instances (appeals) are also indicated, all of those within two years and six months of the creation of the respective task force.

This complexifies the discussion about the real effects that operations such as Lava Jato produce in discouraging the practice of corruption by businesspeople and politicians. Firstly, because it enables selectivity in the criminal punishment of corruption, even allowing specific economic sectors and political groups to be deliberately targeted, with the possibility of signaling the politicization of the



Source: Prepared by the author

justice system and not the improvement of criminal control at the national level. Moreover, the combination of ample discretion and low control makes it difficult and even prevents the identification of MPF's real motivation for choosing priorities, which provides an incentive to the practice of corruption within the institution (Klitgaard, 1998).

The Federal Police is part of the Ministry of Justice and Public Security and performs various functions, such as the prevention and repression of drug trafficking, migratory police activities, control of possession and trade of firearms, and act as the federal investigation police, that is, its officers conduct investigations involving federal crimes. The Federal Police underwent a process of institutional construction, marked by investments made by the Executive Branch since the early 2000s. This process led to its material and human resources equipping, with an increase in its budget, holding regular public tenders, and significantly increasing the number of police officers and other employees, which resulted in a progressive increase in the famous operations (Arantes, 2011a).

The judges enjoy significant independence and autonomy in the 1988 constitutional regime, with prerogatives of office stability and life tenure (dismissal only through judicial proceedings and mandatory retirement at age 75). Access to first-instance courts is only possible through public tenders. The Appeals Courts comprise career judges and reserve a fifth of its seats to members of the Brazilian Bar Association (Ordem dos Advogados do Brasil -OAB) and prosecutors. The Superior Court of Justice (Superior Tribunal de Justiça—STJ) overlaps the court of common justice (federal and state) and the Supreme Federal Court (Supremo Tribunal Federal -STF), which is the constitutional court that combines standard jurisdictional functions (original and appeals). Only Justices nominated by the President of the Republic comprise the STJ and STF, which must be referred by the Federal Senate after a hearing where the nominee responds to questions from members of a committee of senators. It is required that two-thirds of the STJ judges come from appellate courts (federal and state) and one-third from members of OAB and public prosecutors. Figure 2 shows a simplified organization chart of the Brazilian Judiciary, with common justice on a white background and specialized courts on a gray background.<sup>4</sup>

Notes: (1) State and Federal small claim courts are not included, which are competent to judge small value cases and infractions of less offensive potential.

<sup>4</sup> The organization of the Brazilian Judiciary is characterized by the specialization of the branches of justice in a structure that reflects the federative system, with a complex division of powers between the courts, divided by subject (common and specialized), complexity (courts and common justice), instance (original and appeal), federated entity (federal or state), appeal (ordinary, special and extraordinary).

**Fig. 2** Organogram of the Brazilian Judiciary.

Notes: (1) State and Federal small claim courts are not included, which are competent to judge small value cases and infractions of less offensive potential. Also not included are Military Courts of Justice in 3 states and Councils of Military Justice in 27 states. *Source:*  Prepared by the author

Also not included are Military Courts of Justice in 3 states and Councils of Military Justice in 27 states.

Arantes (2011b) argues that the repressive control of political corruption promoted by the Public Prosecution Service in Brazil went through a double shift during the 2000s. First, there was an emphasis on the state sphere that used civil mechanisms (public class action suit and civil investigation); then, there was a concentration of efforts at the federal level through significant criminal operations, comprised of greater articulation between the three institutions involved (PF, MPF and Federal Justice). One may say that this arrangement found its culmination in Operation Lava Jato. The strategic action of the Federal Judiciary was essential for the results achieved, notably in the central unit of Curitiba. Noteworthy are the ad hoc changes in the state capacity of the courts responsible for the operation; judicial decisions with concealment of information that could move the case to another court; and the selective streamlining of cases, according to the selected priority of the defendants to obtain cooperation agreements that would lead to political actors being held accountable, with the former President Lula as the main target (Rodrigues, 2020). Despite the relevance of this strategic action for the production of results, including and especially making Lula's candidacy in 2018 unfeasible, there is a history of institutional development of the Federal Justice that was also essential for Lava Jato, of which we highlight: a significant increase in the number of local courts, which rose from about three hundred in 1996 to almost a thousand in 2016; specialization of judicial units focused on the significant white-collar crimes (money laundering and crimes against the national financial system); substantial investments in the qualification of human resources; and progressive implementation of technological tools for optimization of judicial activities, such as fully electronic proceedings and software for online fulfillment of court orders aimed at producing evidence (Rodrigues, 2020). These technological advances have led to more efficiency in tracking the resources of complex financial operations that are usually associated with white-collar crimes. This process brings to the scene three institutions that have cooperatively integrated into the flow of investigation and punishment of these crimes: the Control Council of Financial Activities (Coaf), the Federal Revenue of Brazil (RFB), and the Department of Asset Recovery and International Legal Cooperation (DRCI).

Coaf was created in 1998 and, as the Financial Intelligence Unit of Brazil, produces intelligence reports disseminated to various national accountability agencies. It also compiles relevant financial data from communications made by multiple entities, public and private, which are part of sectors usually used in money laundering. Despite the lean administrative structure, its reports have been essential for tracking surreptitious financial transactions that lead investigators to the illicit assets of people and companies involved in the practice of corrupt acts. Table 4 shows the amount of exchange information between Coaf and other agencies in the country, from 2003 to 2019, while Fig. 3 shows the number of Finance Information Reports (RIF) produced and propagated annually by Coaf in the same period. This data can be considered a good indicator of close cooperation aimed at screening suspicious financial transactions and thus the possible increased costs associated with the practice of wrongdoings by covert financial mechanisms. One should note that, in 2019, eighty-three percent of the total exchanges were carried out with the police and prosecution' service.

The RFB generally escapes the radar of academics who dedicate themselves to studying the accountability institutions focused on combating corruption, which is explained due to its institutional mission aimed at the Union's tax and customs administration activities. As it occupies a central role in the financial activities related to raising public revenue, the federal government has promoted constant investments in the institution, ensuring effective tax collection. One should highlight the technological advances for more accurate taxpayers' records, tax


**Table 4** Amount of information exchange between Coaf and other agencies (2003–2019)

Source: Adapted from table 4 of Coaf (2020, p. 25)

intelligence, and service agility (Ezequiel, 2018). It is important to notice that, in 2009, OCDE approved a recommendation to strengthen the role of tax authorities against bribery (Recommendation on Tax Measures for Combating Bribery of Foreign Public Officials in International Business Transactions). Recommendations are non-binding legal instruments, but compliance with their provisions is subject to pressure by the OCDE assessment system and may be required as a condition for entry as a member, a claim made in May 2017 by Brazil, which has the key partner status since 2007 (Thorstensen & Nogueira, 2020).

Two critical elements help to explain the relevance of the RFB in the control of illicit activities that involve corruption and embezzlement of public resources but remain unnoticed in some analyzes on the accountability institutions (Power & Taylor, 2011; Prado & Carson, 2014; Machado & Paschoal, 2016). Firstly, the agency is part of the country's list of anti-money laundering institutions, with an explicit provision for prevention activities in the institution's bylaws. The RFB

**Fig. 3** Amount of Financial Information Reports (RIF) produced by year. *Source:* Prepared by the author, based on Coaf (2011, 2020)

occupies one of the eleven seats in the Coaf plenary session, is part of the Integrated Management Office of Enccla, and participates in the National Network of Anti-Money Laundering Technology Laboratories—REDE-LAB, which will be discussed below. Moreover, the Complementary Law 105 / 2001, allowed the RFB direct access to taxpayer's bank transitions without relying on a court order, as is the case in investigations carried out by police and public prosecutors. This shortcut to carry out global financial analyses, with tax and banking data, gives the RFB a competitive advantage, which helps explain its active participation in significant police operations, being present even in the press conferences held by Lava Jato investigators.

The annual report on the activities carried out by the RFB in 2019 confirms that it operates even before the ostensible phases of the operation, "through crossings and analysis of internal data carried out by the investigation sector" (Brazil, 2020a). Cooperation with criminal investigation agencies also led to the development of Sislava, a data management software for processing and analyzing information shared by criminal justice system. In 2017, there were more than 3.5 million pages of documents containing information on 58.7 thousand people and companies (Brazil, 2017). The RFB also formalized special inspection teams to act in other large police operations: Zelotes, which investigates frauds committed by lawyers and companies within the scope of the Administrative Council for Tax Appeals (Conselho Administrativo de Recursos Fiscais—Carf); Calicute, which has as its object the deviation of resources in the management of the government of the state of Rio de Janeiro; and Greenfield, which investigates pension fund fraud (Brazil, 2020a).

The activities of tracking illicit resources that move through more sophisticated financial networks also count on the relevant performance of the DRCI, created in 2004 under the Ministry of Justice. Among its primary functions, the DRCI coordinates the agencies involved in combating money laundering and transnational organized crime, especially the recovery of assets and acts of international legal assistance in criminal matters. The agency exercises central authority functions in almost all international legal assistance agreements and treaties signed by Brazil.5 Intermediation in carrying out acts of international legal assistance allows procedural steps such as depositions and arrests. It also allows the search for evidence of illicit behavior that leaves traces abroad, notably deposits held in bank accounts undeclared to the Brazilian authorities. In 2004, the agency brokered 780 requests for international assistance and 2,439 in 2018. Table 5 shows the values repatriated through the DRCI since 2007, indicating the police operation they refer to (eighty percent of them at the federal level) and the countries involved in the cooperation.

In addition to the essential brokering of requests for international legal assistance, the DRCI hosts the Anti-Money Laundering Technology Laboratory (LAB-LD), created in 2007 to overcome difficulties encountered in money laundering and corruption investigations in the analysis of the large volume of banking, tax, and telephone information. The laboratory works as a model to develop and disseminate the best technological and training practices. This model was replicated to other federal and state agencies in 2009, with the National Network of Technology Laboratories (Rede-Lab), which in 2019 possessed 58 units in operation and five in installation phases (Brazil, 2020b). The network's laboratories analyzed more than seven thousand cases with traces of illegality between 2006 to 2017 (Brazil, 2018), which materializes a background activity that reinforces the effectiveness in producing more solid evidence about complex financial tracings.

<sup>5</sup> International cooperation agreements and treaties provide that the Ministry of Justice occupies the role of Brazilian central authority, but this function is exercised by the Department of Foreigners in requests for extradition and transfer of convicted persons, and by DRCI in other cases. There are two exceptions to cooperation in criminal matters, which envisage the Office of the Prosecutor General as the central authority: bilateral treaties with Portugal (Decree 1.320 / 94) and Canada (Decree 6.747 / 09).


**Table 5** Repatriation intermediated by DRCI (2007–2018)

The control of the legality of acts performed by public officials is not limited to that exercised by the justice system since various organs of the federal structure have their departments of internal affairs. Moreover, since 2003, the federal anti-corruption agency, the Office of the Comptroller General (CGU), began its operation. The CGU reorganized the internal control of the federal administration by making it centralized and started to include activities to promote the quality of public management and monitor national public policies carried out by subnational entities with federal funding.

CGU's mission is to promote the defense of public assets, prevent and combat corruption, and increase management transparency in the federal area by establishing investigations and administrative proceedings or, as the case may be, by pressing charges to the competent authorities to investigate the omission of public officials. Also noteworthy is the innovative function of inducing and strengthening social control through the participation of civil society (Loureiro et al., 2012). Figure 4 shows the number of special operations conducted by the CGU

Source: Prepared by the author, based on DRCI reports and the MPF website

**Fig. 4** Special operations conducted by CGU (2003–2019). *Source:* Prepared by the author, based on CGU (2020)

in partnership with other agencies, federal and state, with their calculated potential losses, which gives an idea of the importance of his role in determining unlawful at the expense of public funds.

The CGU gained even more relevance after the entry into force of Law 12,846 / 2013, known as the Anti-Corruption Act or Clean Company Act. In addition to providing for administrative and civil liability of companies for acts committed against public administration, both domestic and foreign, this law introduced the leniency agreement for the offenses provided for in the law. It also made CGU responsible for celebrating this agreement within the scope of the Federal Executive Branch. These agreements have become an essential tool for identifying those responsible for wrongdoings that are harder to determine and as a means of imposing more efficient compliance mechanisms in the structure of large corporations.

When it comes to leniency agreements in Brazil, one must highlight the Administrative Council for Economic Defense (CADE). Since 1994, CADE has gained status as a federal autarchy linked to the Ministry of Justice, with structure and instruments that allowed it to exercise more effectively the protection of free competition in the market. It carries out its mission through different kinds of activities: preventive, when deciding on mergers and other acts of economic concentration that may affect free competition; repressive, in the investigation and punishment of the practice of cartels and acts harmful to free competition; and educational, through various activities in a civil society focused on the subject of free competition. The main bridge connecting CADE to the other branches of the anti-corruption institutions is the close connection between the practice of cartels<sup>6</sup> to circumvent the competitive principle of contracts signed with the government and corruption pacts in the execution of these contracts. This spurious symbiosis appears in the main charges filed by Lava Jato prosecutors and marks the first antitrust leniency agreement entered into in the operation, in partnership with CADE and the MPF. This first agreement narrated the anti-competitive behaviors involving at least 23 companies, practiced in the industrial onshore assembly works market in Petrobras' tenders, which would have lasted from the late 1990s until at least the beginning of 2012 (CADE, 2015).

Figure 5 shows the history of leniency agreements concluded by CADE, highlighting those that are part of Operation Lava Jato. Since the early 2000s, there are leniency agreements concluded by CADE, but only involving anti-competitive practices, since the leniency linked to corruption was only introduced in August 2013 (Law 12,846). One should note that there was already a growing use of the institute. Still, there is a significant increase after Operation Lava Jato, partially explained by the institutional proximity established between CADE and the MPF, since the first antitrust leniency agreement. This joint experience even guided the role of the MPF in modeling the anti-corruption leniency agreement (Pimenta, 2020).

Finally, mention should be made of the Federal Court of Accounts (TCU), which also is part of the list of institutions that monitor, inspect, and punish deviant behavior of public authorities. Besides occupying the role of auxiliary agency of the National Congress in monitoring the budget and financial execution, TCU performs the external control of the federal government through various competencies. These competencies include the supervision of compliance with federal funds passed on to sub-national entities; the application of sanctions and correction of illegalities practiced by public and private agents who manage the federal assets and resources; and the judgment of the accounts rendered by managers or those responsible for the custody and use of federal public resources. The studies on the performance of accountability institutions in the country usually emphasize that TCU is one of the strongest institutions among its Latin American counterparts. Despite imposing reputational costs to politicians, with repercussions on the elections' results (Pereira & Melo, 2015; Winters & Weitz-Shapiro, 2013), the TCU audits are technically rigorous. However, it has weak enforcement of penalties, primarily because they depend on political and

<sup>6</sup> Cartels, punished as an administrative infraction and as a crime by Brazilian law, can be defined as agreements between competitors to adjust prices, offers, divide markets, and other variables that may have relevance in market competition (Laws 8,137 / 1990 and 12,259 / 2011).

**Fig. 5** Leniency Agreements concluded by CADE (2003–2019). *Source:* Prepared by the author, based on CADE (2020)

legal institutions in subsequent phases, with coordination and cooperation deficits between institutions (Speck, 2011; Taylor & Buranelli, 2007).

Unlike what is observed concerning other agencies described in this section, some evidence suggests that there has not been a harmonious and cooperative integration of TCU in the investigations related to Operation Lava Jato. The institution is not even mentioned in the main portal to disseminate the operation results (MPF, 2020a), which possibly expresses the little acknowledgment of TCU's performance in assessing unlawful at auctions held by Petrobras. This is confirmed by the existence of only one charge of fraud in public bids until December 2018 by the nucleus of Curitiba.

In addition, some Lava Jato investigations reached members of TCU, which may have reduced the trust and accentuated tensions between the institutions, as observed by the following events: i) the investigation before the Supreme Court involving the ministers of the TCU Raimundo Carreiro and Aroldo Cedraz, the latter formally accused in October 2018 (case No. 4075); ii) the charge filed by the MPF in Curitiba against the former Senator and now Minister of TCU Vital do Rego Filho, who allegedly received R\$ 4 million in bribes as Senator and President of the CPI of Petrobras (action n. 5,041,210–45.2020.4.04. 7000); iii) the charge filed by the MPF in Rio de Janeiro about corruption involving TCU auditor Cristiano Albuquerque and the influence peddling against several lawyers hired on the pretext of influencing decisions at TCU (action n. 5,053,463–93.2020.4.02.5101); iv) the Office of the Comptroller General obtains an injunction from the Supreme Federal Court preventing the compliance with the TCU decision determining that all information on leniency agreements conducted within the scope of the Lava Jato were to be sent to TCU (procedure n. 340,031); v) an investigation initiated by TCU to assess the dimension of the parallel budget administered by the MPF due to agreements signed (case records 007.597 / 2018–5); vi) a proceeding initiated by TCU to investigate the alleged irregular purchase of telephone interception equipment by the Lava Jato task force in Curitiba (case no. 025.306 / 2020–0).

This diagnosis finds confirmation in the content of messages that have been exchanged between the members of the task force Lava Jato, obtained by hackers and published by the media under the name "Vaza Jato" (Leak Jet). Despite doubts concerning the content's authenticity, an excerpt referring to the suspicion deposited on the TCU is transcribed in a conversation that allegedly occurred on April 17, 20167 :

Júlio Marcelo (Federal Prosecutor at TCU): My friend, if Gim Argello is going to whistleblow, I would like to make a suggestion/request: include with detail and evidence the TCU ministers. Certainly, Vital do Rego played a vital role in Petrobras' CPI so that it would come to naught, but I suspect that they both had essential articulations at TCU (Cedraz, Nardes and Carreiro, with the participation of Tiago Cedraz and/or Carlos Juliano Nardes).

Deltan Dallagnol (coordinator of the Lava Jato task force): We need to clean up the TCU. TCU's role is too important for it to be infiltrated by bandits. I remember that Nardes supported the departure of Gim Argello to TCU and only changed sides down to the wire.

Except for these more nebulous facts involving TCU, one can say that in recent decades, Brazil has undergone a progressive strengthening of corruption control institutions, which goes beyond the triad traditionally involved with the large police operations that have become routine in the country. There is also a growing integration of other contributing agencies, each with its expertise and focus of action, with the exposure of the intricacies of non-republican practices in the management of public assets and collective interests. On the other hand, Lava Jato shows that the strengthening of these bureaucracies comes up against questions about the legitimacy, limits, and control of anti-corruption institutions, especially

<sup>7</sup> See Veja Magazine. Available at: <https://veja.abril.com.br/blog/radar/em-mensagens-pro cçador-chama-ministros-do-tcu-de-bandidos-infiltrados/?fbclid=IwAR1cX5Gh9qgwb1ZS MgGCVYk5FwRfd5lc\_viR0mMHz11-\_. Accessed on September 13. 2020.

when their activities advance to influence electoral competitions or contribute to the delegitimization of party politics, with damaging reflexes for the functioning of democracy.

#### **4 More Rigorous Punishment and Improvement of Investigation Tools**

Brazilian legislation has undergone a progressive hardening of the politicalelectoral consequences resulting from a criminal conviction. Since 1984, sentences exceeding four years for crimes committed with abuse of power or violation of duty towards the public administration entail the loss of public office and elected mandate. This four-year baseline was reduced to one year in 1996, when the electoral legislation from 1990 was already in force, which established ineligibility, for three years, as a consequence of the definitive criminal conviction for several crimes, including crimes against the public administration and the financial market. In 2010, this legal text was hardened, imposing ineligibility for eight years in convictions involving a long list of crimes, in addition to anticipating these effects to the decision rendered by a judicial panel, even without judgment on all instances.

The legislation has also toughened the punishments for corruption, money laundering, and organized crime, which can be considered the main charges of the central core of Lava Jato. Table 6 shows the main laws, the changes promoted and the politicians involved in their approval, highlighted in bold those that were later affected by Lava Jato.

In 2003, the penalty for corruption was increased from one to eight years to two to twelve years. On the same occasion, the duty to repair damages caused by crimes against the public administration was also imposed as a condition for the progression of the regime.8 The overall effectiveness of criminal sanctions was increased in 2010, eliminating one of the statutes of limitations instances. The advances on the assets of people involved in acts of corruption and money laundering also stand out in the legislative changes between 2012 and 2013. These advances refer to the blocking of lawful assets equivalent to the benefit of crime;

<sup>8</sup> The sentence of liberty deprivation is characterized by the length of incarceration and the initial regime of its serving, which can be closed, semi-open or open. These regimes differ by the type of establishment where the sentence must be served and by the rigor imposed on the prisoner. The sentences are served progressively, allowing the transfer to a less severe regime, which depends on the merit of the convicted and the fulfillment of a fraction of the sentence in the most severe regime.


**Table 6** Main laws related to criminal anti-corruption operations (2000–2014)

(continued)



Source: Prepared by the author

the express authorization to block goods of illicit origin in the name of interposed persons or stooges ("laranjas"); and the maintenance of blockages even with proof of the lawful source of the assets, in amounts sufficient to compensate the victim and to pay the economic obligations fixed in case of conviction. These changes not only lead to asphyxiation of the financial flow that is expected to exist in contexts of systemic corruption or organized crime, but somehow increase the costs associated with the practice of illegal acts, since the blocking measures are enacted in the investigation phase or at the beginning of the criminal action, before the rigorous procedure of verifying the guilt of those involved. As an example, Odebrecht executives, including its president at the time, were the target of a judicial measure to block bank accounts in the amount of twenty million reais, which took place about a month before the first indictment against them.

Brazil criminalized acts of money laundering only when the resources were the product of a closed list of offenses, which changed in 2012 when the list was removed, extending the criminalization of concealment of gains derived from criminal enterprises. The following year was marked by the expansion of the criminal association crime, which instead of four started to require a minimum of three participants. Besides, a criminal organization started to be considered under the terms of the Palermo Convention, accompanied by the provision of a comprehensive list of investigative measures, including controlled action, infiltration of agents, and more comprehensive regulation of the cooperation agreements, which can be considered the apple of the eye of Operation Lava Jato.

The award-winning collaboration ("delação premiada") is one of the bargaining institutions introduced in the Brazilian legislation by Law 8,072 / 1990, which regulated the possibility of sentence reduction for a criminal association member. The benefits may be granted to those who confess and divulge their accomplices involved in heinous crimes, allowing the dismantling of the association. The same institution was later provided on several legal texts regulating specific criminal situations perpetrated by a group of people, namely: crimes against the financial system and tax order (Law 9080 / 1995); extorsion through kidnapping (Act 9269/1995); crimes committed by criminal organizing (Act 9034/1995); the money laundering (Act 9613/1998); the law that regulates protection of victims and cooperating witnesses (Law 9807 / 1999); drug trafficking (Law 11,343 / 2006); crimes against the economic order and cartel practice (Law 12,529 / 2011); and environmental crimes (Law 12,651 / 2012).

The actors in the criminal justice system had already made use of the cooperation agreement involving crimes of corruption since at least 2003, when the first plea agreement was signed in the country, within the scope of the Banestado Case,9 making use of the law to protect victims and cooperating witnesses. Controversies over the possibility of widespread use of the agreement were closed in 2013 with the law on criminal organizations (Law 12,850), introduced in the same year the so-called anti-corruption law (Law 12,846) was passed. This law provided for civil and administrative punishments to companies that practice acts related to corruption, in addition to introducing the leniency agreement on these issues.

The coexistence of cooperation agreements from the criminal area with the leniency agreements of the anti-corruption and the antitrust laws, which are essentially civil and administrative, started to generate some conflicts between the agencies of accountability. These conflicts exist because the leniency agreement can exclude the possibility of filing criminal charges against those who sign it. Still, the law is unclear on how to share the control of institutions that operate in different fields. Besides, it is expected that people and companies interested in concluding a cooperation agreement are also interested in ensuring that the facts described are not used to impose other sanctions beyond those already agreed upon, a possible setting considering the independent and concurrent action of the institutions of the accountability in the country.

<sup>9</sup> The Banestado case had as its object investigations and related criminal actions the alleged illegal transferences abroad of more than US \$ 120 billion, of which approximately 24 billion would have been transferred through the use of non-resident accounts (called CC5) held in several banks, but especially at the Banco do Estado do Paraná (Banestado). The case was processed in the Federal Court of Paraná and was also conducted by Judge Sergio Moro, in addition to the participation of some members of the Federal Prosecutors Service who later joined the Lava Jato task force.

The Odebrecht case illustrates this trajectory of inter-institutional conflicts derived from the multi-agency anti-corruption system. The company's executives met with the swift and rigorous action of the criminal justice system, in addition to the pressure exerted by Citibank in the USA, threatening to block the company bank accounts in the country, due to the application of the FCPA. Faced with little success in court contesting the charges and arrest warrants, in December 2016, the Odebrecht Group agreed to simultaneous leniency with Brazilian, North-American, and Swiss authorities, followed by almost eighty cooperation agreements signed by executives of the group's companies. The Brazilian leniency agreement only counted on the participation of the MPF, which led the Attorney General's Office (AGU) to successfully challenge the legitimacy of the MPF (case no. 5 023,972–66.2017.404.0000). This challenging took the company to a new negotiation process that culminated in signing a leniency agreement with CGU and AGU that provided for the payment of indemnity equivalent to R \$ 2.7 billion (CGU, 2018). The TCU entered the dispute, requesting the suspension of negotiations on the Brazilian agreement signed by Odebrecht, for alleged violation of Normative Instruction 74/2015, produced by TCU itself and which attempted to introduce the mandatory participation of the agency even in the negotiation phase (TC file 035 857 / 2015–3). TCU judged the request for suspension unfounded, but the general dispute entered the judicial and legislative arenas. This dispute generated the questioning of the legality of TCU's normative instruction before the Supreme Court (Direct Unconstitutionality Action n. 5294); obtaining a preliminary injunction from the STF to suspend a decision by the TCU that obliged the CGU to provide information on all ongoing leniency agreements (writ of mandate n. 34,031); and the edition of Provisional Measure no. 703/2015, which lost its effectiveness for lack of approvement by Congress and which provided that the agreements would only be sent to TCU after conclusion and signature.

The process of accommodating institutional disputes over leniency agreements took another step forward with the Technical Cooperation Agreement on the subject signed by the AGU, CGU, TCU, and the Ministry of Justice, on August 6, 2020, with the participation of the Supreme Federal Court. This agreement came at a time of great conflict in the political arena over the direction of Operation Lava Jato, with disputes over narratives about the legitimacy of the criminal justice system's role in fighting corruption. This dispute possibly explains the lack of participation by the MPF in the Cooperation Agreement, which led the institution's Anti-Corruption Chamber to publish a technical note. It questions several points of the agreement, highlighting the depletion of the MPF's role in these agreements (MPF, 2020b). Although the trajectory of interinstitutional conflicts associated with leniency agreements has not ended, it can be said that the competition among the accountability agencies has not prevented its use by some companies, as can be seen in Table 7, which shows the leniency agreements signed by CGU until August 2020, when there were another twenty-two under negotiation. The two reimbursement columns refer to the total amount provided for in the agreements and the amount paid up to August 2020 due to clauses on installment payments.

Leniency and cooperation agreements enable the discovery of illicit facts that would otherwise be difficult for the authorities to know. They also enable the introduction of more stringent compliance mechanisms in the companies' structure since integrity programs are taken into account in calculating penalties. The first leniency agreement signed by Odebrecht, for example, was responsible for


**Table 7** Leniency agreements signed by CGU (until August 2020)

Source: Prepared by the author, based on CGU website

the significant increase in the work of the Lava Jato task force in Paraná. This signing led to creating an Agreement Management Unit (Unidade Gestora de Acordo—UGA) to oversee the more than 250 facts narrated in more than a thousand pages of agreement related to illegal activities practiced in at least eleven countries besides Brazil (MPF, 2020c).

It is crucial to note that the importation of these negotiated justice instruments, typical of the North American common law, has generated intense debates in academia and the legal community, with criticisms that point to their inadequacy to the country's legal culture, generally associated with the already high incarceration rate of the poorest and most vulnerable population, as well as the lack of control over the exercise of the discretion of the Public Prosecution Service.

One should note that the progressive advancement of technological tools has greatly facilitated the investigation of illicit activities related to corruption and money laundering. The changes include the progressive digitalization of court proceedings, fully electronic since 2010 in the Federal Court responsible for the main core of Lava Jato, and the development of several software programs that allow online enforcement of court orders, such as the blocking of resources held in financial institutions (2001), obtaining tax data (2008), and blocking vehicles (2008).

Last but not least, in parallel with this process of institutional advances promoted by the political class itself, the Supreme Federal Court has contributed on its own initiative to shape the anti-corruption institutional design. One can observe this modeling activity when the Court formulates new paradigms deciding concrete cases involving high-level corruption, as occurred in the judgment of Mensalão. It also occurs when the STF carries out new interpretations of the Constitution in subjects that can decisively influence the outcome of significant investigations underway. In Mensalão trial, we can say that STF expanded the possibilities of criminal liability of high-ranking authorities by developing a theory of position domain (Leite, 2014), appropriating in a very criticized way the Theory of Final Domain of Fact developed by jurist Claus Roxin.

STF also acted strategically in two opposing scenarios related to Lava Jato. In the first moment of support for the operation, the Court exercised no veto activity over almost any aspect of the performance of the actors of the Lava Jato main unit. One can see this support by looking at the low number of preventive arrests decisions overturned by the STF and, especially, by the change in the constitutional interpretation of the presumption of innocence. Reviewing its position adopted in 2009, in 2016, the Supreme Court started to authorize the execution of the sentence before the judgment of all appeals, even without the decree of preventive detention (called prison before the res judicata). The change possibly contributed to the results of Lava Jato, as the prospect of anticipating the beginning of the execution of the sentence encourages the search for collaboration agreement, especially given the strategic time management in the processing of actions in the main core of Lava Jato (Rodrigues & Arantes, 2020).

Still in a scenario of strong public support for the operation, in May 2018, the STF remodeled the constitutional interpretation of the "privileged forum". The Court significantly restricted the cases of authorities with the prerogative to be prosecuted before the Collegiate Courts, transferring this jurisdiction to first instance judges. The measure has an immediate effect on reducing the political exposure of the Collegiate Courts, specially the STF, which exercised criminal jurisdiction over several federal authorities. It also has the potential to encourage the protagonism of the lower levels of the justice system through police operations against political agents. This institutional modeling is still ongoing in the Court, which maintained the privileged forum in some cases of crossed mandates in a trial in May 2021 (procedure No. 4846).

In a scenario in which Lava Jato became the target of more intense criticism, with its image tainted when ex-judge Sergio Moro assumed the position of Minister of Justice in Jair Bolsonaro's government, the STF restricted the competence of the Federal Justice, in March 2019, by excluding it in cases of corruption that are associated with an accusation of electoral crime. The Court also resumed the previous position of prohibiting the execution of the sentence before the final decision (*res judicata*), but it did so only in November 2019. Here there was a deliberate choice not to rule the judgment of this constitutional issue when the STF maintained the arrest of the ex-president Lula, in April 2018. In this instance, there was already a majority formed on the thesis that would have prevented his imprisonment for 580 days, derived from a conviction later overturned in two STF collegiate trials. These trials recognized the partiality of Judge Sergio Moro (March 2021) and the incompetence of the Federal Court in Paraná (April and May 2021).

The actors and institutions involved with Lava Jato used the incremental process of improving the mechanisms for controlling corruption and the institutional loopholes that make their activities less accountable. Although this institutional dimension was not sufficient for the operation results, it is difficult to deny that it will continue to impact the strategic choices of politicians and businesspeople operating in the country, as well as modulating the incentives and constraints of control bureaucracy.

#### **5 Final Considerations**

The data presented in this chapter provide a general map of the institutional design that was built and remodeled in Brazil in the years preceding Operation Lava Jato. This design shows the scenario in which businesspeople and politicians decide their strategies on establishing economic relations that interconnect private interests and public morality and resources. Brazil assumed international commitments to fight more effectively not only the practice of corruption but mainly the acts of occultation and concealment of assets from illicit sources. These commitments were internalized by the approval of more rigorous legal rules, strengthening of control institutions, intensification of international legal assistance and domestic inter-agencies cooperation, in addition to the improvement of mechanisms for monitoring the traces left by illegal financial activities.

These incremental advances can make the practice of some types of corruption more costly insofar as they provide signals about the political costs of being involved in the practice of some corrupt behaviors (Levi, 1999). They mainly alter the incentives and constraints imposed on the actors, which possibly occurs in parallel with the reshuffling of the behavior of politicians and businesspeople to enable the realization of their interests. It is also important to highlight that a broader definition of corruption does not only include the traditional exchange relationship (*quid pro quo*) between a public agent and a private agent. This definition also covers several less explicit forms of appropriation of public assets using direct or indirect relations with people in positions within the state structure. These may refer to institutional characteristics of the relations between the State and economic agents, which will not always be subject to the control mechanisms developed in recent decades. In addition, one cannot disregard that the engagement of accountability institutions and the patterns of interaction between them are also a reflection of self-interested movements, aimed at the search for institutional affirmation and preventing their activities from being subject to accountability (Arantes & Moreira, 2019).

The institutional framework summarized in this chapter and the specific issues related to Lava Jato show the complexity of political relations that drive the country's anti-corruption and anti-money laundering institutional design. On the one hand, the progressive strengthening of control institutions involves an apparent paradox since the political class promoted most of these advances. There were politicians affected by Lava Jato who played a decisive role in the legislative process described in this chapter. On the other hand, the interests that drive the members of the control bureaucracies are not very clear. They secure personal, institutional, economic and reputational gains from the strengthening of the accountability mechanisms of the political class. There is also little certainty to affirm that the centrality given to the anti-corruption policy by the criminal justice route adopted by Lava Jato has some effect of discouraging corruption in the future. However, it seems reasonable to believe that the incremental process that preceded Lava Jato will continue to produce the impact derived from the constraints and incentives it generates, in addition to keeping open the possibility of intense activity of accountability institutions in the country, with all the consequences that the deficit of control over some of these institutions can produce on democracy.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Rational Executives and Selfish Politicians: The Cultural Repertoires of Legal Experts**

Maria Eugenia Trombini and Elizangela Valarini

#### **1 Introduction**

Grand corruption affairs, and the arrest of politicians and high-ranking executives, has ignited the debate about the strategic action of law enforcers and the effects produced on democracy and political institutions. Other authors have already focused on the political uses of Law and the consequences of the fight against corruption in Brazil. But after all, what do legal professionals, working inside and outside the administration of justice, think about the people labeled white-collar criminals in the cases they sentence, accuse, and defend? The objective of this work is to advance the research agenda on the point of convergence between Law and Politics by shedding light on the action orientation of members of the justice system.

Rather than looking at the institutional design (Praça & Taylor, 2014; Madeira & Geliski, 2019; Kerche, 2018; Kerche and Marona 2022) or the interaction between agents within the accountability network (Castro & Ansari, 2017; Sadek, 2019; Oliveira et al., 2017; Engelmann & Menuzzi, 2020; Almeida, 2019; Sá & Silva, 2020), our gaze shifts to the actors and their cultural repertoires.

E. Valarini Alumni Max Weber Institute of Sociology, Heidelberg, Germany e-mail: eliz.valarini@gmail.com

M. E. Trombini Heidelberg Center for Ibero-American Studies, Heidelberg, Germany

83

M. E. Trombini (B)

Max Weber Institute of Sociology, Heidelberg, Germany e-mail: maria.trombini@uni-heidelberg.de

Using qualitative interviews with the group of experts on corruption and whitecollar crime, we will reconstruct the collective mindsets that underlie the actions of members of this professional complex. The objective problem of action, to which the interpretative patterns respond, is the relationship between the political and economic sectors. Collective mindsets are enunciated by judges, prosecutors, and lawyers with previous experience in corruption cases. By analyzing the worldviews held by this group our aim is to understand, focusing on the relationship between both sectors, where they locate the problem of corruption and who are its main participants in the Brazilian context.

Sources of empirical material are 24 in-depth interviews in which professionals experienced in major corruption scandals throughout Brazilian history (1985–2021) articulate ideas about the economy and politics. Our investigation of legal professionals' understanding of corruption and the links between the political and economic sectors uses the hermeneutic methodology of Collective Mindset Analysis (CMA). From the epistemological point of view, the stocks of knowledge acquired and built up by the various societies make it easier for the individuals within them to classify situations, problems, crises (Schütz, 1971), as for example the phenomenon of corruption. Mapping such knowledge stocks allows us to understand not only the meaning behind the actions of a given group, but also to identify in which institutions they are anchored.

After this introduction, the second section discusses the theoreticalmethodological choices, followed by the presentation of the results of the Collective Mindset Analysis (CMA). The third section introduces the Brazilian institutional setting, focusing on the anticorruption legal complex. Finally, the results are discussed in the fourth section and at the end research implications are outlined.

#### **2 Theoretical and Methodological Approach**

We know of the moral conception of legal practitioners in modern society (Karpik, 2007; Parsons, 1937, 1939), and that compared to other professionals such as doctors or engineers, lawyers have notable advantages in political mobilization given their ability to exercise moral authority in the name of technical competence (Halliday, 1985). An extensive body of studies labeled "political lawyering" (Karpik, 2007), has suggested that the affinities between jurists and individual liberties is a historical reality, but factors external to the professional *ethos*, such as a class logic, may anchor illiberal decisions by legal professionals (Couso, 2006). We enter the topic of the nexus between legal professionals, politics and the economy by following an agentic perspective on experts from the anticorruption complex.

The understanding legal actors have about the phenomenon of corruption, that is the worldviews they hold of the two poles involved, bribe-givers and bribe-takers, can only be indirectly accessed. Therefore, we will reconstruct the collective mindsets activated by the group under examination using qualitative interviews conducted for the purpose of applying this interpretive method (CMA) based on the social constructivism of Peter Berger und Thomas Luckmann (1967). The study's specific objective is to map on which set of available knowledge, beliefs, and value systems the actions of this subgroup are anchored. More generally, we aim to understand what stock of knowledge is activated in that particular society and context to interpret corrupt behavior. We may hypothesize that, as members of the legal system, the knowledge base and symbolic representations through which the group interprets laws are more uniform than the interpretations of others regarding the same symbols. Our investigation will look concretely at the narratives related to confronting corruption found among judges, prosecutors, and lawyers working in the Brazilian field. We will follow this question with the support of a method that enables to answer which cognitive and normative institutions are among the stocks of knowledge accessed by these agents to understand and locate the phenomenon of corruption.

Within and across institutional settings, including professions, discursive interaction forges meanings, values, commitments, and worldviews, providing individuals with a shared stock of knowledge (Schütz, 1971; Berger & Luckmann, 1967; Schütz & Luckmann, 2003), also understood as cultural repertoires. The existing literature on cultural repertoires typically employs the qualitative paradigm, such as in-depth interviews (Lamont, 1992), case studies (Ravasi & Schultz, 2006), content analysis of communications produced by a specific community (Weber et al., 2013). Drawing on more recent efforts to study repertoires, or collective mindsets, within a specific group (Valarini, 2018; Valarini et al., 2015; Valarini & Pohlmann, 2016; Elias, 2019; Klinkhammer, 2018; Höly, 2018) and a profession (Pohlmann, Höly & Trombini, 2022), and about the public–private (corrupt) relationship (Valarini, 2021), this study will apply the hermeneutic method of Collective Mindsets Analysis (original *Deutungsmusteranalyse*, denoted here by the abbreviation CMA) to interviews made with practitioners in the anti-corruption complex.

Ullrich Oevermann (2001a, b) originally developed CMA as a technique to study shared ways of understanding a phenomenon. The prerequisite for the emergence of patterns of interpretation, or mindsets, is an objective problem of action which is shared by individuals in a society or a sub-group in it. Faced with a specific circumstance, each individual uses socially available collective mindsets that offer appropriate, and thus reciprocally recognized, interpretations of the underlying situation. To allow ordinary or exceptional problems to be dealt with routinely, such patterns of interpretation attain the status of a stand-alone entity, existing independently of agency. In this paper, the action problem dealt with is corruption and where it is located according to the group studied. One of the advantages of the chosen qualitative method is that collective mindsets are relatively temporally stable contents of social knowledge accumulated in a society. This is because in order to deal with the different everyday spheres of interpretation and action of social phenomena and processes,1 societal groups and subgroups store, based on their biographical experiences, common ways of interpreting and evaluating a phenomenon or a situation. Such repositories of historical and social references achieve validity within that group and facilitate knowledge allocation, helping social actors simplify complex situations without having to interpret each situation as if it were unique (Meuser & Sackmann, 1992; Oevermann, 2001b; Sachweh, 2010). The manifestation of collective mindsets takes the form of derivations—explanations, justifications, and narratives about a fact, issue, phenomenon, etc.—when actors are questioned or asked to explain their actions, decisions, beliefs, etc. Thus, methodologically, it is not the collective mindset itself that is directly accessed, but the derivations, that is, the meaning, explanation and justification attributed to a specific problem (Ullrich, 1999). Our purpose with the qualitative analysis is to identify the "recipes" or guidance orientations containing unwritten rules that govern action within the societal group being studied (Bögelein & Vetter, 2019).

Our database for the qualitative analysis is comprised of 24 interviews conducted between 2017 and 2021 stratified by quotas of career, gender, and place of work (Appendix I). The choice of interviewees was based on the definition of experts by Meuser and Nagel (1989) so the specialists who are part of the field of action constitute the research object. Selection of interviewees was based on the

<sup>1</sup> According to Alfred Schütz an individual experiences the social world, *Lebenswelt*, at the subjective level, in his cogitations within the world of everyday life, and at the level of social collectivities, apprehending relations symbolically and generating meaning interpretatively. The origin of the concept is found in the phenomenology of Edmund Husserl and that was used by Jürgen Habermas in his Critical Theory and system concept, "...it is the world of natural attitude, of the relationship between fellow human beings, of everyday existence, of the experience of common sense thinking, natural, social and cultural world, world of exchanges, therefore intersubjective, in which it is necessary to fundamentally understand those who are closer (the associates) and be understood by them" (Fioravante Garcez, 2014, p.75).

following criteria: (1) legal professionals: lawyers, judges, and prosecutors; who (2) have acted and/or were acting in cases of corruption and financial crimes, at the time of the interviews, whose relationships between the economic sector (executives and companies) and the political sector (politicians and parties) are or were in evidence; and (3) that the cases were being handled in the federal jurisdiction.

Although the status of expert is relational, since it depends mainly on the interest and research problem, it is attributed to the actors who have privileged access to information about the object, about specific groups of people, or even about decision-making processes (Meuser, & Nagel, 1989). Thus, experts within an organizational context, as is the case of the interviewees from the anti-corruption complex, are actors who have tasks, responsibilities, as well as unique experiences and knowledge acquired during their trajectory. The actors of the justice system interviewed for this work are understood as experts in the policy area of corruption and money laundering.

The interviews were conducted in Portuguese, recorded in audio and then transcribed using NVivo software and manually corrected. To analyze the qualitative material, the Collective Mindset Analysis analytical method developed by Pohlmann et al. (2014) was employed.

The selected text passages were analyzed by the authors, first individually and then jointly, to reduce subjectivity in the analysis. For the operationalization, different thematic text passages related to corruption and its main participants were coded using the MAXQDA software. Both specific questions from the questionnaire that had the purpose to prompt responses in this regard as well as text passages related to the research problem triggered without the direct stimulus were considered. The deductive-inductive process allowed latent structures of meaning and knowledge accessed by the interviewees to be found. Beyond theoretical assumptions that the group understands corruption as a phenomenon where economic and political participants are equally to blame, these latent structures, when systematically analyzed, provide more trustworthiness to our qualitative research.

Considering legal professionals as experts, we are assuming that their professional action would be guided by group-specific collective mindsets. Because the theoretical approach established that they wield influence in the anti-corruption complex, we also expect them to be playing a key role in the formation and diffusion of mindsets about corruption and its main participants. Against this backdrop, findings of the CMA method are explained drawing on neo institutionalism to explain the mindsets of Brazilian experts in light of other supporting


**Table 1** The eight steps for the Collective Mindset Analysis

structures. The search for knowledge and cultural repertoires activated by the group is part of steps 7 and 8 of the method (see Table 1).

In this chapter, we will present the justifications and narratives of legal professionals about corruption and its main participants.

#### **3 The Anti-Corruption Legal Complex in Recent Decades**

In the last decade Brazil has seen a strengthening of the institutional capacities of existing bodies and the creation and implementation of new ones with respect to fighting corruption (Da Ros, 2019; Rico, 2014; Speck, 2000). In the terminology of the Federal Public Prosecutor's Office, what exists is an anticorruption normative microsystem that encompasses several legal diplomas, disciplining the legal consequences of corrupt practices, notably state sanctions produced in different spheres of accountability, and assigns to public agencies the tasks of control and sanction of illicit practices (MPF, 2017). In terms of efficiency, the coordinated action between prosecutors and federal police officers in the use of mechanisms to investigate and control suspicious activities indicates a maturation in these two members of the microsystem (Arantes, 2011).

The number of political corruption cases prosecuted at the federal level was a result of this accumulation of organizational maturity. According to the creator of the specialized courts for crimes against the national financial system and money laundering "With these courts came the Banestado case, then the Mensalão and now the Lava Jato. The greatest recent achievement of the Brazilian judiciary was the creation of the money laundering courts, which inspired several countries".2 The incentives to concentrate material and human efforts in a single administrative unit within the justice system go beyond structural aspects, and specialized personnel become an isolated advantage. The Lava Jato Task Force, created at the Public Ministry, Brazil's accusatory organ, was composed of some of the characters that made up the CC5 Task Force, responsible for the Banestado Case, one of the largest police operations on financial crimes in Brazil and a pioneer in cooperation with international control agencies. At the time, the Federal Justice of Paraná already had a specialized court, in which the former judge Sérgio Moro accumulated institutional learning3 . Because of the interaction between the various parties in this fit, formal rules and institutional increments not only constrained the actors, but gave them the possibility to significantly alter and reshape each organization and its setting incrementally (Praça & Taylor, 2014).

From a geopolitical point of view, like other initiatives in the context of the new global order, financial crime regulations were led by the United States (Slaughter, 2004; Franco & Wood, 2010; Tourinho, 2018). A multilateral act to sanction bribery of foreign officials was signed under the auspices of the Organization for Economic Cooperation and Development (OECD) in 1997 (Saglibene, 2014). From then on, international organizations such as the IMF and the World Bank established a normative vision of good governance, and played a key role in the dissemination of New Public Management ideas (Brinkerhoff; Brinkerhoff 2015).

<sup>2</sup> Gilson Dipp, president of the Federal Council of Justice in an interview to CNN available at https://www.cnnbrasil.com.br/politica/2021/04/25/lula-sera-julgado-em-foro-no-qual-jafoi-duas-vezes-absolvido

<sup>3</sup> The specialization began with Res. 20, of May 26, 2003, of the TRF of the 4th Region, and Res. 99, of July 11, 2013, of the TRF changed the name of the 2nd Federal Criminal Court of Curitiba, renaming it the 13th Federal Court of Curitiba.

In parallel, professions and globalization have become intertwined, with lawyers, as economists, serving as epistemic brokers of ideas flowing through the transnational and domestic fields (Dezalay, 2004; Dezalay & Garth, 2002; Flood, 1996; Galanter, 1983; Kauppi & Madsen, 2014; Liu, 2008; Trubek et al., 1994). The transnational circulation of tangible and intangible resources and the consequences it produces within the legal complex is an endeavor of a multitude of actors (state bureaucrats, intellectuals, NGOs, etc.) combining uniformity with local differentiation (B. d. S. Santos, 1992; Trombini and Valarini, 2023). Because our study is empirical, addressing intentionality depends on accessing a particular arena and examining it at the actor-level. An example is the analysis by Ribeiro (2017) that shows how actors engaged in anti-corruption initiatives influence policy-making and were key in convincing parliamentarians to approve the Brazilian law on criminal organizations (Law 12.850/13). Acting as an interest group, prosecutors, judges and other members in the anti-corruption complex successfully formed a coalition with actors from the public safety policy subsystem, thereby diminishing the importance of the special investigative tools that would then be used in corruption and white-collar crime investigations.

To gain a visual understanding of how the anti-corruption policy arena was constructed, the chronology of events is helpful. This institutionalization happened around a shared vocabulary (legislation), tools (organizational systems), and spaces (forums and networks). We will call anti-corruption complex the empirical space in which actors from state bureaucracies and other legal professionals specialized in the defense of financial crimes interact.

#### **4 Where Do Legal Professionals Situate the Problem of Corruption?**

The analysis of the knowledge and cultural repertoire activated by the anticorruption group showed that the political sector and especially the political actors would be the core of the issue. Corruption would be located in the political field and would be attributable to personality traits of individuals who comprise it. Selfish and immoral people who lack moral and formal education would be occupying positions in the state administration, so corruption is understood as the "capture" of public interests for personal gains, such as enrichment, power, continuity and advancement in the political career. A second group of interviewees conceives corruption as something instrumental that serves to achieve both political and corporate objectives. The problem would be located within the electoral system, in which interchangeable actors participate, but with the main objective of supplying the high expenses necessary for the functioning of the political-party system. A smaller part of the interviewees localize corruption at the macro level. Corruption would be anchored in the capitalist model, which place the political "class" at the mercy of economic interests.

Next, we will present the results of the empirical analysis according to the frequency in which the collective mindsets were found among experts in the anti-corruption complex.

#### **5 The Moral Problem of Political Actors**

In this collective mindset the problem of corruption is located at the individual level. The actors that make up the political system have individual characteristics that explain their deviant behavior. Thus, corruption consists of a moral and educational problem of personal character, which afflicts the Brazilian political class. To stay in power and in politics politicians need money; therefore, they distance themselves from the purposes and expectations that fall on the holder of public office. Citizens of integrity shy away from this *locus*, characterized by a lack of integrity, and do not want their image linked to an immoral environment. Politics has undergone a deterioration process in the last decades, accentuated by recent governments. This deterioration is implicitly related to the rise to power of a group of "non-elite", people who do not meet the expected criteria of having educational and symbolic capital. With the rise of such political actors with no republican principles, the "public" arena has been distorted. However, politicians are not disconnected from society. Because they are elected by the people, the explanation for the composition of an immoral group mirrors the "bad" choice of a large part of the population, which, like the politicians, is made up of individuals with deviant tendencies, who would identify with the models (candidates) willing to run for elective positions. The actors in the justice system, along with some specific social groups, form the small group of people capable of discerning between a "good" and a "bad" candidate.

"I think that this if I do the cashiering is indispensable, the politicians at least see it that way, indispensable to be able to pursue the career, which would be that City Councilor, Representative, Mayor, Governor. And at the same time, the career is important. Unfortunately, I think that in most cases it's not for the public good etc., but for personal growth, both for enrichment and status. I think that these are two things that, until today, I'll tell you, in the last few years this has been changing, two, three years I don't know, but until yesterday it was a very big status to be a deputy, to be a senator, now I've noticed that it's not. I see some politicians, friends of mine, who are leaving and unfortunately usually those who have the most, rightly, integrity and are ashamed to present themselves as. This was not seen here in Brazil" (Federal Judge in Paraná).

"It is possible to separate two main groups: one of them that in my point of view is the worst, which is the corrupt politician, that politician that for many times already enters public office with the intention of doing wrong, wants to get rich, wants to take advantage of public office so for me this group is the most nefarious and many, in my understanding, Although I am not an expert, they border on psychopathy, they are psychopaths who do not face the situation of misery that we live in, they do not face how half of the Brazilians do not have basic sanitation, do not have water treatment, this is absurd, a rich country like ours, we are not a poor country, we are rich. And they close their eyes, they don't see that, they don't become a mayor, a governor, a president, a congressman, a senator, to do the best for the people... no, they are there to perpetuate themselves in power and to embezzle public values, employ their friends and family members and continue in power, continue stealing and continue that pattern. These are also considered white-collar criminals, right? (Federal prosecutor in São Paulo)"

The normative structure of this mindset reinforces the role of the politician vis-àvis society and the abdication of personal and individual aspirations for the sake of a greater good, the public good. The theory of change is based on individual characteristics, like parts that make up a whole. To render politics "clean", upright and moral, it would be enough to elect new virtuous representatives, eliminating corruption from the Brazilian context. This horizon is not seen by the holders of this *mindset* as a utopia, but a transformation possible through changing the people in power.

#### **6 Corruption: A System that Feeds and Regulates Itself**

The second CM that appears most frequently among the interviewees also locates the problem of corruption in the political sector. Differently from the dominant mindset, this one qualifies the phenomenon as instrumental to both economic and political organizations. Corruption is understood as a system that feeds and regulates itself, obeying its own rules. The competitiveness between parties and candidates for the supply of financial support coming from the market (big companies that need state intervention) and the need to raise votes in increasingly expensive campaigns would be the elements of self-management of this system. The economic system would play a part in this logic, although its relevance has a smaller scale compared to the centrality awarded to the political system in the relationship between both. Such an explanation is not lodged at the micro level, since the dynamics do not depend on specific actors, be they from economic or political organizations, to exist. Although individual interests are not excluded, the focus is on the configuration of the system itself, conceived to accommodate the relationship established between the two poles: state and corporate, public and private.

"I think it's basically a business view, it's the business model in Brazil. The Brazilian State is very big, it is very difficult to have a business with really great potential that you don't have as a client the State, and the State is basically corrupt. So there is no way to do business ethically, you wouldn't compete, you wouldn't win bids, you would have countless problems. Sometimes it seems that I'm saying that businessmen are the victims of a harassment by the political class. It is not true. It is a system that feeds itself, I mean, they are not the same as always, the politicians change, but the system is this of relationship, so for them it is a business model and for the politicians it is a model of political financing and also of personal enrichment and this is the system, its actors only change, it is a business model" (Federal prosecutor stationed in Paraná).

"These companies cost money and there is a certain lack of control over this, the campaigns in Brazil are increasingly expensive and the candidate to become competitive often has to have large sums of resources and in this context some fall into this temptation to accept values of criminal origin, right, eh there are so many cases of corruption involving illicit enrichment, as cases in which values of bribery settlements were intended for electoral campaigns right, there are no justifications for this being presented by the convicted for this behavior, because normally, as I said they deny it, but yes I think there is, if there was a decrease in the costs of these campaigns that would also lead to a decrease in the need for resources, it may be that there would be an incentive less, but see there are some cases in which these values were used for illicit enrichment, were not intended for elections" (Federal magistrate stationed in Paraná).

The analogy of corruption as a business system reinforces the interactionist framework of this explanation. Fighting a phenomenon characterized by reciprocity similar to that of market relations is a complex task, which involves reducing the incentives and trying to increase the costs of deviant conducts for each of the sides involved.

#### **7 Political System "Hostage" to the Economic System**

The last mindset reconstructs the idea of an economic bourgeoisie which holds power over several social spheres, among them, the Legislative and the Executive. Although politicians exert influence in certain domains, a form of power, they are not, in their great majority, the holders of economic capital, the necessary substratum for the realization of political elections. Mere pieces in a larger gear where class logic is determinant, elected representatives would be servants of the interests of the businessmen who elected them.

"When you see in practice what the parties do, what the politicians do, this group of parties and politicians...there are even politicians that are not members of Congress. Roberto Jefferson is no longer a deputy, but he still has power. So, this whole group, what do they do? They are not bourgeois? not necessarily. There are one or two bourgeois. You get the list of the 500 richest people in Brazil, the politicians are not in it, you know? You have one or another" (Rio de Janeiro lawyer).

One of the rules of interpretation of this mindset is that economic power, because it is what matters most in a society like the Brazilian, or any other capitalist one, dictates the rules of the game and controls which players remain or not in the game. Being a politician offers concrete advantages compared to how much power the majority of the population has access to, but the appearance of power confuses the true reality of a country divided into social strata with an economic elite at the top.


Table: Overview presentation of the results

(continued)


(continued)

(continued)


(continued)

#### **8 Discussion**

In this chapter, we consider which stocks of knowledge guide the action orientation of legal professionals about corruption, taking the relation between the political and economic sectors as our focus. The most recurrent pattern of interpretation among the group "Corruption: moral problem of the political actors" uses psychopathological explanations to interpret and justify deviant behavior, attributing to the individuals belonging to one of the poles of the corrupt relationship the character inclination to commit crimes. Locating the problem of corruption, or of any other deviance, at the micro level is not surprising considering that we are dealing with individuals trained in law and accustomed to calculate guilt based on individual conduct. Imbued with classical theories, in which rule-deviation is considered something "abnormal", as in criminogenic approaches, such a way of understanding reality is compatible with daily-life experiences and cognitive frames available in the spaces of the justice system.

However, the valence found in relation to politicians is a surprising element of the mindsets. The studied group are experts on the theme, but their way of understanding and locating corruption does not seem to differ from the way corruption is commonly understood by other societal groups. In other words, this does not seem to be a group-specific mindset associated with a certain expertise, given that blaming politicians for corruption is a common sense understanding of this action problem. Scholars of political culture in Brazil have argued that the concept of patrimonialism, the number of corruption scandals, and the way they are portrayed by the media, breeds the idea that the state is the natural space for vices (Filgueiras, 2009). Public officials working in white-collar crime in Brazil are more distrustful of the government than of the private sector, as supported by our findings. Even though the dominant mindsets do not rule out the responsibility and deviant action of the executives, they do so in a lateral way. Only in the recessive mindset (Political system "hostage" of the economic system) the

greatest amount of power, and of blame, is attributed to members of the economic class. Numerically, this pattern of interpretation is not expressive within the group.

The mindset that appeared as the second most frequent defines corruption as a reciprocal system "that feeds and regulates itself". This pattern of interpretation brings a more elaborate understanding of the phenomenon, which distances itself from personified explanations. Because it depends on the interaction of actors from different spheres, financial criminality, also regarded as a way of making business, requires a greater effort to be investigated and fought against. Considering that the holders of this collective mindset are mainly the members of the bureaucracy specialized in financial crime, we may associate this explanation with a professional vanity for the merits achieved in anticorruption investigations. Recognizing the "ecological" dimension of the phenomenon and contributing conceptually to the fight against corruption by emphasizing the procedural instruments that make criminal prosecution more efficient reinforces the Lava Jato group's attempt to differentiate itself from the rest of the state administration (see Trombini, 2023 for a detailed analysis). While the systemic mindset ("Corruption as a system that feeds and regulates itself") evokes self-conceit, it points out that there is no solution for the treatment of corruption, because it is a phenomenon that exists regardless of the actors and their agency. The existing points of contact between different sectors are not simply undone when certain actors (political and economic) leave the scene, like the moral mindset posits.

We know that the professions wish to aggrandize themselves, constituting a "jurisdiction" through professional knowledge systems (Abbott, 2005) or socially organized perceptual frameworks (Goodwin, 1996). Here, the stocks of knowledge about corruption and its main participants are contrasted with those of other holders of public office. Members to the anticorruption complex are neither unskilled lawyers too close to local power to hold them accountable for misconduct nor immoral politicians with a penchant for disregarding public interests. Experts ascribe knowledge and moral to themselves, and negatively evaluate politicians on the basis of the absence of these features. They claim to be more tuned with the republican principle of a clear separation between the public and the private, something that the two most dominant mindsets consider to be lacking in how corporate-state relations unfold at the Brazilian context.

The results of the CMA show that experts are preoccupied with providing answers to the question of how to best perform in the public interest. Efficiency and ethics are the key concepts guiding the cultural repertoire of legal professionals when inquired about corruption. Searching for normative and regulative institutions at the judicial field where the expert group studied is inserted helps explain the findings. The context in which these professionals are embedded is permeated by a certain vision of good governance and infused with ideas from new public management. To face the challenge of drawing a separation between the public and the private, the old-developmentalist and statist ideas (linked to corruption) should be replaced by a conservative liberal credo (linked to a competitive market). Rather than finding a way between the two extremes, as Bresser-Pereira (2001) recommended, they seem to insist on following the same steps given by Western European countries and the United States.

In common, lawyers, prosecutors, and judges suggest that the problem of corruption can be addressed by changing the "quality" of the political professional with the entry of people of "good character", bearers of morality. By doing so, they draw a distinction (Bourdieu, 1984): on the one hand there is a qualified civil service, recruited on the basis of merit, on the other, "unskilled" elected authorities that are not exposed to competition—either because the electorate cannot vote, or because party dynamics make competition dysfunctional, and corruption self-feeding. It is interesting to point out that these interpretative patterns, anchored in the belief of the advantages of meritocracy, are not shared by other elite groups in Brazil, especially professionals from which technocratic thinking would be expected, such as top executives at STEM (science, technology, engineering and math) careers.

The knowledge stocks of the Brazilian economic elite are more guided by instrumental concerns. State regulation of economic activities is understood as a necessary element for the survival of companies operating in Brazil, and problems of misconduct are seen as something that must be solved within the organizational environment (Valarini, 2018). Unlike the experts in the anti-corruption complex, the interpretation patterns of senior executives are more anchored in elements that symbolize paternalistic relationships, present in the political-economic environment, brought in parts by the neodevelopmentalist model, than elements acquired by the profession. As such, their understanding of efficiency and ethics is closely connected to doing business in a Latin American country.

The timeline of events depicted above showed that changes in legislation and institutional design created an anticorruption legal complex. A double movement of judicialization of conflicts and politicization of judicial institutions (Arantes, 2007) has awarded a central role to the actors employed in these bureaucracies. Their understanding of corruption portrays politicians as selfish, and executives as rational, the State as the place of vices, and the market of merit. If the professional independence of judges and prosecutors allows them to represent citizens' interests, the technocratic basis of their decisions (Pádua, 2008) and the low social control they experience (Bresser Pereira, 2007) are negative factors. The

same elements that grant control bodies necessary autonomy to investigate and judge high executives and politicians offer them a chance to reshape the electoral, social, and economic landscape without having to be elected.

Their advice for how to best perform in the public interest involves professionalism, impersonal selection through entrance competitions, and increased autonomy from political influence. The most dominant mindset vocalizes a resentment against the political elites. A potential explanation is that the interviewees do not come from the same families that traditionally occupied the judiciary before public competitions and the creation of the Public Ministry, in a historical period when political, business and judicial elites all belonged to a single uniform group. According to Grün (2011), the newly arrived individuals in the elites are usually the main operators of scandals, the "champions of justice" and traditional moral principles that would be ignored by the dominant elites. Yet our work (Trombini et al., 2022) with prosopographical data from the anticorruption group has showed that the majority comes from an elite background. Still, the moral mindset agrees that politicians are who to blame.

This brings us to an ambiguous aspect of the mindsets. Not all politicians are considered to be immoral, and the emergence of a systemic pattern of corruption is regarded as a phenomenon from recent history. Legal professionals suggest that greater pluralism (an intrinsic requirement for democratic consolidation) has been responsible for increasing the distances between the "virtuous" senior civil servant and this other less respectable public officials at a "contaminated" political space. There were the politics of before and the politics of now, with skilled people staying in the market instead of running for office. This stance, of elitist arrogance, is opposed to populist resentment, but the two often go together. Writing about such unexpected companions, Galston (2014) cautioned that "public-spirited individuals who genuinely want to promote the public interest can end up longing for a form of leadership that is not regularly accountable to the people".

Blaming out-groups for corruption did not come as a surprise, neither did blaming the political group that was in power in the last decades, while scandals like the Mensalão and Lava Jato took place. Recent episodes that placed members of the high echelons of the political and economic system in prison show that members of the anticorruption complex behave like *elite challengers* (Johansson & Uhlin, 2020). Noteworthy is that rather than allying themselves to other ruling elites, like in the past (Love & Barickman, 1986; Adorno, 1988; Werneck Vianna, 1999; Carvalho, 2003; Almeida, 2010), the elite from the anticorruption complex is more selective when deciding who to challenge. They still exhibit liberal ideologies and conservatism, which might explain why they attribute blame to a certain political group. Together with resentment over the loss of privileges, conservatism is another motivation for antipetism, being against the PT, or Worker's Party (Aquino, 2019). Other investigations into the political uses of law by experts, in public administration particularly, still need to shed light on the interplay between individual and professional characteristics that may affect variance in mindsets. Focusing on class, ideology and career might be a promising route for addressing convergence and divergence in cultural repertoires.

#### **9 Conclusions**

We started from a question about the meeting point between Law and Politics to look at the empirical space of the professionals working in the anti-corruption complex. Our results show that the bearers of the reconstructed patterns locate the problem of corruption among politicians, by looking at the micro (immoral individuals) or meso (political system) levels. Their mindsets criticize patrimonialism and a lack of separation between the public and private domains and recommend efficiency and ethics, two features of the anticorruption complex, to remedy that. Not all politicians are blamed and labeled corrupt, and resentment and arrogance help explaining who is regarded as guilty and unskilled. The fact that the subpopulation of lawyers, prosecutors and judges working on corruption scandals does not recognize itself as agents highlights the importance of using qualitative methods to get a grip at the self-presentations of the group. In representative democracies with electoral systems, the autonomy of legal professionals employed at state careers that enable them to do policy-making must be checked, as other contributions in this volume argue.

One can safely admit that not only in Brazil the dynamics of corruption scandals transformed the legal complex. With "*affaires*" *of* corruption, French, Spanish and Italian judges became protagonists in the public arena. Sometimes against their will, these members of the Judiciary engaged in anti-corruption become "symbols of a tumultuous revolution in the more or less Jacobin conceptions of the Republic and the evolution of a democracy of opinion orchestrated by the media" (Pujas, 2000). Studying the rise of Western liberalism from the 18th to the twentieth century, Karpik and Halliday (2011) had predicted a contest between a "political" strand that emphasizes local democratic practice, and an "economic" strand that emphasizes market transactions, noting that the latter leads. We can, for now, say that the anti-corruption complex in Brazil is also an example of this trend of rising liberal economic thinking. Mapping which cultural repertoires are activated by the group when explaining other action problems, like what logics are at work inside economic organizations, and comparing these results with the findings here is a first step to advance this research agenda. The content of the group's thinking remains to be investigated in more detail and the undemocratic consequences of their behaviors further problematized to avoid harm in the future.


#### **Appendix I**

(continued)

(continued)


#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Moro's Opinions: A Quantitative Analysis of Sentencing in Operation Car Wash**

Luciano Da Ros, Luísa Zanini da Fontoura, Sérgio Simoni Junior and Matthew M. Taylor

#### **1 Introduction**

The literature on legal accountability is playing catch-up to the literature on decision-making in judicial review cases. Relatively little is known about the factors that influence judges as they consider imposing civil and criminal penalties, and even less in countries outside North America and Europe. The gap is especially pronounced with corruption: despite growing numbers of cases around the globe, the literature on judicial decision-making in corruption prosecutions remains sparse. Why do judges convict some defendants in corruption cases and not others? Why do judges apply harsher sentences to some defendants than others? In sum, what explains variation in the severity with which the judicial system judges public figures and private citizens accused of corruption?

These questions are clearly relevant both for public debate and academic theory, but have only recently been raised in the specialized literature. One cause

L. Z. da Fontoura · S. S. Junior Federal University of Rio Grande Do Sul, Porto Alegre, Brazil e-mail: luisazf.zf@gmail.com

S. S. Junior e-mail: sergiojr\_ssj@yahoo.com.br

M. M. Taylor American University, Washington, USA e-mail: matthewmacleodtaylor@gmail.com

© The Author(s) 2024 M. E. Trombini et al. (eds.), *The Fight against Systemic Corruption*, Organization, Management and Crime – Organisation, Management und Kriminalität, https://doi.org/10.1007/978-3-658-43579-0\_5

111

L. Da Ros (B)

Federal University of Santa Catarina, Florianópolis, Brazil e-mail: daros.luc@gmail.com

of this neglect may be data availability: there have not always been sufficient numbers of cases to evaluate judicial behavior regarding legal accountability. Associated with this is a problem of comparison: seldom are cases comparable to one another along a range of dimensions, holding constant legal system, jurisdiction, statutes, case details, or contextual background. One consequence is that most academic research of legal accountability in corruption cases tends toward qualitative analyses of aggregate patterns, using longitudinal case studies or small-N comparative research.

Brazil's Operation Car Wash (*Lava Jato*, in Portuguese), perhaps the "the largest corruption scandal ever to beset at democratic nation" (Fisman & Golden, 2017, 13), offers a unique opportunity to address these questions, partly because of the large number of cases that were heard in a single courtroom, by a single judge.

Operation Car Wash was a criminal investigation into grand corruption and money laundering that began in March 2014 in the Brazilian city of Curitiba. It began as an inquiry into irregularities at the state-controlled oil firm Petrobras, Brazil's largest company. Investigations progressed rapidly, unveiling evidence of corruption totaling billions of dollars, covering a wide variety of government agencies, and implicating politicians, bureaucrats, businesspeople and financial operators. Over the course of nearly seven years, the operation indicted, convicted and imprisoned hundreds of individuals (Rodrigues, 2020; Lagunes & Svejnar, 2020; Da Ros & Taylor, 2022). As Operation Car Wash both fueled and was fueled by severe political and economic crises that brewed simultaneously in Brazil over the 2010s, many studies cast an unflattering light on the investigation's consequences for Brazilian democracy (Sá e Silva, 2020; Kerche & Marona, 2022; Santos & Gallego, 2022).

At the heart of the operation was the 13th Federal Trial Court of Curitiba, specialized in money laundering crimes. It was presided over by Judge Sérgio Moro between March 2014 and November 2018, when Moro resigned from the judiciary to become Minister of Justice. From Curitiba, the operation spread to several courts around the country and even beyond Brazil. But it was in Curitiba that the most consequential decisions were rendered, including the convictions of leading politicians such as the former president of the Chamber of Deputies Eduardo Cunha, former presidential chief of staff José Dirceu, former Finance Minister Antônio Palocci and former President Luiz Inácio Lula da Silva. Prominent business people, long accustomed to impunity, were also convicted, including the scion of the Odebrecht construction empire, and many of his ostensible competitors.

The prominence of the defendants, the political repercussions of the case, Judge Moro's own career choices after he left the judiciary, and the intense public awareness of Operation Car Wash have meant that the trial court's decisions are frequently interpreted through a subjective lens, defined by observers' political leanings or their perspectives on how corruption should be addressed. Operation Car Wash, Moro, and Moro's decisions have increasingly become an object of criticism over the last few years. Many of Moro's decisions would later be overturned by the Brazilian Supreme Court—notably the conviction of former president Luiz Inácio Lula da Silva—due to different allegations of abuse and bias while Moro presided over the investigation (Da Ros & Taylor, 2022; Prado & Machado, 2022a, b).

Despite these important caveats, we believe that analyzing Moro's decisions on their own—without reference to his later career decisions or the subsequent decisions of higher courts—is an important exercise for understanding the factors that determine judicial decision-making in corruption trials. Seldom has a trial court judge in any jurisdiction presided over such a large number of interconnected corruption cases. The extensive dataset of cases compiled from Moro's decisions, moreover, enables this paper to take an arms' length perspective, analyzing Moro's sentences through multivariate quantitative analysis of the effect of two sets of variables on judicial decisions regarding conviction and sentence length: variables related to defendant profiles and to prosecutorial strategies. In so doing, we hope to contribute both to the broader global literature on the judicial role in anticorruption, as well as to the debate within Brazil concerning Judge Moro's decisions and the workings of Operation Car Wash, building upon the predominantly qualitative literature on these topics.

To that end, the next section offers a conceptual framework and lays out the hypotheses that guide the analysis. The third section details the data and method. The fourth presents the results. The paper concludes with a summary of the findings and suggestions for future research.

#### **2 Legal Accountability: Definitions and Hypotheses**

Questions regarding variation in the incidence and intensity of judicial responses to corruption are closely related to the debate on accountability. Accountability refers to the obligation of public officials to explain their actions and to answer for them. Accountability has been classified in a variety of types, which seldom explicitly reference the judiciary (Schedler, 1999; Manin et al., 1999; Taylor, 2019). For the purposes of this paper, we refer to the specific judicial role in accountability as legal accountability, defined as the enforcement of civil or criminal sanctions by judicial agents (Bovens, 2007; Lindberg, 2013; Da Ros, 2019).<sup>1</sup>

The academic literature on judicial behavior has focused predominantly on supreme courts and constitutional review, seeking to understand which factors political, social, institutional etc.—influence the decisions made by judges and justices in the exercise of their duties (Da Ros & Taylor, 2019; Hilbink & Ingram, 2019). The vast comparative literature on judicial behavior has used judicial review cases to theorize a variety of explanatory models, such as the attitudinal, strategic and institutionalist.2

Theorization of how judges decide corruption cases, however, remains incipient. There has been almost no adaptation of theories from other areas of judicial behavior to corruption cases, nor has there been much theorizing derived from corruption cases themselves. In part, this is because of a methodological bias that hinders both the formulation of new theories and the testing of existing theory: most approaches to legal accountability in corruption rely on aggregate data (e.g., the total number of criminal convictions per country or period). The unit of analysis is often the entire collectivity: cases in a given country or state, for example. As a consequence, conclusions are drawn by aggregation: patterns of corruption and accountability in a given jurisdiction. This approach does not allow for analyses in which each judicial decision constitutes the unit of analysis (King, 1997).

The most typical aggregate approach is the longitudinal case study, seeking to explain why the judiciary in a particular country has come to convict a larger or smaller number of defendants for corruption over time. Generally motivated by the occurrence of major investigations that profoundly impacted the political system, the effort of these studies is to understand why prosecutors and judges began to behave more assertively. Examples of this approach include a variety of exemplary case studies, such as analyses of the Mani Pulite investigations in Italy in the 1990s and its consequences over the following decades (della Porta, 2001; della Porta & Vannucci, 2007; Vannucci, 2009; Dallara, 2019; Manzi, 2021), the

<sup>1</sup> It is worth stressing that the literature has found no positive association between rates of corruption convictions and the incidence of corruption. Indeed, "a higher level of convictions per capita is often less an indicator of high corruption than of a more determined fight against it" (Karklins, 2005, 134). The number of criminal convictions in corruption cases thus cannot be considered a reliable indicator of corruption levels proper. Instead, it captures the performance of judicial institutions, including the workings of investigative, prosecutorial, and trial bodies.

<sup>2</sup> Epstein and Lindquist (2017) review existing theories.

corruption trials in France during the 1990s (Roussel, 1998; Garraud, 2001; Adut, 2004), recent prominent cases in Latin America (Conagham, 2012; Michener & Pereira, 2016) and, of course, studies of Operation Car Wash itself (Lagunes & Svejnar, 2020; Rodrigues, 2020; Sá e Silva, 2020; Kerche & Marona, 2022; Da Ros & Taylor, 2022).

A second approach also relies on aggregate data, but does so comparatively. Such studies seek to understand why some polities have produced more legal accountability (i.e., investigations, prosecutions, convictions) than others. Such studies typically draw from a relatively small number of cases, dedicating themselves to comparing different countries, states or governments in Western Europe (Sousa, 2002; Maravall, 2003; Sims, 2011; Guarnieri et al., 2020), Latin America (Da Ros, 2014; Ang, 2017; González-Ocantos & Hidalgo, 2019; Pimenta & Greene, 2020) and Eastern Europe (Popova & Post, 2018).

One of the consequences of the predominance of aggregate approaches is that the use of qualitative methods prevails. Multivariate quantitative studies based on individual data have been less common, despite recent efforts in this direction (Ang, 2017; Gehrke, 2019; Bento et al., 2020; Mancuso et al., 2021). This is because aggregate approaches often reduce the number of units of analysis by treating the available data in aggregate form. That is, because they are concerned with explaining the overall pattern of charges or convictions, these studies often do not inquire into which factors influence individual decisions by prosecutors and judges to charge or convict defendants, or which factors impact the penalties imposed on convicted defendants.

This paper disaggregates the data, using each of the individual judicial decisions against specific defendants as units of analysis. The use of individual decisions permits us to employ multivariate techniques that complement existing qualitative perspectives. Our hypotheses are derived from aggregate research on the subject and from studies on judicial behavior in judicial review cases (an area of research in which individual approaches predominate).3 The hypotheses fit into two pairs, regarding: 1) defendant characteristics; and 2) prosecutorial strategies.

The first pair seeks to understand judicial decisions to convict, and the penalties imposed, based on defendant characteristics, such as profession and, in the case of politicians, their party affiliation. With regard to profession, our hypothesis is that judges may act differently depending on whether they are judging politicians, businessmen, bureaucrats, money launderers or unwitting accessories

<sup>3</sup> This includes the studies most commonly associated to the attitudinal and strategic approaches to judicial behavior (Segal & Spaeth, 2002, Epstein & Knight, 1998).

(*laranjas*). Politicians are expected to be less likely to be convicted, in line with the strategic approach to judicial behavior, because courts are likely to be reticent to move against those who can wield power against them (Maltzman et al., 1999). Indeed, most judicial anticorruption efforts seemed targeted at petty corruption in Brazil until recently (Madeira & Geliski, 2019; Levcovitz, 2020). We assume that more powerful defendants, or those with greater capacity to damage the judiciary or individual judges, are likely to be treated more leniently than less powerful defendants (H1a). The second hypothesis is that judges may decide cases differently depending on defendants' political affiliation (H1b). We test whether there is differential treatment of politicians by party. This hypothesis conceives of judicial behavior as an extension of party competition (Gordon, 2009; Ríos-Figueroa, 2012; Ang, 2017; Popova & Post, 2018), and is closely aligned with the attitudinal model, in which judges' political attitudes are considered major motivators of judicial behavior (Segal & Spaeth, 2002).4 This produces two hypotheses:


The second pair of hypotheses refers to prosecutorial strategies. Even though the cases all involved corruption, defendants faced diverse charges. Brazilian law dictates different sentences for different crimes, ranging between 2 and 12 years for active and passive corruption (roughly, bribe-paying and bribe-receipt), between 3 and 10 years for money laundering, and between 3 and 8 years for criminal organization, for instance. Defendants may also face more severe charges, for example, for higher volumes of money laundering. The first hypothesis of this second pair (H2a) is that some types of crime are likely to lead to more severe treatment, and the higher the number of crimes for which defendants have been charged, the more likely this is to lead to conviction and heavier sentences. The second hypothesis (H2b) is that the type and quantity of evidence will increase the severity of sentencing. Evidence may come from a variety of sources, including prosecutorial witnesses, phone intercepts and electronic data (including text messages, messaging apps, email, and cloud data), audit reports, and information provided by the financial intelligence unit COAF and the Federal Revenue Service, among others. This information allows us to evaluate how the information prosecutors provide influences judges' behavior (Machado & Paschoal, 2016;

<sup>4</sup> More broadly, corruption accusations are often deployed as weapons in political competition (Balán, 2011).

Arantes & Moreira, 2019). We also take into account plea bargain agreements, which likely affect plea bargainers' chances of conviction and the severity of sentencing and other defendants tried within the same case. The two hypotheses are expressed as follows:


#### **3 Data and Descriptive Analysis**

The dataset used to analyze these hypotheses was compiled from the website maintained by the Federal Judiciary of Paraná (JFPR), which encompasses the 13th Federal Court of Curitiba and permits full consultation of all criminal cases tried by Judge Sérgio Moro between March 2014 and November 2018. Similar data has been used previously by other scholars, such as Fontoura (2019) and Rodrigues (2020); we compiled the dataset used here directly from the JFPR website, checking it against the data compiled by those two authors.5 We have downloaded the sentences and decisions on interlocutory appeals (*embargos de declaração*, which are appeals within the same court, akin to requests for amendment of judgement). Having analyzed the text of the decisions, we compiled the database for the variables of interest.

This study does not cover the totality of Sérgio Moro' s behavior within Operation Car Wash. We only analyze his sentences quantitatively. This means that the paper ignores both the judge's off-the-bench behavior, such as interviews, press releases and opinions on legislative proposals (Hilbink & Ingram, 2019). We also leave aside Moro's control over the court's agenda, the timing with which cases were dispatched, and several consequential decisions taken outside the strict limits of the cases, such as the decision to publicize recorded audios of conversations between former President Lula and then President Dilma Rousseff. Similarly, our research does not cover all sentences offered under the umbrella of Operation Car Wash. Other judges also handed down decisions, including Moro's successors at

<sup>5</sup> We are grateful to Fabiana Rodrigues for sharing data on the number of prosecution witnesses.

the 13th Federal Court of Curitiba after 2018, as well as judges participating in related cases in Rio de Janeiro, São Paulo and Brasília, including higher courts.

That said, Moro's role was significant, setting both the tone and the tenor for Car Wash, and representing the bulk of the decisions made in the investigation from 2014 to 2018. During almost the entire period he ran the 13th Federal Court of Curitiba, Moro's court had unique jurisdiction over cases in Operation Car Wash, judging a total of 46 criminal cases filed against 193 different defendants on the merits. Because some defendants appeared in more than one case, Moro actually made 289 decisions. These decisions constitute our units of analysis. We exclude those cases in which the sentence was suspended by the judge because the defendant signed a plea bargain agreement and had already served the sentence established in the agreement (e.g., Alberto Youssef in case n. 5028608- 95.2015.4.04.7000), as well as those cases in which the defendant died before trial (e.g., Humberto Mesquita in case n. 5025676-71.2014.4.04.7000). Finally, we excluded cases of defendants whose indictments were not accepted by the court (*denúncia não recebida*) so that they could be tried in other related cases.6

Because we are interested in what factors lead to conviction and the length of the penalties imposed, the dependent variable was measured in two ways. The first was through a dummy variable for conviction. A zero score might occur either because of acquittal or because the defendant "ceased to be convicted" on account of a lis pendens situation (of parallel proceedings; e.g., Waldomiro de Oliveira in case n. 5083258-29.2014.4.04.7000).7 The second measure of the dependent variable is a continuous measure of the length of sentence for convicted defendants, ranging from 18 to 250 months.

For defendants who signed plea bargain agreements, we use the sentence length stipulated in judgment before the application of the agreements' benefits. We take this approach because most agreements unified the penalties of the defendants not only in cases that had already been tried, but also in cases that were still ongoing. Thus, because the database is comprised only of cases that were actually judged, it would be reckless to stipulate exactly what would be the

<sup>6</sup> In twelve cases the indictments were not accepted. Of these, three were not accepted by the court because the evidence was weak. In the remaining nine cases, the indictment was not accepted because the defendant was facing other charges in other cases, because the case was broken into smaller cases, or because the defendant was a fugitive from justice, awaiting extradition abroad. In four additional cases that were also excluded from our dataset, the indictment was accepted, but the cases were set aside because the defendants were facing extradition.

<sup>7</sup> There are only eight cases of a defendant who "ceased to be convicted," rather than either being convicted or absolved.

size of the reduction obtained, since many cases are still pending trial. Although this may produce a certain artificiality, especially as regards the impact of the plea-bargaining variable on the length of the defendants' sentences, it is the most conservative approach for the quantitative treatment of these cases.8

The first set of independent variables relate to defendant characteristics: their profession, and in the case of politicians, their respective parties at the time. Professions were divided into six groups: politicians, civil servants, business owners, business executives, financial operators and others. The first encompasses both elected officials who did not hold office at the time of the sentence,9 as well as high-level party functionaries (e.g., treasurers); the second includes all public sector officials, including executives at state-owned companies like Petrobras; the third includes owners or shareholders of private sector companies; the fourth concerns salaried executives in the private sector; financial operators includes money launderers colloquially known as *doleiros* and other agents specialized in asset laundering; and finally, the last category is residual and includes unwitting accessories (*laranjas*), family members and advisors. The political parties of those classified as "politicians" in the previous variable were divided into four groups: PT, PP, PMDB, and others.

The second set of independent variables relates to prosecutorial decisions: the type and number of crimes charged; the amount of money laundered; the types of evidence utilized; the timing of indictments; and the existence of a plea bargain prior to sentencing. Crimes were divided into five types: active corruption (bribe-paying), passive corruption (bribe receipt), criminal organization, money laundering, and others. The latter is a residual category that captures crimes that were charged less frequently (e.g., tax evasion, obstruction of justice, use of false documents, international drug trafficking). The number of crimes for which the defendants were accused is a sum of these dummies (including the category "others"). This generated a continuous variable that ranged between 1 and 6. Types of evidence were categorized as follows: phone intercepts, electronic data, search and seizure, and forensics by the Federal Police (PF) and the Public

<sup>8</sup> We also tested a different measure of the dependent variable, incorporating the agreements by dividing the penalty established in the agreement by the number of cases tried after the defendants signed the agreement. This did not significantly affect the results. However, it is not included here because it is somewhat misleading: in some cases, the new "average sentence post-agreement" was higher than the previous sentence. This is because, as noted above, many plea bargainers' cases were still pending, so the actual potential sentence was higher than that reflected in the cases in our dataset.

<sup>9</sup> Under Brazilian law, sitting politicians cannot be tried in the trial court, but instead are tried under the original jurisdiction of higher courts.

Ministry (MPF), as well as evidence produced by other agencies, such as the Central Bank (BACEN), Federal Revenue Secretariat (SRF), the Financial Activities Control Council (COAF), the comptroller general's office (CGU) and the federal accountability agency (TCU), as well as information provided by Petrobras itself and evidence derived from international cooperation. The sum of these dummies accounted for the total number of types of evidence presented in court, ranging from 2 to 10. To account for the accumulation of evidence within the overall operation, the variable "order of indictments" orders the cases according to the date in which indictments were presented. "Number of prosecution witnesses" is a continuous measure ranging from 2 to 29, extracted from Rodrigues (2020). "Plea bargainer" indicates whether the defendant signed a plea bargain before sentencing. "Number of other plea bargainers" sums the number of other defendants within each case who had signed plea bargain agreements, seeking to account for the cumulative impact of plea bargainers on another defendant's chances of conviction and the severity of sentencing.

Table 1 details the dummy variables and Table 2 details the continuous variables. A few descriptive points are of note. The trial data suggest that Moro was a tough judge In Operation Car Wash, especially by comparison to judges in other corruption cases tried in Brazil. On average, 73.7% of Moro' s decisions were convictions, with average sentences of 113.5 months, or nearly nine and a half years in prison. This conviction rate is much higher than found in other studies of judicial treatment of corruption in Brazil: in the case of sitting politicians judged by the Supreme Court (STF), the rate is below 5% (Falcão et al., 2017; Gomes Neto & Carvalho, 2021); in the case of public servants expelled from the federal administration (predominantly for petty corruption), the conviction rate is around 20% (Alencar & Gico Jr., 2011); in the cases of high-ranking public officials tried within the federal court system, the conviction rate is around 10% (Levcovitz, 2020); and in the case of municipal mayors tried by a specialized panel within state courts, the rate reaches 18% (Bento et al., 2020). Evidence from the judicial decision-making in previous national corruption scandals also stresses the severity of Moro's rulings: in very few of the most prominent scandals of the 1990s analyzed by Taylor and Buranelli (2007) were the defendants ever convicted.

Regarding H1a, although Operation Car Wash is often referred to as an investigation into political corruption, "Politician" was the least frequent profession among the defendants judged by Moro, comprising just over 7% of the total, far behind financial operators, who made up 24% of the cases. The fact that there were few political defendants, of course, should not obscure the fact that they were all convicted in all the cases in which they were tried, reaching a



(continued)


#### **Table 1** (continued)

**Table 2** Descriptive statistics, continuous variables


\* Three cases had zero values and were coded as 1 in the multivariate models, so as not to lose these cases

\*\* The maximum value is not equal to the number of judicial cases (46) because a few indictments were presented in court on the same date

perfect 100% conviction rate. This did not occur with any other variable. One consequence of this perfect conviction record for politicians is that there were no differences in conviction rates between defendants from distinct political parties.

Regarding H1b, there was a clear preponderance of members of the PT among the politicians who were tried. While this has been interpreted as a sign of bias on the part of federal prosecutors, who filed charges against PT politicians more often than against politicians of other parties, it is also the case that the PT defendants heard in the trial court were former occupants of federal elective offices and therefore no longer enjoyed the original jurisdiction of the Supreme Court (STF). This was the case for former President Lula and former ministers Dirceu and Palocci, for example. Many politicians from other parties who were implicated in wrongdoing still held public office during the investigations and their cases were therefore not heard by the trial court, but instead by the STF, which moved slowly on cases of politicians implicated in Car Wash. For example, Eduardo Cunha, former president of the Chamber of Deputies and a member of the PMDB, was only tried by Moro after he was suspended by Congress and he was therefore no longer eligible for the original jurisdiction of the STF. Yet, because the number of political defendants is quite small, it is important to be cautious about drawing conclusions within this group of defendants.

Regarding H2a, although Car Wash is usually associated with corruption, the crime most frequently tried by Moro was actually money laundering, present in 85% of his court's decisions, followed by active or passive corruption (60%), criminal organization (35%) and other crimes (17%). Because defendants were often charged with more than one crime per indictment (the average is 2 crimes per indictment), money laundering made up 43% of the charges, followed by active and passive corruption (which together comprised 30% of the charges), criminal organization (18%) and others (9%). Scholars have been broadly aware that much of the expertise invested in fighting "corruption" in Brazil since the 1990s has in fact been derived from increasing capacity to enforce money laundering laws, including specialized police and judicial bodies (Da Ros & Taylor, 2022). Lastly, the residual category "Other crimes" has the highest conviction rate, 93.9%, which may be due to the specificity of several of these crimes, for which only one defendant has sometimes been accused in the entire database.

Finally, regarding H2b, evidence provided by the Federal Police and federal prosecutors is the most frequent type of evidence, and evidence from searches and seizures by these two bodies is cited in more than 90% of cases. There is also intense participation by the SRF revenue agency and the Central Bank in cases involving tax and bank data, respectively, accounting for more than 80% of the cases. There is comparatively little evidence provided by the COAF financial intelligence agency, the TCU accounting body and the CGU comptroller general's office. There is copious evidence provided by Petrobras, which aided the prosecution in several cases, as well as from international bodies, with these two sources providing evidence in more than half of all cases. As a result, various types of evidence have been deployed jointly, averaging more than six types of evidence per case. These descriptive results corroborate something that has frequently been reported in qualitative studies of Car Wash: extensive inter-institutional and international cooperation, unprecedented in previous investigations (Rodrigues, 2020; Da Ros & Taylor, 2022). Finally, plea bargainers constituted a significant group of both indicted and convicted defendants. Indeed, plea bargainers' trials make up almost a third of the decisions, and other plea bargaining defendants within each case average almost 3 per case, reaching a maximum of 10 in one case. This may be a consequence of the widespread and unprecedented use of plea bargains in Car Wash, made possible by the Law of Criminal Organizations enacted shortly before the beginning of the investigation (Rodrigues, 2020; Prado & Machado, 2022a). It may also result, however, from a relatively small set of plea bargaining defendants who were accused in several cases, such as the *doleiro* Alberto Youssef, who was tried in 16 different cases, and former Petrobras executive Paulo Roberto Costa, tried in 10 cases. Still, it is noteworthy that even in cases of plea bargaining defendants, the conviction rate does not reach 100%. That is, even though all plea bargainers were convicted at least once, not all cases against plea bargaining defendants resulted in conviction.<sup>10</sup>

#### **4 Multivariate Analysis**

Because the dependent variable is measured in two distinct yet interrelated ways, the multivariate analysis relies on a two-part model (2PM). The first measure of the dependent variable is a dummy variable of the decision to convict; the second is a continuous variable of the length of the sentences (in months) imposed on convicted defendants. The second decision is dependent on the first: only convicted defendants receive a sentence. Yet, factors affecting the likelihood of conviction need not be the same ones that impact sentence length. Therefore, our identification strategy requires a selection model that estimates two equations

<sup>10</sup> For instance, while plea bargainer Paulo Roberto Costa had already been convicted in other cases, he was acquitted for lack of evidence in case n. 5012331-04.2015.4.04.7000, decided in September 2015.

sequentially, given the presence of censored data in the second measure of the dependent variable.

We adopted the 2PM due to its flexibility in comparison to other sample selection models and its fit with the characteristics of our data (Vance & Ritter, 2014). First, it allows all the independent variables used in the first stage equation to also be used in the second stage. That is, it does not require the specification of an "exclusion restriction" in the first stage that need to be omitted in the second stage. Second, the 2PM permits the results in the second stage to be interpreted as actual outcomes, not just as potential outcomes. That follows from the fact that the 2PM treats the censored observations at the second stage as zeros, not as missing cases, given that they comprise non-convictions.11

The first stage of the 2PM uses multiple logistic regression to analyze the decisions to convict. Coefficients are reported as odds-ratios. The second stage consists of an ordinary least squares regression calculating the marginal effects of the independent variables on sentence length. In each of the two stages, we ran three different models, resulting in six models in all: models 1A, 1B and 1C are the first stage of the 2PM, reporting the results of the multiple logistic regression; and models 2A, 2B and 2C are the second stage of the 2PM, reporting the marginal effects.

The option to adopt three different models in each stage of the 2PM stems from the challenge that the plea bargaining defendants pose to the models. Plea bargainers are essentially confessed defendants, so we use the sentences these defendants faced before the application of the benefits of their plea agreement, as explained previously. Thus, the first model of each stage (the "A" models) excludes *cases* of plea bargainers (reducing N to 196 in Model 1, and to 135 in Model 2); the second model (the "B" models) includes all cases (N = 289 in Model 1, and N = 213 in Model 2), but excludes only the *variable* "plea bargainer"; and the third model (the "C" models) is complete, including both plea bargainers' *cases* and the "plea bargainer" *variable*. In all models, we clustered decisions by repeat defendants and used heteroscedastic-robust standard errors.

In the case of the dummy variables for "Profession", "Political party" and "Crimes", the omitted categories are the variables "Other professions", "Other parties" and "Other crimes," respectively. As observed above, there is a perfect conviction rate for politicians. Consequently, the independent variable "politician" is a perfect predictor of the dependent variable "conviction," and thus is automatically excluded from the first stage of the 2PM. In this case, the sum of

<sup>11</sup> These characteristics of the 2PM make it different, for instance, from the sample selection model proposed by Heckman ( 1979).

politicians with "other professions" is adopted as the omitted reference category. Also because of that, the variable "Political party" was excluded from the first stage of the 2PM, as all politicians (from all parties) were convicted. We used the natural logarithm of the amount of money laundered to normalize the distribution of the variable. The variable "Sum of types of evidence" was excluded from the multivariate models because it was highly collinear with the other evidence variables (since it is their sum). Additionally, our multicollinearity analysis found a high correlation of "plea bargainer" with profession: only one politician was a plea bargainer, and few in the category "other professions" were plea bargainers. Said another way: the vast majority of plea bargainers were financial operators, executives and business owners.

The results in Table 3 suggest that some variables strongly determine the chances of conviction. In addition to the perfect "politician" predictor identified in the descriptive analysis, this includes the profession of "financial operator," which increases the chances of conviction by between 3.4 and 12.6 times, relative to the omitted category. Each additional crime with which the defendant is charged increases the chances of conviction by about 14 times. Given that Moro's court specialized in money laundering, it is interesting that money laundering led to less likelihood of conviction, other things equal. Regarding the types of evidence, two variables are positively associated with increased likelihood of conviction: electronic data and CGU information. The first increases the chances of conviction between 6.9 and 11.2 times, and the second between 45.1 and 175.3 times. Although the magnitude of the coefficients is very high in the latter case, it is worth remembering it is drawing on a very small number of cases (14), so these conclusions should be read carefully. Surprisingly, the number of other defendants who were plea bargainers decreases the chances of conviction, suggesting that prosecutions may become weaker as more plea bargainers are added. Finally, as expected, plea bargainers increase the chances of their own conviction by 2.2 times, as inferred from the only model in which the variable appears, 1C.

The variables mentioned in the paragraph above were statistically significant at the 0.10 level in all models in which they were included. A few other variables were statistically significant in some of the models, but not all. The variable "Active corruption" was significant only in models 1B and 1C, with a negative impact on the defendants' likelihood of conviction. The variables COAF, TCU and foreign cooperation, in turn, were significant only in one of the models (1A in the first two cases, and 1B in the last case). Whereas COAF is negatively associated to the likelihood of conviction, TCU and foreign cooperation are positively associated to it.


**Table 3** Multiple logistic regression, first stage of 2PM

(continued)


#### **Table 3** (continued)

Exponentiated coefficients; 95% confidence intervals in brackets

\* *p* < 0.1, \*\* *p* < 0.05, \*\*\* *p* < 0.01

Table 4, reporting the marginal effects derived from the estimation results of both the first-stage logit and the second-stage OLS regression, suggests that some variables have a substantive effect on sentence length, conditional on the incidence of a conviction. Politicians were treated the most harshly, with sentences between 46 and 50 months longer than other professions. They were followed by financial operators, whose sentences were longer than defendants from other professions by 25 to 33 months. The number of crimes charged leads to increased sentences, with each charge increasing the expected sentence by 47 to 50 months. Regarding the types of evidence, electronic data increased expected sentence length by 41 to 45 months. Contrary to initial expectations, evidence from the Revenue Service in cases of lifted tax secrecy led to a reduction of 40 to 44 months. Also contrary to initial expectations, each additional plea bargaining defendant within each case is associated with a reduction of about 6 months. Finally, as expected, plea bargainers saw average prison sentences rise by 23 months in Model 2C. This increased sanction for plea bargainers must be read with caution, though, because it reflects the sentence before any benefits were conceded by the court. In other words, this longer sentence time is artificial, since at the end of the day, plea bargaining defendants actually saw their sentences reduced rather than extended.

The variables mentioned in the previous paragraph had statistically significant results in all models in which they were included. There were a few other variables that exhibited statistical significance in only two of the three models tested: accused business owners (significant in models 2A and 2B, as compared to other omitted professions); evidence such as the number of prosecution witnesses (significant in models 2B and 2C), information by Petrobras (significant in models 2B and 2C) and by foreign governments (significant in models 2B and 2C); and defendants from the Workers' Party (PT, as compared to the omitted category that includes PTB and SDD). Except for the last one, all these variables were positively associated with the length of the penalties applied, with the increase in prison terms ranging from 1.6 to 25 months. In the case of defendants from the Workers' Party (PT), they had shorter sentences than the omitted category, with average sentences that were about two years shorter. However, we would remind the reader that because there are few politicians in our sample, these findings should be interpreted with care.

Table 5 summarizes the results of our expanded analysis, where we varied the omitted reference categories. Accordingly, we ran 36 additional models in which we adopted as omitted reference categories all possible dummy variables for profession, political party, and crime. That is, in addition to the analysis above where the omitted categories were variables "Other professions", "Other


**Table 4** Marginal effects, second stage of 2PM (only convicted defendants)

(continued)


#### **Table 4** (continued)

(continued)


#### **Table 4** (continued)

Heteroscedastic-robust standard errors in parentheses

\* *p* < 0.1, \*\* *p* < 0.05, \*\*\* *p* < 0.01

parties" and "Other crimes", respectively, we also ran models in which the omitted categories were variables "Politician",12 "Civil servant", "Business owner", "Business executive" and "Financial operator" for professions; "PT", "PP" and "PMDB" for parties; and "Active corruption", "Passive corruption", "Criminal organization" and "Money laundering" for crimes. The purpose of this expanded analysis is to check for the robustness of our findings. Due to space constraints, we do not display all these results here. Instead, we include in Table 5 only those variables that were statistically significant in a majority of the models.

Once we factor in the results from the expanded analysis, the independent variables that most frequently were significant across all of the analysis (both descriptive and multivariate) were politicians, the number of charges, electronic data evidence, whether the defendants were plea bargainers, and the number of other defendants who were plea bargainers. Except for the last one, all variables were positively associated both to the chances of corruption and sentence length; the last variable was negatively associated to both outcomes. The analysis provides particular support for H1a, H2a and H2b.

With regard to H1a, other things equal, politicians exhibit both higher chances of conviction (all were convicted) and longer sentences compared to all other types of defendants. Financial operators, in turn, exhibit only higher chances of conviction, but not lengthier sentences then other non-politician defendants. Lastly, relative to all possible omitted categories, the professions civil servant, business owner, business executive, and the residual "other variables" do not seem to affect either the chances of conviction or sentence length.

<sup>12</sup> In the case of "Politicians", because it is a perfect predictor of convictions, it was included only in the second, sentencing, stage of the models.


**Table 5** Variables associated with severity of sentencing (shows variables with statistically significant coefficients in the majority of the models)

Regarding H1b, because all politicians from all parties were convicted, there are no differences in regards to the chances of conviction. The differences displayed in Table 4 regarding the length of sentences for PT politicians, however, dissipated once we ran the additional models. The negative impact of the variable "PT" was no longer statistically significant once we ran the model adopting as omitted reference categories "PP" and "PMDB'—the "PT" was only significant when the baseline was the residual "other parties" category. This reinforces our conclusion that once politicians were charged in the trial court, there was no significant difference in the severity with which the political defendants were sentenced based on their partisan affiliation.

As for H2a, ceteris paribus, the number of crimes defendants were accused of is consistently associated with an increase in both the chances of conviction and sentence length. Charges of money laundering and active corruption are associated with decreased likelihood of conviction compared to all other categories, but do not affect sentence length. Charges of passive corruption are associated to lengthier sentences, but do not affect the chances of conviction relative to other categories. Lastly, charges of criminal organization and the amount of laundered money did not affect either the chances of conviction or sentence length.

Lastly, regarding H2b, three types of evidence were statistically significant both for the chances of conviction and the length of sentences. Electronic data is consistently associated with an increase in both the likelihood of conviction and sentence length. Plea bargainer is also significant, although it should be read with care, as it was only used in two models (1C and 2C) and is somewhat artificial, as noted earlier, because it is computed before the application of the benefits arising from the plea agreement. The number of other plea bargaining defendants is associated with a decrease in both the chances of conviction and sentence length, suggesting that cases with higher number of plea bargaining defendants may have been interpreted as weaker by Moro. Evidence collected by CGU seems to affect the likelihood of conviction, but not the length of the sentences, although the number of cases is small. Inversely, the number of prosecution witnesses, information from Petrobras and from foreign governments seem to increase sentence length, but not the chance of conviction. Information from the Revenue Service, in turn, seems to reduce the length of the sentences, but not the likelihood of conviction.

#### **5 Discussion and Conclusion**

Several conclusions can be drawn from this analysis. The first refers to the severity of Sérgio Moro' s sentences in Operation Car Wash, which led to very high rates of conviction, especially compared to other courts in the Brazilian judiciary. As highlighted previously, the Curitiba trial court stands out both historically and by comparison to its peers. The most significant demonstration of the trial court's severity relates to its treatment of political defendants: regardless of their party affiliation, politicians in Moro's court faced a 100% conviction rate, with sentences that were 4 years longer on average than those of other defendants, other things equal. This suggests that the Moro's court was less concerned than other courts in Brazil, and especially than the STF, with the costs involved in convicting members of the political class. With hindsight, this finding is tragically ironic: the severity with which Moro and Operation Car Wash treated politicians eventually led to blowback and backlash, and politicians worked assiduously both to close down the investigation and to stonewall Moro's anticorruption initiatives as justice minister (Da Ros & Taylor, 2022).

Secondly, although we tend to think of judicial behavior in corruption cases as a relatively tightly-defined category, the decisions taken by the trial court cover a number of different types of crimes and evidence, with the participation of a significant number of government agencies providing evidence. Corruption charges proper (i.e., active and passive corruption) appear to have been secondary to money laundering, which was the principal crime investigated and prosecuted in Car Wash. The types of evidence brought to bear, in turn, were very diverse, including evidence from the joint Federal Police-prosecutorial task force, but also evidence collected by a significant number of other oversight institutions (COAF, Revenue Service, Central Bank, TCU, CGU), the state-owned Petrobras, and even international partners. This corroborates qualitative conclusions about an increase in coordination within the Brazilian accountability system over the last decade (Taylor & Buranelli, 2007; Power & Taylor, 2011; Carson & Prado, 2016).

Third, this paper suggests that sentencing decisions in corruption cases can be productively analyzed as two relatively autonomous decisions: the decision as to whether or not to convict the accused, on the one hand, and the decision about the length of the sentence imposed on convicted defendants, on the other. Our analysis suggests that the variables that affect conviction do not necessarily affect sentence length in the same manner. In fact, there are a significant number of variables that seem to impact only one or the other decision.

Fourth, despite the importance of taking into account the different factors affecting either the likelihood of conviction or the sentence length, it is important not to miss the more general finding that there are variables associated with both decisions. This is clear in regards to politicians, but also evident in relation to plea bargainers, the number of other defendants who were plea bargainers, and evidence based on electronic data, all of which affect both the chances of conviction and the length of sentences. These last findings are tragically ironic: while electronic data seems to have been fundamental to the severity of the sentences issued by Judge Moro, at least a parcel of the subsequent backlash against Car Wash was made possible because of electronic data. Starting in June 2019, *The Intercept* published a series of news that used hacked electronic data to show ex parte communications between Moro and the prosecutors, casting doubt on the overall fairness of Operation Car Wash.

Finally, it is essential to highlight the centrality of prosecutors, who are gatekeepers in deciding whether or not to move forward with criminal proceedings. Because legal accountability is a sequential process that has distinct phases such as investigation, prosecution and judgment, future research might look at the factors that affect decision-making in other phases of the process. Judicial decisions are of course highly relevant, but without the decision to prosecute, that is moot. Judicial behavior in cases of corruption, after all, is dependent on the behavior of the members of the prosecution, who decide whom to accuse, and which crimes are charged (Gordon, 2009). While our analysis does not address prosecutorial decisions about whether to prosecute and how to frame that prosecution, this paper has demonstrated that these choices are extremely consequential to judges' decision-making, deserving greater attention. Variables that have been seen to be very influential here—such as the number of crimes charged, plea bargainer, and the number of other defendants who are plea bargainers, for instance—are all largely dependent on the behavior of prosecutors as they assemble their cases.

#### **References**


Da Ros, L., & Taylor, M. M. (2022). *Brazilian Politics on Trial: Corruption and Reform Under Democracy.* Lynne Rienner Publishers.

Epstein, L., & Knight, J. (1998). *The Choices Justices Make*. Congressional Quarterly Press.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter' s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter' s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

**Companies and the Interplay with Key Sectors**

## **A Pragmatic Giant: The Logic Behind Odebrecht's Campaign Donations in the 2014 Elections**

Wagner Pralon Mancuso, Rodrigo Rossi Horochovski and Bruno Wilhelm Speck

#### **1 Introduction**

In 2015, Brazil's Federal Supreme Court (STF) banned campaign contributions from corporations. In the previous year, Brazil held its last national election which allowed corporate contributions to political campaigns. In aggregate, the 2014 election moved circa 4.2 billion reais—approximately 1.8 billion US\$ according to the average exchange rate in 2014 (1 US\$ <sup>=</sup> R\$ 2.3534).1 Of this total, Brazilian corporations donated just over 3 billion reais (about 1.3 billion US\$), which represents approximately 73% of the overall amount (Mancuso et al., 2018). The Odebrecht group was among the top donors in that election.

Founded in 1944 as a construction company in the state of Bahia, the largest state in Northeast Brazil, 70 years later the Odebrecht Organization had become, according to Exame magazine's "Biggest and Best" ranking, the fourth largest Brazilian business group in the private sector,2 operating in several areas, such as agribusiness (ethanol, sugar, and electricity), construction, defense, engineering, gas, pipelines, naval industry, investments, logistics, oil, petrochemicals, environmental services (water, sewage, and waste), and transportation (roads, railways, subways, and airports). Among all Brazilian construction companies,

1 http://www.ipeadata.gov.br/ExibeSerie.aspx?serid=31924

2 https://exame.com/revista-exame/o-risco-odebrecht/

W. P. Mancuso (B) · B. W. Speck University of Sao Paulo, Sao Paulo, Brazil e-mail: pralon@usp.br

R. R. Horochovski Federal University of Paraná, Curitiba, Brazil

© The Author(s) 2024 M. E. Trombini et al. (eds.), *The Fight against Systemic Corruption*, Organization, Management and Crime – Organisation, Management und Kriminalität, https://doi.org/10.1007/978-3-658-43579-0\_6

143

the Construtora Norberto Odebrecht ranked first in 2014, according to the traditional "Engineering Ranking—500 Largest Construction Companies", published annually by the magazine *O Empreiteiro*. 3 Marcelo Odebrecht, CEO of the conglomerate, then ranked eighth in Forbes magazine' s list of Brazilian billionaires,4 with a personal fortune valued at 14 billion reais (5.9 billion US\$).

That year, campaign contributions from the colossal Odebrecht group amounted to approximately 88.9 million reais (circa 37.8 million US\$).5 The strategy adopted was to divide donations among different companies belonging to the group, although not equally. Seven Odebrecht Group companies made electoral donations in 2014: Braskem, Construtora Norberto Odebrecht, Enseada Indústria Naval, Odebrecht Agroindustrial, Odebrecht Ambiental, Odebrecht Óleo e Gás, and Odebrecht Serviços e Participações. In total, just over 17,000 Brazilian private companies made campaign donations in 2014. The Odebrecht Group companies occupied prominent positions on the list of donors as all seven ranked among the quartile of largest donors, while Construtora Norberto Odebrecht ranked seventh in the general position; Braskem at 13th; Odebrecht Oil and Gas at 52nd; Odebrecht Agroindustrial at 229th, Odebrecht Serviços e Participações and Enseada Indústria Naval tied at 362nd, and, finally, Odebrecht Ambiental ranked at 3,558th.

In this chapter we discuss the Odebrecht Organization as a source of official resources for the 2014 election campaigns in Brazil. We compiled the information about campaign donations from the Superior Electoral Court (TSE). The intent is to analyze the strategy of the business group in view of the national and international political science literature. To analyze the proposed case, we conducted a quantitative research, using descriptive statistics, social network analysis (SNA), and logistic and linear regression analysis.

<sup>3</sup> Revista O Empreiteiro, Year 52, Nº 533, August 2014.

<sup>4</sup> https://forbes.com.br/listas/2014/10/30-maiores-bilionarios-brasil-em-2014-2/#foto23

<sup>5</sup> We adopted the following procedure to obtain the total sum of political donations made by the Odebrecht group in the 2014 elections. First, we accessed the website https://www. novonor.com.br/sites/default/files/ra\_odebrecht\_2015.pdf to download the 2015 Odebrecht Organization Annual Report (for the year 2014). Second, we identified in the report the companies that comprise the group. Third, we sought these companies' registration numbers in the National Register of Legal Entities (CNPJ). Fourth, we used these numbers to identify on the website of the Superior Electoral Court (TSE) all donations made by the Odebrecht group in the 2014 elections. This information can be found at http://inter01.tse.jus.br/spceweb.con sulta.receitasdespesas2014/. We only considered direct donations made by the Odebrecht group to candidacies, committees, and parties. Therefore, in an attempt to avoid duplicate counting of donations, we did not count indirect donations, i.e., those in which the group's companies appear as original donors of transfers made by parties or candidates.

For this purpose, this chapter has been divided into two sections, on top of this introduction and the final considerations. In the next section, we present a brief overview of the political science literature addressing the decision of private corporations to invest in political campaigns. We then analyze, in the following section, the official donations by the Odebrecht group in the 2014 elections, both to candidates and political parties. We must not forget that, according to the Brazilian law in force at the time, business actors such as the Odebrecht Organization were allowed to make electoral donations both to individual candidacies for different public offices (whether in plurality elections, such as president, state governor, or senator; or in proportional elections, such as federal or state deputy), and to party organizations, such as national and state directorates or common financial committees (the latter set up exclusively during the electoral period to organize fundraising activities and distribute donations among individual candidacies for different offices and/or among party directorates at the national or state level).6 The limit for corporate electoral donations was not set in absolute values, but relative to 2% of the company's gross revenue in the year prior to the election, which enabled large donations by major companies, such as the Odebrecht group conglomerate. Lastly, the concluding section summarizes our findings and proposes some ideas for future investigations.

#### **2 The Political Science Literature on Corporate Electoral Financing**

In capitalist and democratic countries such as Brazil, major business groups may engage in politics at different levels, whether at the macro level, in which major national issues are discussed; the meso level, which addresses sectoral issues; as well as the micro level, which debates specific issues pertaining to the specific interests of each company (Lowi, 1964).

The major business community has a myriad of tools to engage with politics at these different levels (Schneider, 2010). Some of these tools are collective and institutionalized, such as participation in business associations within formal spaces for debating public policies, often created and maintained by way of government initiatives to gather feedback from society in the policy-making process, among which consultations, hearings, or even a diverse range of collegial bodies such as councils, committees, commissions, chambers, workgroups, forums, etc.

<sup>6</sup> The legislation at the time allowed candidates to form committees for their own campaigns. Donations to these individual committees were accounted for as donations to candidacies.

(Schmitt, 2020). Other tools for political action are individual, often activated by the initiative of the business community itself, such as lobbying, mobilizing personal influence networks, and financial contributions to electoral campaigns, which form the specific object of our study.

The subject of corporate electoral donation raises important normative and methodological issues, already thoroughly explored in other works and which we will not explore in detail here. Generally speaking, the main normative issue concerns the discrepancy between the democratic value of political equality and the asymmetry of influence engendered by political investments from major corporations, especially when there is no clear rule in place to impose transparency and restrictions to corporate donations, potentially leading to negative consequences such as corruption, distortion of political competition in favor of better-financed candidates, control of the government agenda by the wealthiest, invisibilization of the poorest, and erosion of public trust in democratic institutions (Przeworski, 2011; Norris & van Es, 2016).

Methodologically, the main issues concern mutual determination or reciprocal causation. These issues refer to the two goals which, according to Mueller (2003), political donors may pursue when opting to support certain candidacies: to affect the electoral outcome or the behavior of victorious candidates. Hence, the first question is whether the correlation between funding and performance at the polls exists because donations increase the candidates' odds of winning or because the candidates' prospect of winning attracts more electoral contributions. In other words: if we adequately consider all factors affecting electoral performance and campaign financing, would election donations still carry any significant effect on the outcome of the polls? To address this issue, political scientists have continually strived to improve their research designs and explanatory models so as not to neglect relevant variables. Thus far, the literature has indicated a series of interfering conditions in the electoral effect of campaign financing, such as the candidate's condition of being an incumbent or challenger (according to Jacobson (1978), the effect is greater for challengers), man or woman (for Speck and Mancuso (2014), the effect is greater for women), or the nature of the electoral competition, whether more or less fierce or whether or not there are incumbents in the race (according to Przeworski (2011), the effect seems to be greater in fierce competitions as well as those without incumbents).

Another methodological issue concerns the potential correlation between electoral financing and the behavior of the winners and, if such correlation does exist, whether business donations influence the behavior of elected representatives or whether the projected behavior of elected representatives determines business donations. Once again, to adequately address this challenge we must rely on research design and models that do not neglect relevant variables. A prime example of the divergence surrounding this issue is the controversy between Ansolabehere et al. (2003), for whom campaign donations have minimal effect on how legislators vote, and Stratmann (2005) who argues in the opposite direction, i.e., that electoral financing does indeed affect the vote of parliamentarians.

Without disregarding the implications of the normative and methodological issues described above, in this chapter we shed light on another challenge, namely, deciphering the logic behind corporate electoral financing, or, in other words, understanding a corporation's decision to engage or not in electoral donations and, when doing so, how much to contribute and how to distribute donations among candidates and parties running for different offices.

The literature on the Brazilian case has cast attention to some possible determinants in the donor behavior of corporations. We will explore the possible impact of these factors on Odebrecht's electoral contributions in this chapter as we seek to characterize the profile of this business group as a source of political campaign revenue. We will focus on five elements, two of which linked to candidates (political capital and region) and three linked to parties responsible for launching the candidacies (ideology, belonging to the government's ruling coalition in Congress, and size of the political party).

Starting with the individual factors, some have argued that a candidate's political capital is an important element in a company's decision to finance political campaigns (Araújo et al., 2015). The reasoning is that companies would rather direct resources to candidates with more political capital, who have verified political competitiveness and a more familiar political activity profile. Prior electoral success is an important indicator of political capital as well as a relevant—albeit not foolproof—predictor of future electoral success (Speck & Mancuso, 2013; Avelino et al., 2018; Araújo Júnior & Pires, 2020). As a result, companies would rather invest resources in incumbents than to bankroll challengers (Lemos et al., 2010; Marcelino, 2010; Mancuso & Speck, 2015). Brazilian legislation allows indefinite re-elections for legislative offices, but limits re-elections for executive offices (president, governor, and mayor) to a single subsequent period.7 Thus, when a previously re-elected president, governor, or mayor can no longer run for re-election, the expectation is that the company will invest resources in candidates who "inherit" the political capital of the incumbent, i.e., candidates who belong to the same party as the incumbent or supported by the same political group. In theory, the strategy of targeting candidates with more political capital would allow a corporation such as Odebrecht to make safer political investments,

<sup>7</sup> Constitution of the Federative Republic of Brazil, article 14, paragraph 5.

minimizing the risk of wasting resources on defeated candidacies and maximizing the proportion of money allocated to successful candidacies, thus paving the way for important channels of political influence.

The second individual factor addressed in this work is the candidates' region of origin. The hypothesis is that the companies' geographical origin matters, as they tend to foster stronger political ties with candidates in their region, thus making them favored beneficiaries of campaign contributions. The Odebrecht group is originally from Northeast Brazil, and thus the company would have stronger ties with candidates from this region, favoring them when it comes to making donations.

Regarding political parties, the first factor is the ideology of the political group, whose impact on electoral financing has been discussed by the available literature (Samuels, 2001; Mancuso et al., 2016). Candidates and collective organizations belonging to right-wing parties would be more likely to receive donations from Odebrecht, since parties that adhere to this ideological line are traditionally more favorable to business interests.<sup>8</sup>

The literature on political financing has also addressed another variable regarding political parties, namely if the party belongs to the government's ruling coalition in Congress (Samuels, 2001; Lemos et al., 2010; Mancuso, 2012). The expected effect is ambiguous. On the one hand, the company may favor ruling candidates and political parties out of genuine affinity with the government's agenda, or simply to please a political coalition which, in the company's view, is likely to remain in power. On the other hand, the company may consider candidates and parties from outside the ruling coalition, whether due to the company's own rejection of the government's line of action or due to suspicions that government supporters will fare poorly in the next election. Thus, we also included this variable in our study to verify whether Odebrecht favored one particular side of

<sup>8</sup> To operationalize this variable, we measured party ideology from surveys conducted with Brazilian congressional representatives by political scientist César Zucco Jr., whom we thank for providing the information. Based on the score attributed to the parties in the surveys closest to 2014, we used cluster analysis (with the k-means clustering method, where K=3) to identify left, center, and right parties. Parties without scores in any survey comprise the group without ideological classification. Left parties include PC do B, PSB, PSOL, PT, and PSTU. Center parties include PDT, PPS, PROS, PV, and SD. Right wing parties include DEM, PMDB, PP, PR, PRB, PSC, PSD, PSDB, PTB, and PTN. The following parties were not classified: PCB, PCO, PEN, PHS, PMN, PPL, PRP, PRTB, PSDC, PSL, PTC, and PT do B.

the political competition or if it divided resources between ruling and opposition parties.<sup>9</sup>

Finally, the scholarly literature has also considered the size of the political party as another explanatory factor (Mancuso et al., 2018; Mancuso et al., 2021a, b). Several indicators exist for measuring the size of a party, such as the number of affiliated members or the number of municipal and state directories, among others.10 The size indicator adopted in our research is the number of seats held by the party in the Chamber of Deputies, on election day.11 The representation of Brazilian states in the Chamber of Deputies is not exactly proportional to the population in each state, as the Constitution establishes a minimum of eight and a maximum of 70 federal deputies per state, which leads to overrepresentation of sparsely populated states, mainly from the North region of the country, and underrepresentation of the most populous states, especially the state of São Paulo (Nicolau, 2017). Nonetheless, we justify this methodological decision since, among all existing elections, this election provides the best comparison for the level of support obtained by different political parties at the national level.12 By including this factor in our analysis, our objective is to assess whether Odebrecht preferred to donate to candidates and collective organizations belonging to larger parties. As stated earlier, past success in politics often serves as a predictor of future success. In other words, if a party was strong enough to win a significant number of seats in an election, one would reasonably expect it to have advantageous conditions to remain among the largest parties in the

<sup>9</sup> To classify candidates under this criterion, we used information from the Legislative Database organized by the Brazilian Center for Analysis and Planning (CEBRAP). Candidates from parties that controlled ministries in the Dilma Rousseff government on October 5, 2014, the day of the first round of the presidential election, were classified as belonging to the ruling coalition. Seven parties made up Rousseff' s ruling coalition: PC do B, PDT, PMDB, PP, PR, PRB, and PT. The other 25 parties did not control any government ministry, and they could either behave independently towards the government or act in open opposition.

<sup>10</sup> Regarding indicators for party organizational strength as applied to the Brazilian case, see Bolognesi et al. ( 2019).

<sup>11</sup> We separated parties into size categories by means of a cluster analysis, with the K-means clustering method, where K=3. The major parties are PMDB and PT. Medium-sized parties are DEM, PP, PR, PSB, PSD, PSDB, and SD. The small parties are PC do B, PDT, PEN, PMN, PPS, PRB, PROS, PRP, PSC, PSOL, PT do B, PTB, and PV. Ten parties had no parliamentary representation in the Chamber of Deputies, and were classified under "no seat". These are: PCB, PCO, PHS, PPL, PRTB, PSDC, PSL, PSTU, PTC, and PTN.

<sup>12</sup> Berlatto et al. (2016) measured the size of Brazilian parties using a similar strategy to our own.

next election. As a result, candidates and collective organizations from these parties would attract more funding from pragmatic companies willing to invest their resources in the most promising alternatives.

In addition to these variables, the international literature has also explored the relationship between corporate political financing and the driving motivations and strategies of the business community during election periods. Political scientist Iain McMenamin (2008, 2009, 2012) has provided important contributions to the study of the logic behind corporate political donations. In general lines, the author identifies two possible motivations for the behavior of corporate donors: (i) ideological motivation, which prompts companies to concentrate donations on candidates and parties with similar ideological positions to the companies' owners, generally further to the right of the political spectrum; and (ii) pragmatic motivation, in which the company defines its donations not according to the ideological convictions of the owners, but rather two criteria: political (to establish ties with those already in power or leading the polls) or economic (to improve the company's relative position in economic competition). Both motivations manifest themselves through different donation patterns. In the first case, the company donates to parties regardless of whether they are ruling or opposition or their likelihood of winning elections. In the second case, financial support favors ruling parties or parties with likelihood of electoral success. Other authors have explored the same division between pragmatic and ideological donations (Snyder Jr., 1992; Francia et al., 2003; Stratmann, 2005; Goerres & Höpner, 2014). Speck (2016) associates the concepts of ideological versus pragmatic donations to two different moments of exerting influence on the electoral process and elected representatives. Given that it is impossible to maximize both objectives at the same time, the author argues that companies face a dilemma, which manifests itself in the financing strategies during plurality elections. Ideological donors favor the strategy of influencing the electoral process and, to achieve this goal with maximum efficiency, they invest their financial support solely in one party. This strategy, however, risks failing to obtain good relations with the future government in case of defeat of the party supported by the company. Pragmatic donors, in turn, specifically prioritize access to and influence over the future government, regardless of the party's ideology. Consequently, they finance all feasible parties during the electoral campaign. The cost of this strategy, i.e., supporting opposing parties, is that companies refrain from influencing the electoral outcome. We' ll return to the donor dilemma when we explore Odebrecht's donation pattern for president and governor elections.

In summary, our analysis explores to what extent characteristics such as the candidates' political capital, party size, and the relationship with the ruling government influenced the Odebrecht Organization's donation strategies in the 2014 elections, which would indicate the company's pragmatic rationale. In turn, if other motivations have dictated the Odebrecht group's donor logic, we would also find a significant impact of other factors, such as ideology, the company's relationship with the Rousseff administration, and geographical region. Likewise, when focusing on plurality elections, we interpret the balanced allocation of donations among several candidates as a preference for pragmatism, while unilateral financing of a single party signals a more ideological stance.

#### **3 The Odebrecht Group's Political Donations in the 2014 Elections**

In the 2014 Brazilian elections, the last to allow corporate campaign financing in the country, official campaign contributions from the Odebrecht group amounted R\$ 88,895,700.00. Of this total, R\$ 22,575,700.00 was allocated to individual candidacies (25.4%) and R\$ 66,320,000.00 (74.6%) to collective party organizations, such as national or state-level directories, and common financial committees.

Campaign contributions create ties between donors and recipients, forming relationship networks between them. SNA allows us to observe and interpret such relationships.13 The actors of a financing network are nodes (or vertices) in a graph. Here, the nodes are the Odebrecht group, in the capacity of financing agent, as well as the candidates and party organizations that received donations from the company. The donations form the edges (or ties). This configuration enables us to comprehend some network measurements employed in this research, which we present and discuss below.

The *degree* of a node equals to the number of relationships it establishes with other nodes in the network. Therefore, the Odebrecht Group has a 285 degree, equivalent to the number of actors to whom it donated in the 2014 elections. Given that this is a directional network, it is an out-degree as the company only donates. In turn, the 207 candidates and 78 party organizations, given that they only receive in this network, have different in-degrees equivalent to the number of deposits they received (Table 1).

<sup>13</sup> We based our definitions and metrics of the networks used in this work from Dégenne and Forsé ( 2007).


**Table 1** Frequencies and measures of centrality, by office and party organization

*Source:* Data from the TSE, prepared by the authors

The degree, however, is an insufficient measure to analyze the network since, for example, a greater number of donations received by an actor may add up to less than the amount of resources received by another actor from a single donation. We must therefore also consider the *weighted degree*, which is the sum of the donations that exit from or arrive at a given node. Thus, the weighted degree of the Odebrecht Group is the total sum of donations, i.e., R\$ 88,895,700.00. Among recipients, national and state directories have the highest weighted degree, followed by presidential candidates.

#### **3.1 Odebrecht's Donations to Candidacies**

As shown in Table 2, the Odebrecht group's donations to individual candidacies were divided as follows: 56.7% for national candidacies for the presidency of the republic and 43.3% for statewide candidacies, whereas among the latter, candidates for governor received 17.8% of the total; for federal deputy, 13.7%; for state deputy, 9.4%; and for senator, only 2.4%.


**Table 2** Odebrecht donations to candidacies, 2014

*Source:* Data from the TSE, prepared by the authors

Table 3 shows, in aggregate, the distribution of Odebrecht's total donations to candidates according to their ideology and relationship with the federal government (i.e., whether they belong to the government's ruling coalition in Congress). We may already identify some typical indicators of political pragmatism: the Odebrecht group distributes its eggs among various baskets, establishing ties with government and opposition candidates, without prioritizing ideological lines.

We may better visualize Odebrecht's relational dynamics with political actors through graphs of the financing network. Figure 1 shows the general graph of the


**Table 3** Measures of the Odebrecht Group's Financing Network—Candidates

*Source:* Data from the TSE, prepared by the authors

relationships between the company and the candidates it finances. The chosen layout is the Circle Pack, which distributes nodes from a hierarchical structure of the research variables. Here, the order of the hierarchy is as follows: 1) rulingopposition; 2) ideology; 3) proportion of the weighted in-degree.

The size of the nodes represents the weighted in-degree, i.e., the sum received from the Odebrecht Group. On the left side of the graph are the actors who comprise the support base of the Dilma Rousseff administration; on the right side, the opposition. The hues of the nodes express the positions in the left–right spectrum, as indicated by the caption.

The visualization reinforces the information from the tables above, i.e., we find a balanced distribution of resources between ruling and opposition candidates/parties. Furthermore, the company clearly has few ideological restrictions, cautiously balancing its donations to parties and candidates from the left, center,

**Fig. 1** Odebrecht Group Financing Network—Candidacies. *Source:* Data from the TSE, prepared by the authors

and right. It is worth adding that candidates from the left, center, and right are present both in the ruling coalition and in the opposition.

We shall now address the Odebrecht Group's financing pattern according to political office. We will begin with the plurality elections, i.e., for president, governor, and senator.

Table 2 shows that donations to presidential candidates represent 56.7% of the total sum donated to candidacies. This data reveals the primacy that the group attributed to elections for the highest office of the republic. Table 1 and Fig. 1 show that the Odebrecht group was selective and donated only to three of 11 presidential candidates: Dilma Rousseff (PT), Aécio Neves (PSDB), and Eduardo Campos/Marina Silva (PSB).14 Broadly speaking, the candidacies represented different positions in the ideological spectrum: Dilma and Eduardo Campos/Marina Silva further to the left, while Aécio was further to the right. As for size, the PT was the largest party in the country while the PSDB and PSB were medium-sized parties. The distribution of resources was not equal but incorporated all candidates with some prospect of electoral success: R\$ 8,250,000.00 (64.5%) went to thenpresident Dilma Rousseff, candidate for reelection; R\$ 4,000,000.00 (31.2%) to Aécio Neves, the main opponent; and R\$ 550,000.00 (4.3%) to Eduardo Campos/ Marina Silva. The group aimed accurately and the ranking of three candidates paralleled their funding: Dilma won the election against Aécio in the second round, and Marina Silva finished in third place in the first round. Odebrecht pragmatically divided its investments among the three main competitors, with different ideological positions, but concentrated the largest donation on the victorious campaign. When distributing its investments, Odebrecht was mindful of the electoral polls, which consistently showed Dilma Rousseff leading the race,15 and knew that every president running for reelection in Brazil until then had won the electoral race.16 The story was no different in 2014 and, as a result, nearly two-thirds of Odebrecht' s resources went to the winning candidacy.

The Odebrecht group allocated 17.8% of their total sum of investments to elections for governor. Once again, the group was very selective, donating to only 22 of the 176 candidacies across 27 Brazilian states which held elections for governor (12.5% of the total candidacies). In this scenario, the donations

<sup>14</sup> Eduardo Campos died in a plane crash during the 2014 campaign. He was replaced by Marina Silva.

<sup>15</sup> The results of the electoral polls for the 2014 presidential race, performed by the Datafolha and Ibope Institutes, can be found at: https://eleicoes.uol.com.br/2014/pesquisas-eleitorais/ brasil/1-turno/

<sup>16</sup> Fernando Henrique Cardoso in 1998 and Luiz Inácio Lula da Silva in 2006.

followed a more proportional distribution than in the presidential race—18 candidates (81.8%) received R\$ 200,000.00; three candidates (13.6%) received R\$ 100,000.00, and one candidate (4.5%) received R\$ 120,000.00. Of the 22 candidates funded by Odebrecht, nine were elected, which corresponds to 41% of victorious candidacies, a considerable level of success. Nine other candidacies came in second place (41%) and four (18%) finished in third place. The nine victorious candidacies amassed R\$ 1,620,000.00, equivalent to 40.3% of the total sum invested in races for governor. The 22 gubernatorial candidacies funded by Odebrecht were in 12 different states. In four states, Odebrecht financed only one candidate for governor. In the other eight states, the company simultaneously financed two or more opposing candidates. The group's prevailing strategy of distributing eggs across several baskets is once again clearly observable in the governor campaigns. If the company risked undermining its influence over the electoral process by recurrently backing both sides of the contest, on the other hand it improved the likelihood of securing good relations with the newly elected government, whichever it may be, by ranking among its campaign funders.

Seven candidates for governor who received financial donations from Odebrecht (31.8%) were either incumbents or belonged to (or were backed by) the party of the re-elected governor in 2010. The remaining 15 (69.2%) did not fit any of these criteria. Therefore, Odebrecht did not seemingly follow political capital as a criterion for distributing resources among gubernatorial candidates—at least not according to the indicator adopted in our work.17 Eight candidates financed by Odebrecht were from the Northeast (36.4%), six from the Southeast (27.3%), five from the Midwest (22.7%), and three from the South (13.6%), which suggests a relatively balanced geographical distribution, even with the absence of financed candidates in the North. Regarding ideology, 13 candidacies funded by Odebrecht were launched by right-wing parties (59.1%), eight by left-wing parties (36.4%), and one by a center party (4.5%). This distribution pattern shows that Odebrecht donated to gubernatorial contenders with diversified and sometimes opposite ideological profiles. Regarding affiliation to the federal government's ruling coalition, exactly half of the candidacies were either situationists or oppositionists. As for party size, eight beneficiaries belonged to large parties (36.4%), 11 to medium-sized parties (54.5%), and only two were affiliated to small parties (9.1%). Therefore, we find that the group has a clear preference for candidates affiliated to larger parties to the detriment of smaller ones.

<sup>17</sup> Future researches may assess whether state-level election polls affected Odebrecht's donor behavior in gubernatorial elections.

Odebrecht invested only 2.4% of their total donations in elections for senate. Brazilians elected 27 senators (one per state) in 2014.18 In a more selective approach, Odebrecht benefited only 11 out of 185 senatorial candidates (5.9% of the total), in eight different states, with the same amount of R\$ 50,000.00 for each candidacy. Seven funded candidates were elected—a high percentage of 63.6%. Three funded candidates arrived in second place (27.3%) and only one candidate finished in third (9.1%). In five states, Odebrecht financed only one senatorial candidate. In the three remaining states, the group financed two opposing candidates concurrently. In this case, we identify a certain predilection for unilateral financing. However, given the high success rate, the company seems to have opted for candidacies with a high prospect of success.

Among the senatorial candidates who benefited from donations by Odebrecht, five (45.5%) were either incumbents or belonged to (or were supported by) the same party of the senator elected in 2006, while the other six (54.5%) did not have such characteristics.19 Therefore, the political capital criterion, as adopted in this chapter, does not seem to have determined the business group's donations in this particular case. As for region, five of the funded candidates were from the Northeast (45.5%), three from the Southeast (27.3%), two from the North (18.2%) and one from the South (9.1%), which indicates a certain regional dispersion of the group' s contributions, albeit still somewhat concentrated in Northeastern candidates. Six of the funded candidacies were launched by leftwing parties (54.5%) and five by right-wing parties (45.5%), revealing a balance in the ideology category. Three candidates were members of parties comprising the ruling coalition in Congress (27.3%), while eight belonged to parties outside the coalition, whether independent or opposition (72.7%). Regarding party size, three beneficiaries belonged to large parties (27.3%), seven to medium-sized parties (63.6%), and only one candidate belong to a small party (9.1%). Hence, Odebrecht clearly favored the larger parties.

Tables 4 and 5 provide a more detailed look at proportional elections, i.e., for federal and state deputies. As for selectivity, the group donated to only 97 of the 5,813 candidacies for the Chamber of Deputies ratified by the Electoral Court in 2014 (1.7% of the total candidacies). The company was even more selective in the state deputy elections, as it donated only to 74 of the 15,139 candidacies

<sup>18</sup> Each state holds three seats in the Federal Senate. Elections for senate take place every four years, sometimes for 1/3 of the seats (as in 2006 and 2014), sometimes for 2/3 of the seats (as in 2010 and 2018).

<sup>19</sup> The impact of state-level election polls on corporate donations to senatorial candidates also merits further analysis in future works.


**Table 4** Logistic regression, candidates for federal deputy, 2014

Nagelkerke: 0,244

*Source:* Data from the TSE, prepared by the authors

for the legislative assemblies (0.5% of the total). Therefore, Odebrecht carefully selected candidacies before allocating resources.

As for parity, Odebrecht donated the same amount to most candidates for each office. In the elections for Federal Congress, 90 of the 97 candidates (92.8%) received exactly the same amount: R\$30,000.00. Only five candidates (5.1%) received above the modal value. Among them, Samuel Moreira stands out, a candidate launched by the PSDB for the state of Sao Paulo, who received R\$ 133,300.00, over 4.4 times the typical amount. Only two candidates (2.1%) received less than usual. Jander Filaretti, from the PTB in the state of Minas Gerais, raised the least funds, earning only R\$ 3,100.00, i.e., 89.7% less than most candidates. In the elections for the state legislative assemblies, 64 of the 74 candidates (86.5%) obtained R\$20,000.00 from Odebrecht. Nine candidates (12.2%) received over this amount, among them veteran Barros Munhoz, from the PSDB of Sao Paulo, who raised R\$ 200,000.00 for his campaign for state deputy, i.e., 10 times more than the most common value. Only one candidate (1.3%) received less than the modal value: Carlos Avalone, from the PSDB in


**Table 5** Logistic regression, candidates for state deputy, 2014

Nagelkerke: 0,195

*Source:* Data from the TSE, prepared by the authors

the state of Mato Grosso do Sul, raised only R\$1,000.00 from Odebrecht, 95% less than the most common value.

As for the company's "good marksmanship", that is, their bet on victorious candidates, 63 of the 97 candidates for federal deputy who received funding from the Odebrecht group were elected—a considerable success rate of 64.9%. Regarding the company's total investment in political campaigns for a seat in the Chamber of Deputies, 67.5% were allocated to winning candidates (R\$ 2,083,300.00, from a total of R\$ 3,086,400.00). As for candidates running for state deputy, 46 of the 74 candidates financed by the group were elected—a high success rate of 62.2%. The elected candidates amassed 68.3% of the sum invested by Odebrecht in electoral campaigns for a seat in the legislative assemblies (R\$ 1,448,300.00 from a total of R\$ 2,119,300.00).

Table 4 shows the results of a logistic regression in which the dependent variable is whether the candidate received campaign contributions from the Odebrecht group in the 2014 elections for federal deputy. The 97 candidates who received donations from the company were marked with a value of 1. The remaining were marked with a value of 0. We included five independent variables in the model, discussed in the previous section, that could influence the company's donor behavior. The reference profile is an incumbent candidate, from the Northeast, right-wing, member of the government's ruling coalition, and launched by a large party.

The results in table 4 reveal that three explanatory variables included in the model are significantly associated with candidates for federal deputy having received campaign contributions from the Odebrecht group (we consider the association between variables to be statistically significant when the p value is equal to or less than 0.100). First, the company has a clear preference for experienced candidates, as the likelihood of incumbents having benefited from the company's campaign contributions is almost 8.9 times greater when compared to challengers. Second, candidates from the Northeast were more likely to receive contributions from Odebrecht than competitors from all other regions of the country. The likelihood of Northern, Midwestern, Southeastern, and Southern candidates to receive donations from the business group correspond, respectively, to 3.8%, 11.3%, 36.3% and 44.3% of the odds for Northeastern candidates. Third, the company favors candidates from the largest party associations: candidates from large parties received more donations than candidates from smaller parties, with a statistically significant difference when compared against candidates from parties without seats in Congress (on average, the odds of a candidate from a party without parliamentary representation receiving a donation from Odebrecht corresponds to only 16.9% of the odds of a candidate from a large party), and against candidates from small parties (in this case, the likelihood of obtaining resources from the corporation corresponds to 21.7% of the odds of a candidate from a large party). The ideology professed by the candidates' parties as well as belonging to the government's ruling coalition do not seem to affect the company's donating behavior.

From the data above, the emerging profile of the Odebrecht group, as a source of funding for candidates running for federal deputy, reveals a politically pragmatic business group which allocates the same amount of resources to the vast majority of selected candidates, not singling out beneficiaries by ideological criteria nor by affinity with the incumbent government, but rather by factors associated with the likelihood of electoral success and the geographical origins of the corporation, favoring candidates for reelection from major parties, especially from Northeastern states. Such pragmatism is probably the most important explanatory factor for the group's high investment in successful candidacies who, having benefited from its resources, may, at least in theory, during the term of office become more accessible and open to company representatives or more receptive to its interests.

Table 5 presents the results of another logistic regression in which the dependent variable is whether candidates were beneficiaries of campaign donations from the Odebrecht group in the 2014 elections for state deputy. The 74 candidates who received donations were marked with 1, while those who were not beneficiaries marked with 0. The explanatory variables included in this model, as well as the reference profile, are exactly the same as in the previous model.

Once again, running for reelection is the factor most closely associated with receiving donations from Odebrecht: incumbents were 8.4 times more likely than challengers to rank among the company's list of beneficiaries. Once again, the likelihood of candidates from large parties obtaining electoral contributions from the group exceeds that of candidates from smaller parties, as indicated by the sign of the coefficient, but in this case the difference is only statistically significant with candidates from small parties. The sign of the coefficient also suggests that the regional effect continues to favor Northeastern candidates, albeit less obvious in the elections for state legislative assemblies, as the difference only gains statistical significance when we compare against candidates from the Midwest. Contrary to elections for the Chamber of Deputies, party ideology seems to have some effect in the state legislative elections, as centrists are 2.2 times more likely than right-wingers to gain electoral funding from Odebrecht, while the opposite occurs with candidates from parties without ideological classification, whose likelihood of receiving resources corresponds to only 26.1% of the odds of right-wing candidates. As in the previous model, the candidates' membership to the government's ruling collation is not statistically associated with receiving donations from the construction company.

In summary, while the statistical association is less substantial in this case, whether due to fewer recipients or the greater number of non-recipients, Odebrecht' s pragmatic behavior nevertheless remains conspicuous, flavored by regional criteria. Once again, the construction company offered the same amount of money to the vast majority of financed candidates, predominantly consisting of incumbents launched by larger political associations, especially in the Northeast, regardless of their status as supporters or opponents of the Dilma Rousseff administration, yet in this case with a certain predominance of a centrist ideological profile. Although the company invested relatively small amounts in elections for the state legislative assemblies, allocating less than 10% of the total donations to these candidacies, the company's political investment once again seems predominantly guided by the prospect of electoral success, combined with regional criteria. This seems to have been the group's leading strategy to raise the proportion of contributions allocated to successful candidates, who could be eventually called upon as a potential channel of political influence.

#### **3.2 Odebrecht' s Contributions to Parties**

Table 6 shows the partitioning of donations by the Odebrecht group to collective party organizations (national directorates, state directorates, and common financial committees), aggregated by political party.

According to the data above, the company concentrated 74.2% of its donations to the three largest parties in the country at that time—PMDB, PT, and PSDB which received, respectively, 29.3%, 23.8%, and 21.1% of the total donations. The company distributed the remainder among 19 parties. Ten political parties received nothing from the group.

Table 7 shows the distribution of donations among parties, aggregated according to their relationship with the government's ruling coalition and ideology. Although we find a predominance of donations to ruling and right-wing parties, opposition parties and left-wing parties also received a significant sum of electoral contributions.

The graph in Fig. 2 provides a visual representation of the relationships between the Odebrecht company and party organizations, with the same previous layout for candidates (Circle Pack) and the same hierarchical distribution for the nodes, i.e.: 1) ruling-opposition; 2) ideology; 3) proportion of the weighted in-degree.

The figure reveals a clear predominance of the largest and most important political parties in Brazil at that time. The figure shows the allocation of resources between ruling and opposition parties and according to their positioning in the left-right spectrum. Ruling parties are to the left of the graph while opposition parties are to the right, and the different hues indicate ideology.

Table 8 presents the results of a multivariate linear regression analysis, in which the dependent variable is the amount Odebrecht donated to political parties (cf. Table 6) and the explanatory variables are ideology, relationship with the government's ruling coalition, and party size (see criteria in Tables 4 and 5). The reference profile is a right-wing, ruling, large party.

The results once again indicate the Odebrecht group's political pragmatism. The only explanatory variable with a statistically significant impact is party size. When compared against large parties, medium-sized parties received, on average, circa R\$ 15.4 million less from the construction company, with further disparity in the case of small parties and parties without parliamentary representation, which received on average, respectively, R\$ 18.3 and R\$ 18.7 million less than large parties. The status of being a ruling or opposition party, as well as the party's ideological position, were not factors that significantly affected the donor behavior of the business group. In other words, the construction company's strategy


(continued)



*Source:* Data from the TSE, prepared by the authors

**Table 7** Measures of the Odebrecht Group Financing Network– Political Parties


*Source:* Data from the TSE, prepared by the authors

concentrated its stakes on the largest Brazilian parties during the 2014 elections, wagering on their current strength as the optimal predictor of their future strength, hopeful that the company's donations would prompt the country's major parties to be more receptive to the interests of the business group.

#### **4 Final Remarks**

In sum, we may characterize the Odebrecht group's donor behavior in the 2014 elections as largely pragmatic. The group was in fact extremely selective when choosing recipients for its campaign contributions, prioritizing candidates and parties with greater odds of electoral success, with a view to enhance the company's future prospects of political influence. As for the criteria investigated in

**Fig. 2** Odebrecht Group Financing Network—Party Organizations. *Source:* Data from the TSE, prepared by the authors



*Source:* Data from the TSE, prepared by the authors

this chapter, Odebrecht mainly adopted the criterion of party size, as the company preferably donated to candidates and collective organizations belonging to large and medium-sized parties rather than smaller political associations. Another pragmatic criterion adopted by Odebrecht was their clear-cut preference for financing incumbent candidates in presidential elections as well as in elections for the Chamber of Deputies and for state legislative assemblies. Political pragmatism yielded good results. When financing candidacies, the company revealed exceptional marksmanship and obtained high success rates across all elections for different offices, consistently allocating a large share of donations to winning candidates. The company had a uniform approach when apportioning resources to candidates in statewide plurality and proportional elections, allocating a similar sum of resources to competitive gubernatorial candidates (many of which competing against each other), to the vast majority of candidates for federal and state deputy, and to all senatorial candidates. In the presidential election, the company concentrated its resources on the victorious candidacy of Dilma Rousseff, while also supporting the main contestants, allocating significant amounts to the runner-up candidate. In the plurality elections for federal and state executive offices, faced with the dilemma of influencing the electoral process or ensuring good relations with elected governments, Odebrecht mainly opted for the latter.

Other motivations had a lesser impact on Odebrecht's donor behavior. The group generally did not prioritize factors such as the ideological profile of contestants or the parties' association with the federal government. In most cases, the construction company financed ruling, independent, and opposition candidates and parties alike, positioned across different ideological alignments within the left-right spectrum. The regional criterion only becomes clearer in proportional elections, especially for federal deputy, when the group showed a clear preference for Northeastern candidates.

In this chapter, we sought to contribute to the political science literature on the logic behind the use of corporate donations to fund electoral campaigns. We hope this work will encourage similar studies on other business groups in Brazil as well as in other countries.

Our investigation focused exclusively on the Odebrecht group's official donations in the 2014 elections. However, we remain tasked with the challenge of investigating illicit corporate financing of political campaigns, i.e., funding procedures that do not comply with the current legal system. This type of financing is at the root of major political scandals in Brazil and worldwide (Mendilow & Phélippeau, 2021). In fact, 2014 also marked the launch year for Operation Car Wash, a large-scale investigation by the Federal Prosecutor's Office into a massive corruption scheme involving Petrobras, Brazil's giant oil state company. The scheme involved, on the one hand, major companies supplying goods and services to Petrobras, especially in the engineering and construction sector, among which the Odebrecht group, and, on the other hand, some directors of the state-owned company appointed to office through referrals by political parties belonging to the federal government's ruling coalition in the National Congress. Petrobras directors would then surreptitiously transfer a share of the unlawfully appropriated money to political parties, which referred their nomination, to use at their discretion, including funding electoral campaigns. Analyzing major scandals such as the Car Wash scheme allows for a better understanding of the intricacies and modus operandi of illicit electoral financing. In this same book, the chapter by Costa et al. (2020) provides an important contribution in this direction as they address how illegal political financing, as much as its relationship with Brazilian democracy, has been described by former executives of the Odebrecht group, who signed plea bargain agreements amid the Operation Car Wash investigation.

Furthermore, illicit electoral financing not only lies at the heart of major corruption scandals, but reemerges, through different prisms, amid customary practices in election periods throughout the country. Hence, our current research agenda also seeks to analyze the impact of this issue on the judicialization of electoral competition in Brazil (Mancuso et al., 2021a, b). We believe this is another promising path to overcome the persisting shortfalls in the study of corporate campaign funding.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter' s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter' s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **The Public and the Private in Odebrecht's Vocabulary**

#### Maria Eugenia Trombini and Raphael Bischof dos Santos

#### **1 Introduction**

It was announced in Congress that a "criminal society" formed by construction companies was in charge of public sector decisions, directing bids and overcharging expenses in order to divert resources to political campaigns. The year was 1993, but the staging could have taken place in several other times in Brazil's history. Like many other episodes in which malfeasance was reported, the Parliamentary Commission launched to investigate the construction sector "ended up in pizza", a common slang to say nobody was punished. Odebrecht was at the center of it. This same luck in weathering scandals was gone a few years later when *Erga Omnes* Operation, named after the latin brocard meaning "For Everyone", arrested the company´s then president, and heir of the family business, Marcelo Bahia Odebrecht. The 14th phase of the Car Wash probe, the set of investigations on graft that hit Brazil and Latin America, was a fallout for the economic and political elites that had been until then successful in securing wins at a biased game in the tropical setting. But what are the ways, largely missed by scholars, that enabled this pattern of state-firm relations from emerging on the first

R. B. dos Santos Heidelberg, Germany e-mail: raphael.bischof@mwi.uni-heidelberg.de

M. E. Trombini (B) Heidelberg Center for Ibero-American Studies, Heidelberg, Germany e-mail: maria.trombini@uni-heidelberg.de

© The Author(s) 2024 M. E. Trombini et al. (eds.), *The Fight against Systemic Corruption*, Organization, Management and Crime – Organisation, Management und Kriminalität, https://doi.org/10.1007/978-3-658-43579-0\_7

171

M. E. Trombini

Max Weber Institute of Sociology, Heidelberg, Germany

place? The empirical argument put forward here is that the boundaries between private and public are far more than blurred in the Brazilian context, and that from the standpoint of the second largest private group operating in the country,<sup>1</sup> the imbricated relationship with the state was a deep-rooted strategy, one that would potentially still be employed, so long as the desired outputs continued to be secured.

An economic organization very active in public procurement and one of the few enterprises who could provide services and products to the state administration in the construction and chemical sectors, the story of the Odebrecht group offers relevant lessons for understanding state-business relations. We draw on institutional polycentrism, its understanding that institutions originate from and are configured by multiple sources, to analyze the interaction between Odebrecht and the state, locate the public and private spheres, as well as the place where they are interwoven. A qualitative approach will be employed to retrace episodes and narratives when business and statesmen relate while trying to detect stable patterns in their interaction across time. To this end, we intend to gauge, using the empiric material available from criminal investigations and historical sources with corporate communication, both the formal and informal exchange practices so as to identify patterns of action. The underlying assumption is that political and economic actors choose the formal rules to follow and couple them with informal ones in a process that may generate a mismatch between the written and unwritten norms.

To bypass a normative view of corruption, we will rely on data produced by the organization in two different contexts: first a historical one, tracing episodes when *Odebrechtian* interests were enmeshed with the government's, then a judicial one, when its executives spoke before the judiciary and the malfeasance had already been exposed. Instead of following a definition of corruption or tracing its manifestation in an inductive exercise, the research agenda ahead lies on taking stock of elements from the organizational knowledge repertoire that express how, for the Brazilian case, the public and private sectors are permanently in contact.

The objective of this paper is to scholarly inquire the boundaries between the public and private spheres as understood by the Odebrecht organization. Rather than following a normative approach, empirical material from the enterprise and its members touching on the topic of corporate-state relations from throughout its history (1943–2019) will be used to derive conclusions about the meaning of interactions and boundaries between the private and the public. Over the long

<sup>1</sup> Data from shortly before Car Wash Operation was launched, in 2012 Odebrecht was the second largest private group operating in Brazil, only behind Vale.

term, what have those who repeatedly interacted within a set of institutional arrangements learn about how to relate to one another? What does that tell us about predicting decision-making in the future, when similar actors engage in similar problem-solving activities? To scrutinize domestic determinants for systematic corruption taking the company as an example, the text has been structured in three main parts. The first section will present a conceptual and methodological scheme. The second section will enter the Brazilian setting and Odebrecht´s place in it through a historical rescue of episodes of contact between the infrastructure and petrochemical industry and the government followed by a scheme of exchange of favors framed as a strategic game. In the third section, findings of actors' preferences will be discussed with the aid of an interpretive method. We finish the chapter with concluding remarks.

#### **2 Conceptual and Methodological Scheme**

With respect to the extent of meddling in the economic affairs, states can be more or less intrusive. The literature on regulation outlines that capitalist states can be classified according to the kinds of priorities in regards to government– business relationships. There are those in which a developmental orientation, or substantive matters, is key, and others where a regulatory orientation, or formal matters, predominates. Traditionally, research efforts have focused on the rise of the regulatory state associated with the US and European Madisonian models, characterized by its market-rationality and clear ambitions of modernization, standardization and good governance. At the flip side, the developmental state, coined as a typology to capture inductively the characteristics of the Japanese state in the era of high growth, would give leeway for the bureaucracy to "guide" business using its discretion (Johnson, 1982). Because this governance system would still be based on trust and tacit agreements between business and governmental elites, it would conserve the discretionary style of a positive state, rather than having evolved to a rule-bound, legalistic policy style (Majone, 1997).

Together with Japan, Brazil has been positioned as a developmental state where powers shift towards the Executive and discretion takes precedence over legality. In a stark contrast with the US model, rather than the Legislative, the Executive would manage the regulatory activities, exerting influence through budgetary provisions (Prado, 2010). For scholars from the legal field, the Brazilian setting would be marked by "a public–private relationship with greater organizational verticality, in the form of pacts and bargaining between the state and interest groups" undermining the emergence of a pluralistic representation found in adversarial legalistic governance (Schapiro, 2018, p. 583). An orientation towards development has been the priority of Brazil's administration for seven centuries, but not only the governmental elites participated in this endeavor. Anchored in the notion of "institutional complementarities", Matthew Taylor (2020) argues "developmentalism" survived until these days unfolding complementarities across the private and public spheres. In a self-reinforcing strategic choice game, such complementarities offer incentives for economic and political actors to select individually optimal but collectively suboptimal choices. The combination of the developmental state; autonomous government bureaucracy; weak regulatory control and coalitional presidentialism makes of Brazil a "developmental hierarchical market economy" (Taylor, 2020). Non-market and hierarchical relations in and across business is one of the features of the variety of capitalism found in Latin America. Rather than liberal market economies (LMEs) of Anglophone countries or the coordinated market economies (CMEs) of northern European countries, the distinctive type of Ibero-American States would be hierarchical market economies (HMEs) (Schneider, 2009).

The frame that shifts attention from institutional changes to continuities was a timely rebuttal to the canyons of economic liberalization. Casting aside the claim that "in developing countries, the history of developed countries is being repeated" was an exercise proposed by one of Brazil's former president, when still a sociologist (Cardoso & Faletto, 1970, p. 34). His ideas challenged modernization theory introducing concepts like path dependence, critical junctures and punctuated equilibrium to explain the Brazilian historical experience. In terms of the relationship between state and economic development, the state assumed the role of a "great entrepreneur" in developing countries. Besides the positive externality, the stimulus offered by the state, an entity with far more resources to attract and secure investments in a nascent economy, a negative one would be that "somehow, the private entrepreneur ends up depending on official institutions to maintain the pace of growth of their businesses" (Cardoso, 1961, p. 160). If the effects of institutions on firm actions is interdependent, then market behavior in Brazil has historically been based in a reliance on a strong public sector. Before deeming a country more corrupt only because it has a relatively large public sector, the specific types of political influence in place merit more scientific attention (Scott, 1972). This chapter intends to be one step in this direction concerning Brazil.

Contrary to the idea that economic marketisation curbs corruption, the incentives for corruption offered by transition economies (Tonoyan et al., 2010) and a co-development of the market and corruption (Gong & Zhou, 2015) have been empirically proven. Using data from 26 transition economies between 1998 and

2009, interaction between formal and informal institutions—such as customs, traditions, social norms and deeply entrenched beliefs—positively influenced high-growth firms (Krasniqi & Desai, 2016). Evidence from North Africa shows that trade openness, in the short run, reduces local firms' profits on the domestic market, favoring corruption (Delavallade, 2012). Studies of modern days Eastern European economies suggest that the difficulty to make public administration more rule-based and less corrupt should be understood in light of legal cultures and political traditions already existing before the transition to a market economy (Solomon & Gadowska, 2018). New regulation would be a necessary, yet insufficient condition for preventing corruption in public procurement in countries where the state relies on a very narrow group of people to perform various functions (Ateljevic & Budak, 2010).

Informal institutions affecting entrepreneurs in transition countries were the object of careful investigations (Horak & Klein, 2016; Williams & Vorley, 2015; Williams et al., 2016). Evidence that interpersonal relations, social capital, and establishing trust with government officials has a positive effect on entrepreneurial activities has been found in a multitude of emerging economies (Cao et al., 2016; Wales et al., 2016). It is undisputed that in some contexts informal practices whereby government officials demand or receive gifts, bribes and other payments from private sector firms and provide a service in return are an efficient strategy at the firm level. Even in countries like the USA and France, with a strong rule of law, networks still offer the potential for positive returns to entrepreneurs (Batjargal et al., 2013). In regards to the mechanisms by which business and political elites interact there is no consensus, and different factors should be accounted for by careful analysts of the topic. A distinction between administrative and electoral corruption is a good place to start (Scott, 1972). Some efforts point out to differences between the economic incentives of corruption for public procurement and fiscal matters. Evidence from firms in 75 developing countries supports that in public procurement, corruption is a supply-side phenomenon, with bribe transactions generally initiated by firms to secure public contracts, rather than being requested by public officials. The opposite occurs with tax administration, which tends to be mainly a demand-side phenomenon (Gauthier et al., 2020). The demand and supply sides of bribery orient the frame of the game sketched for the Odebrecht executive and public officer.

According to democratic theory, "private interests contaminate public purposes in a democracy", or corrupt exchange happen, "when they influence the government without the warrant of the democratic process" (Thompson, 1995, p. 28). The early liberal project we inherited that had to do with securing the dividing lines between state and society, public and private, "has been molded, ex post and with a certain clumsiness, to democratic politics and institutions". (Warren, 2004, p. 329). Even if the concept still bears the marks of these origins, considering system malfeasance as a plague "in a way tantamount to a cancer" or a product of morally unconscious executives and public officers oversimplifies the economic and political landscape that is context-specific. In terms of definitions, if benefits, or expected gains, are accounted for at the firm's behavioral dispositions, even if they did not necessarily follow in the course of the events, then the private-regarding interests were reputed to have entered the other-regarding realm and will thus be considered here.

Brazil is an intrusive state with a developmental orientation, yet the role played by the economic elites in response to state centrism remains to be further explored. How did the relations between business and statesmen grow deepseated and linkages between private and public become rife? What argumentation schemes are employed by enterprise managers to represent "typical" aspects of their job roles when interacting with governmental bodies and officers? In regards to Odebrecht, the hypothesis is its executives share a similar orientation to the one of governmental elites. Rather than passively, we expect them to actively engage in reshaping, evading, and channeling the orientation in pursuance of their own goals. The discourse-historical approach used to treat the available data distinguishes between the discourses' content, or topics, its strategies and its linguistic means (Wodak, 2011, p. 38). Instead of operationalizing the method in detail, the intent of the data processing was to focus on conventional evaluations and usages of language. Therefore, categorization devices related to bribery employed by members of the organization under study are at the core of the data analysis.

This chapter wishes to empirically explore private/public boundaries across the history of a big player of the infrastructure and petrochemical sectors in the world economy in its interaction with the Brazilian government. First, we will use historical sources to retrace Odebrecht's trajectory following the steps of the three generations ahead of the family business. Second, based on a strategic action theoretical framework following Ostrom and Ostrom (2004), a configuration of transactions between businessmen and statesmen will be put forward where an executive from the company is interacting with a public official with the aim to secure a contract. Third, outputs will be discussed in light of the interpretive method of qualitative content analysis (Mayring, 2000), where categories are formal and informal practices, dealing with the private (Odebrecht) and the public (the state). Apart from historical and bibliographical sources, evidence from a dataset comprised of 8 sentenced criminal cases from which testimonies2 of 48 members of Odebrecht (at the time of the investigations) is presented in the sections that follow. Not having been produced for scientific purposes, they belong to the judicial genre, and should be interpreted in light of their limitations (Kersting & Erdmann, 2014). As speakers, executives are responsive to the particularities of context as well as content, so before accusations they structure the readability of their testimonies in one fashion whereas when giving a newspaper interview for a profile piece they do so in another. This property of the "act-text-act" sequence (Smith, 2006) will be considered for the analysis.

#### **3 The Brazilian Setting and Odebrecht's Place in It**

This section is dedicated to a historical rescue of episodes of interaction between Odebrecht, a representative of the infrastructure and petrochemical industry, and the Brazilian government. We will proceed referring to episodes from the firm's early childhood, teenage years and as a grown up, an aging which accompanied the consolidation of democracy in Brazil, or the periods before, during and after the military rule (1964–1985). By a twist of fate, Odebrecht steered towards progress once Petrobras, state-controlled Brazilian oil group and largest company, started to grow. That was decades before both companies ran into the pitfalls of a well-equipped and carefully planned investigative probe. Car Wash, sometimes referred to as *Petrolão* (Portuguese for "big oil"), stroke hard both giants: at its outset, in 2015, Odebrecht lost almost \$88 million compared with a net profit of \$210 million in the previous year while Petrobras recorded a loss of almost \$8.5bn on revenue of \$97bn.

Almost half a century before both were embroiled in the corruption investigation, Petrobras' main office, a 26-story building in downtown Rio de Janeiro, was built by Odebrecht. The construction firm founded in 1944 in Salvador, state of Bahia, by Norberto Odebrecht, the son of a German immigrant, had "fingerprints on most of the continent's developmental touchstones" (McManamy & Schmid,

<sup>2</sup> For the collection of data, the keys provided at the repository organized by the Federal Prosecution were retrieved to access each criminal case at the website eproc.jfpr.jus.br/ and gather the relevant audiences, transcribed as PDF files. Testimonies range from audiences carried out in October 2015 to October 2018, therefore before and after the plea agreements, entered in December 2016 (both in Brazil and with the FCPA). This is a sample of a larger dataset assembled by the "Corporate Crime and Systematic Corruption in Brazil" project supported by the *Deutsche Forschungsgemeinschaft* (DFG).

2001) But in the years when only southeastern companies were granted infrastructure works, it had not always been like that. The reason? Lack of political support, in Norberto's own words—that would later become a gospel for the firm's team.<sup>3</sup> As the prospects of exploring oil at the coast made the administration of Petrobras decide to build its headquarters in Rio, the winds of fortune finally blew for Odebrecht and they won the public contest. Its market expansion enabled the company to move to the Southeast, Brazil's most industrialized region. Apart from a long-pursued goal, the geographic sprawl also meant being closer to the centers of power, and overcoming the threshold of political support that had stopped them from participating in the construction of Brasilia, planned to be the new capital, in the 1960s. Years later, Odebrecht S.A., a holding company that conducted business in multiple industries, including engineering, construction, infrastructure, energy, chemicals, utilities and real estate, was at the top of the Latin American ranking (Odebrecht, 2015a, b).

#### **4 The days of the Forefathers: Cracking the Doors Open**

Norberto inherited from his father a bankrupt construction company, but the lessons learned while recovering it became an important part of the corporate storytelling for the generations that followed. In order to do so, the first pact the young entrepreneur established was a political one, with Banco da Bahia, the company's main creditor, which became an ally in the conquest of new clients. The second pact, an economic one, promised the clients a better performance in shorter time and for less money. The third was a social pact, a "partnership" entered with the construction workers of his father's firm (Moura, 2014) whose wages would be compensated through profit-sharing. This feature was the embryo of the decentralization that defined the company in the years to come. Norberto juggled his studies in engineering with efforts to regain footing, and after graduating in 1943 founded his own company "Norberto Odebrecht" the following year. In his own words, looking back on the early days: "Those men decided to respect the newly graduated engineer and, in the course of time, came to trust him" (ibid).

<sup>3 &</sup>quot;Do you know why we didn' t participate in the construction of Brasilia?" he asked his subordinates, already with the answer on the tip of his tongue: "I didn't have political support (Gaspar, 2020, p. 42).

Son of a German immigrant, Norberto moved from Southern Brazil to the Northeast as a child. In Salvador, at the age of 12, he stopped being educated by a Lutheran pastor and first started to learn Portuguese in a school. Tuned with the protestant ethos, Norberto was perplexed with the manners of the local elite, differentiating himself by saying "I came to the world to serve and not to be served" (Moura, 2014). So that no doubts remained, the boss made clear to his subordinates that building a profitable relationship with decision-makers was a resource-intensive task. At the Northeast macro region, the firm's expansion was facilitated by the Northeast Development Superintendence (Sudene) in the 1960s. Along with the support of fortuitous macroeconomic conditions, the ties with the oil giant enabled the *Baianos* to finally move towards the South. In what history would later see as a longer spurt, Odebrecht was first associated with Petrobras by building a pipeline in the year of its foundation, in 1953. Besides doing Petrobras' main office and the capital city's Galeao International Airport and Federal University, Odebrecht won in 1972 a competition to build the Angra I Thermonuclear Plant in the same Rio de Janeiro state. The regional differences were not negligible, nonetheless, and affected the organizational culture a great deal: "When we went to Angola, the shock was much smaller than when we went to the South", said Norberto in retrospect (Lucena, 2014).

Ernesto Geisel, Petrobras president (1969–1973) and later president of the Republic (1974–1979), was a crucial partner at the time. During the Geisel administration, a breakthrough in prospecting resulted in absolute success for oil exploration in the Brazilian territory and placed Petrobras at the top of the rank in deep waters oil production technology worldwide. This promoted a boost at national key industries ahead of large petrochemical and refining enterprises, and the construction and assembling of offshore platforms. To translate it with the aid of numbers, the real growth of the group was 212% between 1973 and 1977 (Fantine & Alves, 2008).

Not only Odebrecht prospered from the economic scenario in the "miraculous" years that followed with the country steadily growing 10% every year. In 1985, an entrepreneur reported to the Senate's Commission of Inquiry on stateowned companies that Geisel, the head of state, had blatantly disregarded the law so as to favor a number of businessmen from the petrochemical sector, among whom Norberto Odebrecht. The matter surrounded Petroquisa, a project put forward during the dictatorship to develop the petrochemical sector by encouraging Brazilian entrepreneurs to enter the business, having the government's partnership as an incentive. "The oil is ours" had been the slogan for Petrobras' foundation in 1953, a symbol of Brazilian sovereignty. In a country where the industry's autonomy was only relative, given the share of foreign capital, exploring sovereign oil was a statement against the drain on the income created in Brazil towards the economies of developed countries. Ever since the days of Getúlio Vargas, "father of the industry", matters of national interest continued to play an important role when business conglomerates decided the fate of the domestic economy.<sup>4</sup>

Ernesto Geisel's decisions also paved the way for Odebrecht's internationalization. His administration was the first worldwide, in 1975, to establish diplomatic relations with Angola, where the company would take charge of all civil construction works involved in the Capanda hydroelectric plant project, soon to become the largest Brazilian investment in the African continent (Cardoso, 2015a, b). Due to the size of the project—the company's services involved constructing the dam itself, access roads, industrial plants, maintenance facilities, office buildings and residential compounds—unless the Brazilian state had given its blessing, Odebrecht's budget would be earmarked. The solution envisioned was a bold one: Petrobras took on the responsibility of paying upfront for the goods and services provided by Odebrecht in exchange for crude oil shipments from Angola.<sup>5</sup> Roberto Dias, the company's director of institutional relations at the time, was convinced that Angola was a promising investment and engaged in the bilateral negotiations on behalf of the proposal to trade works for oil. Since then, Odebrecht's international strategy, aligned with the domestic one, relied on "real connections with politicians and politics" (Calixto & Gama, 2016). Thanks to the firm's relationship with home country institutions, it was already becoming "a transnational fed by the State" (Vigna, 2013).

After departing from the presidency, the relationship between Ernesto Geisel and the businessmen he had worked with would endure, especially the ones from the petrochemical sector, where he coupled technical and practical credentials. In the 1980s, Geisel's appointment to the boards of Northeast Chemistry (Norquisa), and the Northeast Petrochemical Company, (Copene) was a proposal from several businessmen, among them Norberto Odebrecht. As Norberto, Geisel was born to a family of German origin and educated in the Lutheran tradition. According to historical sources, after meeting in 1969, the two men maintained periodic

<sup>4</sup> Secret documents from 2012 given to Glenn Greenwald by former NSA contractor Edward Snowden reportedly show that Petrobras was at the top of a list of targets for intelligence gathering, confirming foreign interests in the oil company.

<sup>5</sup> Gaspar (2020) explains the transaction in the following terms "Each stage of the work completed would correspond to a shipment of oil to Brazil. As soon as the commodity arrived in the country, Petrobras would deposit the corresponding amount with Banco do Brasil, which in turn would pass it on to Odebrecht. In exchange for the \$400 million in construction work, the Angolan government would provide over five years something like \$617 million in oil at the time".

(Maranhão, 1980). Norberto Odebrecht used to say: "I get in the mud with the pigs, but come out clean at the other end in a white suit" (Gaspar, 2020, p. 48), one among many of his lessons to the generations to come, all gathered in the *Survival, Growth and Perpetuity* series with Odebrecht's Entrepreneurial Technology. Before the family empire was hit by Car Wash, Norberto passed away. In 2014, he would be remembered saying: "We are all products of the choices we make and the opportunities that life gives us. We have to turn the crisis into an opportunity. It takes more courage than analysis, because if you just keep analyzing, time passes" (source). The management blueprint of courageous leaders accompanied the second generation ahead of the family business.

#### **5 The Boomers' Generation: Keeping the Doors Open**

When his father stepped off in the early 1990s and Emilio took over the stewardship of the family business, a lot had changed. Some continuities, though, remained, because horses should not be changed in midstream (*time que tá ganhando não se mexe*, the Portuguese popular verb with soccer). Excellency "in Brazil's rough-and-tumble regional patronage game" (Millman, 1993) was one of the informal rules whose validity was retained. Shortly after Norberto had been in a nine-page story at Playboy Brazil entitled "Blood, gold and mud" (Barros, 1992), Forbes magazine published a profile piece on his son, Emilio Odebrecht. Because of the critics to the company's unorthodox market practices, Norberto contacted the Brazilian outlet's director with a protest: "Do you think I like having to pay thugs to release what the governments owe me?" Emilio, for his turn, was quoted at Forbes saying "A man has to look far into the future to withstand the present. You have to be an optimist" (Millman, 1993). As a businessman, he was praised for "A principled way of life and business" (Javetski et al., 2014), even if his habits were less ascetic than that of his father and son. Expansive and charismatic, Emilio enjoyed food, wine, women, and art (Gaspar, 2020). Born in 1945, Emilio experienced the Tropicalist movement and the birth of Bossa Nova. Idealist, like other Baby Boomers, Emilio was engaged with problems of national concern, publishing in several newspapers his ideas "to build a more citizen, egalitarian, and productive country in which future generations can have a better life" (Odebrecht, 2008).

For Emilio, Odebrecht's second generation in office, the interests of the company and those of the country overlapped. In Emilio's vision, among the many duties of the entrepreneurs there is a civic one. To "give their contributions on how to solve shortcomings, difficulties", regardless of whether inside or outside the organization's direct goals. Asked more recently about what he meant saying that the executives had to offer solutions to the country's problems, Emilio Odebrecht replied the Federal Prosecution stating:

The educational process, how to solve the educational problem in the country? This has always been a concern of ours in this sense. We were responsible for contributing to the Congress in several, let's say... What do you call it... Public policies, for example, the concession law, we were an organization, when it came up, that worked abroad, in the United States, and had the most information capable of making contributions in this sense, this is what I wanted to say, these are the types of contributions (Emilio Odebrecht, Criminal proceeding 5054932-88.2016.404.7000).

His nationalist vision was also present in regards to the petrochemical sector, in which Odebrecht managed to secure a massive share. The organization spent the 1980s buying stakes in petrochemical companies owned by foreigners who had withdrawn from the country, and the 1990s taking advantage of the privatization rounds, controlled 47% of Brazil's petrochemical production in the early 2000s, Odebrecht Braskem S.A. was founded in 2002, gathering all the petrochemical assets from the group into one publicly traded company. Listed in São Paulo, New York and Madrid, its equity was held principally by Odebrecht (38.3%), Petroquisa (Petrobras Quimica SA 35.8%), and BNDES 5 (5.5%), with an open float on the market of about 20.4%. Odebrecht's voting rights surpassed 50%, making it the controlling shareholder of Braskem (Glemser & Lelux, 2011).

When a debate surrounding Petrobras' monopoly changing to a model of open concessions started, Emilio was cited by Fernando Henrique Cardoso (1995– 2003), then president:

He [Emilio] works in petrochemicals and has a certain vision of Brazil, he is not simply a money maker. He also thinks that if Petrobras doesn't enter the competition, nobody will hold it [i.e., it is necessary to break the monopoly in order to have competition and thus give parameters to the company] (Cardoso, 2015a, b, p. 134).

Not only Emilio and former president FHC conceived the interaction between businessmen and statesmen as such. The justification of the proximity between the economic and the political elite grounded on instrumental reasons, to make Brazil more productive, was reciprocated by elected authorities. Before the audience, Luiz Inácio Lula da Silva (2003–2011) confirmed the understanding that corporate responsibility encompassed social interest. Quoting Lula, at the ceremony launching the Paulinea Petrochemical complex in 2007: "The laying of this cornerstone for a new polypropylene plant goes far beyond the symbolic act that marks the beginning of a project. Here, the public and private sectors have joined hands to meet the real demands of Brazilian society" (Da Silva, 2007). The proximity among the two leaders, the political and the entrepreneurial, was no secret, but Car Wash investigations would reveal the stability of the relationship, and attribute a negative valence to the depth of governmental alignment between them. In a speech at Odebrecht's 60th anniversary, Lula called Emilio a *companheiro* and friend. At a famous statement in the Petrobras investigation Emilio admitted, with a sense of pride "I had the same relationship with all the presidents that preceded Lula" (Emilio Odebrecht, Criminal proceeding 5021365-32.2017.4.04.7000).

Emilio led the company through the emerging-markets crisis in the early 2000s and laid the groundwork for succession. Pedro Novis, a nonfamily member—but long-standing member of Odebrecht and close friend of Emilio—was CEO from 2002 to 2008 to help ease the transition to the next generation. Before lawenforcement, Mr. Novis adopted a defensive stance: "What was always clear is that these expenses (associated with political relations) were incurred and necessary. And as far as I remember and was aware, in much smaller volumes (in my days) than they became over time" (Pedro Novis, Criminal proceeding 5054932-88.2016.404.7000. Resorting to a strategic argument, Novis grounds the separation between an initial phase, in which he took part, and another, that followed, according to the necessity and scope of "unaccounted payments". Novis partook of Emilio's view that infrastructure in Brazil was "so backward, outdated" that creating the conditions for the company to survive was simultaneously "in the interest of the country" and an enabler for Odebrecht's development. "I tried to explore in these dialogues, the opportunity to make the government aware of what was needed" (ibid. Consequently, the dialogue between businessmen and statesmen was not only necessary, but it was wished and also recommended as a macro-political orientation.

The company's expansion beyond national markets is another narrative of the second generation. In conformity with the group's economic liberalization, such strategy would stand (on the verge of) disconformity with following formal rules. Apart from Angola, Odebrecht secured many projects abroad in the 1970s, mostly in countries that were aligned with Brazil's geopolitical base of influence—in the Latin American and lusophone African regions. In 1979, with the construction of the Charcani V hydroelectric plant in Peru, the internationalization process continued. Later, the detour works of the Maule River and the construction of the Colbún-Machicura Hydroelectric Power Plant, both in Chile, were carried out. Soon after, Odebrecht began its operations in Argentina with the construction of the Pichi-Picún-Leufú Hydroelectric Power Plant in Patagonia and the Santa Elena Irrigation System in the Guayaquil region of Ecuador. In a country like Brazil, where many preferred to safeguard their advantageous positions in the domestic market over improving market access opportunities abroad, Odebrecht steering to become internationally competitive at the engineering market within Latin America was quite an achievement. In 1991, Odebrecht became the first Brazilian company to undertake a public work project in the United States, winning a public bid for the expansion of the Metromover, a surface subway serving the central area of Miami, Florida (Melo et al., 2016).

From the standpoint of Pedro Novis, the internationalization of the organization "started to generate certain opportunities" at the same time it subjected the organization to constraints:

At first, (the internationalization of the organization) aimed at maximizing results, through tax planning, through the creation of companies where the countries in which we operated had more flexible tax legislation and allowed us to send results abroad, and this process was, on the one hand, expanding to the extent that the organization expanded its foreign operations. (...) And, on the other hand, the requirements, the requests, the demands, I would even say, for resources of this type, extra-accounting, *Caixa dois*, were also expanding, for reasons that are not up for evaluation here... politics, the system here and in other countries (Pedro Novis, Criminal proceeding 5054932-88.2016.404.7000).

Interestingly, framing internationalization as the culprit for more informality is a narrative that suits the idea of the paradoxes of modern capitalism. Such claim that unaccounted expenses went from being bounded and "necessary" to ample as a result of the activities performed by the business does not stand on its own, nonetheless, given that investments in foreign markets were no novelty for Odebrecht's second generation. Pedro Novis and Emilio Odebrecht's defense strategy, portraying the illegal practices in a general frame contingent on the economic needs, were more successful in securing a lenient treatment at the negotiations with law-enforcers than the claims of the generation than replaced them.

#### **6 Generation X: Minding the Gap**

When Marcelo Odebrecht took over the leadership of the family business in 2009 the company he encountered had fulfilled the strategic goals set by Emilio and was a fast-growing business. His forefathers' recipe book had been tested repeatedly over the 66 years, and even if "they had set a very high bar for him", Marcelo "also knew he could rely on the entrepreneur-partners he inherited" (Glemser & Lelux, 2011). More than any other regular Odebrecht employee, his grandfather and father's corporate attitudes when interacting with the state impacted his own way of acquiring, processing, retaining, and using knowledge contingencies and information. A few remarkable differences, though, existed. Marcelo's business role model was not Emilio, but the grandfather, Norberto Odebrecht, with whom he shared affinities and similarities. In the eyes of the governments of Lula and Dilma Rousseff (2011–2016), Odebrecht's new CEO had a reputation of being a troublemaker and far more inflexible than his father. He liked to think that because he was more consistent and responsible than his predecessor, the survival and perpetuity of the group were his responsibility (Gaspar, 2020).

While for Baby Boomers leading is the same as commanding and controlling, given their respect for hierarchies, generation X tends to value competencies when taking on leadership (Yu & Miller, 2005), something that matches with Marcelo's trajectory. In furtherance of the family's Lutheran upbringing and "onthe-job" education, he combined practical6 with technical experience to acquire the necessary skills for becoming the steward of the family business. After an engineering degree, an international MBA and passages at Odebrecht plants in Brazil and at a British subsidiary, Marcelo took the helm of the company aged 42. More skeptical and self-centered than the father, he was far less sociable (Gaspar, 2020). Even if their strategies were not always identical, the company continued its route to prosperity. Odebrecht was celebrated as the World's Best Family-Owned Business in 2010 by Switzerland's Institute for Management Development (IMD). The company's commitment to social responsibility led Braskem to build a manufacturing plant in Southern Brazil to develop "green plastic" the year

<sup>6</sup> When Marcelo arrived (in Corumbá, a Brazilian big city from the state Mato Grosso) he made sure never to be addressed as Emílio Odebrecht's son, but rather as Marcelo Bahia. He did not even allow the use of his last name Odebrecht, not even when he opened his bank account he let Odebrecht be used, it was Marcelo Bahia, not even his email appeared Odebrecht, he wouldn't admit that in any moment, in any document that appeared any relationship between him and the family because he didn't want any privilege, he didn't want people to treat him differently because he had this last name (Roberto Paraíso, (Criminal proceeding 5036528-23.2015.404.7000).

Marcelo stepped in as CEO (Calixto & Gama, 2016). When honored as 2012's Engineer of the Year, he paid homage to Odebrecht's almost 200,000 workers, spread in more than 500 locations, on five continents. By 2014, Odebrecht was considered the most internationalized Brazilian company (Fundação Dom Cabral, 2014).

Among the 27 countries the company operated in 2016 was the U.S, where, thanks to the FCPA (Foreign Corrupt Practices Act), non-American multinational enterprises like Odebrecht could be prosecuted. Facing the Car Wash probe and substantial evidence of the firm's involvement in wrongdoing, the Brazilian family-owned business waived indictment and pled guilty to a conspiracy charge in violation of anti-bribery provisions. By then, Marcelo was in prison in Paraná, Southern Brazil, and had been dismissed from his duties as CEO. Pursuant to the agreement entered with the U.S Department of Justice, between approximately 2001 and 2016, Odebrecht corruptly made millions of dollars in payments to, and for the benefit of, foreign officials, foreign political parties, foreign political party officials and foreign political candidates to secure an improper advantage and to influence them in order to obtain and retain business in various countries around the world. The bribery was concealed through a "secret financial structure" known as the Division of Structured Operations ("DOE", in the Portuguese acronym) that was used by the executives and officers at the highest levels of the company taking part in the bribery. Thanks to a subversion of formal rules, Odebrecht and its co-conspirators received approximately \$3.336 billion in ill-gotten benefits.7

Rule-subversion by means of channeling money into politics was one of the preferred ways of companies exposed in the *Petrolão*. When law-enforcers asked if he would have donated funds to electoral campaigns without expecting to get the tax exemption measures for Braskem, Marcelo Odebrecht admitted: "I probably wouldn't have donated 100 million reais if the advantage [REIQ8 ] hadn't come out" (Criminal proceeding 5021365-32.2017.4.04.7000). This example casts light on two important characteristics the private-public framework before us. The first is reciprocity, and an expectation that holders of public office, the other side of the engagement, will act in conformity. The second is its rational-theory backup from the standpoint of the briber.

<sup>7</sup> Plea Agreement, United States v. Odebrecht S.A., Cr. No. 16-643, Statement of Facts (E.D.N.Y., filed Dec. 21, 2016).

<sup>8</sup> The Special Regime of the Chemical Industry, or Regime Especial da Indústria Química in the Portuguese acronym.

In accounting terms, contributing was an optimal strategy for the firm. As one of the attorneys asks Marcelo Odebrecht: "If this tax debt installment plan was not approved, would Braskem's situation be a pre-insolvency situation?" and he responds "It could have gone bankrupt, it probably was going to go bankrupt. The liabilities could have reached I think 4 billion." (Marcelo Odebrecht's testimony in Criminal proceeding 5054932-88.2016.4.04.7000, p.55) The concrete advantages are made clear in this example of the suitability between the legislative measures and the corporation's needs. Also, "the money came out of the petrochemical industry's coffers, the biggest beneficiary of the president's tax exemption package" (ibid). We may predict the same economic-driven orientation found at the top echelons of the hierarchy to be present across the organization, regardless of its particular contours. The richness in details provided by Marcelo Odebrecht when explaining the workings behind the transaction shed light on an important difference with his predecessors. As far as our empiric material goes, no evidence has been found of a counterpart matching the "dialogues", or requests made by Norberto or Emilio's team towards political acquaintances to get things done as the company expected to. We will return to this point later, using the narratives of other executives before law enforcers after the game is sketched in the section that follows.

Once illegal activities were uncovered, the initial efforts from the company were somewhat similar to other episodes in which accusations of corruption were made. Norberto had spoken at a hearing before the Congress in 1979 to defend himself against accusations of embezzlement, overpricing and favoritism in the construction of the Angra nuclear complex and countered other accusations like the one regarding the North-South Railway in 1987. Emilio had outmaneuvered investigations that destabilized Brazil in the outset of the return of democracy, when Fernando Collor was ousted from presidency. The president's brother, Pedro Collor, expressly denounced undercover negotiations on behalf of Odebrecht for the Bank of Brazil to release a loan of 82.5 million dollars for the Trasvase Santa Elena in Ecuador. Those recollections anchored the belief held by Odebrecht father and son they would manage to prevent the family's tradition of excellence from being tainted when the money-laundering probe evolved to uncover a major scam involving Petrobras. Early on, when search warrants targeted the company offices, a public note dismissed the accusations saying "the company and its executives, since the beginning of Car Wash, have always been at the disposal of the authorities to collaborate with the investigations".9 The investigation had

<sup>9</sup> Available at: https://www.infomoney.com.br/mercados/em-nota-odebrecht-afirma-que-pri soes-sao-desnecessarias/

already put other competitors from the construction sector behind bars, and the episodes that unraveled replaced Odebrecht's tradition to weather scandals with a sense of surviving the storm it was at the center of.

"The need to tackle corruption is real", wrote Emilio in newspaper *Folha de Sao Paulo* "Just like it is necessary that society, the media and leaders act so that Brazil stops tolerating incompetence, irresponsibility and a lack of preparation in public administration". His op-ed followed the idealistic tone he became famous for: "It is essential that the energy of the nation be channeled into the debate of what we need to do to change the country".10 Within a couple of days, the company's CEO Marcelo Odebrecht was arrested at the aforementioned *Erga Omnes* operation. By late 2016, having already entered into negotiations, the conglomerate recognized having "participated in improper practices in its business activities" (Odebrecht, 2016). Convinced that the horse race was way past midstream and a change was in order, because things took a turn for the worse, "Sorry, Odebrecht erred" was the title of the note issued. It stated:

It doesn't matter if we gave in to external pressures. Nor does it matter if there are vices that need to be fought or corrected in the relationship between private companies and the public sector. What matters most is that we recognized our involvement, we colluded with such practices, and we did not fight them as we should have. It was a big mistake, a violation of our own principles, an aggression to consecrated values of honesty and ethics. We will not allow it to happen again. Therefore, Odebrecht apologizes, including for not having taken this initiative sooner (Odebrecht, 2016).

When a negotiation round was entered with the Brazilian authorities, 147 employees of Odebrecht provided information on illicit activities, of which 77 formally signed agreements with the Public Prosecution as whistleblowers in exchange for a more lenient treatment.

Because the acts were widespread and the misconduct entrenched at the organizational level, Odebrecht centralized its defense strategy before U.S, Brazilian and Swiss authorities. In Brazil alone, the 77 agreements costed the company roughly 260 million USD (Valenti, 2020). Odebrecht was subject to penalties in the various jurisdictions totaling approximately USD 2.6 billion.11

<sup>10</sup> Odebrecht, Emilio (2015). "I Decided to Break the Silence and Talk About Corruption in Brazil". Folha de São Paulo. Available at: https://www1.folha.uol.com.br/internacional/ en/opinion/2015/05/1624358-opinion-i-decided-to-break-the-silence-and-talk-about-corrup tion-in-brazil.shtml

<sup>11</sup> Reduced in 2017 from a larger penalty level of approximately \$4.5 billion based on an inability to pay analysis at Odebrecht's request.

The above biographical descriptions retelling the historically salient episodes of Odebrecht's rise to the top of the Brazilian market can be divided into four clusters: i) before the company had access to the political arena, in the "prehistory" of the organization ii) when the access was first established (Norberto: 1st generation), iii) when the access was secured and leveraged (Emilio: 2 cd generation) and iv) as the access started to be challenged and became exposed by law-enforcement (Marcelo: 3rd generation). From the standpoint of an Odebrecht looking back on successful endeavors prior to Car Wash, informal relations with a high level of trust had increased the likelihood of expected outcomes and decreased the uncertainty risks posed by excessively bureaucratic regulations. The memory of being left off from strategic contracts like Brasília in the 1960s was fresh, and the antidote was clear at the business blueprint: sustained political support.

#### **7 A Private–public Configuration**

This section will approach the narratives of Odebrecht's executives after exposure by investigative authorities. The focus is on the enterprise managers, those "entrepreneurs" or "leaders" in charge of the execution of an "action plan", because of their vantage point to speak about private–public interaction. According to the Odebrecht Entrepreneurial Technology (OET), the processes led by the entrepreneurs are instituted with a focus on the clients' needs. "Dreaming the Client's Dream" was the firm's *Vision 2020* set a decade before. Qualitative studies with company employees suggest "there is no vision of the organization-client interaction, but rather a need to develop a leader-client interaction, a perspective which personalizes the organization in relation to the customers" (Lamb et al., 2017). Nonetheless, this managerial setup does not exclude the centrality of the meso level, as there is broad evidence of a collective, or *odebrechtian* social capital, often activated by the entrepreneurs when conducting business and applying their action plan.

By "client", Norberto actually meant the governor, a personal entity sitting in the position to deliberate, not the state, an impersonal one (Gaspar, 2020). Instead of bureaucratizing, or overly restricting the relations among the almost 200 thousand members of Odebrecht, infusing values like "Client needs are at the top of our hierarchy" was a more effective way for solving coordination problems. Besides, the existing system of autonomy for CEOs to develop action plans "imposes responsibilities on every component of the organization", enables looking at executives and public officers to ascertain how "client-company" relations were construed. We now focus on these two sets of actors, representative of the private and business sectors, foreseeing how they would choose in a setting involving a public contract. The proposed model is following (Ostrom & Ostrom, 2004).

To describe the structure of the action situation, interaction between a company member who wishes to secure a contract and a public official, we are considering:


In a first stage the game is sketched, and in a second one the potential outcomes are understood in light of the organizational setting, i.e. the fact that Odebrecht executives are facing repetitive social and economic dilemmas when interacting with public officers while trying to secure a project. Instead of using a game-theoretic model with rigor, we are considering that in interactive settings individual decisions are influenced by the decision of other agents in ways that are not entirely reducible to price mechanisms. For example, that the leaders themselves considered the Entrepreneurial Technology as "intangible values" which the representatives of the family should secure is an indicative of this.

The reason for focusing on Odebrecht playing first at this sequential game is twofold. First, looking at company members as the first movers offers a theoretical advantage. In principle, learning in organizations is a collective process whereby the cumulated volume of knowledge stored over time reaches the individual, who refines and exploits the beliefs he holds and occasionally sets new ones (Simon, 1991). Thanks to organizational routines, corporate actors can dodge coordination problems that state cadres in administrative offices often struggle with. Second, it offers an analytical advantage that looking at holders of public office would not. In the state bureaucracy, a multitude of organizations and sectors were involved in the uncovered activities—ranging from the Legislative, like the House of Representatives' Commission of Mines and Energy, to the Executive, with financial bodies like the Brazilian Development Bank (BNDES) and technical ones like Petrobras' refining and supply department, only at the federal level, not to mention state interactions. Should the principal-agent model be applied and self-interest be at the center of a traditional frame "citizen vs. bureaucrat", the choices could be attributed to externalities outside the interactive setting.12 To bypass that, the Odebrecht executive has the chance of moving first in the game below.

In the Odebrecht bribery game, the executive (Player 1) needs to complete a task of securing a contract using the minimum amount of resources possible. At his disposal, besides the parallel budget of the company that allows for "Indirect General Expenses" (*Despesas Gerais Indiretas*, DGI in Portuguese), he accesses the stock of knowledge of the organization, in parts established at the company's philosophy, described in the Entrepreneurial Technology, expectations assigned to fulfil institutionalized roles, information channels and like apparatus. He can either not offer any bribes or start upfront by making a payment. Should he choose the latter, it is irrelevant whether or not the officer will accept, even if our frame is of non-benevolent players, and the only relevant piece of information is whether the bribe has a counterpart from that officer or not, such as a particular request that this authority sees a license to be issued more expeditiously or a given public bid to be publicized prior to the initiation of the contest.13 In other words, the public officer (Player 2) only has a chance to play at the second stage if the bribe was not offered. The game ends if the executive decides right from the start to provide "help", "support", "back-up" (more likely for political/electoral transactions) or "toll", "prize", "kickback" (for administrative transactions). After the executive's offer, the public official (Player 2) may sequentially acquiesce and refrain from moving, or, instead, demand a bribe in order to let the executive (P1) have his prize.

The executive may subsequently accept or reject the bribe demand (3rd stage of Subgame 1 and 2 cd stage of Subgame 2). This third stage of Subgame 1, when the executive moves again, is architecturally similar to that of his first move, but strategically different—since when Player 1 chooses again he knows his choice of trying to get the contract without bribing was not successful at first, as a request from Player 2 followed. We expect the probability of the executive not to pay (s2: s2') to be fairly small under this circumstance. In a slightly modified representation of another sketch following the demand-side approach, Subgame 2,

<sup>12</sup> By traditional we are referring to the bribery game where a "Citizen" performs a task and earns a prize if successful and a "Public Official" may demand a bribe in order to let the Citizen have her prize—the Citizen may subsequently accept or reject this bribe demand.

<sup>13</sup> This is also relevant for the Judiciary given that the penalties increase when the counterpart is matched by the bribe-receiver, a criminal conduct also called "privileged passive corruption" in § 2º, of Article 317 of the Brazilian Penal Code.

the executive (P1) has not already tried and failed at securing the contract without offering a bribe, but the public officer (P2) moves first and requests a bribe. Faced with the officer's request, the Odebrecht executive decides whether to pay or not assessing the risk of saying no to the "toll" or "kickback" demanded. He is more likely to pursue the bribing strategy because a) formal incentive mechanisms in place are supportive to do cost allocation of present and future bribes; and b) informal incentives recommend the executive to keep a *trust relationship* and to secure *networks* (Fig. 1).

The unlikelihood of no bribery in both Subgame 1 and 2 is supported by the speech of the executives: "No, no, it was the following, it was systemic, we knew that by signing the contract there would be an improper advantage for Engineering and an improper advantage for the Supply area" (Rogério Araújo, Criminal proceeding 5015608-57.2017.404.7000). Before law-enforcement, another colleague also clarified the configuration of the private-public intercourse with the state-owned oil company: "Payment of bribes at Petrobras, either at a certain

**Fig. 1** Strategic action game between Odebrecht and public officer. *Source:* by the author

moment, provoked or not, it always happened, as a rule it always happened. In this case, the counterpart was that once the bidding was cancelled, it didn' t go into rebidding" (Marcio Faria, Criminal proceeding 5015608-57.2017.404.7000). The second part of the executives' explanation touches on the output of the bribe interaction: whether the expected behavior was matched with a specific conduct by the holder of public office or rather configured a long-term investment on a quality relationship with State authorities (often through electoral donations). In that particular criminal procedure, the executive is explaining kickbacks paid as a result of a contract that Odebrecht gained together with a consortium comprised of other construction companies, namely UTC and Mendes Júnior, organized as a cartel to secure that particular work. Both Mr. Araújo and Mr. Faria confirm a payment of 15 million reais14 to the Supply and Services departments in exchange for the recipients to refrain from launching an invitation for rebids on that contract. The bid had been launched by Petrobras under normal (competitive) conditions, but the lowest price was not in line with the company's price fluctuation, and Petrobras cancelled the bid. Then the cartel agreed on the price and proceeded to negotiate directly with Petrobras, whose executive directors did not restart the bidding competition having reached an agreement upon an unofficial payment, which expresses a bribery situation initiated by the firm, or the supply side (s1: S1´).

A different configuration explains a bribery situation that took place after the outset of Car Wash, when, according to an Odebrecht executive, the company was being pressured to pay 17 million reais as "toll" in exchange for a loan from *Banco do Brasil* to the firm's Agroindustrial branch. Initially the demand was met with a denial, but Aldemir Bendine, the former bureaucrat ahead of the public bank in question, took over Petrobras' presidency. Instead of risking to spoil the relations with its most important client, Odebrecht settled with paying a smaller amount: 3 million reais:

The idea was that it should be a value that was not so small that it would look like something to be unhappy about and create more anger or risk of some retaliation, or create a show of power, but also that it was not anything that would give the impression that there would be a payment of the 17. (…) The 3 million was actually, in my understanding that Marcelo authorized me to do, was actually to pay a debt that they (public officers involved) understood, they thought they were creditors of a debt that

<sup>14</sup> In a direct quotation of Márcio Faria: "A 15 million payment was negotiated, the contract was worth 1.8 billion, that is, it was less than 1% for each of the 3 events, in that order. A payment of 15 million for the supply directorate, 15 million for the service directorate, and 15 million for 1 agent who worked there in the conduction of the contract (at Petrobras)" (Criminal proceeding 5015608-57.2017.4.04.7000).

they had with the president of Petrobras, that is, in fact the intention was that we would not have a pendency of a debt that they thought they were creditors of (Fernando Reis, Criminal proceeding 5035263-15.2017.404.7000).

The bribery acts of this particular agreement are expressed at Subgame 2, with the public officer from the financial institution requesting a payment upfront, the Odebrecht executive responding negatively, but later changing its original decision to end up paying. Aside from the low probability of no bribery, even when the company did deny a request for kickback "the no was always an evasive 'no', in the sense of deflating the request, so as not to create enmity with the public authorities" (ibid). This evidence that the organizational institutionalizes knowledge prescribed a particular way to avoid negative externalities that could potentially damage the company's reputation confirms how operative the informal rules were. So does the delegated autonomy with which Mr. Reis considers himself to be invested when anchoring his ultimate decision to pay an amount in terms of the CEO's implicit authorization to settle the dispute that way.

Departing from the game-theoretical approach, we will now categorize the results in reference to the organizational knowledge stocks related to the relationships. Using an interpretative approach, the underlying patterns behind the content of the speech of Odebrecht executives will be discussed in reference to a multilevelled field connecting individuals, organizations and state. To introduce the next step, a quote from Emilio Odebrecht synthesizes the complex surrounding entrepreneurial leaders when selecting a preferred course of action:

The executives were not oriented to take (orientations) to anyone, they were oriented to build their dialog agendas with the authorities taking contributions of what was important for the country and not selfishly taking only their own interests (Emílio Odebrecht, Criminal proceeding 5054932-88.2016.404.7000).

While specific and immediate demands from the executives would fall under "selfish", for their pertinence with the concrete action plans, broader and mediumterm investments are closer to the "altruistic" contributions. For Emilio, born an *Odebrechtian* leader, the latter should trump the first.

Recognizing that the accounts of how, and why, executives decide as they do are contingent on the context, three temporalities have been identified in the business-politics narratives: (1) the first addresses the past, (2) the second the present, and the (3) third the future (See Appendix I). They express different strategies, put in place by different speakers, but simultaneously available for the organization. In other words, since Odebrecht, together with other construction companies, had anti-competitive activities regarding Petrobras in a cartel and engaged in several electoral donations throughout the democratic period. The case study is one of systematic corruption, therefore Odebrecht could resort to multiple strategies aimed at long, medium, or short-term goals, or accomplishments, at a time. Its members, nonetheless, stood at an inferior dimension in the organizational ladder. Their speech as enterprise manager participating in a typical bribery situation for earning a public should fit in one: either activating a remote source of social capital (managed by the high man on the totem pole), a moderate source (a pecking order that relied on institutional relations he was responsible for), or a specific source (coming from the lowest rung of the ladder, the led, conditional on his blessing as leader).

First, the relationships inherited are retrospective, they refer to the past and its agents are the first and second generation of Odebrecht. It is a state verb, to inherit, or rather a past participial adjective "inherited", for relationships are something received by succession, therefore gained without special exertion of efforts. Apart from Marcelo, another noteworthy example of the legacy principle is Claudio Melo Filho. He replaced his father, the group's director of institutional relations in Brasília, who had to step away due to an illness. Melo Filho was recruited because they shared "the same profile, that of a hard-working young man, an enterprising engineer, innovative, creative" and also since he was "a man of the Odebrecht Group's absolute confidence precisely because of his ability to work within the line Norberto Odebrecht drew up of full dedication to the Group's business goals" (Paes Landim, 2012, p. ?). Years later, in his plea bargain, Melo Filho used the word "inherited" to describe many political acquaintances symbolically transferred from his father.

Second, best expressed with "to replenish", in the indicative form of the verb, vocals the business-state relationship is now undergoing maintenance: that executives are making sure the overall quality of the bond between economic and political affairs stays in its optimal state (one of trustworthiness). Among the job attributions of a business leader is the maintenance of personal proximity with players capable of hampering or helping the business under his responsibility. Their performance at building and strengthening informal relations would take precedence over technical matters, especially when the employee did not have an accumulated social capital prior to job entry:

The role of a superintendent director is much more strategic politics, than ... everything that concerns the work, we ... even the works, let's say, formal ones, we didn't go into much detail. At that time my concern, I remember well, was that the governor was changing… I had lived 17 years abroad, I had almost no relationships in São Paulo, I needed to speed things up in some way, somehow. To establish networks... Because they were going to change secretaries and governor and I had three very big contracts in São Paulo, with many problems (Carlos Paschoal, Criminal proceeding 5021365-32.2017.404.7000).

Should the ties be "healthy", apart from securing open doors with decisionmakers and gatekeepers, the leader would occasionally be spared from requests for bribes from public officers. This form of exchange often took place through electoral donations or other things of value to an elected official or a political party, but not always. Replenishing the relationship is defined by a loose correspondence between the bribe and the expected benefit, which differentiates it from the next action orientation.

Third, also an action verb, "to channel" expresses an activity carried out in the present intended to generate effects in the future. When the executives either requested the Division of Structured Operations by their own hand or authorized the demands of payments for the benefit of public officials and political party officials coming from subordinates, they did so to secure a specific business advantage, conscious of a correspondence between the beneficiary and the desired act. At Odebrecht's vocabulary, those would be "the transaction costs of a project" and the so-called "rule of the game" that Mr. Faria and Mr. Araújo's aforementioned accounts present as "inevitable" when interacting with Petrobras.

#### **8 Coupling Formal Rules and Informal Rules for Optimal Results**

In regards to other emerging economies, where informal institutions are of special significance, Brazil is characterized by "accommodating" informal rules that get around the effectively enforced but restrictive formal institutions and reconcile varying objectives that are held between actors in both formal and informal settings (Estrin & Preverzer 2011; Eunni & Manolova, 2012). Not long ago, enforced prescriptions about certain prohibited acts were fairly loosely grounded, and the boundaries between public and private were more elusive than the written norms regulating corrupt behaviors wished. The practices uncovered at the *Petrolão* scandal suggest that Odebrecht, and the construction sector, was mimicking compliant behavior to divert regulatory efforts. Even if public tender contests resembled market competition, the cartel-setting confirms it had not embodied the substance of what antitrust and competition law would expect. Inspired by the new public management movement (NPM), this set of market imperatives, carried just as many perils as promises in the Brazilian context. For them to make past ink and paper, norms need more than enactment. The Portuguese expression "*a lei não pega*", alludes to the notion that written rules, though part of statutory law, are not enforced and often end up being overlooked. Gaining social legitimacy is an extralegal factor demanded from enacted legislation that is at odds with customary practices, without which conflict situations between the law in books and law in action frequently arise.

Domestic and public life in Brazil were shaped by a colonial and imperial past when the realms of the state, the market and the society were not fully separated. Like this chapter has tried to do with Odebrecht, contemporary Brazilian sociability is better understood if its constitutive nuclei are retraced historically. In the confines of the Latin language and Roman law, the word *privus* (single, individual) gave rise to two variants *privatus* (private) and *privus-lex* (law for a private individual) or *privilegium* (privilege), which in Brazil often came to mean the same thing.15 Informal institutions have been, before and since formal norms emerged, capable of providing answers to conflict situations within and across the different spheres. Reminiscent of the French expression *plus ça change, plus c*' *est la même chose* adapted to Odebrecht's vocabulary: "Everything changes, on a base that never changes" (Odebrecht, 2015a, b). Thanks to contiguity and access to politics, Latin America's largest construction and engineering conglomerate profited from written and unwritten rules and was actively engaged at agenda building and agenda-setting. Refraining to change horses midstream in what was incontestably a winning race, having the government as its main partner, was the most rational course of action available to company members with a successful recipe like the one developed by Norberto, the forefather of the Odebrecht family business.

Back to where this chapter started, in 1992 when the *(CPI das Empreiteiras)*  "ended up in pizza". Documents found at the house of one of Odebrecht's directors proved that in order to secure projects, the firm manipulated, along with other eight companies, the Brazilian budgetary provisions. The evidence was robust: in the 1992 budget alone, Odebrecht had managed to get sixty-three amendments and twenty subprograms allocated to projects of its interest. To quote Paulo Bisol,

<sup>15</sup> A historical incursion to the days of the colony will remind of the different meanings ascribed to the private order in Brazil, as the following passage from the History of Private Life in Brazil illustrates: "In the context of modern slavery these variants (of *privatus* and *privilegium*) merge into a single meaning: in which the right, the privilege, to own slaves directly affects the concept of private life. As in the Colony, Brazilian private life merged in the Empire with family life. During the process of national political and juridical organization, the slave private life unfolded into a private order full of contradictions with the public order (Alencastro, 1997).

the senator conducting the works of the Commission of Inquiry: "It is not a parallel power. It is superior. The Brazilian state is (an) instrument in its hands". Now, the landscape of control and oversight mechanisms was very much altered in regards to the early 1990s, particularly at the federal level (Praça & Taylor, 2014). Institutional changes enabled Car Wash to happen, a breakthrough in the fight against corruption (Castro & Ansari, 2017), and regulatory control by the Judiciary improved in regards to construction companies. While unenforced prescriptions add another layer to the institutional configuration that makes the Brazilian state more accessible to interest groups, some progress has been made. Framing unaccounted electoral donations (or *caixa dois*, literally, an "unofficial cash till") as part of the "rule of the game" has long worked as a defense strategy before public opinion and courts. *Petrolão* revealed that the practice strengthened the ties between business and governmental elites, particularly the ones responsible to monitor, undertake inquires and conduct investigations on the industrial activities of Petrobras. The Public Prosecution, alongside other institutions in the web of accountability, managed to sustain a claim that many, if not most, electoral donations had business interests behind them.16 Marcelo's own account that he would not have donated one hundred million reais to the campaign if the expected tax exemption to the benefit of Braskem was not granted confirms how structured the gambling actually was, and the compromises donations entailed for the incumbents.

Rather than eventually wearing out, as do most tangible forms of productive capital, well-established, yet informal, relations grow to constitute a form of "durable capital" that becomes more refined and ingrained through repeated use (David, 1994). Its costs are "sunk" for the ones who learned them without investing efforts in building them. There are strong reasons to believe that, throughout its history, Odebrecht's interaction with the state has been a cooperative, instead of a competitive game: where all strategies took precedence over the one in which the executive does not bribe, irrespective of whether he offers or the public officer

<sup>16</sup> That was not only true in regards to the party in government, but also others like Gim Argello (Brazilian Labour Party—PTB), a sitting congressman at the Petrobras Parliamentary Commission of Inquiry. According to lawsuit 5022179-78.2016.4.04.7000, "Jorge Afonso Argello allegedly asked Marcelo Bahia Odebrecht, head of the Odebrecht Group, for five million reais so that the Group's executives would not be summoned to testify. The payment of R\$ 200,000.00 in the form of registered electoral donations to parties nominated by Jorge Afonso Argello was confirmed". The same criminal proceeding mentions other companies like OAS and UTC that also "paid" bribes through official electoral donations, and other construction companies, namely Andrade Gutierrez, Engevix and Camargo Correa, that were summoned, but did not pay.

solicits first. Thanks to the coupling of formal rules that allow a margin of discretion to the public officers conducting public procurement at the demand side, of Petrobras, and informal rules in vigor among the members of Odebrecht with undisputed effectiveness before Car Wash, bribery was a cooperative game in the Brazilian setting. Shared historical experiences and conscious perceptions of a shared past provide shortcuts for groups of people to form a system of consistent mutual expectations when they are not readily able to arrive at a common course of action via direct discussion of the problem before them (Lewis, 1969). Thus, the eventual outcome of a bribery game between executives and bureaucrats is codetermined by early and, quite possibly, adventitious choices from both players and their organizations.

Research on multinational corporations impacted by Car Wash Operation have shown an increase in formal corporate governance for preventing and fighting corruption (Barboza, 2020). The perils of restricting the analysis to official documents and above the surface behavior, however, are the elements that lie underneath, often the bulk of the behavior's mass. Like the iceberg analogy suggests, the role of unwritten rules below the tip of the organizational façade should not be underestimated. To some extent, the interaction between the company and its members and the state and public officials were more open in the past, and became less open (and more informal) as institutional constraints started to be implemented—be it in the legislation or in the kind of market imperatives being pushed at the sector, like compliance measures. Nonetheless, even if the narrative might have changed, the actors continued to be accessing the organizational repertoire and the same knowledge bases that regarded an interlock between the private and the public as beneficial to achieve a shared purpose of development.

Informal organizational arrangements have been, for Odebrecht and the Brazilian State, the key to success of the 70-year-old firm before its CEO, Marcelo Odebrecht, was arrested. Irrespective of reforms at the front-stage, exogenously imposed by law-enforcers trustworthy of compliance and integrity systems, the company's backstage may retain a different configuration (MacLean et al., 2015). The argument used by Marcelo Odebrecht´s successor in the company´s presidency, Newton Souza, after having unsuccessfully tried to request a meeting with Aldemar Bendine, Petrobras' then president, is illustrative. Mr. Souza, trying to regain reputation for the firm downgraded to a BBB minus rating, Standard & Poor's lowest investment grade (Brandimarte, 2015), decided to take an alternative course of action, through informality:

I tried to find out internally, within Odebrecht, who had knowledge, who had relations with President Bendine so we could make him aware of the importance of us telling him our version (in regards to Petrobras' revocation of a contract with Odebrecht Oil and Gas) and see what could be done to prevent these developments from happening. At this point we identified Fernando Reis among the leaders of the organization, who said he had a relationship with him and was willing to call him, this was the information he gave me, to try to set up this meeting. After a few days Fernando returned saying that yes, President Bendine was willing to meet with us (Newton Souza, Criminal proceeding 5035263-15.2017.404.7000).

Activating personal and informal channels is not irrational if it achieves the ends it aims to. In a rule-bound, legalistic system, formal rules should take precedence over substantive matters, but not always do they secure optimal outcomes for the entrepreneur. It is plausible that a symbolic adoption of ethical programs is decoupled from actual implementation, given that decoupling is not a binary choice (i.e., say vs. do) but bears nuanced organizational actions (Fiss & Zajac, 2006).

For Odebrecht at Brazil's modern setting, both historical examples and discourse analysis indicate informal relations tend to often display more "hillclimbing" features as compared to following formal rules. As any other study of a modern economic organization was expected to depict, the Odebrecht group has acted on the spectrum of legitimacy, occasionally crossing the boundaries as an illegal sideline for an otherwise legal enterprise across its history. Contact between members of the economic and political realms is not something that can, or should, be completely avoided, though. Private incursions may contribute, rather than tarnish the common good when made in conformity with regulation. The notion of 'publicness' from the republican tradition, when combined with a skepticism on the effectiveness of formal rules alone, can provide an initial answer to the puzzle.

#### **9 Conclusion**

Rather than advancing traditional approaches of intertwinement between entrepreneurs and politicians, we have argued corruption, and encroachment of private interests in the public sphere, is not a mal adaptive reaction to regulation, but rather an adaptive one because of the competitive advantages it provides to its agents, business or statesmen. Quoting Emilio Odebrecht, with a structural account for the overlapping between the construction and the oil giants: "Without a doubt Petrobras has always been, since my time as an executive and afterwards, a multifaceted company with the organization because at the same time it was a partner, it was a client, it was a supplier" (Emilio Odebrecht, Criminal proceeding 5035263-15.2017.404.7000). Other contributions in the present volume will be dedicated to elements that may enhance our understanding of issues on organizational culture and collective mindsets surrounding corporate-state relations in Brazil.

Despite having tried to sketch configurations for strategic and communicative action, the complex cognitive systems available for individuals render it difficult for the analyst to ascertain what working rules underlie an ongoing situation, like bribery, especially when considering both formal and informal rules. We have not attempted to predict organizational behavior but to provide a more general framework to understand the interaction between business and statesmen in a bifold fashion, where the main divide is across the private and public sphere. For the modelling, having as second player a public officer that may be a technocrat or a politician, mandate holder, comes at a cost. Even if it oversimplifies the relationship that often involves intermediaries (or brokers), other designs have shortcomings for explaining the research question developed here: how Odebrecht entrepreneurs depict a typical situation of interaction with the state, either through its elected representatives or civil servants. As the specialized literature has already established: heuristics like shared histories, organizational routines and corporate language may permit concise articulation, but its consequences in behavior often depend on a much richer background of understanding that is not fully articulable (Cohen et al., 1996).

Drawing lessons from Odebrecht, Brazil's institutional arrangement appears to have been too flexible to cope with the conditions posed by the firm, as well as the government, in pursuance of their developmental goals. Such arrangements, nonetheless, were unable to meet the needs of a set of actors at the justice system, part of the state bureaucracy, for whom anti-corruption grew to become more than salient, but a spearhead against political malfeasance. Just how much continuity and change the regulatory landscape is undergoing, and its impact on opportunity structures available for the economic and governmental elites is a question that remains unanswered. In terms of rule-setting, findings suggest that the classical liberal frame of a radical separation between private and public affairs is doomed to fail, and insights from a republican tradition are welcome. Preventive measures from ready-made templates that worked for other polities that share little with Brazil except the fact that they were, sometime in the far past, as underdeveloped as the stage the country finds itself today are abundant, rhetorical, yet of questionable effectiveness. To advance, understanding the competitive advantage of blurring the boundaries between private and public from the standpoint of a big firm is educational, and merits further investigation across sectors.

Culturalists have argued Brazil has many structures capable of uniting the house (the kin, the private) with the street (the others, the public), like Carnival, "in which everything and nothing is simultaneously represented and, apparently, solved" (Da Matta, 1997, p. 135). It is unquestionable that the long-lasting pattern of morally questionable relations between the business and governmental elites revealed in the Car Wash plot looted public coffers. But its usefulness in securing state-business relations that had historically been guaranteed through interpersonal bonds remains an open field for further explorations. Comparative studies could help unpacking the effect of culture within the Brazilian institutional landscape and across other in emerging economies.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter' s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter' s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Businessmen, Political Financing, and Corruption: Odebrecht in Operation Car Wash**

Paulo Roberto Neves Costa, Silvana Zulmira Ferreira and Marta Pontes de Campos

#### **1 Introduction**

The political behavior of businessmen in Brazil has been a matter of interest since the early days of the Social Sciences in the country. One of the persistent research issues that permeates this literature is political culture, i.e., how businessmen think, evaluate, and position themselves within political processes.1 In general, most studies have focused on the analysis of political practices and values within the framework of institutional legality despite the wariness, often found in the common sense, the media, as well as in the academic environment, as to the obscure and illicit nature of the multi-dimensional relationship between businessmen and politics. In recent years, many of the investigations into corruption and other crimes in Brazil have, on the one hand, attracted the attention of Brazilian and foreign researchers to this matter and, on the other hand, have produced several documents and data that have opened new horizons for studies of the business community and politics, allowing for new perspectives on the behavior and ideas of businessmen as well as a novel point of view for the analysis of democratic institutions in Brazil.

<sup>1</sup> Regarding the relationship between businessmen and politics in Brazil, see the article by Wagner Mancuso (Mancuso, 2007). The issue of political values and attitudes dates back from classical studies on the business community and politics in Brazil, such as Fernando Henrique Cardoso (Cardoso, 1964), Luciano Martins (Martins, 1968), and Renato Boschi (Boschi, 1979) to more recent studies (Costa, 2005, 2012; Costa & Borck, 2019).

P. R. N. Costa (B) · S. Z. Ferreira · M. P. de Campos Federal University of Paraná, Curitiba, Brazil e-mail: paulocostaufpr@gmail.com

M. E. Trombini et al. (eds.), *The Fight against Systemic Corruption*, Organization, Management and Crime – Organisation, Management und Kriminalität, https://doi.org/10.1007/978-3-658-43579-0\_8

As is widely known, Operation Car Wash (OCW) is the most complex and impactful investigation into illegal practices stemming from the relationship between companies and politics in Brazilian history thus far.2 Among those investigated were businessmen linked to important construction companies, who, after controversial legal procedures and criminal charges, admitted their involvement in such practices and produced reports pursuant to the investigation. Initially set up to investigate crimes against the financial system by black-market money dealers (known as *doleiros* in Brazil), OCW evolved to encompass crimes against the public treasury and money laundering, thus revealing a complex scheme involving bribes and illegal financing of politicians from a wide political spectrum, operating at all levels of the Brazilian political system.

The scope and magnitude of issues arising from OCW demands a focused analytical framework. As such, our underlying research question is the *representation*  produced by businessmen linked to Odebrecht and implicated in OCW regarding political financing, more particularly the legal/illegal financing of political campaigns. From this analytical category, we hope to grasp how these businessmen described, analyzed, and evaluated this issue, thus allowing us to verify, *from their point of view,* not only their relationship with authorities,3 politicians, and political parties, but also the inner workings of this fundamental dimension of the democratic political system within a recent period of Brazil's history. Lastly, the goal is to understand how democracy operated from the standpoint of these businessmen who found themselves in a situation of confessed illegality and in the condition of defendants.

Thus, *representation* refers not only to a set of normative and prescriptive ideas, values, or propositions, but also to a relevant dimension of the *definition*  and *justification* behind *concrete actions,* in this case, the defense of business interests through illegal political financing.4 We do not intend to explain the

<sup>2</sup> As opposed to an idealistic point of view about this operation" the procedures adopted by the OCW prosecutors and legal authorities have been widely denounced as questionable and may lead to important repercussions in Brazil's legal and political environment. Suspicions of bias and illegality fell on both the Federal Prosecution Service and former judge Sérgio Moro from the disclosure of private messages by *Intercept Brasil* ( https://theintercept.com/ series/mensagens-lava-jato/) as well as questionings by lawyers (Carvalho, 2020), in addition to criticisms of "judicial activism", "car-washism","Jacobinism" etc., expressed both by the press (https://opiniao.estadao.com.br/noticias/notas-e-informacoes,combate-a-corrup cao-nao-tem-dono,70003384283), and academia (Kerche, F. & Feres Junior, 2018; Lagunes, P. & Svjnar, 2020; Pimenta, 2020; Rodrigues, 2020).

<sup>3</sup> Here we are considering the holders of public office or members of the State administration.

<sup>4</sup> The *representation* analytical category was initially used by Costa and Borck (2019) to verify the way Brazilian industrialists analyzed and evaluated their relationship with the

political system from these documents or from the discourse therein found, but rather to verify what they reveal about an important segment of the Brazilian *economic elite* and their relationship with this crucial dimension of democracy, from their own analysis about their concrete experience, in this case *illegal,* with politics. Hence the importance of this issue for a *Political Sociology* of democracy in Brazil.

We shall then briefly describe Operation Car Wash, how Odebrecht was involved in the investigations, and the plea bargain agreement between the Judiciary and Odebrecht executives, given that within this context we find the documents germane to our research object. Subsequently, we present the analytical categories and address the nature and particularities of these documents, explaining the methodological procedures used for analyzing this empirical material. Finally, we analyze the results of the research and conclude with our final considerations.

#### **2 Odebrecht and Operation Car Wash**

Without a designated name at the time, Operation Car Wash began in 2009 with the investigation of Federal Deputy José Janene, from the city of Londrina, in the Brazilian state of Paraná.5 The police investigation set out to investigate alleged money laundering crimes practiced by the parliamentarian and others involved in the scheme, such as black-market money dealer Alberto Youssef.6 The operation was named "Car Wash" because the individuals under investigation used a network of gas stations and jet washers to, according to the Federal Prosecution

Federal Legislature in the process of defending their interests and falls within the so-called "ideational turn" (Perissinotto & Stumm, 2017). We will return to this category later.

<sup>5</sup> We extracted this information as well as related information for this paper from the page of the Federal Prosecution Service: http://www.mpf.mp.br/grandes-casos/lava-jato/faq, accessed on 10/16/2020.

<sup>6</sup> Banestado was an important state bank in the state of Paraná in operation until 2000, when it was privatized and acquired by Banco Itaú. Regarding the case of Banestado Bank and the participation of black-market money dealer Alberto Youssef, the Federal Prosecution Service informed that: "Alberto Youssef, central character of the Car Wash case, was already investigated, prosecuted and arrested, in 2003 as a result of his activity in the clandestine dollar market, after an investigation into one of the largest criminal schemes that ever existed, the 'CC5 Scheme', also known as 'Banestado Case'. Youssef was one of the top black-market money dealers in Brazil, operating in the wholesale market, where he provided dollars to other black-market operators and some special customers".

Service (MPF in the Portuguese acronym), provide a "semblance of legality" to illegally obtained resources.

In July 2013, the investigation wiretapped Carlos Habib Chapter, a blackmarket money dealer, from which four criminal organizations were identified as well as their leaders, all of them money operators in the parallel exchange market with prior associations to one another: Mr. Chater, Nelma Kodama, Alberto Youssef, and Raul Srour. Each organization was investigated separately and each operation was given a name: Car Wash, Dolce Vita, Bidone, and Casa Blanca, respectively. However, the name "Operation Car Wash" became increasingly popular and was used to refer to any phase of the investigation.

In 2014, the Federal Court of Curitiba, Paraná, launched the first phase of the operation, which implicated black-market money dealers and Paulo Roberto Costa, former Supply Chain Director of Petrobras, who allegedly received a Land Rover Evoque as a donation from Alberto Youssef. According to information from the MPF, on this day "81 search and seizure warrants, 18 pretrial arrest warrants, 10 temporary arrest warrants, and 19 bench warrants were served in 17 cities of 6 states and in the Federal District".

The information gathered from the plea bargain agreements7 established between the MPF and Paulo Roberto Costa, Alberto Youssef, and other individuals under investigation, revealed that parliamentarians, public agents, companies, and businessmen were allegedly involved in a "major corruption scheme". Thus,

<sup>7</sup> According to Law 12.850, which defines Criminal organizations, the plea bargain agreement is a legal benefit granted to the defendant who decides to cooperate with the investigation and criminal prosecution: "Art. 4 The judge may, at the request of the parties, grant judicial pardon, reduce by up to two thirds (2/3) the custodial sentence, or replace it by the restriction the rights of the person who effectively and voluntarily collaborated with the investigation and criminal proceedings, provided one or more of the following outcomes from such collaboration: I—the identification of the other co-authors and participants of the criminal organization and the criminal violations practiced by them; II—the disclosure of the hierarchical structure and division of tasks of the criminal organization; III—the prevention of criminal infractions resulting from the activities of the criminal organization; IV—the total or partial recovery of the product or profit from the criminal infractions practiced by the criminal organization; V—the location of an eventual victim with their physical integrity preserved. § 1 In any case, the granting of the benefit will consider the personality of the collaborator, the nature, circumstances, gravity, and social repercussion of the criminal fact, and the effectiveness of the collaboration. § 2 Considering the relevance of the collaboration rendered, the Prosecution Service and the chief police officer may, at any time, request or file a representation to the judge, in the police investigation records, with the acquiescence of the Prosecution Service, to concede judicial mercy to the collaborator, even if such benefit was not foreseen in the initial proposal, applying, as appropriate, *art. 28 of Decree-Law 3.689, dated October 3, 1941 (Code of Criminal Procedure).* (...)".

on November 14, 2014, search warrants were executed at several construction companies, among them the Odebrecht Group.8 According to the MPF allegations, the Odebrecht Group had established a cartel with other companies in the civil construction sector and defrauded the public bidding system for the construction of major Petrobras units and offices.

We will then provide a brief overview of the plea bargain agreement between the Judiciary and Odebrecht and contextualize the documents comprising our empirical research.9

During the 14th phase of OCW, in May 2015, the Federal Police arrested Marcelo Odebrecht and Otávio Marques de Azevedo—presidents of Odebrecht and Andrade Gutierrez, respectively—as well as other construction company executives. In July of that same year, the executives of these companies were charged with organized criminal activities, national and international money laundering, and active and passive corruption. In the 24th phase of OCW, in March of 2016, the Federal Court convicted Marcelo Odebrecht of active corruption, money laundering, and criminal association. In December of 2016, the MPF established leniency agreements with the Odebrecht, in which they pledged to reveal illicit acts practiced with Petrobras.

In the 37th phase of OCW, in January 2017, the president of the Supreme Federal Court (STF, in the Portuguese acronym), Minister Carmen Lúcia, ratified the 77 statements by Odebrecht executives, recognizing the legal validity of these denunciations. In April 2017, in the 39th phase of OCW, minister Edson Fachin, rapporteur of Operation Car Wash in the STF, authorized the Office of the Prosecutor General (PGR in the Portuguese acronym) to investigate 98 politicians, 8 ministers, 3 governors, 24 senators, and 39 federal deputies. The requests were based on the so-called "Janot list" a reference to the Prosecutor General of

<sup>8</sup> On the abovementioned MPF webpage, we find the following statement: "On November 14, 2014, 85 warrants were served by the Federal Police in conjunction with the Special Department of Federal Revenue, including 4 for pre-trial detention, 13 for temporary imprisonment, 49 for search and seizure proceedings, and 9 for bench warrants in several cities across the country, especially in major renowned construction companies such as Engevix, Mendes Júnior Trading Engenharia, Grupo OAS, Camargo Correa, Galvão Engenharia, UTC Engenharia, IESA Engenharia, Construtora Queiroz Galvão, and Odebrecht Plantas Industriais e Participações."

<sup>9</sup> The newspaper *O Estado de São Paulo* published a helpful infographic with information regarding the Odebrecht executives and activities. See https://infograficos.estadao.com.br/ politica/a-maior-delacao-da-lava-jato/ accessed on 09/18/2020.

the Republic at the time, Rodrigo Janot, compiled by statements of former Odebrecht executives under the plea bargain agreements.10 Subsequently, Minister Fachin publicly disclosed the content of the recordings in the plea bargain agreements of Odebrecht executives, which were then widely publicized in the press and on the internet.

#### **3 Analytical Categories and Methodological Procedures**

On a different occasion, we resorted to some of the analytical categories hereby proposed to analyze the political actions of businessmen.11 Below, we indicate how they were appropriate to the objectives of this chapter. As we have shown above, the notion of *representation* does not refer to political representation, but rather to how agents describe, analyze, and evaluate the situations they *experience*  and upon which they *act* as a member of a group or institution. Our goal is to grasp the "*insider perspective*" as found in the discourses of businessmen involved in OCW, to approach *action* through the actors' own descriptions and justifications and verify what was effectively collective in this *representation* and locate the frameworks of tacit, explicit, accepted, and shared knowledge.12

Therefore, this analytical category aims to verify how the actors described, characterized, and evaluated the events, processes, institutions, characters, and themselves as well as the justification for their actions. The *representation* under study here, as well as the corresponding *action,* refer to the Odebrecht executives involved in OCW as much as to politics and the functioning of democracy, more particularly political financing. We grounded this analysis from three other analytical categories which implicate three different dimensions of this same *representation:* (1) *management culture of the company,* (2) *institutional format,* and (3) *political culture.* 

The *representation* about the *management culture of the company* refers to the *principles, guidelines, criteria,* and *values* in the discourse of businessmen

<sup>10</sup> The Prosecutor General of the Republic is the head of the MPF, and therefore of all prosecutors of the Republic. For more on the "Janot list" see https://brasil.elpais.com/brasil/2017/ 03/14/politica/1489522721\_207980.html.

<sup>11</sup> In another work, we analyzed the relationship between an organization that represents the industrial business community and the Brazilian parliament (Costa & Borck, 2019).

<sup>12</sup> We refer here to the propositions by Markus Pohlmann and others (Pohlmann et al., 2014). However, we do not develop or explore all the theoretical implications and analytical procedures that the authors associate with these propositions.

as to how the company internally managed illegal practices related to political financing. Therefore, most references to *management culture* also reference the action of the company itself, i.e., what it did.

The notion of *institutional format* aims to locate the *representation* of the *institutional dimension* and the *internal organizational particularities* established by Odebrecht to manage and account for external issues and challenges. The *representation* of Odebrecht executives about the *institutional format* is relevant as it concerns the *internal actions of the organization,* taken to address its relationship with a specific dimension of the political regime, in this case political financing, which in turn became a significant part of its strategy to defend its interests.

In turn, *political culture* refers to the way businessmen describe, analyze, and evaluate politics, the decision-making process and the political system, and its institutions and actors, including themselves. The objective is to verify how they view forms of legal and illegal political financing as part of the *management culture of the company* concerned with the defense of their interests within the scope of political authorities and institutions. This allows us to analyze how these business executives considered and reflected upon the functioning of democracy in Brazil, specifically from the standpoint of our research issue, i.e., political financing.

In short, we stem from a single set of statements, ideas, and opinions from which we outline the *representation* expressed by the deponents in the documents comprising our research object, albeit analyzed from a conceptual framework within our objectives in this chapter. Finally, it is important to understand how businessmen *represent* how democracy works, particularly when they undertake illegal practices. Despite having some diagnoses and evaluations, and even sources of information available to ordinary people, businessmen *experience* politics as well as corruption in different ways because they are part of an *elite* (Paz & Costa, 2016). Additionally, their ideas and behaviors have different impacts on reality. As underscored by Claus Offe, "owners, managers, and representatives associated with capital…" have "…the ability to *define reality* in a highly consequential way, in such a manner that what is *perceived by them as 'real'* may have a very real impact on other classes and other political actors" (OFFE, 1984) (emphasis added).

#### **4 Nature and Particularities of Our Research Object**

As we have seen, our research object consists of documents produced within the dynamics and objectives of *justice operators,* such as individuals responsible for the investigations, the prosecutors, the judge and defense lawyers, and the interests of the deponents themselves. Our investigation of this empirical material began with a documentary analysis (Cellard, 2008). Secondly, we considered the constraints posed upon Odebrecht executives within the context of the production of the analyzed documents. Therefore, we weighed the fact that the defendants' situation and their commitment to tell the "*truth, nothing but the truth*", as a condition for maintaining the plea bargain agreement, are part of the circumstances in which such documents were produced.<sup>13</sup>

Thirdly, there is the obvious fact that such constraints do not allow us to consider the statements contained in these documents as "true" or free from contradictions, inconsistencies, incompleteness, and alterations. As an example, we quote an excerpt from Cláudio Melo Filho's statement, former Institutional Relations Manager at Odebrecht: "First of all, I'd like to clarify that I bring forth fully redone and revised reports, which resulted from substantial efforts to detail and recover facts and data. I also bring relevant facts to your attention, which I know may be relevant sources to develop new lines of inquiry. I am fully convinced that, with my reports, I will clarify the pragmatic manner by which politics worked behind the scenes in the National Congress" (D9).<sup>14</sup>

In one of his statements, Marcelo Odebrecht provided an interesting characterization of the discourse hereby analyzed. Marcelo argued that the judiciary should bear in mind that other deponents could commit "errors, omissions, lies,

<sup>13</sup> Most of the deponents are under the condition of collaborators, and thus formally obliged to "tell nothing but the truth" without the right to remain silent, without exaggerating or concealing information about the facts at stake, and under the risk of perjury charges and losing their right to the plea agreement. To illustrate the formal condition of the deponents, we quote an excerpt from Marcelo Odebrecht's testimony, the central character of our research, in which the judge declared: "You have signed a collaboration agreement with the Prosecutor General of the Republic through which you have renounced this right, and you have pledged to speak only the truth before the judiciary. Is that right, Mr. Odebrecht?" The question is followed by his affirmative answer (D8).

<sup>14</sup> Cláudio Melo Filho produced this document himself, as part of his defense strategy, as he reported to investigators his experience regarding the relationship between Odebrecht, politics, politicians, and political parties. Although allegedly secretive, this document came to public attention as one of several "leaks" that characterized OCW. We extracted this excerpt from Cláudio Melo Filho's declarations from the document identified as D9 in the list presented in Appendix II. This same procedure will be adopted throughout this chapter.

obstructions", and therefore such statements could be "flawed" since "they did not provide elucidative information". To conclude, Marcelo Odebrecht declared that he hoped to bring forth "elucidative" information, while "trying, within my possibilities, to go beyond what (…) I knew at the time to help clarify all these illicit acts. And I think it is important, Your Honor, and I'm saying this to everyone, people need to turn over a new leaf, that is, upon leaving an environment of defense towards an environment of collaboration (…) they need to be committed to the truth, and often, by not acknowledging their responsibilities, they make it difficult to identify the facts." Thus, he positions himself as having "turned over a new leaf" to the condition of collaborator, that is, not exempting himself and striving to present novel and "elucidative" information (D5).

In the same vein, in another testimony Marcelo Odebrecht declared to the judge:

"I ask (...) that you do not to interpret me. As a collaborator (...) people sometimes find it odd (...) that I constantly try to clarify things. But that's because I'm pursuing an optimal adherence to the reality of the facts, to the actual truth. And so many times I try to clarify facts which in the legal complaint are not in harmony with how I see the reality of past events. I acknowledge that, and not only do I acknowledge it, but from my experience with this criminal lawsuit people were not trying to elucidate facts. And that's what I'm going to try to do, even if unbeknownst to me at the time, I'm going to delve deeper and that's what I'm trying to do in my statements to the Federal Police and the Public Prosecution Service, not only now but also at other moments. I hope to elucidate and identify possible illegalities, as I have already identified in this lawsuit. I think there are omissions, I think there are lies, I'm trying to clarify this with the Federal Police and with the Federal Prosecution Service. And I'll strive to be even more effective than when seeking information that I believed other collaborators would provide, and unfortunately from what I gathered in this criminal lawsuit, they are not doing so. And so I'm offering myself to (...) go beyond my initial collaboration and promote greater effectiveness to the company's entire collaboration." (D7).

For our analytical category of *representation,* as described above, the procedures are more relevant than the actual veracity or the pretended "optimal adherence to the reality of the facts", given that this reality is unlikely to be located and precisely defined. In short, we seek to interpret the individual subject. In one of the documents, Marcelo Odebrecht refers to what we understand by *representation* when referring to the "…way we rationalized" (D5). That is, the objective of this analytical category is to verify how the actors described, characterized, and evaluated events, processes, institutions, characters, etc. This is significant as it relates with the explanation and *justification* of his actions, in this case his relationship with politics. This is the path through which we hope to comprehend how he and other Odebrecht executives described the workings of illegal political financing, in which they acted as important business executives and important political actors.

We find in the excerpt above the external constraints and the internal aspirations to tell the "truth", to contribute to the investigations, and to report the events on behalf of the Odebrecht executives. While we must bear in mind that his answers follow the strategies of defense lawyers to reduce culpability, there is an assumed effort and commitment to detail procedures and actions, not least because the plea bargain agreement was conditional on his and any other collaborator´s provision of information to operators. This is the element we seek in these testimonies. After all, calculations also exist whenever individuals participate in interviews and *surveys,* but constraints of this nature do not.

That being said, our interest lies not in a reliable description of how these facts and events unfolded, nor in the "*truth*" as it interests investigators and judges, nor even in the public opinion and the press. We understand that these testimonies allow us to verify the *representation* pertaining to political financing as constructed by business executives who, before justice operators, acknowledged that they undertook illegal practices within this political dimension. We therefore seek *sociological* validity, and not the "truth" or legal validity, although our analytical purpose benefits from the particularities of the conditions in which the deponents found themselves. We sought this *validity* in discourses unlikely to be produced under other research conditions, given their core content, i.e., illegal practices, and from which we can analyze, in a very particular and unique way, the relationship between businessmen and politics.

#### **5 Methodological Procedures**

The research sought to locate the Odebrecht business executives who held prominent positions in the company's top management and who were simultaneously entangled in government relations, more particularly, political financing. Our verification confirmed the central role played by the then chairman of the holding company and descendent of the founders, Marcelo Odebrecht. According to the particularities of these documents, our analytical cohort of the other executives followed a logic akin to "snowball sampling", a technique used in questionnairebased surveys, namely the mention and attribution that the deponents themselves

made to each other, with particular emphasis to Marcelo Odebrecht.15 The list of business executives whose testimonies we addressed in this chapter, as well as information regarding their offices, are summarized in Appendix I.

As shown above, the empirical material used in our research stems from a set of documents relating to the plea bargain agreements of Odebrecht executives. We collected these documents in the complex and intricate information system of the Brazilian Federal Court, mostly by consulting the electronic case files available in the online system of the Federal Court of the 9th Region.16 These documents refer to different litigations between the years 2015 and 2017, as seen in Appendix II. Our analytical selection was based on the presence of information germane to the objectives of our research and processed through the methodological procedures of content analysis, with the use of an appropriate software for this type of research.

We codified the documents according to the abovementioned analytical categories and the objectives of this chapter. We also resorted to Marcelo Odebrecht's declarations from Electoral Court documents produced during the trials dealing with the financing of the 2014 presidential campaigns, given the convergence of their contents with our research to complement and compare them with the findings obtained in the OCW documents.

We were cognizant of the accuracy of the sources and methodological procedures for verifying the documentary analysis: on the one hand we were attentive to the adequacy of this empirical material to our research objectives and analytical categories, while on the other hand we verified the reliability of our analyses by cross comparing the codifications of the documents. All these steps were carried out and executed exclusively by the authors.17 Finally, as noted above, we numbered the cited documents to facilitate their use throughout the text, as identified in Appendix II of this chapter.

<sup>15</sup> Marcelo Odebrecht is admittedly the most significant character for our purposes, which is in line with the perception of the judiciary and the press. As an example of the multiple and reciprocal references during his dealings with political financing, Marcelo mentioned Alexandrino Alencar, Hilberto Silva, Emílio Odebrecht, Cláudio Melo Filho, Fernando Migliaccio, and Luiz Eduardo da Rocha Soares, who in turn mentioned Fernando Reis and Hilberto Silva, and many of them mentioned Marcelo himself.

<sup>16</sup> Some of the OCW criminal cases are still confidential. Minister Luiz Edson Fachin of the Supreme Federal Court mentioned in dispatches the names of whistleblowers whose lawsuits became public. See https://www.poder360.com.br/lava-jato/despachos-de-fachin-citam-50 dos-77-delatores-da-odebrecht-saiba-quem-sao/, accessed on 12/01/2019.

<sup>17</sup> On the topic of validity and reliability in content analysis, see Sampaio and Lycarião (Sampaio & Lycarião, 2018).

#### **6 Odebrecht and Illegal Political Financing: Analysis of the Results**

We present below our main research findings based on our research problem and the analytical categories detailed above, namely, the *company's management culture, institutional format,* and *political culture,* all related to illegal political financing.

#### **6.1 A** *representation* **of the company's management culture**

As previously mentioned, according to Marcelo Odebrecht, since the late 1980s the company made undisclosed payments to politicians and political parties, the so-called *slush fund,* but this management culture changed in the 1990s when the company stopped using false invoices and fictitious contracts and began separating resources from undisclosed payments (D1).18 It is worth mentioning that Marcelo himself alludes to this change in the administrative management of the company, and his statements will be fundamental to address the *representation*  about the *management culture of the company* regarding illegal practices (D1). We detail below our most important findings for the purposes of this chapter.

A first characteristic is the effort to differentiate *slush fund* from "bribe" as found in the statements addressing undisclosed payments to political authorities and parties.19 This distinction was possibly an attempt from the defense's legal strategy to characterize the company's illegal political financing practices as a *slush fund,* a less serious and less precise offense than the crime of active corruption, as is the case of bribes involving public agents or political authorities.

<sup>18</sup> This practice made use of "third-party companies" for political campaign financing, by way of *slush funds* and other illegal practices. The use of "third-party companies" was not seen as illegal, that is, other companies would make legal donations and would be reimbursed by Odebrecht, which remained anonymous. According to Marcelo Odebrecht, the problem is that this allowed "some businessmen" to use these resources to "make bribes" (D1). We will address this issue below.

<sup>19</sup> As we address this subject, we reinforce that we are not referring directly to historical phenomena and processes, but rather to how individuals belonging to this group describe, analyze, and evaluate such phenomena and processes and justify their actions. For an analysis of bribes and slush funds in Brazil's recent political history, see Alaor Leite and Adriano Teixeira (Leite, A. & Teixeira, 2017a, b).

Nonetheless. Marcelo Odebrecht's statements reveal something more complex and interesting for our purposes.20

Marcelo Odebrecht characterized "bribes" as resources paid directly to individuals, without any connection to the company's agenda or interests (D1). Thus, a bribe would be the private appropriation of a resource, legal or illegal, donated by the company. Although Marcelo openly admitted that the group had paid bribes to politicians and parties, he insisted that they differ from *slush funds*  (D3). According to him, slush funds would be a "natural" contribution to political parties and electoral campaigns without any ties to a "specific compensation", as with the "bribe" (D2), even if he acknowledged that the practice is an "electoral illegality" as it impacts electoral disputes. According to Marcelo Odebrecht, despite "wrongful" there would be a distinction between illegal campaign financing and payment of "bribes"—that is between paying to have "political support" and "paying the person directly" (D5).

Marcelo Odebrecht also expressed the concern that since bribes are easier to trace—as was the case in the OCW investigations—they could harm and compromise other "undisclosed payments" within the agenda of the company's interests, and thus in turn compromise and "contaminate" this agenda. Furthermore, bribes would have an impact on the management of payments, listed among the reasons why Marcelo personally coordinated the financing of presidential campaigns (D1).

Marcelo Odebrecht refers to an institutional and effectively political dimension in the use of *slush funds,* that is, the pursuit for "political support" targeted institutional figures, a political party or an important politician, or even the Federal administration. Such practice was not a mere personal act even if, as we will see below, it was based on personal trust. Thus, while acknowledging the "wrongful" (D7) and illegal content of the practice, he also saw it as a way to safeguard business interests, just like any other interest within legal boundaries.

With this line of reasoning, Marcelo Odebrecht attempts to place himself in the position of someone who had a "broad agenda" related to the interests of the company, of the sector, or even of other sectors, albeit never clearly defining this "broad agenda" and how it diverged from "specific projects". And although he mentions his protagonist role as a supplier of "credit", he does not detail what he expected out of this, that is, what this "agenda" comprised. The relationship of illegality with the government is thus naturalized, and it would be up to the

<sup>20</sup> As an example, during a statement to the Electoral Court, the court repealed an attempt by Marcelo Odebrecht's defense attorney to conceal the fact that his client had spontaneously acknowledged the relationship between the allocation of resources, even if not fully used, and the sanction of a provisional measure by Antonio Palocci (D2).

businessman to establish a procedure, in this case a line of credit, to a representative of the party in government, who, according to Marcelo, was inadvertently the same middleman for the legal and illegal financing of the political party campaign. The relationship of illegality with the government is thus naturalized. In this case, it would be up to the businessman to establish a procedure, a line of credit, to a representative of the President of the Republic's political party. We will return to this matter later.

Thus, the procedure consisted of assigning a monetary value and allowing government party leaderships to decide on what and how much to spend. In this "monetization" of a political relationship, an expression used by Marcelo Odebrecht, after establishing a value, the company made it available, politicians used it as they pleased, and the corporate demands would supposedly be, if not complied, at least more thoughtfully considered by the competent authorities. Marcelo states that, following his father's guidelines, he only controlled the use of credit to preserve the "relationship" and encourage the receiving end of the credit line to treat him "more thoughtfully" (D5).

Another dimension within this distinction between bribes and *slush funds* illustrates and corresponds to the *company's management culture,* that is, between the "specific compensation" and the "broad agenda" of the company's interests. The idea of a "specific compensation" was not only associated with bribes, but also with a "specific agenda" as in Marcelo Odebrecht's statements regarding the "Refill Crisis" and a line of credit in 2009 (D1, D3, and D7).21

According to Marcelo, the payments related to the "Refill Crisis" was a legal settlement from a set of compensations agreed between the company and the government, which were "contaminated" by other illegal financing activities. And Marcelo declares that, despite having regretted paying these two government procedures, which would not correct his mistake, they were a way to defend the "legitimate interest of the sector, of the company", and had not caused any drain to the public purse, but which, still according to him, was in no way an attempt to justify his actions. Finally, in his understanding this was a successful action, or at least a "reasonable" one, involving both legal and illegal payments (D1).

<sup>21</sup> The 2009 "Refill Crisis" was a long-term installment agreement to pay off the tax debt with the Union, established by Law 11.941/09 by way of Provisional Measure 449, for mitigating the impacts of the 2008 crisis on taxpayers and companies. According to Marcelo Odebrecht, a specific financing made by Odebrecht facilitated the enaction of this provisional measure (D5 and D3), which supposedly allocated R\$ 50 million for this demand (D1). As for the credit line, there are no references in Marcelo Odebrecht's documents that allow us to verify its nature or content.

In turn, the idea of a "broad agenda" emerges as an attempt to present illicit political campaign financing as a legal activity for safeguarding a "broader" set of legitimate interests, even if through an illicit activity. Marcelo Odebrecht describes the "broad agenda" as a domain of the Finance Ministers during the Lula and Dilma presidential administrations, and which concerned the interests of other companies such as Vale, Itaú, and Unica as they sought to benefit from Marcelo's relationship with the federal government and asked him to relay their demands, which cost him many hours of negotiation with the Ministers (D5).

Thus, Marcelo Odebrecht's relationship with the federal government, which involved prior confessed illegalities, especially *slush funds,* would have paved the way for forwarding demands from other companies. In this regard, he did not see this practice as an illegality, but rather an approach with the government which also served as a representation channel for supposedly legal interests of other companies in other sectors. Ultimately, Marcelo became an important representative for the business elite and took on this task, even if it gave him a lot of work.22 Whether or not a defense strategy, Marcelo sought to place his relationship with the government as a "broad relationship with all sectors" which he did not associate with any illegality, quite the contrary, he saw it as praiseworthy due to the wide-ranging agenda and his personal dedication.23 And even when acknowledging past illegal payments, he claims that if "electoral illegality" did exist, it mostly referred to the source of financing than its use (D7).

According to Marcelo, Antonio Palocci, then Minister of Finance, had agreed not to "discuss specific projects" and to annul requests from other party members. Hence, Marcelo was not particularly worried about "expenditures" as he had already established the line of "credit": "It was as if the account belonged to them". The "illegitimacy" or "illegality" of this relationship between Palocci/PT/ federal government would lie in "the definition of credit" (D7).

In short, we find Marcelo Odebrecht's efforts to characterize certain "specific compensations" and the "broad agenda" as the defense of legitimate interests, not only of the company, but also of the construction sector, and that the public purse was untainted. This behavior is very similar to the legal and explicit actions of businessmen in the pursuit of their interests, that is, by alleging a diffuse and legitimate interest, even when there is an explicit private and often exclusive dimension.

<sup>22</sup> According to Marcelo Odebrecht, even politicians such as the then Governor of Rio de Janeiro Sérgio Cabral resorted to him to access the Federal Government, more particularly the Finance Ministers (D3).

<sup>23</sup> In this statement, Marcelo Odebrecht describes how he organized the gathering of demands with his businessmen to relay to the federal government (D5).

Furthermore, Marcelo's efforts to distinguish *slush funds* from bribes, without denying that Odebrecht resorted to these practices, reinforces the idea that the company sought to react and promote readjustments in the *management culture*  of illegal practices insofar as new facts emerged. Secondly, the resort to the idea that the *slush fund* is natural and inexorable resembles the findings of Paz and Costa, who analyzed businessmen who were free and under no legal charges or suspicion, but who acknowledged the possibility of an illegal practice, even if they condemned it morally, justifying their actions under management, cultural, or even customary reasons as to how corruption works in Brazil (Paz & Costa, 2016).

A third aspect of the *company's management culture* is the relationship between *centralization* and *decentralization* of the management of illegal activities, both of which stemming from Odebrecht's intense relationship with the federal government, at the time under the Workers' Party administration (from the Portuguese, *Partido dos Trabalhadores,* PT). Centralization was a twofold movement: on the one hand, as we saw above, a "checking account" was created with a "limit", that is, an amount previously agreed upon with the Finance Ministers.24 On the other hand, a fund collection was set up with the group's executives to compose this "account" while Marcelo controlled the use of this fund by the party representatives. According to him, the legal part of this amount was divided between the *official* contribution to the party and the political campaign to "minimize the impact" of the contribution. Conversely, the use of the illegal amount was agreed upon between Marcelo and the Ministers. With this procedure, Marcelo hoped to avoid excessive "expectations" or a "delicate situation" with politicians and avoid pressure upon him or having to think about how and where the resource was used (D1 and D3).<sup>25</sup>

*Centralization* coexisted alongside managerial *decentralization.* While Marcelo focused on presidential elections and the relationship with the federal government, the group's executives had autonomy in their relationships with

<sup>24</sup> Marcelo Odebrecht asserted that this relationship was initially established with Antonio Palocci and resulted in the "Italian Spreadsheet", and afterwards with Guido Mantega, when the spreadsheet came to be called "Post-Italy". Marcelo mentioned that he allocated R\$ 64 million in this account (D1, D3, and D5). According to Marcelo, one of the reasons to accept the values requested by the PT were the company's good results with the Petrobras contracts, and that this would be the only area yielding positive results for the company, but he claims that he was unaware of the number of contracts with Petrobras. (D5).

<sup>25</sup> This strategy of rationalizing campaign financing expenses in the group by stipulating maximum amounts would also have been used in relation to disputes for the state government (D3).

politicians at the municipal, state, or federal level. According to Marcelo, Cláudio Melo Filho handled relations with the National Congress and the group's companies that would eventually be favored by the political maneuvers articulated by Cláudio Filho. Subsequently, such companies would assume campaign financing expenses, legal or illegal, arising from such agreements (D1). "Decentralization", which Pedro Novis sometimes referred to as "planned commission" (D12), also meant a fragmentation of the financial "limits" that each group company had for campaign financing (D2) and the financial amount raised for illegal activities (D4).

When asked about the meaning of the expression "Leaking campaign donations" found in one of the documents seized by the Federal Police, Marcelo Odebrecht declared that his goal was to follow the "*Group culture* (…) based on the planned commission and decentralization" (*emphasis added*), meaning donations would be fragmented by the group's companies, but once the news related to OCW came to light, he decided to "proactively" (*sic*) disclose the values donated, legally, by the companies in hopes to "avoid certain implications" (D1).

While Marcelo Odebrecht mediated some arrangements with companies regarding presidential campaign financing, such companies were free to compose further agreements with politicians who defended their particular interests. Therefore, the CEO of the group managed the illegal and legal political campaign financing of the main federal executive authority, the President of the Republic, not limited to a specific party or president in office, but encompassing different presidential candidates (D1).26

In short, *centralization* and *decentralization* serve as another example of the rationalization behind fundraising and the use of resources for illegal political financing which characterizes the *company's management culture* for illegal expenditures, marked by the division of responsibilities in the pursuit of resources for payments within the interests of each company. This occurred for both legal and illegal forms of political and campaign financing, the former deliberately disclosed and presented to the press, while the latter evidently only came to light after the disclosure, or eventual leaks, of the testimonies derived from the OCW investigations and litigations and their subsequent developments.

To conclude, we mention a few other dimensions of the *company's management culture.* Emilio Odebrecht, former president of Odebrecht and Marcelo's father, declared that he instructed Marcelo and Pedro Novis to observe "three

<sup>26</sup> In another statement, Marcelo Odebrecht reaffirmed that he handled the relationship with the executive authorities and national politicians, while his "subordinates" dealt with lower political levels (D7).

vectors" when donating to political parties: "avoid discrepant (values) among parties, (…) negotiate to the maximum" and make payments "stretched out over the longest possible period of time" (D14). Marcelo Odebrecht informed that the group also paid for political campaigns in other countries and that other group executives were perhaps less worried about his warnings about potential "contamination" as the campaigns were not in Brazil.

Another interesting aspect were the different ways for routing illegal financing. Companies contributed through cross donations, i.e., a company from one state could not justify donations to a politician from another state, so it would ask a different company to do it and later reimburse them. Marcelo was aware of this but said: "I never got involved". Alternatively, payments were sent to campaign marketers, such as João Santana (D3). Another strategy involved donations, both legal and illegal, to coalition parties led by the government party, in this case the Worker's Party (PT), and to election campaigns for different offices (D3).

*Trust* was another vital element of the *company's management culture,* which is not surprising given the nature of illegal political financing. Marcelo Odebrecht indicated this was an important element in the relationship between the company and the politicians as well as with other companies that provided services to political parties which Odebrecht also illegally financed. Marcelo referenced Hilberto Silva's "historical" trust in political publicist João Santana, a well-connected individual who represented the interests of the Worker's Party (PT) and its representatives alongside Odebrecht (D1 and D3).27 Pedro Novis also expressed the importance of trust when describing his relationship with "Dr. Palocci" as being "… personal, very close" (D12).

Although such representatives were Ministers, their role in the relationship was primarily as party emissaries holding top-level government offices. This was not a direct relationship between the company and Ministers, but rather related to the office that Mr. Palocci and Mr. Mantega held in government. The same occurred, for example, in the case of the "Refill Crisis". Thus, the relationship of trust entailed a personal agreement and/or settlement, ascertained through a relationship of expected behaviors from each side, rather than agreed upon in a formal contract.

Another interesting aspect is the existence of the "internal godfather", i.e., people within the company responsible for contacts with a specific politician. For example, Emílio Odebrecht would be the "godfather" for former President Lula, and Marcelo for former Senator Aécio Neves as well as presidential candidate

<sup>27</sup> João Santana was responsible for the presidential campaigns of Lula (2006) and Dilma (2010 and 2014).

Eduardo Campos, who died in 2014 (D5). Marcelo Odebrecht described the relationship between ministerial and state company offices and political parties: such offices, when "not technically appointed", would be "divided" by the political parties and designated to their "political godparents," in other words "every office had a political godfather."28 The "act of becoming a godfather" referred to a relationship of knowledge and proximity between company executives and politicians: "within Congress, each politician had a godfather, and nothing was agreed upon about any given person, even if related to another business or whether by another business, if not aligned with and communicated to this internal godfather: this was our own internal procedure" (D5 and D7).

In sum, for Odebrecht the most common form of undisclosed payments was the *slush fund.* This method was appealing as politicians would not raise questions about contributions to other politicians, since official donations are public information (D3). Marcelo Odebrecht admitted that the company, "unfortunately" regarded the *slush fund* as something "natural" and as a "donation" and, consequently, there would be no "illegality" in this activity. When Marcelo argues in his testimony that he would be "rationalizing how we rationalized" (D3 and D7), he references, simply and bluntly, the *representation* category, that is, how someone describes their own actions, how things operated, and his own assessment regarding the *management culture of the company* regarding illegal political financing.

#### **6.2 The** *representation* **of the institutional format**

As we have shown above, the particularities of the *company's management culture* refer to how the Odebrecht company internally managed its relationship with illegal financing, which impacted the *institutional dimensions* and the *organizational particularities* regarding this matter within the company. Thus, one of the main aspects in the *representation* about the *institutional format* lies in references to the company's "Structured Operations Sector", the name given to the sector

<sup>28</sup> Marcelo Odebrecht declared that, in addition to the company's "internal godfather", there was the "political godfather", responsible for suggesting names for public offices. Hence, the understanding was that these positions were occupied by people sponsored under their godfather not on behalf of "technical merits", but rather "political influence". Such would be the criterion for holding important offices within public management, something akin to how common sense and the press interpret the relationship between parties and public offices, commonly referred to as cronyism. Hence, according to Marcelo, the company would be able to influence politicians by way of their sponsors (D5).

responsible for the control and execution of illegal payments, especially political financing, which attracted a lot of attention from justice operators as well as the national and international media.

Marcelo Odebrecht initially denied the existence of such a "Sector" (D1). He later stated that it had been created in 1993 unbeknownst to him and other company executives. Marcelo declared that he only became aware of this sector after his arrest and the company's leniency agreement. But he acknowledged that this was a widespread procedure among Odebrecht executives for managing information regarding "undisclosed payments" (D3). Subsequently, he acknowledged that since 1980 there had been a "team", but not a "sector", responsible for these payments and for supporting the group's business executives (D5 and D7), and that such payments were destined to financing political campaigns (D2).29 Pedro Novis, Marcelo's predecessor in the presidency of the Odebrecht Group, explains in a statement that the structure created for "tax planning" had eventually become a mechanism for "paying public officials" (D12).

At any rate, Marcelo Odebrecht reported that, in 2006, Hilberto Silva suggested to change the name to "Structured Operations Team" since this was the denomination "given to financing operations with specific guarantees, and it would be way to support all businessmen without people suspecting that we were making undisclosed payments" (D5 and D7).

An important tool in the operationalization of illegal payments was called *MyWebDay,* a software which, according to Marcelo Odebrecht, was an improvement envisaged after the "Budget Scandal" (D5).30 This same information is found in the testimony of Cláudio Melo Filho (D9). The goal would be to avoid, on the one hand, undisclosed payments through fictitious contracts and, on the other, "to create a process where (*sic*) generation was separate from distribution". In all types of undisclosed payments stemming from "commitments" acquiesced by 200 to 300 businessmen of the group, including "bribes", a person from the company's financial sector became responsible for sending the information to the "undisclosed payments team", coordinated by Hilberto Silva, since, although unaccounted for, the payment "had to be managerially allocated to the cost of

<sup>29</sup> As a result of the OCW revelations, the Brazilian electoral justice questioned Marcelo Odebrecht. The documents pertaining to these interrogations are D1, D2, and D3, and were also under the conditions of the plea bargain agreement.

<sup>30</sup> In 1998, an investigation by a Parliamentary Investigative Committee (CPI in the Portuguese acronym) revealed that shell corporations and contractors had diverted funds from parliamentary amendments enacted by Parliamentarians belonging to the Budget Committee. The investigation became known as "The Budget Dwarfs".

that project" (D5).31 Hilberto explained, in one of his statements, that he adapted the official system of the federal treasury and the company's prospective payment system, to "systematize and organize the records" (D15).

This was not a mere bookkeeping, but a thorough rationalization of the resources used for each "project", an expression used to refer to each undisclosed payment akin to, we may presume, when managing the legal version of these "projects". This system allegedly remained operational until 2004, when Marcelo, then President of the Construction Company Odebrecht, decided to centralize this process around a single individual as he believed there were too many people involved. This person would have a spreadsheet and exclusively manage payments, checking whether there was "positive balance" and enabling payments to the "Structured Operations Team" with a nickname for each recipient (D5).32

Thus, here lies a significant example of the procedures adopted in the defense of the company's interests which, on the one hand, stemmed from the *management culture of the company,* and, on the other, revealed an effort towards rationalization, professionalization, and institutional improvement, both for legal and illegal financing practices, which transpired in an articulated and organized method. We find here more concern with the administrative and financial management of illegal payments than with the possibility of public exposure.

These statements also reveal that the managerial activities of the *holding* president, in this case the CEO, included the administration of fundraising and illegal payments within the company's relationship with politics, as he had to supervise the organizational and institutional dimensions of the company and promote changes and solutions to problems emerging in the previous forms of this same practice, not least because there is no training course or graduate education for such activity.

In short, we are less interested in addressing Marcelo Odebrecht's natural inconsistencies, not least because he was in a condition previously unattained by anyone of his business and political standing, than we are in rationalizing and

<sup>31</sup> Marcelo Odebrecht stated that a share of the undisclosed resources was used for "business matters", i.e., without any involvement of the "public sector" (D1). Marcelo detailed in his statements some organizational elements of these accounts. Seemingly, Hilberto Silva managed the payments, and Marcelo controlled the balance and the general guidelines. However, Hilberto allegedly changed this pattern and compromised the "check and balances" system. In short, there was a managerial problem regarding undisclosed payments (D7).

<sup>32</sup> These nicknames drew a lot of attention from the press and the general population when they were publicly exposed. According to Marcelo Odebrecht, this procedure allowed them to refer to people in front of others without disclosing their identities (D5).

covering the illegal practices found in the managerial and administrative procedures typical of a large company. Rationalization and professionalization also appear in the statements by Cláudio Melo Filho, when he declares that the political monitoring system was internally elaborate and well-structured through the formation of a qualified team and the definition of agendas of interest, as well as by monitoring the behavior of politicians within the National Congress. Furthermore, there was a protocol of sorts, with well-established steps for monitoring the legislative process and the use of the *Drousys* system by the "Structured Operations Sector" (D9).33

The *representation* about the *institutional format* suggests a combination of: 1—striving for efficiency, managerial gains, and rational business management, for which these businessmen had studied and amassed professional experience; 2—perfecting ways for tackling challenges inherent to the political game; 3 acknowledging the conditions imposed by the admitted criminal nature of the practices concerning illegal political financing. This suggests that the economic power and managerial sophistication of large companies may favor rationalization and optimization when managing not only normal and legal business, but also illegal activities, which were just as common.

Lastly, the company's efforts reported in the *representation* about the *institutional format* reveals that the management of illegal payments reacted and learned from events of the political world, as indicated by the reference to the "Budget Scandal". In other words, adjustments to the *institutional format* took place not only between the individual and the company, a dimension analyzed by Pohlmann and others (Pohlmann et al., 2016), but also as a consequence of the dynamics of the political system and the practice of illegal political financing.

<sup>33</sup> The *Drousys* system is one of the most complex aspects of the OCW investigations, consisting of the information and management systems for undisclosed payments together with *MyWebDay.* According to Marcelo Odebrecht, unlike *MyWebDay,* no one knew of the existence of *Drousys,* unintelligible even to the executives within the Odebrecht group. Marcelo stated that *Drousys* was a "control system" created by Hilberto Silva in which not all undisclosed payments were recorded, while in *MyWebDay* all payments "generated a managerial allocation, and in theory all undisclosed payments made by the company would be in *MyWeb-Day*", which explains it being more detailed (D5).

#### **6.3 The** *representation* **and political culture of Odebrecht**

Regarding the *political culture* of Odebrecht and its illegal practices, a first pillar of how these businessmen *represented* politics is the characterization of *illegal* corporate political financing. On the one hand, Marcelo Odebrecht claimed that "every public notice has an owner", as defined by the capacity of a given company to shape the public notice according to their interests together with a public authority, an ongoing and recurrent practice for the past 30 or 40 years (D7). Such was the relationship of the contractors with government authorities. The "trust" between businessman and politician, as discussed above, comprised a "tripod" together with the company's ability to deliver the public construction work desired by the politician as well as the financial support to the politician's project.34 According to Marcelo, this would happen across many different economic sectors, upon which he concluded that a "relationship grounded exclusively on money" just as a relationship "exclusively comprised of good projects, good investments" would not work, because the politician had the "expectation" to receive money for his political project, or to use it as a pretext to ask for financial support (D7). In the words of Pedro Novis, these would be "extra-accounting (…) requirements" (D12).

As we have shown above, even though he admitted that *slush funds* contained an "electoral illegality", Marcelo Odebrecht tried to differentiate it from "bribes". One of the arguments behind this differentiation also referred to the particularities of the political system. First, he claimed that 3/4 of election spending in all campaigns in the country is undeclared. Secondly, the parties themselves as well as the candidates requested *slush funds,* as they did not want to reveal all their expenditures to the Electoral Court. And thirdly, as we saw above, the companies also preferred *slush funds* so as not to publicly reveal the amount donated to each candidate. If a company legally donates to a party, another party, or even the same party from a different region in the country, could ask for money. In the words of Emilio Odebrecht, they had to find ways to circumvent "jealousy and problems" with political parties (D14). Finally, Marcelo concluded that, if the company had

<sup>34</sup> According to Marcelo Odebrecht, there would be a "tripod": "you have to perform as a company, you have to present important things, which means, the good side is you have to create the second tripod, the relationship of trust established with a person, and the third tripod is the financial support to that person's political project. I don't believe that a relationship grounded exclusively on money would work, unfortunately, nor a relationship comprised exclusively of good projects, good investments, based on a relationship of trust. It wouldn't work because they also expect you to support their political project or act on behalf of their political project" (D7).

a privileged relationship with a politician, this was by virtue of a major donation through a *slush fund* (D1).

This is an important characteristic of Odebrecht's *representation* of the political system. On the one hand, how the CEO and other senior figures of the Odebrecht *holding* described corporate political financing. On the other hand, the way they positioned themselves or saw themselves as passive agents for politicians to fulfill their requests and demands, even though admitting that this relationship also interested their company. This cannot be reduced to a mere attempt to justify illegal activities, not least because they were ultimately confessed in Court.

This *representation* reinforces the idea that this was how these businessmen described not only themselves, but the entire business community in the country within the political process. Furthermore, Marcelo Odebrecht stated that "he felt used by the government" and the outcome of this political relationship would be detrimental to his person and his business position, because, as he himself argued, these activities resulted from problems that the government itself created. Finally, even though he accepted the idea of being an "ambassador" to business interests, an expression used by the Electoral Court judge, he stated that he really felt "used" (D3).

According to Marcelo Odebrecht, politicians would not honor agreements and promises, furthermore, they would create constraints that forced him to "plea" like "a beggar" for things the government requested the company. Marcelo attributed his efforts to solve problems to his "spirit to serve" the company, which he believed had been misinterpreted by the Federal Prosecution Service. And when asked by a judge if he felt like the "owner of the government," he replied that he saw himself as a "beggar": "Deep down, I was not the owner of the government, I was the government's sucker" (D3). Interestingly, a similar discourse is commonly found among ordinary Brazilian citizens regarding the political class.

In our understand, this *representation* means more than simply trying come off as a victim of the government and politicians. It refers to the way these businessmen thought and acted in relation to politics, politicians, and political institutions. As much as this was part of the defense strategy, it is also a justification for the passivity and inertia in the face of institutional forms, both legal and illegal, of the political system, in this case the particularities of *illegal* corporate political donations.

Bruno Carazza (Carazza, 2018) addressed some aspects of what we call the *representation* of Odebrecht, defining it as the company's "internal donations policy". However, we understand that this dimension cannot be analyzed as mere "pragmatism" or "lack of ideological criteria" as he does, since, on the one hand, we would need a precise definition of "pragmatism" and, on the other hand, analyze the agent's behavior from a normative criterion against such "ideological criteria", especially since the Odebrecht executives themselves claimed that this was a recurrent practice for the past 30 or 40 years. Be it as it may, Marcelo Odebrecht described the company's behavior, as well as his own, by characterizing and distinguishing "bribes", "specific interests", "broad agenda", "fluid agenda" etc., and assigning at least some of the blame to Brazil's political culture and system.

Emílio Odebrecht adopted a similar discourse, saying he advised his executives against bringing "problems of the organization", but rather "solutions to the country's problems", arguing that "companies have a duty to contribute to solving the shortcomings" of the country. However, he also declared that the executives would bring "their own problems" and that he was "aware of" future donations, suggesting that he knew about them (D13).

We find other interesting characteristics of the company's political culture regarding illegal activities. Marcelo Odebrecht expressed discomfort and criticism in his relationship with the PT administrations due to the intersection between the party's fundraiser and the Ministers with whom he and other businessmen addressed their demands, referring more specifically to Antonio Palocci and Guido Mantega. Thus, there would be a direct relationship between fulfilling or considering these demands and the prospect of political funding. As we have seen, he described this dynamic as a "fluid agenda", since the interlocutor receiving the demand was also the fundraiser for the government party (D7). Thus, the problem would emerge in the resolve and primacy of the fundraiser's expectations when meeting these demands as a Minister. For Marcelo Odebrecht, these demands had to be considered and preferably accepted, without the need for a counterpart in political, legal, and illegal financing, even more so when dealing with the "broad agenda".

In Marcelo's assessment, this was a "huge mistake" and a "major blunder" of the PT administration, that is, to have finance Ministers as government interlocutors when addressing the "legitimate demands" of business executives. These same Ministers were also responsible for the political campaign fundraising, undermining an important value for him: the separation between these two dimensions. Marcelo claims that he discussed "legitimate issues" with the Ministers, but this relationship would be "skewed by the fact that the person with whom I talked (…) was the same person asking for resources and with whom I negotiated the financial contributions", a fact that would have contributed to the eventual discovery of the entire scheme (D5).

Marcelo's description of this situation is particularly interesting, that is, as something supposedly new and devised by the PT administration in contrast to the longstanding relations between businessmen and governments. This means that, for Marcelo, this intersection should not exist in the relationship between businessmen and authorities in a democracy. For Marcelo, the mistake was in how the PT mediated the relationship between campaign financing and meeting the demands of businessmen, and not the relationship itself, which he had naturalized and accepted as *legitimate,* even if admittedly *illegal.* 

The testimony of Fernando Reis, former President of Odebrecht Ambiental, expressed a different point of view, even though it corroborates Marcelo Odebrecht's discourse. Fernando declared that campaigns contributions, legal or illegal, were seen by the company as a way to intervene in the electoral process as well as in the candidates' discourses and proposals, even if it not decisive: "This was done in an erroneous, distorted manner and ultimately caused Odebrecht the problems that it has today". Odebrecht's extensive work in the field of investments was seen as a "way to protect" the group's companies' interests. Such "exaggeration" would have turned Odebrecht into an "open door for any politician" looking for funding. In addition to the volume of demands, as the company had activities throughout the national territory, the campaign financing system, both legal and illegal, was "geographically spread out" given that the company would have become a "major gateway" for political demands (D11).

According to Fernando Reis, this led to a trivialization and difficulty to distinguish the legitimate from the illegitimate: "We went deep into this vicious circle". Donations were made for the purpose of circumventing mayors threatening contract breaches. Thus, they adopted the "concept" of making donations with the "institutional purpose to uphold contractual regularity": "it was close to absurd", but it was a way to force the mayor to "comply with the law" and not resort to artifices to breach the contract. In the 2012 municipal elections, Odebrecht´s superintendent directors provided a list of candidates and donations were evenly distributed, to avoid future problems with whoever was elected. However, this did not mean benefits would be guaranteed. Hence, "in an erroneous and distorted manner" the company sought contractual regularity and a way to safeguard their interests (D11).

In short, our intent is not to consider these discourses as an objective, reliable, and thorough account of how things worked, but in addition to converging on the characterization of illegal political funding, the discourses reveal that they were more than just businessmen acting as fragile victims of a political system, as the company clearly enjoyed many of the advantages of this type of relationship with politicians, and only confessed their illegal actions due to external circumstances.35 The complexity behind the *representation* of this process indicates that it cannot be understood as mere "pragmatism". Moreover, what would be the need or even the possibility of following "ideological criteria"?

What ultimately led them to regret how things worked was not solely their condition of defendants, but also the acknowledgement that this form of defense of their political interests had a cost. Reducing this behavior and point of view to "pragmatism" does not allow us to understand the nuances of its construction as *representation* and how this allows us to reflect upon concrete actions, *cultures,*  and *institutional formats.* 

Thus, this *representation* reveals the tendency of businessmen to brush aside the construction and change process in how they engage with politics and decision-making, as well as the lack of a clear project for a political system existing beyond the direct relationship between money and politics, although it is easier to play the victim, the passive actor, or even the "sucker", when the illegal, very profitable things being done come to light.

The notion of pragmatism could conceal the fact that businessmen acted as they usually do when faced with the complexity of the political game: passively, even more so considering it was a profitable situation for them. This was therefore a choice and points to the limits of the very constitution of businessmen as political actors. When it comes to defining the mechanisms of the political process, when faced with the professionalism of politicians in this field, especially politicians in government, the lack of preparation of businessmen for this dimension of the political game blatantly manifests itself, even if we do not literally commit to Marcelo Odebrecht's self-definition as a "sucker" or "court jester".

After all, we cannot analyze these businessmen as merely greedy above all else or simply feigning when they express dissatisfaction with the outcome of these political relations. This seems to be how they dealt with politics and the decisionmaking process of the State. Furthermore, Marcelo Odebrecht's figure reveals how difficult it can be for businessmen to act in politics when this relationship extends beyond money and fulfilling demands—the dimension for which they are most prepared—towards the requirements of the political game, for which, as we observed above, there was no training course.

When the Electoral Court questioned the purpose of the electoral contribution payments, Marcelo Odebrecht explicitly spoke of a "culture" that existed since the 1980s and in which he was raised. According to Pedro Novis, this "has been going on since forever" (D12). For Emilio Odebrecht, besides being "a

<sup>35</sup> Bruno Carazza analyzed the data regarding the positive results obtained by the company in the period before the outbreak of the investigations (Carazza, 2018).

dominant model in the country" (D13), it dated "way back to my father's days" (D14). When questioned by the Minister, Marcelo argued that it was their relationship with politicians, the political system, and democracy in Brazil. Hence, when politicians supported a bill that could be of interest to the company and the sector, an "expectation" emerged.36 And Marcelo saw this behavior by both politicians and businessmen as natural, because it was on the company's interest to elect someone close to their interests, as it was the politician's interest to receive the contribution.37 This dynamic would be inherent to the existence of the electoral contributions. The problem would lie in the uncontrolled dimension of *slush funds,* in the "smuggling" of bribes, in the non-legalization of *lobby,* in how legal contributions work, and the way the press associated campaign financing with potential shady benefits granted to companies. According to Marcelo, Because of the way the press associated campaign financing with potential shady benefits granted to companies, Brazilian society had a different *representation* as that of business and "criminalized legal campaign financing" (D1).

Marcelo Odebrecht also had a stance regarding the press. According to him, the media would have distorted the legitimacy of safeguarding business interests, that do not necessarily involve "bribes", and reduced everything to *slush funds,*  which should not always be the case. In other words, this is Marcelo's interpretation about the problems in the political system, the media, or society's reactions. The company's mistake would have been to convert all political financing into *slush funds,* a choice he justifies due to the "fear of criminalization" of regular campaign financing. Defending interests through clear contributions would be the best thing to do, and by "looking forward" a lot would have to change, hence the importance of OCW according to him (D3).

Marcelo Odebrecht indicated that the relationship between meeting demands and a "political project" (D5) was naturally accepted by the company and regarded as a mere characteristic of the political system. This may sound obvious, but it is interesting when explicitly mentioned by a major businessman,

<sup>36</sup> In his statement to the Electoral Court, Marcelo Odebrecht expressed how he saw the relationship with the government, namely, bringing claims "of public interest", which led to an expectation by the federal authority and the idea that they were helping the company: "That's it. That was the relationship that, unfortunately or fortunately, took place between politicians and businessmen" (D3).

<sup>37</sup> Marcelo Odebrecht reinforced that he understood the expectation of donations by politicians without any "specific compensations" as legitimate, and that this occurred not only with him, but with "every politician who dealt with businessmen […] we saw it as natural, and thus if that politician was someone who would help, who would support his company or the sector, a legitimate expectation emerged. I thought this was natural, like everybody else" (D3).

because, at least from a common sense or naive perspective, politics could easily exist without such a relationship. The point here is not to suggest inconsistencies on Marcelo's discourse, but to emphasize one doubt: "whether requests were prioritized because they were legitimate or because money was involved" (D7).

#### **7 Final Thoughts**

Our goal was to present the results of an analysis of the testimonies of Odebrecht executives involved in Operation Car Wash, particularly regarding illegal political financing. For this approach, we used *representation* as our main category to verify a specific and underexplored dimension of the *political culture*  of businessmen, upon referencing individuals who belong to an important segment of the *economic elite,* as well as their own political actions—in this case admittedly illegal—in democracy. In addition, we considered that the description, understanding, actions and justification for such actions, and characterization of politics are complex objects even for the actors who produced the *representation.*  Lastly, we identified that the practices reported by the Odebrecht executives refer to what is commonly called *grand corruption,* as it involved major political characters and institutions and a significant misuse of the public purse for the benefit of a specific and reduced group.<sup>38</sup>

The analysis of the *representation* of these businessmen regarding illegal political financing indicates a certain homogeneity or a certain sharing, such as, for example, when they sought to characterize the political practice itself and its relationship with the political system as taken for granted, that it could not exist in any other way. This raises the hypothesis that there is a widespread lack of effective interest and attitude on the part of businessmen in changing the political system, despite commonly professed dissatisfactions. Furthermore, both within the legal and illegal scope of corporate actions, there is a tendency to resort to the justification of *impotence* so as not to engage towards change in how politics operate and their lack of interest in doing so. However, we must also remember that, both legally and illegally, they may have their interests met, even if expressing dissatisfaction, and therefore do not feel motivated to promote changes in the functioning of the political game. Another hypothesis is that, if they ever try to

<sup>38</sup> On the issue of *grand corruption,* see the work of Inge Amundsen (Amundsen, 2019) and the characterization by https://www.transparency.org/en/our-priorities/grand-corruption, accessed on 10/18/2020. The *World Bank's Global Report* draws attention to Odebrecht's case as a recent example of *grand corruption* (World Bank, 2020).

promote such changes, they will have to face, for one thing, their own way of thinking about politics and themselves as political agents.

Despite their differentiated economic and political status, the *representation*  by Odebrecht business executives about political financing does not always differ from what we find within *common sense,* that is, the idea that politicians only think about resources, legal or illegal, and that political parties are mere machines that superimpose their fundraising and electoral interests over social interests, and that this dynamic would be marked by corruption and influence peddling.

That is not to say that this interpretation is misguided, but rather that it is partial and simplified, such as when we consider the other side of this *representation.* From business executives' point of view, agents with great economic power experience illegal political financing in a more realistic and less normative manner, precisely due to this combination between their own condition of major business executives and political activity in defense of the company's interests within an illegal context, albeit overlapped and articulated by legal forms. From this standpoint, this condition places business executives in experiences and practices that are quite different from the experience of ordinary citizens, the bearers of what we call *common sense.* 

The *representation* of illegal practices as "natural" and recurrent for the past 30 or 40 years served as a way of justifying them, even more so when before the judiciary, for businessmen sitting as defendants. However, more revealing to us is how this worked perfectly well for the company's interests, given that it never sought to change this framework of relations with authorities, and only began to consider them an "illegality" or "mistake" when under trial for such practices. This suggests that a rupture with illegal habits in the defense of the company's interests was only possible through an external and dramatic factor, as was the case of OCW, even if questions and doubts remain regarding the institutional and even legal validity of its methods.

Marcelo Odebrecht stated the law should punish whenever "some illegality" exists, such as "bribes". The problem would lie in the "agreement", in the exchange of favors, whether legally or illegally. Therefore, for Marcelo, more important than criminalizing *slush funds* would be to verify the existence of an "agreement", which he claimed to have perpetrated only in rare occasions. Therefore, "illegality" would exist only in form, i.e. *slush funds,* and not in content and substance, i.e. exercising the defense of legitimate corporate interests, the *lobby*  (D1). Therefore, we find not only the articulation between legal and illegal practices, but rather a certain way of considering the promotion of corporate interests of a major company through corporate political financing as it has operated in Brazil since the 1980s.

This *representation* shows that one of the main traits of this perspective is the fact that the CEO of a major group such as Odebrecht, regardless of his family background, held a position that required, among other things, aptitude to implement and refine a management *culture* in addition to administrative procedures and illegal institutional practices, on top of negotiating with politicians and managing conflicts between them, while often having diverging interests.

As our interest lies not in legal or moral judgments, it is worth remembering that an act is considered legal or illegal according to the legislation in force in a given society, its desires, and historical context. Many acts may cease to be criminalized or become crimes throughout history. While this does not justify the crimes committed, it may help us understand some of the statements by the businessmen.

Investigations such as OCW, as well as many others still underway in Brazil, do not lead us to believe that enough constraints exist to end such practices, since the testimonies by Odebrecht executives suggest a tendency towards the persistence of illegality as long as it continues to be one of the easiest paths for goal achievement. Therefore, the Odebrecht case suggests that only through a combination of more effective laws and investigative tools and a change in *management culture,* and especially in the *political culture* of businessmen when defending their interests, would we see effective long-term results in the fight against political corruption. There is an underlying generational dimension in all this, and ruptures with longstanding practices only occurs by way of a very relevant, dramatic, and external reason.

Lastly, we champion the need to expand this research front of the *Political Sociology of democracy* in Brazil: to reflect upon about the functioning of democratic institutions from the perspective of relevant social and economic agents, their ideas, their values, and their actions, within and beyond the confines of legality.

#### **Appendix I**

The table below presents information about the Odebrecht executives whose documents we used in this paper.


#### **Appendix II**

The table presents the numbers of the documents used throughout the paper, followed by the identification of the deponent, the date of the document, and the legal identification in the information system of the Brazilian Judiciary.


(continued)


(continued)

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Compliance Implementation Challenges in the Shadow of Corporate Cime: A Case Study of Odebrecht S.A**

Mario Helton Jorge Jr.

#### **1 Introduction**

When it comes to corporate crime, recent studies show that bribe givers are usually high-ranking, well-educated, and well-paid managers, which undertake risks in the name of the company and generally orientated toward growing and achieving results as a unit with the company. Until there is some sort of accountability, illegal actions are seen as useful to the organization and themselves (Pohlmann et al., 2016; Klinkhammer, 2015). Corporate Crime encompasses the supply-side of corruption, which is associated with a firm's interaction with the public sector, with cases of grand corruption, in contrast to petty corruption, being the most commonly studied, given its high-profile (Mahmud et al., 2022).

There are several approaches to analyzing corporate crime, or more generally, organizational wrongdoing, given the multi-level complex phenomena that it is, which has a reflection on the compliance policy suggestions that are made. If the focus is on individual rationality and greed, an effective web of accountability (monitoring—investigating—sanctioning), ethical formal standard setting, and moral training are presented as viable courses of action (Root, 2019; Soltes, 2020; Valentine et al., 2019). If the organization has a bigger role in the criminological genesis of corruption, then structural prevention should be considered alongside measures that target accountability and ethical training, such as an opening up the career system for outsiders, job rotation between company units, diversity management, whistleblower protection, prolonged bonus terms, detach

M. H. Jorge Jr. (B)

Alumni Max Weber Institute of Sociology, Frankfurt, Brazil e-mail: mariohjorge@gmail.com

243

the ombudsman from management, among others (Tanner et al., 2019; Van Erp, 2018; Yagmur, 2020). Although anticorruption research benefits largely from macro-level analysis, both industry and country wide, it has less of an influence into compliance research due to it being largely out of the hands or extremely demanding for organizations to focus and lobby for collective action at that level. So moving from the decision-makers to the private sector seems to be present an efficient strategy in combating corruption (Mahmud et al., 2022), due to the need for active cooperation and participation from the supply-side at its organizational level.

In order to understand compliance system's features and implementation challenges, this chapter aims to use a particular case study, in which the company suffered previous prosecution for wrongdoings, and present its findings considering the timeline of implementation. This will allow us to understand, in this particular context, how are anticorruption practices integrated into processes and structures and how is it handled and perceived by management, highlighting important factors for a successful system implementation.

After this brief introduction we will first explain our research methodology and data collection process (Sect. 2), the concept and structure of compliance standards and of Compliance Management Systems (Sect. 3), go over the case study (Sect. 4) considering the timeline of the standards before investigations started (Sect. 4.1), the agreements and deals with law enforcement (Sect. 4.2), the implementation of compliance measures (Sect. 4.3) and a discussion over our findings (Sect. 5).

#### **2 Methodology and Data Collection**

As mentioned before, corruption is a complex multi-faceted phenomenon, where quantitative analysis fails to grasp the particularities of each case (Cardoni et al., 2020). A qualitative single-case study provides a more "hands-on" practical depiction of how a multinational business handles and experiences the issue (Bhattacherjee, 2012). Odebrecht S.A., now rebranded as Novonor S.A., fits the purpose of this chapter, being the most high-profile prosecution of corporate crime in the history of Brazil, where the rate of corruption is particularly high (rank 96/180 Transparency International CPI, 2021), with repercussions in several other countries where it had business. Additionally, the construction sector is one of the most prone to corruption in the world (Risk Advisory's Corruption Challenges Index, 2020).

Odebrecht S.A., the holding company, parent to several companies in the construction, oil, gas, and chemical industries, has its headquarters in Salvador, Brazil and had hundreds of projects across more than 27 countries. According to its case files from Car Wash Operation and the FCPA investigation, between 2001 and 2016 Odebrecht made approximately \$788 million in corrupt payments to foreign political parties, foreign officials, and their representatives in 14 countries around Latin America and Africa, in order to secure an improper advantage to obtain and retain business, generating more than \$3.3 billion of profit to the company as a result of these corrupt payments. These were possible, among other factors, due to inadequate controls, dysfunctional leadership and incentive structures and an inexistent compliance program during the period of the criminal conduct.

Legal proceedings against Odebrecht S.A. arising from allegations of bribery were concluded in 2016 in the USA, as well as allegations including money laundering in Brazil reached 1st degree verdicts in 2016 and the following years, having the company agree to pay more than \$3.5 billion in penalties and executives being criminally liable. The Odebrecht case is unique in Latin America given the magnitude of the case, sum of payments and key public politicians involved, resulting in more than three presidents or ex-presidents being arrested. It has also changed the compliance landscape in Brazil, kickstarting the practice of leniency agreements and monitorships, influencing how law enforcement and other companies handle and evaluate internal controls and budget allocation for integrity programs. Nowadays, executives from companies of all sizes have to be aware that noncompliance can lead to criminal liability as well as put the existence of the company in risk. Also, prosecutors, regulators and monitors now have deeper knowledge of corruption structures, making compliance standards rise even in the absence of official action.

Odebrecht has been implementing its compliance program in the years since their leniency agreements and we will review the past and present context surrounding measures taken, bringing organizational features into the spotlight. Odebrecht publishes annually a report that summarizes the business performance of the organization in the previous year. It provides consolidated economic-financial results, presents the main achievements and most significant performance indicators, besides updated information about the macro structure of the organization.

For this chapter our dataset consists of the annual public reports published by Odebrecht between 2016–2020, the Code of Conduct, previously published research into Odebrecht's compliance management system implementation, the case files from Operation Car Wash and the Odebrecht bankruptcy filing, news agency reports, and lastly several (undisclosed due to anonymity requests) semistructured interviews conducted with current and former company employees pertaining to key leadership positions and the compliance department, former monitors and investigative authorities. Triangulation of data sources helps diminish the risk of data unreliability.

#### **3 Compliance Standards and Management Systems**

A compliance program is the means by which an organization can prevent and detect illegal conduct, and also promote behavior that encourages *compliance*  with regulations and ethical standards. The institution of compliance management systems (CMS) and integrity programs in general are part of a set of macro-institutional changes in the anti-fraud and corruption landscape worldwide (Cherepanova, 2021). These trends are commonly kickstarted by a general international agreement regarding fairer business practices, economic development, and governance (e. g. the Organization for Economic Cooperation and Development's Anti Bribery Convention, 1997, see Tourinho, 2018; Spahn, 2013), alongside a reaction to an impactful event. The Enron fraud scandal and company collapse in the early 2000s, for example, affected thousands of stakeholders and gave way to several compliance demands. With fake holdings and off-the-book accounting, by means of special purpose entities (SPEs), they hid huge amounts of debt from investors, creditors, and regulators (Segal, 2021). Consequentially, in order to promote accurate financial reports and punish corporate fraud, the Sarbanes–Oxley Act1 was enacted, alongside higher levels of ethical scrutiny from the Financial Accounting Standards Board (FASB). To avoid a similar disaster (Enron's shares plummeted from \$90.75 to \$0.26),2 company boards of directors became more aware of the importance of independent monitoring and auditing.

Another key event in the history of compliance was the 9/11 terrorist attack in 2001, which made stricter money laundering and *KYC* (know your client)

<sup>1</sup> Also known as the SOX Act of 2002 and the Corporate Responsibility Act: H.R.3763— Sarbanes-Oxley Act of 2002, 107th Congress (2001–2002) : "Title I: Public Company Accounting Oversight Board—Establishes the Public Company Accounting Oversight Board (Board) to: (1) oversee the audit of public companies that are subject to the securities laws; (2) establish audit report standards and rules; and (3) inspect, investigate, and enforce compliance on the part of registered public accounting firms, their associated persons, and certified public accountants".

<sup>2 &</sup>quot;Enron shareholders look to SEC for support in court" (WEB). The New York Times. May 2007. Retrieved October 8, 2020.

legislation a key focus, in order to undermine the financing of terrorism. AML (anti-money laundering) tactics fulfill the state's primary goal of counterfinancing terrorism (CFT) and not to allow the proceeds of crime (dirty money) into the legal economy, an agenda shared by the International Monetary Fund (IMF) and the FATF (Financial Action Task Force), institutions formed by multiple countries meant to care about the damages financial crimes can have on the integrity and stability of the financial sector and the broader economy. More objectively, AML tactics serve to make sure the corporation is compliant with regulations that require them to monitor and subsequently report suspicious activity and allows them to achieve corporate goals such as protecting shareholder value, avoid penalties due to noncompliance or negligence and reduce costs and capital reserved for risk exposure (FATF, 2012–2022).

So, there are major political, social, procedural, and economic facts, as well as previous major fraud and corruption cases, that gave rise to the anti-corruption laws and frameworks, respectively being enforced, or promoted. These frameworks, or "international accountability standards", are soft law regulations, and constitute "a large institutional infrastructure designed to address the transparency expectations of society towards business" (Cherepanova, p. 224). Globalization, of course, is also one of the catalysts for the emergence of these standards, i.e. global policymaking, due to the fragmentation of regulative authorities on the global markets. That is not to say there aren't problems that arise from the number of standards in existence, such as overlapping requirements, confusion as to which should be implemented, management overload, increased costs, and a dilution of focus and impact. Despite these issues, standards such as the ISO 37001 were developed by over 80 experts with the involvement of 59 participating or observer countries and 8 liaison organizations, giving this multi-stakeholder approach its due legitimacy and credibility regarding its effectiveness, a global collaboration between public and private actors (ibid, p. 234).

These factors lead to the understanding of why companies feel compelled to invest in integrity programs not only to comply with the law—and gain legal and reputational protection against wrongdoing, but also to have a competitive edge—against companies that don't. A common problem, however, is "window dressing". The adoption of these practices as a brand improvement strategy, formally creating positions and requirements on paper without practical or encompassing implementation of these internal controls (Scherer et al., 2013). Another is the "check-off' approach, where hotlines, codes of conduct and training are also perceived as very formal technical tools, which could lead to their depreciation over time if not bounded into the mindset of executives, managers, and employees, not independent from the company's system and also not only to imbue legitimacy, but as effective tools for change (Cardoni et al., 2020).

By current standards, compliance lies under the umbrella of corporate governance, which itself is one of the three main aspects behind ESG (Environment, Social and Governance). In the investment world, an ESG investment is one that incorporates environmental, social, and governance issues as criteria in the analysis, going beyond the traditional economic-financial metrics and thus allowing for a holistic evaluation of companies.

The standard of compliance requirements in each country is set by its anti-corruption laws and international soft laws. There are the starting point principle-based standards (such as the OECD Good Practice Guidance on Internal Controls and the Transparency International Business Principles); certification standards that require external validity through independent audits (e.g. TRACE certification); reporting standards for disclosure of impact on the economy, environment and society (e.g. Global Reporting Initiative standards); and process standards, which have more practical directives into how to achieve the previous standards checklists and aspirations (e.g. ISO 19600:2014 CMS). There are also relevant aspects of how compliance programs are interpreted by different regulatory agencies.

In general, however, the Compliance Management Systems' success in Brazil depends on the following key issues, based on the Brazilian legislation (Law 12.846/2013 and Decree 8.420/2015), and Integrity Guidelines from, among others, the Comptroller General's Office (CGU, 2015), the Brazilian Institute of Corporate Governance (IBGC, 2015) and the International Organization for Standardization (ISO 37001—Anti Bribery Management Systems, ISO 19600 Compliance and ISO 31000 Risk Management):


of the company, the partners, third parties, and suppliers, measuring the likelihood and severity of potential violations, identifying mitigation actions, an effective timeline, and a responsible area/employee.


Firms specialized in Compliance or professionals from the field, when asked to create a CMS for a particular company, will do so according to a masterplan and follow particular integrity pillars. The objective is to implement effective integrity management, where there are enough prevention measures, but if infractions do occur, they can be detected and corrective actions can be taken immediately: prevention, detection, and correction. Implementation could be staged, e.g., in three phases:


There is no single formula for a compliance program, being fundamental that it be developed and adapted so as to properly and proportionally address the level of risk and the peculiarities of each line of business and each jurisdiction in which the company operates, as well as other factors unique to each organization. This flexibility also implies less completeness and comparability possibilities. There is also the caveat about empirical evidence regarding specific compliance measures or the general effectiveness of compliance programs being contested. In that regard, the literature and research in this field is still being developed and it is also complex to prove a negative, i.e., that something that never happened would have happened without compliance measures. Additionally, given the costs of organizational change and the true incorporation of these standards into daily operations, it is not surprising that many companies may prefer a formal "paper program" (Cherepanova, p. 241). Only certification standards are audited, meaning that companies will mostly only face accountability regarding this implementation if faced by market demands (such as during due diligence controls, funding rounds, etc.) or ex post while being prosecuted.

Nonetheless, compliance, being the creation of an incentive field towards business ethics and accountability, should receive political support for implementation from parties in all sides of the spectrum. For those who believe in alternative means of control for crime, or that we rush to jail, compliance provides for ways to reach agreements and focuses on prevention, rather than repression. On the other hand, to protect the free and fair market competition, to avoid monopolies and cartels from forming, to prevent slave or children labor, to make sure merchants and suppliers haven't been corrupted or have a conflict of interest, compliance provides mechanisms for control and monitoring. Regarding mainly anticorruption measures, given that corruption encompasses "hard-to-observe" illicit activities in complex degrees and ways, Davis (2019) argues that evidencebased regulation might not bring us further than judgement based regulation, if some conditions are present: difficulty to collect data on interventions or outcomes, causal inferences are difficult to draw, there is no reason to believe that same causal relationships will apply in a new context, and/or decision-makers lack the capacity to perform one of these tasks. He writes that "In these settings, feasible types of research on the impact of past interventions will tend to be of limited value in predicting the impact of future interventions. As a result, careful thought is required about whether and how to use research as opposed to judgement in making decisions about regulation, assuming the goal is to maximize regulatory effectiveness" (p. 50). This is certainly not an incentive to abandon evidence-based regulation, but an argument in favor of multi-stakeholder developed international accountability standards as the more pragmatical approach in dealing with such a complex problem (see also Almond & Van Erp, 2020). Until evidence-based research is further developed, these seem to be indeed the "best practices" possible. Current research into compliance focuses on firm level analysis (which is why this chapter will attempt to bring multi-level elements of analysis into the spotlight) and research into corruption focuses and is limited by the polarization between bad apples and bad barrel metaphors as well as by putting different phenomena all under the same umbrella of organizational wrongdoing (Chaves & Raufflet, 2022).

With the knowledge of how Compliance Management Systems work and are implemented, we will now analyze the compliance landscape inside Odebrecht S.A. before Car Wash Operation and their judicial prosecution.

#### **4 The Odebrecht Corruption Scandal and the Previous Compliance Standards**

In the company's 2014 report there is no mention of the words "compliance", "integrity" or "transparency", although it presents a set of guidelines in the form of Odebrecht Entrepreneurial Technology (TEO) and a Code of Conduct. According to the report, TEO is the basis of a corporate culture created by the founder, Norberto Odebrecht, focused on education, work and on humanistic values. Its principles, concepts and criteria provide to the members of the organization the ethical, moral, and conceptual foundations which allow them to act with unity of thought, common strategic direction, and coherence of action (Odebrecht, 2014, p. 10). The Code of Conduct had additional concepts and guidelines, which incorporated the legislative evolution, albeit being vague.

While there is a brief mention of public and privates' relations being the object of public scrutiny, the 2015 report follows the pattern of the previous year and the company stood its ground affirming their business exchanges were nothing but legitimate (Odebrecht, 2015, p. 11), even after investigations started. A message from the former President of the Board, Emilio Odebrecht, gives a description of the companies' self-image at the time: "For all this, we have had the discernment to make decisions guided, always, by the public interest, convinced that only what serves society serves shareholders. Based on the two forces that have brought us this far and will lead us to perpetuity—Trust in People and the Spirit of Serving—, we seek the convergence of tangible and intangible results, working with simplicity, humility, detachment and willingness to share, as conscientious businessmen, who do not abandon their social and environmental commitments and seek to act as partners of the State in building solutions that enable the development of the countries where we are" (p. 18). The company had, however, deeper, and illicit ties to the State.

This relationship, according to Carazza (2020), is not particular to Odebrecht, as Brazilian companies have profited from and developed relationships with government officials in order to expand and subsidize their business (in more detail, Campos, 2012). Construction companies around the world, but especially in Brazil, given its demand for infra-structure projects, became highly dependent on public contracts and only a smaller part of its revenue came from private projects. Odebrecht, even in 2018, still had 81% of its revenue derived from government contracts.3 The previous internationalization of the company also reflects this, since Odebrecht turned to Latin America and African countries, where public–private partnerships along with the same characteristics as the ones in Brazil were possible: "closing the market against international competition, exclusivity contracts with state-owned companies, and subsidized credit lines in public banks, in addition to rigged bidding documents, cartels with would-be competitors, and tax benefits" (Carazza, 2020, p. 2).

Published in 2018, Fernandes also related events in Brazil's political scandals with data from Carazza's doctorate thesis: "the sectors that contributed most to campaign financing were, in descending order, construction, food and beverages, financial, steel and metallurgy, mining, and pharmaceutical".4 However, the weight of the donors holds no relation to their participation in the GDP and the construction sector, for example, contributes less than 8% of national wealth and was responsible for 28% of campaign donations. She continues: "the discrepancy is therefore largely attributed to the impact of regulation and taxation on these sectors. The intermediation of congressmen in these activities justifies the funding and explains the frequency with which leading companies in these segments have appeared in recent years in operations such as Car Wash and Zelotes, which

<sup>3</sup> Revista O Empreiteiro, Ano LVI, Julho/Agosto 2018, N. 569, p. 92.

<sup>4</sup> Until 2015, companies were still allowed to make campaign donations.

investigated the sale of decisions in the Administrative Council of Tax Appeals (Carf)".5

According to the confession statement of a former Institutional Relations executive from Odebrecht,6 his area supported institutional agendas of critical interest with public entities and agencies, including the National Congress. He declares "it was common knowledge that the legislative support offered by political agents to companies happened, in practice, at least in exchange for contributions in election periods, when not in exchange for more immediate financial arrangements. Because of this, several political agents tried to approach me, and I selected certain agents with political relevance, who preferably exercised strong leadership in their party and in their peers, and who would have better conditions to generate positive results for the company. Odebrecht had an interest in the permanence of these parliamentarians in Congress and in the preservation of the relationship, since historically they supported projects of their interest and had the ability to influence others. To do that, they maintained a financial relationship with these politicians. Additionally, I sought to identify and support promising politicians, who, besides defending converging interests, demonstrate the capacity to exercise leadership in Congress and in their respective parties, thus becoming part of the list of strategic politicians. The payments I indicated based on this list were approved by Marcelo Odebrecht, by the presidents or directors of the respective businesses. The political agents knew the weight of my favorable opinion within the company, and I used this in my favor. Still, without wanting to evade my responsibilities, I think it is important, just for contextualization, to point out that, inside my company, other people kept their own agenda in the National Congress.". He goes on to declare that he preferred to avoid discussions in the House of Representatives because of the number of agents and interests, which made negotiations difficult, the Senate being much more preferable for his negotiations. When, eventually, he needed to deal with some deputy, he went directly to his contact person within the company, since there were many involved in this networking process.

All this was possible through *sophisticated financial engineering* in order to generate billions of dollars in slush funds and by distributing it to authorities, which made bribery the *standard operating procedure* from Odebrecht S.A. and the corruption schemes were not restricted to certain businesses from the Group, but pervaded the whole company (Braskem, Odebrecht Ambiental, etc.).

<sup>5</sup> Fernandes 2018.

<sup>6</sup> https://www.migalhas.com.br/quentes/250418/veja-a-integra-da-delacao-de-claudio-melofilho--ex-diretor-da-odebrecht

The investigations also discovered a specialized area inside Odebrecht, the Department of Structured Operations (DSO), which was responsible for keeping the "parallel" accounting, moving the money and processing illicit payments. During the period before the prosecution, compliance also popped up in some events, *but as something to be circumvented instead of applied,* one of them reported by Bloomberg Businessweek with the captivating title of: *No One Has Ever Made a Corruption Machine Like This One* (2017): the Antiguas Overseas Bank was used to funnel money, via a shell company called Klienfeld Selvices Ltd. When the bank's account was frozen due to near bankruptcy, Klienfeld and his partners tried to acquire it, but the transaction failed due to an already filed report from the bank's compliance office about suspicious activity. The solution found by DSO was to acquire 51% of the Antiguan branch of the Meinl Bank, which was turned into their facilitator for corrupt transactions. According to Odebrecht's Plea Bargain (p. 11), they utilized banks with features like strict bank secrecy and that weren't cooperative with international law enforcement, paying high remuneration fees and rates to the institutions and percentages of transactions to the bank executives. To bypass compliance inquiries and backstop the transactions, they would use fictitious contracts. At another instance, Odebrecht used an intermediary in order to open an offshore account at the Swiss Julius Bar Bank, having a Brazilian state-executive, a politically exposed person (PEP), as it's beneficiary. The same intermediary, with his European connections, managed to acquire an authorization from the bank after compliance flagged the beneficiary (Indictment N. 5036528–23.2015.4.04.7000/PR, pp. 55, 92).

During the interrogation of one of Odebrecht's main executives, it was asked if the company pursued any internal investigations about the facts that came to light while the scandal was breaking out. The executive answered that "as soon as the first notes that mentioned Odebrecht in the industrial engineering case were published in the media, we set up a totally autonomous and independent investigation committee for this purpose" (Case File n. 5036528–23.2015.404.7000, Deposition R.A., p. 36). However, this did not correspond to the reality presented by other company members. Compliance, at the time, was defended as a new company policy, in order to turn the company more professional and respected abroad, but good practices never should reach the top floor (Cabral & Oliveira, 2017, p. 11). Due to the *involvement of the senior management,* there was only a paper structure "ready" to prevent corruption. Cabral & Oliveira (2017, p. 154) documented the aforementioned internal investigation: after the accusations of irregularities in contracts came to light, the internal audit was triggered to start investigative work, but it was initiated by the same executives being investigated. The chief compliance officer and the internal committee would have their independence, but a very limited scope of investigation was set, and investigators would report their findings directly to those involved in the larger scandal in the first place. The executives encouraged that the audit be done only on the basis of public data alone, so nothing but what was known could come to light.

A recent survey-based study (PWC, 2020) shows that the corporate crime rate in Brazil declined slightly over the last two years, now with 46% of respondents alleging at least one case, very close to the global average of 47%. From the respondent enterprises, 43% (compared to 40% around the world) intend to expand their budget for their compliance programs, which indicates a consolidation of international tendencies of law enforcement in Brazil. The most common corporate crimes at Brazilian enterprises are bribery and corruption (41%), accounting fraud (40%) and asset theft (24%). Worryingly, however, is that about a third of respondents (36% compared to 29% globally) informed having received some form of bribery request and 48% believe they have lost a contract or deal due to bribery from other companies. This goes to show how early it was in 2014 for any sort of compliance framework to be present, for all companies in Brazil across all sectors, but also how needed some form of internal control was, and how the landscape at Odebrecht was especially malicious, given its combativeness when faced with the investigation, its structure to process illicit payments and its modus operandi centered around influence networks within the government.

From the interviews, we were able to extract some main takeaways about the state of compliance before and during implementation. According to more than one interviewee, the company did not create a compliance structure from the ground up, but from an even deeper starting point since the company had a previous negative attitude towards any regulatory implementations. Project managers, Directors, and other executives at Odebrecht thought the Code of Conduct was sufficient to address their needs and did not want to make changes. In the 2010s there were multiple engineering business units, without process unification, which was supposed to "preserve egos" and "avoid conflict". In this way, decision-makers had each their own space to handle business as they willed it, and those who tried to change or contest how things were done were seen as troublemakers or non-trusting, losing space and projection. It was fundamental not to create an ill situation among peers and to foster relationships. Those who actively knew about wrongdoing or suspected it, had all the incentives to leave or stay silent. Their systems were seen as self-sufficient, where external input was not desired. Odebrecht had a culture based on decentralization, trust (especially among leader and team), and favoring client's interests—based on the founding family's influence. One interviewee argues these pillars were misinterpreted, leading project managers to be able to act without supervision—also an attempt to shield executives from liability—, auditing initiatives to be seen as lack of trust and client's interests being attended even through illegal means.

Up until 2014 there was no hierarchy structure and compliance demands had to go over several positions before—if even—reaching anyone from top management. When the scandal broke out and the company took a defensive posture, top management alienated employees themselves into thinking the company was a victim being politically targeted by authorities. Odebrecht paid approximately \$788 million in illegal bribes while securing *benevolent* treatment and securing public contracts, across more than a dozen countries and undetected for at least a decade. Given the proportions of the scandal, strengthening preventive measures, which includes a structural and mindset change, is fundamental, especially if those involved have a low risk of suffering legal consequences, due to the complexity of white-collar crimes, investigation deficits or prosecutorial leniency.

#### **4.1 The Agreements and Enforcement Guidelines**

Odebrecht and Braskem settled in 2016 with the Brazilian Federal Prosecutors Office, the US Department of Justice and Switzerland's Prosecutors Office to pay a combined total penalty of at least \$3.5 billion.7 In 2018, they settled a deal with CGU and AGU to pay R\$ 2,72 billion over 22 years. Beyond these, another seven agreements have been settled with authorities from governments of Brazil, Equator, United States, Guatemala, Panama, Dominican Republic and Switzerland.8 Deals with the World Bank, authorities in Peru and CADE (Brazil's Antitrust authority) have been settled only recently, in early 2019.

Collaboration agreements were also concluded with 78 of their executives. At the time, the DOJ classified the agreements as "the largest foreign bribery case in history" for violation of the FCPA (ibid.). In Brazil, the agreements became known as the "end of the world whistleblowing" (Rossi, 2016) for its national and international reach and for the implications to the political system. For its collaboration with American authorities, Odebrecht received a 25% reduction of the bottom to the applicable U.S. Sentencing Guidelines fine range.

<sup>7</sup> Official note from the DOJ at: https://www.justice.gov/opa/pr/odebrecht-and-braskemplead-guilty-and-agree-pay-least-35-billion-global-penalties-resolve.

<sup>8</sup> Chapter 11 lawsuit, p. 2 e 12.

To understand why Odebrecht was also and fundamentally prosecuted by American authorities, one needs to understand that faced with the need to seek capitalization alternatives, without resorting to loans, many companies find in the stock market a low-cost form of capitalization. This option entails that companies whose shares are listed on the stock exchange adapt to Corporate Governance standards and codes, adapting their ethical posture to be consistent with international standards, and in the case of Odebrecht it meant that by having their shares exchanged in the NY Exchange, they were also under the jurisdiction of the American justice system. U.S. federal prosecutors enjoy broad discretion, including being able to halt or defer prosecution, if the defendant complies with their agreement demands, which could entail compliance and self-reporting obligations, as was the case for Odebrecht (Scollo & Winkler, 2017).

As Brazilian prosecutorial authorities also came to the last rites of their leniency agreement with the company, Odebrecht S.A. signed an FCPA Plea Bargain with the Department of Justice (DOJ), in 2016. For the deal, the DOJ considered Odebrecht's "failure to voluntarily disclose the conduct that triggered the investigation; the lack of an effective compliance and ethics program at the time of the conduct" (DOJ FCPA Plea Bargain, Attachment C Details the Corporate Compliance Program, pp. 56–74) and required Odebrecht to commit to a series of obligations under the agreement, the ones most important in the context of compliance being:


. risk-based due diligence and compliance program towards third parties, including M&A targets;

An independent monitor was appointed to overview Odebrecht's compliance with the terms of the deal over a temporary period of time, having the company guarantee his access to all documents and resources that may be needed for said assessment, reporting directly to the DOJ and taking into account the commitment of the board of directors and senior management, what can be described as a very broad set of powers, especially since the company waived certain legal rights and signed a "muzzle clause", preventing the company from contradicting the Statement of facts from the agreement (Scollo & Winkler, 2017).

The leniency agreement with Brazilian authorities, for its own worth and given the instrument's novelty in Brazil, consolidated it to serve multiple functions. Particularly, public interest is prioritized, beyond provisions on effectiveness and broadening investigations, to preserve the existence of the company and the continuity of its activities to make sure reparations are met and to ensure the effectiveness of the company's integrity practices (Pimenta, 2020).

Furthermore, from the 78 executives who entered the individual plea bargains, 51 had their employment terminated and 26 received a demotion to non-managerial positions besides anti-corruption and business ethics training (Odebrecht Annual Reports, 2018–2019).

#### **4.2 The Implementation of Compliance Programs**

Compliance areas, which take care to prevent companies from breaking laws and regulations, are no longer a novelty in the corporate world. The question is to how to make sure that they are actually infused into the fabric of the business model. Bureaucratic demands aren't always adequate or fair and need to be revised on a constant basis. It is understandable if long term insiders think some bureaucratic hurdles are excessive, but not something to be skipped or overcome in order to preserve their business. At Odebrecht, both soft law and binding legislation were seen as a nuisance, and illustrates the corporate culture that the monitors appointed by the US Department of Justice (DOJ) and the Federal Public Prosecutor's Office (MPF) came across when they arrived at the company four years ago. Later, however, the monitor appointed by the Public Prosecutor's Office, Otavio Yazbek, a lawyer and former director of the Securities and Exchange Commission, CVM, presented his report to prosecutors and gave a statement that Odebrecht no longer is the same and that its systems and rules are ready to detect criminal operations. According to his report, the company centralized previously loose processes, created compliance and internal audit departments, and unified purchasing systems. It hired independent advisors for the companies, dismissed almost all whistleblowers and members of the old guard, alongside removing owners Marcelo and Emílio Odebrecht. Over these four years, Odebrecht spent 68 million USD to make the changes required by the agreement deals (Goulart, 2020).

The difference between the annual reports from 2014–2015 and those published from 2016–2020, when they adopted the Global Reporting Initiative (GRI) directives is notable. GRI is an international organization that has developed one of the most widely used sustainability reporting models in the world, to confer organizational transparency and enable companies to measure and communicate their performance in the economic, environmental, and social dimensions. While there were zero mentions about "transparency" in the 2014 report, the 2016 report finds the term being mentioned 26 times. There are also 34 mentions of "ethics", 17 about "integrity" and 20 about "governance". The annual reports from 2016 forward demonstrate how the company applied the mandatory changes:


(continued)

<sup>9</sup> Collective Actions, Third Party Compliance, Communication and Training, Risk Assessment and Controls, Policies and other Guidelines, Governance and Compliance, Ethics Channel Management, Risk Monitoring, Risk Remediation, Disciplinary Measures.


(continued)

Another notable difference, Odebrecht had climbed 47 positions in the ranking of best practices in transparency of the information disclosed, according to the Transparency in Corporate Reporting report presented by Transparency International, moving from 97th position in 2013 to 50th in 2016. The study, which was based on 2015 data, and evaluated the disclosure information of the 100 largest multinationals in 15 emerging countries. Specifically in anti-corruption programs, Odebrecht went from 0 to 77%, ranking 22nd, well above the overall industry average of 43%. The study also highlighted the points in which Odebrecht needs to improve, such as adopting policies that explicitly prohibit facilitation payments and prohibit political contributions in all countries in which it operates. According to the company, the Policy on Compliance with Acting Ethically with Integrity and Transparent Actions, approved in 2016, was created to addresses these two aspects, among others.

Additionally, Silva and Monteiro (2019) published a study about Odebrecht which had three objectives, namely: to present the history of corporate governance and critically analyze the compliance structures created (*sufficient autonomy, resources,* etc*.);* to verify which were the actions of the top management for sponsoring/reinforcing the implementation of the compliance program *(tone at the top);* to identify the risk methodology used by the organization *(risk assessment).*  They used internal policies, external documents and interviews with employees to make their evaluation.10 The study found the compliance structures to be sufficient, having "verified the support of the management in providing a compliance structure, with professionals of the appropriate hierarchical level, resources, and direct access to the highest governance levels of the company" (p. 431), with the exception of the quota for independent advisor seats, which weren't filled according to the statute. Top management support was also stated to be present and acting. For risk assessment, Odebrecht's policy was based on Enterprise Risk Management Framework of COSO, and on ISO 31000:2009, considering each business has a different risk profile, although there is an inherent risk to all of them: violating anti-corruption laws. The risk portfolio for the holding company will tend to be more simplified since it does not have any operations. At the time of the study, both the tools and the controls that would evidence the risk management process would still be defined and implemented by the organization and couldn't be analyzed, leading to the obvious conclusion that they were still insufficient.

According to De Araujo (2020), in 2017 the Department of Justice issued the Evaluation of Corporate Compliance Programs (ECCP), soft law guidance to help authorities to evaluate the effectiveness of compliance programs, considering the Justice Manual's three question-pillars for remedial actions: "1. Is the corporation's compliance program well designed? 2. Is the program being applied earnestly and in good faith?; 3. Does the corporation's compliance program work in practice?" (p. 9). Odebrecht was assigned an independent monitor by the DOJ and was also monitored directly by Brazilian authorities (CGU). The U.S. monitorship was set to end Feb. 2020 but was extended for another 9 months given that Odebrecht had not settled its financial dues to the monitors and could possibly have failed to fulfill obligations under the plea agreement (Sun, 2020). On the Brazilian side, CGU recommended several—publicly undisclosed—improvements and required an action plan, including ISO 37001 certification. Remarkably, due to Odebrecht's 22–year payment plan, CGU will perform continued monitorship for the whole period. In general, however, the author—a public prosecutor—mentions "it is possible to perceive how corporations have improved its compliance programs, from cosmetic compliance in the first years to sophisticated programs within the last two years".

Farias et al. (2019) published a similar case study in order to analyze the process of restructuring compliance in a construction company involved in a

<sup>10</sup> Deliberação 15/16, Política da Odebrecht S.A sobre Conformidade com Atuação Ética, Integra e Transparente; Resoluções do Diretor Presidente da Odebrecht S.A. 01/17 e 02/17; Questionários CGU e anexos; CC-ODB, R-Conformidade e o Líder de Conformidade da Odebrecht S.A.

corruption scandal. Although not disclosed, it selected Odebrecht S.A. for its evaluation. The research concluded that the actions taken by the company have proven to be adherent to the guidelines established by Federal Decree 8420/ 2015, and by ISO 19600/2014 and ISO 37001/2016 standards, although it was not yet certified by these standards. The points not met by the company were the performance of audits to verify the adherence of the practices developed, the non-identification of treatment of eventual non-conformities identified, and the approach and critical analysis performed by top management regarding compliance. Risk management also presented deficiencies to be highlighted, such as lack of methodology and the non-application of the due diligence when hiring specific professionals. A limitation of this study is the fact that the data collection was carried out in an intermediate phase of the implementation of the Compliance Management System.

Our data from the interviews showed further development in the day-to-day operations, certification requirements and risk management practices, although some critical factors still need be addressed. Since the agreements with the authorities and the start of the monitorships, Odebrecht also had a high turnover in leadership, going through seven different administration council compositions and different conformity committee leaders. By 2021 a lot of the prejudice against the compliance department had disappeared—the dismissals helped—and the implementation of the compliance program was taken seriously to the point of being able to be verified by U.S. and Brazilian authorities. In 2020 the DOJ certified that Odebrecht had a robust program and now the Brazilian Public Prosecutors Office (MPF), which was an important achievement. Odebrecht, now rebranded as Novonor, was also able to receive the ISO 37001 anti-bribery certification, having implemented all of its main requirements.<sup>11</sup>

According to interviewees, some old timers, which could represent resistance, are now a minority and don't have the means to oppose an overwhelmingly compliant culture, which is corroborated by internal surveys conducted by the monitors. Another contributing factor is the fact that currently 50% of the companies' employees joined after Car Wash Operation started and around 45% are under 35 years old, i.e. less accustomed to the "old ways" of doing business.<sup>12</sup> From that minority, however, some are still in leadership positions and have been through the process of normalization, rationalization of wrongful behavior,

<sup>11</sup> Novonor, former Odebrecht, announced (6/23) that after an audit by an international company, it was awarded with the ISO 37001—Anti-Bribery Management System Certificate, a milestone in the Group's transformation history.

<sup>12</sup> The data points were presented by the interviewees and cannot be corroborated by the authors.

and socialization inside the company for many years, still presenting different forms of resistance, such as asking for a dismissal of new implemented tools or questioning their need. Another form of influence from top tenured leadership is cutting down on communication processed between teams and opening discussions for outside input. One of the interviewees narrates: "Do you think he will invite you to a meeting that he knows may have a potential conflict? No, because he knows what you are going to say and he doesn't want to hear it. And I say that compliance is a structure that should not exist, because compliance is in day-to-day life. It is in your capacity to make decisions, in your capacity to perceive a dilemma. We should exist in a transitory way, you know, until that is absorbed by people. For them to be able to perceive the dilemma, for them to perceive how much they are rationalizing, for them to perceive how much the group is influencing them". A key insight into how different it is to implement a checklist and absorb compliance into corporate culture.

Another crucial complaint made was that due to the speed with which the leniency agreements were handled, leaders with conflict of interest were still plenty, leading to a company to make deals with executives in order for them to collaborate with the justice department. A company which later downsized from almost 200 thousand employees to less than 40 thousand and declared Chap. 11 Bankruptcy was until recently still paying hefty sums to crooked former executives.

Additionally, but more successfully, we heard reports about how where previously there was no control, but an incentive to informal institutional relationships, now there is a three way sector of controlling the interaction between public and private sector: the demands and meetings are set as a group, represented by the representative entity in the construction sector; there are new policies in place regarding illicit payments; and most importantly, there is strict financial control into expenditure and diminished possibility for project leaders or managers to find financial sources to make any type of facilitation payments. The cultural shift is also related to the Action Plan from employees, not only communication and training measures adopted by compliance policy. The Action Plan is tied to the macro and micro qualitative and quantitative goals of employees, which are modeled by their career plan and financial bonifications, leading the implementation of the compliance measures—included as goals—to be discussed, evaluated, and absorbed by all on a constant basis.

#### **4.3 Discussion**

Odebrecht filed for Chap. 11 Bankruptcy or rehabilitation bankruptcy to reorganize its debts in June 2019.13 According to the suit, Odebrecht's involvement in the corruption scandal uncovered by Car Wash operation brought several economic setbacks to the company, including the lack of access to sources of financing and the possibility to pitch and secure new projects in Brazil and abroad. Moreover, several contracts have been suspended or rescinded and assets blocked. Furthermore, the group's investment in remedial measures and Compliance have also taken a toll on their budget but have been improving given the successful creditor negotiations.14

Rebranded to Novonor, Odebrecht could be considered a rehabilitation success case. In light of the agreement, Odebrecht agreed to terminate the employment of 51 individuals who participated in the misconduct, to discipline and train further 26 individuals (suspensions, penalties, demotions) involved, to create a Chief Compliance Officer (CCO) position that answer straight to the administrative council, adopt heightened controls and anti-corruption compliance protocols, to allow double independent monitoring and increase the budged and human resources for compliance, amongst other more specific measures, all of which happened.15 In 2018 the holding company replaced most of its board of directors and Emilio Odebrecht stepped down as chairman after almost 20 years on the board. During this move, it has also been established that members of the Odebrecht family will no longer be eligible for the position. Odebrecht's 2018 compliance budget was \$20.45 million, compared to \$3.19 million previously (Russo, 2018). Among compliance and governance measures, Odebrecht has implemented or started implementing all of the DOJ's 10 hallmarks and Brazil's Transparency Ministry 17 recommended initiatives, as well as a global advisory council with national and international members. One of the boldest initiatives is that up to 30% of executives bonus payments are now conditioned to reaching compliance targets (Estadão, 2019). Each business unit now should have its own board of directors while the holding provides uniform governance and guidelines. At least a fifth of board members will be independent and hired by external consultants. Other promising initiatives are outsourcing whistleblower hotlines and

<sup>13</sup> File noº 1050977-09.2019.8.26.0100, 1st Court of Bankruptcies and Judicial Rehabilitation from the State of Sao Paulo.

<sup>14</sup> Chapter 11 lawsuit, pp. 10–14.

<sup>15</sup> FCPA Agreements, pp. 3–11.

strengthening due diligence checks of suppliers. If Odebrecht manages to rehabilitate itself economically, it could be well on track to change its previous incentive system that fostered deviance.

One could argue that corporations previously involved in scandals and who had to go through monitorship periods and pay hefty fines might now be mostly dealing with the reputational damage and stigma than with a need for cultural change. After Siemens Scandal had its offices raided in 2006, by 2009 the company had invested a heavy amount into the compliance department and ensuing investigation and changed its compliance systems. Around 1750 interviews with Siemens employees were conducted and more than 14 million documents reviewed. The company employed more than 500 compliance officers in their global operations, renewed their policies, created internal reporting lines, a compliance hotline, etc. For their main concern, to deal with scrutiny into potential partners, they developed the IT-based Business Partner Tool (BPT).

Although strengthened and better prepared to deal with illicit conduct, the company did not become impervious to it. Recently, the monitoring reports commissioned between 2009–2012 were disclosed and show that US and German authorities found \$1 billion of bribes paid to foreign government officials in return for business. According to the report "Siemens was allowing some business partners to sidestep vetting through the BPT, thus allowing what were, according to Siemens's own definition, "high-risk" entities to conduct the company's business in China without proper due diligence". The employee responsible for pointing out the compliance issues in China was fired in 2010 (Knight, 2021, based on a report made by 100R.org).

Coming back to Odebrecht and the characteristics of what is an effective compliance system, we could review some of the points raised during the previously published works and the interviews:

#### a) **Institutional Relationships and Borderline Positions**

Borderline positions lie between the organization and the environment, which makes them susceptible to disruption. They are considered "semi-external institutions" (Luhmann, 1995, 229; Bergmann, 2016, p. 14). In the company they structure make it possible to undermine formal rules and exploit control gaps because of the vague definition of their tasks (acquire new business, handle relations with government representatives or customers) and because of their relative distance from internal organizational processes (third-party collaborators, intermediaries, different offices around the world). Bergmann (2016, p. 16) cites a Siemens executive who reported even after the scandal that in sales in particular it was still hardly possible hardly possible to create global standards of behavior, because in the treatment of business partners they had to be differentiated according to country and custom (Lamparter, 2007).

In January 2019, when Jair Bolsonaro's government arrived in Brasília, Odebrecht hired a new director of institutional relations, thus reopening the Brasília office (in the capital) that had been closed at the height of the Operation Lava Jato process. Having an office in Brasília is fundamental for a large company, especially when it is a construction company that thrives on public works, which cannot give up having dialogue with governments or politicians. According to the magazine *Veja* (Goulart, 2020), one of the reasons that weighed in favor of hiring the new director was that he had good relationships with army men, which are now in key positions in the government. All in all, hiring someone thinking about their ease to push an agenda is not uncommon, but it reminds us of Odebrecht's old culture of keeping themselves close and having access to power.

Of the 77 executives who turned collaborators, most of them are either institutional relations directors or company/sector leaders, such as CEOs or Project Directors, since Odebrecht had a very decentralized operating structure regarding their projects. The answer to the dilemma of institutional relations and borderline positions at Odebrecht is by control of the cashflow, accounting and financial aspects of the company. If employees can't find, hide, scatter or produce the money, they can't make facilitation payments. This brings up another aspect of compliance management systems and very discussed in the field of organizational deviance, structural incentives.

#### b) **Structural Incentives**

Google was famous for having the motto "Don't be evil" in the preface of its Code of Conduct. However, instead of believing in free-will or pursuing ethical standards or expecting moral behavior from employees, one of the possible solutions is to take the choice away from them. Instead of "don't", we move to a system of "can't be evil". And that is what the financial scrutiny put in place by Odebrecht could achieve.

Again, using Siemens as an example, even though all the changes had been made internally and the CMS implemented, the company still had very aggressive growth and expansion goals. This resulted in pressure to perform, increasing internal competition and heightened expectations of economic efficiency (Bergmann, p. 15). This pressure creates a conflict, because it can be hard to achieve the economic goals without resorting to "alternative" methods, making compliance concerns go out the window. For Siemens, illegal bribery of public officials for the purpose of acquiring contracts was regarded within the Group as an efficient response to this conflicting incentive constellation (ibid).

Odebrecht's idea to include compliance targets into executives' bonus payments (up to 30%) and adding macro (to be achieved by all) and micro (to be achieved individually or by sector) compliance goals into everyone's Action Plans leads to a general and specific positive incentive to follow through with all measures and policies implemented, be it related to communication and training or risk assessment.

#### c) **Power Struggles and Corporate Governance**

Corporate Governance relates to the allocation of power among the owners, board, management and shareholders, and encompasses actions taken by management from the most senior levels to all administrative instances (IBGC, 2015). The objective is to create an efficient set of mechanisms, internal and external (incentives and monitoring) that aim to harmonize the relationship between ownership (shareholders) and management. One of the ways to do is to create an administrative council.

This type of structure arose from the need to reconcile the interests of company partners and those responsible for corporate management. Boards are the consequence of the professionalization of business management. The shares of a company, when traded on the stock exchange, mean that an organization has many owners, and the management of operations is in the hands of other executives. When it comes to family businesses, decision-making can be complicated, as it is difficult to separate personal intimacy and interests from what is best for the company.

Therefore, the administrative council is the organizational structure that serves to act in the alignment of interests between the executive management and the owners of the company, being the link between them all. Odebrecht is a family company. It turned multinational in the 1980s and later an open company with shares in the stock market. Marcelo Odebrecht was a third generation CEO. Odebrecht's former struggles are related to the complete dominance of the Odebrecht family over the families' business practices, their ties to power and the lack of supervision over their actions, choosing to grow "by any means". Having removed Emilio and Marcelo Odebrecht from key positions, diminishing their influence and voting power, as well as introducing independent positions in the administration council are all favorable marks of a more professional direction and conformity orientated company.

#### d) **Effective Communication and Cross-Functional Collaboration**

According to some of the interviewees on our dataset, an important factor in any organization is transparent communication, allowing access between sectors and information flow without penalties for divergence, creating an environment in which employees will feel comfortable raising their hands and saying "Look, there's something strange here" and not be rejected by the group. A facilitating factor allowing for this environment to foster and communication to take place is an integrated approach into CMS supervision and implementation (Cardoni et al., 2020). Multitasking teams dealing with internal and external stakeholders also demand a multi-competent leadership (Stiles & Uhl, 2012). "Process owners (marketing department, CFO, production department, etc.) are no longer the unique parties responsible for the risks affecting their areas because today world complexity makes it difficult for process owners to identify all the risks. That is why companies must enable synergies between process owners and the anticorruption supervisor (FN1). Together they can define the key risks and response strategies. The anticorruption supervisor establishes the best types of control tools for monitoring the risks in the process owner field and shares the risks with the process owners; they become not only a controller but also a co-owner of the risks" (Cardoni et al., p. 1182). If an anticorruption mindset is to take place and a corporate cultural shift happen, a cross-functional integration would benefit this movement greatly. At least where top management is concerned, Odebrecht does not seem to have adopted this approach yet, with lack of communication and division of competencies being some of the issues raised.

#### **5 Conclusion**

This chapter presents a case study approach to understand compliance system's features and implementation challenges inside a company which has previously been prosecuted for wrongdoings. Our results provide some theoretical and practical insights to the literature of the field. This allows us to explain, in this particular context, how anticorruption practices are integrated into processes and structures and how is it handled and perceived by management, highlighting important factors for a successful system implementation.

Rebranded to Novonor, Odebrecht could be considered a partial rehabilitation success case. In light of the agreement, Odebrecht agreed to terminate the employment of 51 individuals who participated in the misconduct, to discipline and train further 26 individuals (suspensions, penalties, demotions) involved, to create a Chief Compliance Officer (CCO) position that answer straight to the administrative council, adopt heightened controls and anti-corruption compliance protocols, to allow double independent monitoring and increase the budged and human resources for compliance, amongst other more specific measures, all of which happened. In 2018 the holding company replaced most of its board of directors and Emilio Odebrecht stepped down as chairman after almost 20 years on the board. During this move, it has also been established that members of the Odebrecht family will no longer be eligible for the position. Odebrecht's 2018 compliance budget was \$20.45 million, compared to \$3.19 million previously. Among compliance and governance measures, Odebrecht has implemented or started implementing all of the DOJ's 10 hallmarks and Brazil's Transparency Ministry 17 recommended initiatives, as well as a global advisory council with national and international members. One of the boldest initiatives is that up to 30% of executives bonus payments are now conditioned to reaching compliance targets. Each business unit now should have its own board of directors while the holding provides uniform governance and guidelines. At least a fifth of board members will be independent and hired by external consultants. Other promising initiatives are outsourcing whistleblower hotlines and strengthening due diligence checks of suppliers. If Odebrecht manages to rehabilitate itself economically, it could be on track to change its previous incentive system that fostered deviance. A worrying factor will still be incentives arising from the public sector and the characteristics of the construction industry (handling bidding competition). It is possible that in the future, taking the state-of-the-art approaches recommendations that also take sustainability into account (Asif et al., 2011; Cardoni et al., 2020), compliance implementation will be further developed from the check-off formal approach into a more integrated seamless system. Other than cross-functional collaboration, transparent communication, structural incentives and the handling of borderline positions and power struggles, technology will also play a huge role in allowing for system transparency. Monitoring around the world should change from quarterly or annual controls into an ongoing continuous tool, by the use of IT tech and Big Data, meaning that sampling methods could be a thing of the past, allowing us to "question the underlying assumptions and governing principles of the current anticorruption model, thus leading to its modification" (ibid, p. 1183). A system capable of adapting to particularities of the business, sector and country will always be preferable and have more chances of lasting.

#### **References**


June 8. https://www.bloomberg.com/news/features/2017-06-08/no-one-has-ever-made-acorruption-machine-like-this-one. Accessed 3 March 2022.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Epilogue**

## **Judicial Institutions and Institutional Corruption: Some Lessons of Operation Car Wash for a Further Research Agenda**

Conrado Hubner Mendes and Vanessa Elias de Oliveira

#### **1 Introduction**

The issue of corruption, which has always been present in Brazilian politics, has taken on a new centrality since Car Wash Operation (Lava Jato). But the operation didn't just shed light on that issue. It itself adopted shady methods to deal with the cases it was investigating. The most emblematic event occurred when Sérgio Moro, then the Lava Jato judge, made public the unauthorized interception of Lula and Dilma's phone calls. According to Judge Pizzolatti, of the Federal Regional Court, "the investigations and criminal proceedings of the socalled Operation Car Wash constitute unprecedented cases, bring unprecedented problems and demand unprecedented solutions". Moro carried out an illegal act, but the illegality was cleared by his own peers on the basis of an unjustifiable exceptionality within legal parameters.

The intimate affairs between judge Sérgio Moro and federal prosecutor Deltan Dallagnol in the Car Wash Operation, among other almost normalized practices in the justice system, help raising a series of relevant questions to critically analyze the performance of judicial institutions in the struggle against large-scale corruption. The central issue that arises from this series of events is *institutional corruption.* Very little is said about such kind of corruption practiced within the

C. H. Mendes (B)

University of Sao Paulo, Sao Paulo, Brazil e-mail: conrado@hubner.org.br

V. E. de Oliveira Federal University of ABC, Sao Paulo, Brazil e-mail: vanessa.oliveira@ufabc.edu.br

© The Author(s) 2024

M. E. Trombini et al. (eds.), *The Fight against Systemic Corruption*, Organization, Management and Crime – Organisation, Management und Kriminalität, https://doi.org/10.1007/978-3-658-43579-0\_10

279

public administration—and, more specifically, within the organs of the justice system. It existed before Lava Jato, for sure, but the operation exacerbated it.

We seek to discuss the concept of institutional corruption with a specific focus on the Judiciary and the Public Prosecution Service. We assume that the wide discretion of judges and prosecutors, and the lack of external accountability mechanisms and transparency about their actions and behavior, facilitate institutional corruption within the justice system. There are scarce institutional instruments for preventing or punishing them (Kerche et al., 2020).

In addition to the fact that the Brazilian judiciary has no external control, as Kerche et al. (2020) discuss, Lava Jato had become, during its heydays, the shield of the "magistocracy" (Mendes, 2023), protecting it from any attack on its illegal, immoral and corrupt acts. The National Council of Justice (CNJ), an internal control body, left the cases against Moro unresolved. This case is just one of the examples of what we will be dealing with in this epilogue.

To do this, we start from the concept of "judicial corruption" presented by Gloppen (2014), for whom it includes "all forms of inappropriate influence that may damage the impartiality of justice, and may involve any actor within the justice system, including lawyers and administrative support staff" (Gloppen, 2014, p.69).

This is our starting point, but not the end. The international theoretical debate on the subject, which is still not widely disseminated in Brazil, will be essential for our subsequent analysis of the types of judicial corruption seen in the country, especially in the case of Lava Jato. Based on this, we present a typology of judicial corruption we can detect in Brazil and certainly elsewhere. As we argue, there are two distinct types and variations within them: judicial corruption *for institutional purposes* and judicial corruption *for individual purposes.* We then take an exploratory look at the categories presented in the Lava Jato case. We conclude with some final considerations about the implications of judicial corruption for the quality of Brazilian democracy.

This epilogue attempts to cast some new light on an important research agenda that is both conceptual and empirical. It requires theoretical elaboration that informs the analysis of a number of practices that still lack the proper lenses to be institutionally, ethically and legally problematized.

#### **2 Institutional Corruption and Judicial Corruption**

Corruption can be defined as the inappropriate use of public resources for private gain (Rose-Ackerman & Palifka, 2016). This definition, however, looks at corruption as illegality. Another conception of corruption is one that goes beyond the idea of "corruption as illegality", but understands it as "corruption as the breach of duty", which includes improper use of authority, deviation from accepted or formal norms, lack of integrity in the conduct of public duties, including both intentional actions and inactions—when failing to do what should be done (Underkuffler, 2013, p.15).

But since theories that see "corruption as a breach of duty" disregard cases in which the established norm is unjust, and therefore breaking it may be the path found by the bureaucrat to achieve more justice, without the intention of misuse or personal gain, it has been criticized for not being able to capture the essence of the act of corruption. Thus, some theorists added to the debate the idea of "betrayal", "abuse of power", "exploitation" (Underkuffler, 2013, p.21).

These are important concepts for understanding some of the types of judicial corruption we are dealing with here, since many of the corrupt acts that take place in the judiciary are not just "breaches of duty", but involve, in essence, an abuse of power.

Moving on in the discussion of corruption to the specific topic of "institutional corruption", Thompson (2018) differentiates it from individual corruption (Rose-Ackerman & Palifka, 2016), as well as from the structural corruption seen in Latin American countries (Acemoglu & Robinson, 2010). The author questions why institutionalists look more closely at the issue of institutional corruption. According to him, firstly because understanding it can be a way of gaining a deeper understanding of the very institutions in which corruption takes place. Secondly, it reinforces the need to look at corruption beyond individual gain, also considering the issue of conflict of interest. Finally, from an institutionalist conception of corruption, it is possible to better understand corruption that is not individual, which is generally more difficult to capture. According to the author,

"Because it is so closely connected to the legitimate procedures and practices of the institution, its agents often are not seen, and do not see themselves, as participating in corruption at all." (Thompson, 2018, p.3).

This is the inquiry we intend to develop here, looking at institutional corruption within the judiciary, generally not treated as "corruption". We are not going to look at the "varieties of institutional corruption" listed by Thompson (2018), but discuss judicial corruption from a deductive analysis of reality, returning later to the theoretical debate (on another occasion).

For the purposes of the debate here, it is worth pointing out that, like Corcioli Filho (2013), we do not take the term "corruption" in its strict criminal sense, but as an unrepublican practice, undue because it corrupts the democratic principles of zeal for public affairs. In the author's words,

"(...) if the term corruption is taken not in a way strictly linked to the respective criminal types, but rather in the sense of something distorted from its announced attributes (republican) and objectives (democratic), perhaps it is not a mistake, or at least an exaggeration, to observe aspects of corruption in the bowels of a Judiciary that is still very much tied to the owners of power. It is in this sense, then, that it seems pertinent to speak of the corruption of the Judiciary." (Corcioli Filho, 2013, p. 434).

In this context, analyzing the case of the judiciary, we can say that judicial corruption goes beyond the payment of bribes to judges, which would be the private appropriation of public resources—although this is the case we first imagine when we talk about corruption within the justice system.

As Siri Gloppen contended, such corruption does not only include "all forms of inappropriate influence that may damage the impartiality of justice", like even petty corruption, such as bribery, it actually hinders access to justice, since for the poorest the amounts can be prohibitive, incorporating a class bias into the justice system (Gloppen, 2014). Judicial corruption generates damage that goes beyond unequal access to justice.

Another form of judicial corruption is political influence. This can involve illegal aspects (bribes, blackmail, threats), but it can also occur through the influence of other powers over the judiciary, for example through the appointment of judges to specific positions, the definition of criteria for career promotion or the regulation of financial gains, including salaries and benefits (Gloppen, 2014, p. 71).

Finally, Gloppen points out at a kind of influence that does not come from politics, but from the internal hierarchical structure of the judiciary:

Such influence may be the result of direct pressure from superiors; more subtle incentives based on judges' anticipation that a 'wrong' decision in an important case could have career consequences; or selective allocation or cases to judges who are likely to rule in a particular manner. Besides, internal procedures can be misused to limit individual judges' ability to voice criticism, for example by refusing dissenting judgments. (Gloppen, 2014, p. 72).

Analyzing the causes of judicial corruption, Li (2015) states that a number of institutional factors are used to explain judicial corruption in China:

"the most frequently mentioned factors are insufficient salaries for judges (Zou 2000; Li 2002), insufficient funding for courts (He 2009), insufficient legal training of judges (Zou 2000), local protectionism (Zou 2000), and lack of judicial independence (Zou 2000; Li 2002; Henderson 2007) and accountability (Gong 2004)" (Li, 2015, p. 849).

Regarding the "low salaries", the author states that this does not seem to be a plausible explanation for the case he analyzed (from Wuhan province), given that these are officials at the top of the judicial hierarchy, who receive high salaries and privileges that are not available to most other bureaucrats. The same can be said of Brazilian magistrates, although we do not intend to discuss here what explains judicial corruption.

Judicial corruption is therefore seen as the result of a set of factors, both internal (institutional and cultural) and external (social and political).

A series of measures can be taken to combat, prevent or punish institutional corruption in justice system bodies. Some measures can be taken to reduce political influence on justice system actors. We will not discuss these measures here, and some studies have already explored this topic (e.g. Gloppen, 2014). Nor do we adopt an anthropological perspective to understand corruption, as in Ferreyra (2017). We are not interested in understanding the cultural dimensions of this phenomenon. On the other hand, like Ferreyra (2017, p.143), we understand it as a social process, not an isolated or infrequent event. It is a process that is deeply present in the Brazilian judiciary and with serious consequences for the performance of justice, especially in relation to the issue of impartiality, according to the definition used explicitly here.

What is important to us, in this initial approach to the subject, is to make it clear that Lava Jato not only sought to combat political and corporate corruption, but also made judicial corruption explicit, a subject that is little discussed in Brazil.

#### **3 Judicial Corruption in the Brazilian Justice System**

*'if most people who take bribes go to hell, I think that no judges will be let in heaven'.* 

The phrase above, uttered by an actor in the Indonesian justice system (Butt & Lindsey, 2010, p. 190), exposes the type of corruption most commonly known within the justice system: the payment of bribes to judges to serve the interests of litigants. However, this is not the only or main form of institutional corruption in the judiciary, as mentioned above.

Based on the concept of institutional corruption, we can differentiate between certain types of corruption that occur not only in the judiciary, but in the justice system as a whole. This is based on an analysis of the unrepublican practices found in the Brazilian judiciary, here called "judicial corruption"—in its noncriminal sense. The types of judicial corruption can be grouped into two broad categories: judicial corruption *for institutional purposes* and judicial corruption *for individual purposes.* What drives the actors in the justice system is the institutionalization of corporate gains, in the first case, or personal gains, in the second. Both manifest themselves in different ways, referred to here as specific subtypes of judicial corruption.

It is important to clarify that the ideal-types presented are an analytical construction based on a deductive logic—from the observation of factual reality, we highlight elements that are considered central in the cases analyzed, creating the typology presented. This is an analytical tool to guide us through the complex reality of judicial corruption. Although, like the Weberian ideal types, reality is more complex than the typology is capable of expressing, these are useful categories to classify and capture specificities.

The table below classifies these ideal types of judicial corruption, as well as their specific subtypes (Table 1).

#### **4 Judicial Corruption and the Car Wash Operation**

As we contended, the types of judicial corruption listed above help us to understand several of the practices verified in Lava Jato. Based on an exploratory analysis of some exemplary cases, we demonstrated how Lava Jato not only incurred in several cases that, in our categorization, are considered as judicial corruption, but also made these practices explicit within the Brazilian Judiciary and Public Prosecutor's Office.

In particular, Lava Jato used various practices that can be taken as, according to our denomination, judicial corruption for individual ends. Let's take a look at each of them.

Firstly, the promiscuous activity. The Sérgio Bermudes law firm defended Eike Batista, a target of Lava Jato. Gilmar Mendes' wife works for this firm, which is why the justice would be prevented from judging cases related to it. Even so, Justice Gilmar Mendes granted a writ of habeas corpus to the businessman,


(continued)



**Table**

who did not consider himself prevented from judging the case. According to the judge, there was no impediment in the Brazilian Code of Criminal Procedure because there was no direct relationship between Bermudes' office and the habeas corpus, which was in the criminal sphere (and not commercial and labor, which are handled by the office) and presented by another law firm. Article 144 of the New Code of Civil Procedure states in its item VIII: "In which a client of the law firm of his or her spouse, partner or relative, appears as a party, even if sponsored by a lawyer from another firm".

The then Attorney General of the Republic (PGR), Rodrigo Janot, asked the Federal Supreme Tribunal (STF) to consider the justice recused. This was the PGR's first request for the impeachment of a STF justice. The request was judged by justice Carmém Lúcia, who decided to deny Janot's requests, arguing that "sufficient documents to analyze the case had already been attached to the file, demonstrating that it was unnecessary to take the steps requested". Gilmar Mendes' counsel argued that when the justice denied habeas corpus in another case involving the same businessman (HC 141.478), there was no question about his actions in the case. However, this does not make his actions in both cases legally adequate.

What's more, in the case of the justice's promiscuous actions, it's worth noting how close he is to lawyer Sérgio Bermudes. In an article in Piauí magazine in September 2010, the lawyer told the reporter: "Gilmar and I are brothers, we talk twice a day".1 In addition, Gilmar and his wife had already used the lawyer's apartment in NY and Rio de Janeiro. Asked if there was no conflict of interest in this relationship, Justice Gilmar Mendes replied: "not at all". He also said that Sergio Bermudes does not act in criminal cases, as is the case with Eike Batista, and therefore there would be no impediment. However, the lawyer defends the businessman in several other cases.

The second type of judicial corruption for individual purposes is instrumentalizing one's position for business matters. Prosecutor Deltan Dallagnol, the coordinator of Operation Car Wash, began to focus on the business world, giving dozens of lectures and earning at least R\$580,000 (US\$2000) as of 2017, according to conversations obtained by The Intercept Brasil (Vaza Jato). The prosecutor suffered a disciplinary complaint at the Public Prosecutor's Office and defended himself by claiming that he sent most of his earnings to charities. However, the prosecutor refused to disclose the list of companies and organizations that paid

<sup>1</sup> https://veja.abril.com.br/blog/radar/os-lacos-que-unem-gilmar-mendes-E-sergio-ber mudes/. Accessed on 13/05/2020.

for his lectures, as well as the amounts received for each one. The sum of all the lectures given by the prosecutor since 2016 exceeds R\$1 million.

According to the prosecutor's Telegram messages, leaked by The Intercept Brasil, Dallagnol outlined a business plan to profit from Lava Jato's fame, which included opening a company in his wife's name. He claimed that the lectures could be classified as a teaching activity, which is allowed by law, as well as having the purpose of fighting corruption, with the funds going to philanthropy. However, as of 2017, these funds were no longer earmarked for philanthropy. As the prosecutor himself argues in his defense,

"In 2017, after deducting the amount of 10% for personal expenses and taxes, the values of the lectures on corruption and ethics at major events are being allocated, to date, to a fund that will be used, in due course, for expenses or costs arising from the work of public servants in anti-corruption operations, such as Operation Car Wash, to fund initiatives against corruption and impunity, or even for initiatives aimed at promoting citizenship and ethics in general."<sup>2</sup>

In August 2019, the National Public Prosecution Council (CNMP) internal affairs department opened a new investigation into irregularities in one of the lectures given by the prosecutor to a group of bankers during the 2018 election campaign. At the time, eight complaints had already been opened to investigate misconduct by Dallagnol.3 By December 2019, there were already 23 complaints against the prosecutor at the CNMP,4 6 of which were closed on the last day before the 2019 judicial recess.

One of the cases concerns the issue of political action, the third subtype of judicial corruption for individual ends. Prosecutor Deltan Dallagnol took a stand, in a series of posts on his Twitter account, against the election of Renan Calheiros to the presidency of the Senate. Among other things, the prosecutor wrote that "we will hardly see an anti-corruption reform approved".5 The prosecutor also asked his followers to campaign for an open vote, in an attempt to force senators not to vote for Renan Calheiros. In the end, after a closed session, the senator lost out to Davi Alcolumbre (DEM-AP). Senator Renan Calheiros then filed a request

<sup>2</sup> https://www1.folha.uol.com.br/poder/2019/08/deltan-mudou-contratos-de-palestras-E-fil antropia-ficou-de-lado-apontam-mensagens.shtml.

<sup>3</sup> https://oglobo.globo.com/brasil/cnmp-apura-supostas-irregularidades-em-palestra-de-dal lagnol-23851735. Accessed on 13/05/2020.

<sup>4</sup> https://www.conjur.com.br/2020-jan-04/cnmp-arquiva-processos-deltan-ultimo-dia-prerecesso.

<sup>5</sup> https://agenciabrasil.ebc.com.br/justica/noticia/2019-12/conselho-do-mp-abre-novo-pro cesso-disciplinar-contra-dallagnol.

for a PAD (Disciplinary Administrative Proceeding) with the CNMP against the prosecutor, claiming that he was engaged in acts of a party political nature, which is forbidden to members of the Public Prosecutor's Office.

Another explicit case of political action is that of former judge Sérgio Moro, who gained widespread public visibility in the Lava Jato case, which allowed him to be appointed to Jair Bolsonaro's Ministry of Justice. Although he entered and left the government claiming to be a technician and not a politician, the fact is that his actions during Lava Jato made his politicalparty preferences clear. The evidence presented by Fabiana Rodrigues in her master's thesis (2022) and in her chapter at this collected volume confirm that official justifications shroud personal interests and advantages of the main protagonists of the Lato Jato Operation. The multivariate analysis made by Da Ros, Fontoura, Simoni Junior and Taylor in their chapter "Moro's Opinions: a Quantitative Analysis of Sentencing in Operation Car Wash" also show the severity with which political defendants were treated in that court.

Finally, another example of political activity comes from another member of Lava Jato, federal judge Marcelo Bretas. He took part in two public events in Rio de Janeiro in February 2020, alongside President Jair Bolsonaro and other Bolsonarist politicians.6 Not only did Bretas take part in the event, he received the president at Santos Dumond airport and drove to the event (the inauguration of an access road to the Rio-Niterói bridge) in an official presidential car. Considering the ban on political party activities imposed by the Brazilian constitution on members of the judiciary (art. 95), the OAB asked the CNJ, the organ designated to control the judiciary, to investigate Bretas' behavior. However, as Kerche et al. (2020) have shown, the CNJ rarely punishes its magistrates and clearly does not serve as an instrument for controlling members of the institution.

The issue of political expression by judges, also prohibited by the statutory regulation, was the subject of a CNJ rule in 2019, which recommends that judges avoid "expressing opinions or sharing information that could harm society's concept of the independence, impartiality, integrity and suitability of the magistrate or that could affect public confidence in the Judiciary". The measure aims to discipline the use of social media by members of the Judiciary, in view of the various cases in which judges publicly express themselves on party political issues, based on the argument that social media are for personal use, and not as public servants of the Judiciary.

<sup>6</sup> https://www.nexojornal.com.br/expresso/2020/02/17/Quais-os-limites-de-um-juiz-ao-seaproximar-da-pol%C3%ADtica.

The last subtype of judicial corruption for individual ends is acting in public debate. Gilmar Mendes' anti-Worker's Party crusade is an example of this. In 2015, the Worker's Party (PT) published a booklet denouncing the justice's anti-Party stance, in which the justice acts with "maneuvers and statements incompatible with the impartiality and modesty required of a judge".7 In response, Gilmar Mendes stated that he would continue to act in the same way.

In 2016, the justice even called for the political party to be dismissed on the grounds that it had been financed with illicit funds. This certainly doesn't justify the annulment of a party, but it does make public the justice's view of the party and therefore makes him impartial in judging cases related to it, as happened in Lava Jato. As Guilherme Boulos, a left-wing congressman (PSOL) recalled in an article published in Folha de São Paulo on August 11, 2016, the Lava Jato allegations did not only affect the PT, but also indicated illicit funds for José Serra's presidential campaign (with the Brazilian Social Democracy Party, PSDB) in 2010, delivered to Michel Temer (president from 2016–2018, with the Party of the Brazilian Democratic Movement, PMDB). However, Gilmar Mendes did not file a request for the impeachment of the PSDB or PMDB parties. This stance demonstrates not only how taking part in the public debate makes the judge impartial for present and future cases, but also how, by publicly expressing his party-political preferences, he becomes equally impartial.

In all the cases presented, judicial corruption for individual purposes is seen by its practitioners as part of their judicial performance, which is institutionally legal. According to Thompson's statement mentioned earlier, "because it is related to legitimate procedures and institutional practices, agents are not seen, and do not see themselves, as participating in acts of corruption" (Thompson, 2018, p.3). There is always a justification related to acceptable institutional procedures for their actions, hence the difficulty of punishment, despite the visible breach of republican principles of zeal for public affairs. In these cases, they have individual ends since they originate from personal interests, not institutional or corporate ones, as in judicial corruption with institutional ends. These are acts of judicial corruption practiced in the name of preserving personal friendships, for individual political gain or self-enrichment.

The fact is that Lava Jato made them explicit because, unlike the ordinary actions of judges and prosecutors, the operation gained unparalleled public visibility, putting all its actors at the center of public attention. If, on the one hand, it gave notoriety to the judicial actors, who gained individually from it, on the

<sup>7</sup> https://politica.estadao.com.br/noticias/geral,vou-continuar-atuando-da-mesma-forma- diz-gilmar-mendes-sobre-cartilha-do-pt,1794882.

other hand it put them in a delicate situation, exposing their unrepublican actions, excessively focused on individual interests and gains.

#### **5 Conclusions**

The expansion of the concept of corruption, so as to make it capable of capturing non-criminal activities that deeply harm institutional integrity and capacity, is an urgent analytical demand of the anti-corruption enterprise, as both an intelectual and political project that must be pursued by democratic polities. As we demonstrated through a rich collection of anecdotal examples of Brazilian judicial practice, there are several ways of perverting an institutional function by instrumentalizing the office for non-republican ends. This can be for petty individual purpose, but also for non-republican corporatist purpose.

This book is product of a valuable effort of describing, explaining and criticizing organizational crime and anti-corruption operations in Brazil. The protagonism of the justice system, which embarked on a messianic posturing, also needs to be sided by a level of attentive scepticism. Because such anti-corruption endeavor might ironically become deeply corrupted in itself. And, as such, it must become visible.

This is obviously important for refining the analytical parameters to gauge the quality of democracy, to imagine democratic innovation and reform, and to improve the tools of comparative politics.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.