Imprimir Republish

Pesquisa Fapesp 20 Years

Deconstructing myths

Pesquisa FAPESP followed the development of pioneering research on the relationship between the executive and legislative branches of government in Brazil

The National Congress building, photographed by Marcel Gautherot (1910–1996), circa 1960

Marcel Gautherot / Moreira Salles Institute Collection

The formation of coalitions, usually majority coalitions, has been the predominant way of governing in Brazil. Although elected to exercise an independent mandate, with the ability to freely appoint his or her ministers, the strength of coalition presidentialism is not based in the office of the president, but in the powers that give the occupant of the Planalto Palace the capability to influence the process of defining public policy. “The institutional basis of coalition presidentialism from 1988 onwards is the power to define the legislative agenda, which is conferred by the constitution, complemented by the centralization of decision-making within the Congress.” This conclusion comes from political scientists Argelina Cheibub Figueiredo and Fernando Limongi, in a discussion recently published in Dados – Revista de Ciências Sociais (Data: Journal of social sciences). According to their research, in Brazil rulers cannot escape what they classify as an imperative: “If they want to pass laws and change existing policies, presidents will be forced to seek support from the parties in the legislature.” Since the early 1990s, Figueiredo and Limongi have been taking turns coordinating a team of researchers investigating relationships between the executive and legislative branches of federal government. In December 1999, the initial results of their research were the subject of the magazine’s cover story.

The idea of investigating the relationship between the two powers came from Argentine political scientist Guillermo O’Donnell (1936–2011), based on interest expressed by The Andrew W. Mellon Foundation in funding scientific research, in Brazil, on the country’s National Congress. “When I joined CEBRAP [the Brazilian Center for Analysis and Planning] in 1990, O’Donnell was already there, and he asked me to prepare a preliminary project. In Brazil there had been little research regarding Congress, seen then as a kind of realm of the individual parliamentarian and treated as a single actor. All previous studies had dealt with parliamentary behavior, and thought about the subject from the perspective of electoral legislation,” recalls Figueiredo, currently a professor at the Institute of Social and Political Studies of the State University of Rio de Janeiro (IESP-UERJ).

Named by O’Donnell, the study “Terra incógnita – Funcionamento e estrutura do Congresso Nacional” (Terra Incognita: The operation and structure of the National Congress) began in 1991 without his further participation, receiving US$200,000 from the Mellon Foundation. Emphasis was given to studying the activities of the Câmara dos Deputados, Brazil’s House of Representatives. “We decided to collect all data on the processing of legislation from 1988 onwards. Our guideline was to not use what the Representatives said as a source, but rather what they did,” says Figueiredo. For this reason, there were very few interviews. Only a few House leaders who played relevant roles in drafting internal bylaws were questioned.

“The president remains the country’s principal legislator, with control of the budget and bureaucracy, and with the ability to issue Provisional Measures,” Limongi explains.

At that time, researching at the National Congress was no simple task. In the beginning, the House documentation department employees would do searches as requested by the study team, send the printed material to São Paulo, and, at CEBRAP, the researchers then systematized the information into spreadsheets. With the arrival in 1993 of Limongi, a professor at the School of Philosophy, Languages and Literature, and Human Sciences of the University of São Paulo (FFLCH-USP), roll call voting became part of the scope of the investigation, making it necessary to photocopy thousands of pages of the daily Congressional digest. These were used to map the activities of each of the Representatives. The discovery of previously ignored topics regarding the legislative process, such as vote abstentions, and guidance from the party leader, for example, brought about extra trips to Brasília to collect data. “With this research, we fell into the center of the Brazilian institutional debate that followed post-military democratization, i.e., whether presidentialism worked or not,” says Limongi, currently a professor at the São Paulo School of Economics of the Getulio Vargas Foundation (FGV-EESP). Within academia, then current opinion was that the political system would not work.

Among scholars on the subject, such as American political scientists Barry Ames and Scott Mainwaring, and Spanish sociologist Juan Linz (1926–2013), the viewpoint established in 1964 remained: that the highly conservative Legislature in Brazil would be an obstacle to action by the Executive. The problem would be governability. “The idea of Congress as a blackmailer of the Executive was predominant,” says Figueiredo. “Linz, for example, believed that since presidents could choose their ministers, they wouldn’t have incentives to make coalitions.” FFLCH-USP political scientist Regis Stephan de Castro Andrade (1938–2002) was one of the first to shift the focus of the investigations. By analyzing the power of committees, he began to bring attention to the innovations brought about by the then recently-approved Constitution. The research coordinated by Figueiredo and Limongi would ultimately show that the existing relationship between the Executive and the Legislature derives from the constitutional text.

The cover story of issue no. 49 of Pesquisa FAPESP reported on the study of patterns of interaction between the executive and legislative branches of government

Party discipline
Without a stated hypothesis, the initial concern of the scholars was to understand the functioning of the Brazilian political system and what role the Legislature played. They decided to follow the procedural paths of governmental proposals. “If the executive branch is supposedly unable to get its measures approved because the Legislature is uncooperative and blocks them, let’s look at the Executive’s bills to see what actually happens,” Limongi proposed. The general perception was that individualism prevailed in Congress and that there was a lack of discipline between political parties.

“I didn’t have any expectation it would be otherwise, until our first statistics indicated that party discipline did exist. It was surprising,” he reports. Since 1989, the percentage of members of the governing party who voted according to their party leader’s recommendation has been above 80%. During the research, they also came to the realization that House and Senate bylaws favor the political parties. “At the institutional level, parties are privileged political actors,” he adds (see Pesquisa FAPESP issue no. 114).

The comparison between the constitutional texts showed that all the reforms made by the prior military regime to reinforce the power of the Executive were maintained in the current Constitution. For example, the Provisional Measure (a mechanism similar to an Executive Order in the United States), is an adaptation of the former military Decree Law to the democratic system. Unlike the 1946 Constitution, according to Figueiredo and Limongi, the political model implemented in Brazil from 1988 onwards combines “rules for diffusing power in the system of representation, and for the concentration of power in the decision-making system of government.” “The President remains the country’s principal legislator, with control of the budget and bureaucracy, and with the ability to issue Provisional Measures,” summarizes Limongi. In his view, it is a tendency observed in the democratic constitutions approved since World War II. “Such a feature, with a ‘rationalized parliament’ [a legislative branch weakened to give more power to the Executive] is not at odds with modern constitutional theory. If I were to write a constitutional text, I would write it that way,” he says.

The institutional foundations of coalition presidentialism lie in the legislative power of the president, who has mechanisms to control the congressional agenda, says Figueiredo.

This reflects, for example, the pattern of success and dominance the Executive has in passing national legislation, which is very similar to that observed in parliamentary governments. The executive branch’s pattern of success refers to the percentage of the President’s proposals that become law. Dominance, in turn, refers to the percentage of the total number of approved laws that were put forward by the President. Since Brazil’s return to democracy, the average has been 77.3% and 83.5%, respectively. “During the course of our research, we identified that the institutional foundations of coalition presidentialism lie in the legislative power of the president, who has constitutional mechanisms to control the congressional agenda,” Figueiredo explains.

The power of the agenda
As they progressed in their understanding of how the Legislature functions, Figueiredo and Limongi ended up deconstructing the prevailing precepts in the literature, including international preconceptions, involving political agreements and interparty alliances. “The literature said that presidents would not form a coalition to govern. That was quite prevalent, but our findings are beginning to destroy this myth,” says Limongi (see Pesquisa FAPESP 50-Years of FAPESP Special Edition). “The current stage of our research allows us to say that today, from an institutional point of view, coalition presidentialism works in Brazil.”

In the article “A crise atual e o debate institucional” (The current crisis and the institutional debate), published in 2017, they summarize the operational logic of such a government. “Initiating, formulating, and proposing the agenda rests with the Executive, either with the President or his party. The coalition parties generally collaborate in implementing this agenda. This respects the popular mandate that emerges from the ballot box. The final version of the agenda, the one that is approved and that will be implemented, is up to the coalition.” It is the researchers’ understanding that in coalition governments, “rather than a vertical relationship of conflict between the executive and legislative branches, there is more frequently a horizontal relationship of bargaining and cooperation between the Executive and members of the coalition.”

Because it developed in an empirical and comparative way, reviewing the previous model and clearly showing positive aspects of coalition presidentialism, the study ended up challenging a certain theoretical interpretation that has divided the world of politics into the “developed” and the “underdeveloped.” “From an institutional point of view, there is no reason for Brazil to have an inferiority complex. Brazilian democracy is just like any other, it works the same way, our people are no different, and the political rationality is no different,” says Limongi, who, like Figueiredo regards Adam Przeworski, a political scientist at New York University, as his intellectual mentor.

Both assume that presidents act rationally, have plans for their governments, and want to implement public policy. “In this institutionalist view which we’re affiliated with, the rationality of the ruler is central. But it isn’t deterministic. Institutions do not determine the behavior of the political actor. The politician’s actions will be the link between the institutions and the results that must be produced,” explains Figueiredo. “The political actors react strategically to changes in the institutions,” adds Limongi. In light of the updated data, they intend to reexamine the workings of coalition presidentialism in the book they are currently preparing.

Inflection Point
Andréa Freitas, coordinator of the CEBRAP Center for Studies of Political Institutions and Elections, and a political scientist at the Institute of Philosophy and Human Sciences at the University of Campinas (IFCH-UNICAMP), is part of the second generation of researchers on the project. “I was trained to think and do research within this group,” says Freitas, who has been studying the relationship between the executive and legislative branches for 17 years. Freitas currently seeks to understand how the negotiation process that leads to the Executive’s success takes place. To ensure the project’s continuity and vitality, the Center under her responsibility brings together researchers from all stages of education. Concern with technical training remains fundamental. “The types of courses being offered have changed over the years. Today, for example, there’s no need to master the Access tool anymore, but it is necessary to know how to structure a database and use the R and Python programming languages,” she says.

Fabiano Santos, coordinator of the Congressional Studies Center (NECON) at IESP-UERJ, and founder of the Brazilian Legislative Observatory (OLB), believes methodological rigor is one of the project’s greatest merits. “Their work opened up new terrain by drawing the attention of the scientific community to the need to develop more sophisticated, highly-substantiated theories regarding the Legislature,” says Santos, who has been investigating the Brazilian Legislature for three decades. Recently, in partnership with Acir Almeida, Santos adapted hypotheses from information theory—which posits that the Brazilian Congress would become organized to seek information on how to develop public policy—to complement Figueiredo and Limongi’s findings regarding legislative dynamics in Brazil under coalition presidentialism.

A specialist in political institutions in Latin America since the 1990s, political scientist Timothy Power, director of the Oxford School of Global and Area Studies in Oxford, England, also identifies an inflection point in Legislative studies in Figueiredo and Limongi’s findings. “They changed the way we understand parties, and corrected a major misinterpretation by Brazilian and foreign scientists who made inferences about party strength based only on the characteristics of the nation’s electoral system,” he says. “By measuring for the first time the number of representatives who followed their party leader’s direction, they facilitated empirical analysis, disseminating the use of this metric, and have thereby influenced research in other countries.” Power only regrets that this knowledge hasn’t reached the press. “Journalists still repeat that the parties are weak, and prefer to emphasize the frivolousness of the political class. They don’t understand the institutional incentives behind it, and that undoubtedly impacts democracy.”

Political institutions, patterns of Executive-Legislative interaction, and governance capacity (nº 16/14525-6); Grant Mechanism Thematic Research Grant; Principal Investigator Fernando Limongi (CEBRAP); Investment R$953,208.97.