ABSTRACT The rise of Bolsonaro is the latest episode of what seems to be a Brazilian tradition of electing rightwing populists as a response to corruption, bad economic management, and disenchantment with mainstream political parties. In previous opportunities, Brazilians elected Jânio Quadros (1960) and Fernando Collor (1989). However, Bolsonaro, although very similar to both Quadros and Collor on their hostility to parties and legislative institutions, brought novel elements. To identify the distinguishing features of Brazil's new president, we systematically compare Quadros, Collor, and Bolsonaro along 13 dimensions encompassing political context and attributes, policy choices, and institutional relations. Our findings indicate that throughout 2019, Bolsonaro led an unstable minority administration that governed for a minority in society and emphatically excluded some relevant sectors. Moreover, he established close connections with Evangelical groups and the military, and attacked the press and the Supreme Court at an unprecedented level.

RESUMEN El ascenso de Bolsonaro es el último episodio de la que parece ser una tradición brasileña: elegir populistas de derecha como respuesta a la corrupción, la mala gestión económica y al desencanto con los principales partidos políticos. En oportunidades anteriores, los brasileños eligieron a Jânio Quadros (1960) y Fernando Collor (1989). Sin embargo, Bolsonaro, si bien muy similar tanto a Quadros como a Collor en su hostilidad hacia los partidos y las instituciones legislativas, aportó elementos novedosos. Para identificar las características distintivas del nuevo presidente del Brasil, comparamos sistemáticamente a Quadros, Collor y Bolsonaro a lo largo de 13 dimensiones que abarcan el contexto y los atributos políticos, la orientación de las políticas gubernamentales y las relaciones institucionales. Nuestros hallazgos indican que durante todo el año 2019, Bolsonaro lideró una administración minoritaria e inestable que gobernó para una minoría de la sociedad y que excluyó enfáticamente algunos sectores relevantes. Además, estableció estrechas conexiones con los grupos evangélicos y los militares, y atacó a la prensa y al Tribunal Supremo a un nivel sin precedentes.

RESUMO Este artigo contribui para a crescente literatura nacional sobre os determinantes do sucesso em pleitos legislativos por meio da mensuração do impacto eleitoral de fatores financeiros, políticos e pessoais associados aos candidatos à Assembleia Legislativa do Ceará em 2010. Escolheu-se um estado de porte mediano para minimizar o potencial problema de endogeneidade e maximizar a homogeneidade da amostra, de modo a se obterem resultados estatísticos mais robustos. Da aplicação empírica baseada em modelos com dados contáveis, destacam-se as seguintes constatações: 1) os impactos na votação diferem drasticamente de acordo com a localização geográfica dos municípios; 2) características pessoais dos candidatos são relevantes para a obtenção de votos, sobressaindo a condição de casado; 3) dentre as variáveis políticas, a filiação a um partido que integra a coligação do governador é importante apenas para candidatos eleitos com votos obtidos na região metropolitana, não tendo nenhuma delas se revelado significante para os não eleitos; 4) candidatos incumbentes possuem vantagem significativa na obtenção de votos; 5) as despesas declaradas de campanha são decisivas para o desempenho eleitoral dos candidatos. Essa última verificação empírica revela um grave problema relativo ao funcionamento das instituições representativas de um país profundamente desigual do ponto de vista socioeconômico.

ABSTRACT The following article contributes to the growing body of national literature on the drivers of success in legislative elections by gauging the electoral impact of financial, political, and personal factors associated with the candidates running for Ceará Legislative Assembly in 2010. Medium in size, the state was chosen in order to minimize any potential issues of endogeneity and to maximize the homogeneity of the sample, in order to obtain more robust statistical results. A subsequent empirical application based on models for discrete data was carried out, highlighting the following facets: 1) the impacts of the vote varied dramatically according to the geographical location of the municipalities; 2) the candidates’ personal characteristics were relevant to their obtaining of votes, particularly their marital status; 3) within the political variables, an affiliation to a party in the governor’s coalition was only significant for the candidates elected via votes from the metropolitan region, with none of these variables proving significant for those not to have been elected; 4) incumbent candidates enjoyed a significant advantage in the obtaining of votes; 5) the campaign expenses declared influenced the candidates’ electoral performance. The final empirical finding reveals a grave problem in the workings of representative institutions in a country profoundly unequal from a socioeconomic perspective.

RESUMEN Este artículo contribuye a la creciente literatura nacional sobre los aspectos determinantes del éxito en las elecciones legislativas por medio de la medición del impacto electoral de factores financieros, políticos y personales asociados a los candidatos a la Cámara Legislativa del Estado de Ceará en 2010. Se optó por un estado de medianas dimensiones para minimizar el potencial problema de endogeneidad y maximizar la homogeneidad de la muestra, con objeto de obtener unos resultados estadísticos más fiables. De la aplicación empírica basada en modelos con datos contables, destacan las siguientes constataciones: 1) los impactos en la votación difieren drásticamente en función de la localización geográfica de los municipios; 2) las características personales de los candidatos son relevantes para obtener votos, y destaca sobre todo el hecho de que estén casados o no; 3) entre las variables políticas, la afiliación a un partido que integra la coalición del gobernador es importante solo para los candidatos elegidos con votos obtenidos en la región metropolitana, pero ninguna de ellas se reveló significativa para los no elegidos; 4) los candidatos en el poder tienen una gran ventaja en la obtención de votos; 5) los gastos declarados de la campaña son decisivos para los resultados electorales de los candidatos. Esta última confirmación empírica revela un grave problema en lo relativo al funcionamiento de las instituciones representativas de un país profundamente desigual desde el punto de vista socioeconómico.

RÉSUMÉ Cet article veut contribuer à l’essor de la littérature nationale sur les déterminants des succès électoraux grâce à la mesure de l’impact électoral des facteurs financiers, politiques et personnels associés aux candidats aux élections législatives de l’État du Ceará en 2010. Nous avons choisi un État de taille moyenne pour minimiser le problème potentiel de l’endogénéité et potentialiser l’homogénéité de l’échantillon, et ce afin d’obtenir des résultats statistiques les plus fiables possibles. L’application empirique basée sur des modèles pour données discrètes nous a permis de faire les constatations suivantes: 1) les impacts sur le vote diffèrent drastiquement en fonction de la localisation géographique des communes; 2) les caractéristiques personnelles des candidats sont pertinentes pour l’obtention de voix, et en particulier le fait d’être marié; 3) parmi les variables politiques, la filiation à un parti intégrant la coalition du gouverneur n’est importante que pour les candidats élus avec des voix obtenues dans la région métropolitaine de la capitale, et n’est significative pour aucun des candidats non élus; 4) les candidats sortants disposent d’un avantage significatif dans l’obtention des voix; 5) les dépenses de campagne déclarées ont une influence décisive sur la performance électorale des candidats. Cette dernière vérification empirique révèle un grave problème de fonctionnement des institutions représentatives d’un pays profondément inégal du point de vue socio-économique.

Resumo Neste manuscrito, procuramos identificar os aportes do novo método histórico-comparativo à ciência política e à administração pública. Tentamos contribuir com o fortalecimento das pesquisas de orientação qualitativa dessas duas disciplinas no Brasil, ao mostrar como os recentes avanços daquele método podem gerar ganhos analíticos por meio do estabelecimento mais preciso de condições causais e da utilização dos testes associados ao chamado process tracing. Para dar conta desse objetivo, o trabalho, em primeiro lugar, disseca os principais elementos do referido método. Depois, relemos e, por assim dizer, reorganizamos Sessenta e quatro: anatomia da crise, livro clássico de Wanderley Guilherme dos Santos, à luz dos preceitos do referido método. Em seguida, também relemos e reorganizamos alguns estudos sobre a política externa argentina sob a mesma luz. Destacamos os ganhos gerados pela aplicação disciplinada do método histórico-comparativo. Na conclusão, identificamos as potencialidades e limitações do método, oferecendo também considerações a respeito da importância que sua aplicação pode ter para a ciência política e a administração pública brasileiras.

Abstract This article seeks to identify the contributions that the new comparative-historical method (CHM) has made to political science and public administration. By showing how recent advances in CHM can generate analytical gains through establishing causal conditions and tests associated with process tracing, we hope this will contribute to strengthening qualitative-based research of these two disciplines in Brazil. To achieve this, the paper begins by dissecting the main elements of this method. Then we review the classic work entitled Sessenta e quatro: anatomia da crise (Sixty-four: anatomy of the crisis) by Wanderley Guilherme dos Santos in light of the method's precepts. The paper also reviews a number of studies on Argentine foreign policy and highlights the gains that can be generated by the disciplined application of CHM. In the conclusion, the potential and limitations of the method are identified and some considerations are made with regard to the importance of its application to Brazilian political science and public administration.

Resumen En este manuscrito, buscamos contribuir al fortalecimiento metodológico de la ciencia política y de la administración pública de orientación cualitativa en Brasil, mostrando cómo los recientes avances del nuevo método histórico-comparativo pueden generar ganancias analíticas a través del establecimiento más preciso de las condiciones causales y el uso de los criterios de evaluación asociados a lo llamado process tracing. Para alcanzar este objetivo, el trabajo, en el primer lugar, analiza los principales elementos de este método. Luego leemos y, por así decirlo, reorganizamos Sessenta e quatro: anatomia da crise, libro clásico de Wanderley Guilherme dos Santos, a la luz de los preceptos de ese método. En seguida, también releemos y reorganizamos algunos estudios sobre la política exterior argentina bajo la misma luz. Destacamos las ganancias generadas por la aplicación disciplinada del método histórico-comparativo. En la conclusión, identificamos el potencial y las limitaciones del método y ofrecemos consideraciones acerca de la importancia que su aplicación puede tener para la ciencia política y la de administración pública de Brasil.

The years between 2005 and 2014 were one of the best decades of political science (PS) in Brazil. The growth of the economy, the consequent increase in public spending on universities and scientific activity in general, and the policies pursued by the Ministry of Education played a key role in bringing this outcome about. Effective collective action within the Brazilian PS community was also a vital contributing factor. Consequently, Brazilian PS is now closer to the potential generated by the country's population and economy, as far as the Latin American context is concerned. In addition, the discipline has been nationalized, and its research agenda is now less parochial nowadays. On the negative side, however, the output of Brazilian political scientists still stands out for its relatively low international visibility. Moreover, Brazilian PS remains characterized by a scarce systematic application of well thought-out qualitative methods. Finally, a considerable uncertainty over the resources available to the discipline has recently surfaced, given the years of fiscal adjustment lying ahead.

El período 2005-2014 fue una de las mejores décadas de la Ciencia Política (CP) brasileña. Para que esto ocurriera, contribuyeron, en primer lugar, el crecimiento de la economía del país y el consecuente aumento del gasto público dedicado a las universidades y a la actividad científica en general y las políticas del Ministerio de Educación. También fue vital para este éxito la acción colectiva eficaz en el seno de la comunidad de politólogos brasileños. Consecuentemente, la CP brasileña se encuentra hoy, en el contexto latinoamericano, más cercana al potencial generado por la población y la economía del país. Además, la disciplina ha sido nacionalizada y su agenda de investigación hoy es menos parroquial. Por el lado negativo, sin embargo, se mantiene la relativamente baja visibilidad internacional de la producción de los politólogos brasileños, se constata la escasa aplicación fundamentada de métodos cualitativos y, recientemente, ha aflorado una considerable inseguridad sobre los recursos disponibles para la disciplina, teniendo en cuenta los años de ajuste fiscal que se avecinan.

The politics of intergovernmental transfers in Brazil. This article examines the political economy of public resources distribution in Brazil's federal system in 1985-2004. We propose an empirical exercise to analyze how the country's federal governments deal with the tradeoff between the provision of material wellbeing to sub-national governments (the states in our study) and the pursuit of political support from the latter. To identify the determinants of the transfer of resources from the federal government to the states, a set of economic, political, and institutional variables is econometrically tested. Based upon instrumental variables estimation for panel-data models, our estimates indicate that in Brazil the pursuit of political goals prevails over social equity and economic efficiency criteria: higher levels of per capita transfers are associated with the political makeup of governing coalitions, while larger investments in infrastructure and development by the states are associated with a lower amount of per capita resources transferred to sub-national governments. Our findings also suggest a trend toward the freezing of interregional inequalities in Brazil, and show the relevance of fiscal discipline laws in discouraging the use of the administrative apparatus for electioneering.

This article argues that the political heft of a specific state (Rio de Janeiro) within the Federative Republic of Brazil has to do with how state party politics relates to the national party politics. The article assesses the decline in Rio's role in national politics before and after transfer of the national capital to Brasilia in 1960 and the fusion of Guanabara with the former state of Rio de Janeiro in 1975. The article draws on previously unpublished data and proposes two hypotheses: (1) the share of cabinet posts held by politicians from Rio de Janeiro is largely a function of the state's relative weight in the main governing parties' delegations in the Chamber of Deputies and (2) that the strengthening of the Federal government since the promulgation of the 1988 Constitution prompted Rio to adjust to the party alliances forged by the various Presidents, thereby contributing to a relatively recovery of the state's national political influence in the first decade of the 21st century. Finally, the article provides a normative reflection on the consequences, for Brazil, of Rio de Janeiro's political and economic decline.

Dans cet article, on montre que la trajectoire politique d'un État brésilien particulier, celui de Rio de Janeiro, au sein d'une république fédérative, adopte la façon dont les contours plus généraux de sa politique sont reliés au cadre politique et à celui des partis, à l'échelon national. On évalue le déclin du rôle de l'état dans la politique nationale avant et après le transfert de la capitale à Brasília en 1960 et à partir de la fusion de l'État de Guanabara avec l'ancien État de Rio de Janeiro en 1975, par le biais de vérifications empiriques inédites, alliées à deux hypothèses: (1) la participation d'hommes politiques de Rio dans les cabinets présidentiels est, pour une large part, fonction du poids relatif de l'État dans les groupes de députés fédéraux des principaux partis présents au gouvernement; et 2) le renforcement du gouvernement central depuis la Constitution de 1988 a stimulé Rio à s'ajuster aux alliances entre partis passées par les présidents, ce qui a contribué à un relatif rétablissement de l'influence nationale de l'État dans les dix premières années du XXIe siècle. On arrive, enfin, à une réflexion de type normatif concernant les conséquences, pour le Brésil, du déclin politique et économique de l'État.

Este artigo oferece uma proposta de reforma incremental do sistema eleitoral da Câmara dos Deputados. A natureza incremental da proposta se funda na suposição segundo a qual o sistema político é uma arquitetura complexa e delicada, sendo a possibilidade de piorá-lo com mudanças ambiciosas e intempestivas bem maior do que a de aperfeiçoá-lo. A proposta mantém o sistema de representação proporcional com lista aberta, mas altera duas variáveis-chave, uma vez que reduz a magnitude média das circunscrições eleitorais e estabelece uma regra proporcional de distribuição de cadeiras entre partidos coligados. A operacionalização da reforma leva a um novo desenho do mapa eleitoral do país, com circunscrições eleitorais menores dentro de 12 estados. O artigo também apresenta os resultados de um exercício de simulação feito com base nos dados das eleições de 2006 que recalcula a composição partidária da Câmara dos Deputados a partir das regras aqui preconizadas.

This article presents a proposal for an incremental reform of the electoral system of the Chamber of Deputies. The incremental nature of the proposal rests on the assumption that this political system is a complex and delicate architecture, and therefore, abrupt and ambitious changes are more likely to make it worse than to improve it. The proposal advocated here maintains the current system of open-list proportional representation, but it changes two key variables of this electoral system, since it reduces its average district magnitude and establishes a proportional distribution of seats among parties that form electoral coalitions. The way the reform is operationalized leads to a redesign of the electoral map of the country, with smaller districts within 12 states. The article also presents the results of a simulation exercise based on data from the 2006 elections, whereby the party makeup of the Chamber of Deputies is recalculated according to the rules proposed here.

This article provides the reader with a synthetic perspective of what was 2006 for Brazil, in terms of its economic, social and political aspects. Initially, the piece offers preliminary information on what has resonated in Brazil in the aforesaid year. Growing urban violence was one of the hallmarks of 2006. Detailed statistics on the general elections held by the end of the year are the backdrop to the analysis of Lula's re-election. The latter was to a great extent a consequence of his macroeconomic management and targeted social policies. Such policies also resulted in a shifting of the candidate's electoral base to the poorer parts of the country. The political crisis endured by Brazil in 2006 is explained by Lula's governing style, which produced a highly fragmented and poorly monitored executive coalition. Data on legislative output demonstrates the executive branch's control over the policy process. Finally, the ideology of political parties, and the organization and workings of the judicial branch and the federation are briefly described

El presente artículo tiene como objetivo dar al lector una perspectiva sintética de cómo fue el año 2006 para Brasil en términos de su aspectos económicos, sociales y políticos. Mediante el análisis de hechos claves y sus consecuencias, el artículo intenta permitir una comprensión preliminar de lo que tuvo resonancia en Brasil durante el mencionado año. La creciente violencia urbana ha sido una de las marcas de 2006. La presentación de detalles estadísticos de las elecciones generales de final del año es la cortina de fondo del análisis de la reelección de Lula. Esta fue en gran medida resultado de sus políticas sociales. Tales políticas también han resultado en la mudanza de la base electoral del candidato hacia las zonas más pobres del país. La crisis política experimentada por Brasil en 2006 es explicada por el estilo de gobierno de Lula, que produjo una coalición ejecutiva altamente fragmentada y mal monitoreada. Datos de la producción de leyes demuestran el control la rama ejecutiva tiene sobre el proceso de decisión. Finalmente, la ideología de los partidos, la organización y funcionamiento del Poder Judicial y la federación son brevemente descritos

Abstract Brazil's political science managed to establish itself in the 1970s under a military regime. From the 1980s till the late 1990s the discipline grew and became institutionalized. However, there now exist within the country's political science write out community two schools that simply do not communicate with each other, namely, an empirical school and a normative-philosophical one. We propose that a political science community that still needs to expand must adopt a, so to speak, consociational model to organize the interaction between its main schools. To do so, there must be an ample consensus on the community's lines of expansion, a consensus that should rest upon the improvement of the teaching of methods in humanities.

Resumen La ciencia política brasileña logró establecerse en la década de 1970 bajo un régimen militar. A partir de la década de 1980, creció y se institucionalizó. Sin embargo, existen en el país dos formas principales de escuelas que no se comunican entre sí, una empírica y otra normativo-filosófica. Proponemos que la ciencia política, una academia que necesita todavía expandirse, encuentra en un modelo consociativo su mejor forma de organizar la interacción entre las escuelas. Para esto, hay que generar un amplio consenso en torno a las líneas de expansión de la comunidad y que se debe basar en el fortalecimiento de la enseñanza de métodos y humanística.

Shugart and Carey posit that presidential democracies in which legislators have a parochial focus of representation are electorally inefficient because voters are not offered highly identifiable choices on national policies. Such systems are driven by an inefficient secret, which is essentially a nonpartisan representation of the policy process. To check the propositions of the inefficient secret model (ISM), this article investigates the aggregation level, effect, and subject of Brazilian deputies' legislative input and output. Our empirical analysis indicates that, although some ISM-related factors drive legislative output, there is partisanship in deputies' legislative input. This result means that the ISM underestimates the prospects for programmatic parties (especially in the opposition) to emerge within systems where the electoral and constitutional rules encourage particularism.

Shugart et Carey croient que les démocraties présidentielles où les législateurs ont une conception restreinte de la représentation politique sont inefficaces d'un point de vue électoral car elles n'offrent pas aux électeurs de possibilités de choix faciles à reconnaître parmi des politiques à l'échelon national. Ces démocraties sont régies par un secret inefficace, c'est-à-dire une conception de la dynamique politique indépendante d'un parti. Afin de vérifier les propositions du modèle du secret inefficace (MSI), on examine dans cet article le niveau de rassemblement, l'effet et les thèmes des projets et des lois approuvées par les députés brésiliens. L'analyse empirique montre que, bien qu'on trouve une certaine influence des partis dans les projets présentés par les députés, les lois sanctionnées par l'activité législative sont guidées par d'autres types de facteurs, ce qui, en partie, vient confirmer le dit modèle.

Valendo-se das hipóteses formuladas por Wanderley Guilherme dos Santos para explicar o golpe de 1964 no Brasil, o texto procura identificar as condições que propiciaram a ruptura institucional havida na Venezuela em 1999. A análise de dados relativos à produção legislativa e à distribuição de cadeiras parlamentares e de preferências ideológicas, entre 1959 e 1999, indica que as hipóteses de Santos têm um considerável poder explicativo da crise política que atingiu a Venezuela no final da década de 90, oferencedo oferecem predições mais precisas a respeito dos motivos da ruptura institucional ter ocorrido em 1999.

Based on the hypotheses of Wanderley Guilherme dos Santos to explain the Brazilian 1964 coup d'état, this article seeks to identify the leading conditions to the Venezuelan institutional rupture in 1999. Data about the legislative production and the distributions of parliamentary seats and ideological preferences between 1959 and 1999 point that Santos hypotheses have significant explaining capacity of the Venezuelan political crisis at the end of the 90's, specially the causes of the institutional rupture in 1999.

This article discusses the relationship between the Executive and Legislative Branches in Brazil with regard to the issuing of Provisional Measures (PMs) by the former. The article contends that the legal ability of the Brazilian Chief of State to issue decrees with the force of law and taking force immediately (i.e., PMs) does not imply that members of Congress have abdicated from their legislative prerogatives in favor of the Executive Branch, as proposed by some authors analyzing the issue. After presenting the most important analyses of the dynamics in the Executive-Legislative relationship, the article aligns itself with the interpretation emphasizing the process of mutual delegation and control between the two powers, contending that the formation of a solid coalition government generates incentives for the Executive to issue PMs that meet the interests of the legislative majorities, and that Congress is also capable of inferring whether such measures meet its interests, by way of an indirect supervisory mechanism known as the "fire alarm". The article proposes that it is possible to view PMs as part of a complex and changing context of delegation within Brazilian Presidentialism, involving neither usurpation nor abdication.

Dans cet article, on discute les rapports entre les Pouvoirs Exécutif et Législatif au Brésil concernant la promulgation de Mesures Provisoires (MPs) par l'Exécutif. On y montre que la capacité légale du chef du gouvernement brésilien à promulguer des décrets ayant force de loi et application immédiate, les MPs, ne signifie pas que les députés et sénateurs aient abandonné leurs prérogatives législatives au profit de l'Exécutif, tel que l'ont prétendu certains auteurs qui se sont penchés sur la question. Après avoir présenté les analyses les plus importantes de la dynamique des rapports Exécutif-Législatif, l'auteur adopte une interprétation qui distingue le processus de délégation et le contrôle mutuel entre ces pouvoirs, en soutenant que la formation d'un gouvernement de coalition solide stimule l'Exécutif à produire des MPs qui tiennent compte des intérêts des majorités législatives; en outre, le Congrès est capable de juger si telles ou telles mesures répondent ou non à ses intérêts par le moyen d'un mécanisme indirect de contrôle, nommé alerte d'incendie. L'auteur propose qu'on peut comprendre les MPs comme partie d'un cadre de délégation complexe et changeant à l'intérieur du présidentialisme brésilien, sans pour autant y avoir d'usurpation ni d'abdication.

The article provides an explanation for the enormous variation in party discipline during the Brazilian democratic experience from 1946 to 1964. The argument is that party discipline is a function of the Presidents’ strategies in forging coalitions in Congress. One of the main consequences of such strategies was the creation of two factions within parties, one for and the other against the President. The premise is that the size of each faction affects its level of discipline. According to an econometric analysis of 982 roll-call votes, crucial determinants of discipline in the larger parties were the amount of budget resources allocated to them by the President and the time remaining in the President’s term.

Dans cet article, on propose une explication à l’énorme variation observée dans les taux de discipline dans les partis politiques pendant l’expérience démocratique de 1946 à 1964. On part de l’affirmation que la discipline de parti dépend des stratégies du président dans la formation d’alliances à la Chambre législative. L’une des principales conséquences de ces stratégies a été la formation de deux groupes à l’intérieur des partis: l’un favorable et l’autre défavorable au président. On a proposé que la taille de chaque groupe influence le niveau de discipline. L’analyse économétrique de 982 votes nominaux montre que les déterminants principaux de la discipline dans les grands partis ont été l’ampleur des moyens budgétaires que chaque président leur a accordés et le temps qui lui restait à la tête de la nation.

All presidential cabinets formed in Brazil between 1985 and 1998 included politicians from more than one party and could, in principle, be defined as coalition cabinets. The article evaluates to what extent this is a valid statement as far as the legislative behavior of the cabinet parties is concerned. An indicator is proposed for measuring the rate of cabinet coalescence, based on the ratio between the percentage of ministerial posts held by each party represented in the cabinet and the percentage of legislative seats held by each of these parties. Regression analysis is used to ascertain what impact the rate of cabinet coalescence has on the legislative discipline of cabinet parties. In addition to this coalescence index, the study also tests the effect of electoral cycles and the ideological range of cabinets on legislative discipline. Results indicate that the legislative discipline of cabinet parties is only consistent with what would be expected of coalition governments in parliamentary regimes when the cabinet displays a relatively high rate of coalescence. Findings also show that the party support to president tends to decrease over the course of the president’s term and that the greater the ideological distance between the president and a given party, the lower the support rate this party will display for the head of government.

Tous les gouvernements formés au Brésil entre 1985 et 1998 comprennent des politiciens d’au moins deux partis. Tous pourraient être définis comme étant de coalition. Cet article cherche à évaluer la véracité de cette affirmation sous l’angle du comportement législatif des partis qui composent chaque gouvernement. Pour cela, il propose un indicateur qui mesure le taux de coalescence des gouvernements sur la base du rapport entre le pourcentage de ministères occupés par chaque parti représenté et le pourcentage de sièges législatifs détenus par chacun de ces partis. Par le moyen de l’analyse de régression, on vérifie l’impact de ce taux de coalescence sur la discipline législative des partis qui intègrent le gouvernement. Outre ce taux de coalescence, on vérifie aussi l’effet des cycles électoraux et de l’ampleur idéologique des gouvernements sur la discipline législative. Selon les résultats, ce n’est que sous des gouvernements avec des taux élevés de coalescence que les partis au gouvernement présentent une discipline législative en accord avec celle qu’on attend d’un gouvernement de coalition. Les tests démontrent aussi que pendant la durée du mandat présidentiel se produit un affaiblissement de la discipline législative des partis représentés au gouvernement et que plus un parti est éloigné idéologiquement du Président, plus son d’appui à ce dernier est réduit.