The author argues that the conjugation of certain democratic, socioeconomic and political factors has led to the "nationalization" of presidential elections in Brazil since 1960. The expansion of the electoral market resulting from urbanization and the growth of the electorate, together with the progressive removal of the obstacles to voting - due to income, gender, age, and education - has democratized the electoral process by diversifying the social structure of the electorate. Since then, the president's election has ceased to depend solely upon rural political forces and now involves multiple combinations of rural and urban political forces. Candidates can no longer count on specific social groups in order to guarantee their election, and need to widen their appeal. Based on the results of the direct presidential elections of 1960, 1989, 1994 and 1998, the author contends that the "nationalization" of the vote is part of the political integration of Brazilian society and that this, together with the social complexity of the electorate, means that the candidate's appeal and political stance need to be more wide-ranging in order to prevent potential conflicts from arising during the election campaign.

O artigo procura demonstrar que a conjugação de certos fatores demográficos, socioeconômicos e políticos conduziu à nacionalização das eleições presidenciais no Brasil a partir do pleito de 1960. A expansão do mercado eleitoral, como conseqüência da urbanização e do crescimento do eleitorado, e a queda progressiva das barreiras ao voto - de renda, de gênero, etária e de escolaridade - teriam democratizado a participação eleitoral, diversificando a composição social do eleitorado. Desde então, a eleição do presidente deixou de depender apenas das forças políticas rurais e passou a requerer uma infinidade de possíveis combinações de forças rurais e urbanas. O candidato não poderia mais contar com segmentos sociais específicos e numericamente fortes para se garantir, necessitando de bases sociais ampliadas. Com base nos resultados das eleições presidenciais diretas de 1960, 1989, 1994 e 1998, o autor argumenta que a nacionalização do voto, definida pelos indicadores acima, é parte da integração política da sociedade e que esta, associada à complexidade social do eleitorado, exigem que a mensagem política e o apelo do candidato que pretenda maximizar o voto sejam de natureza ampla, de modo a evitar a manifestação, na campanha, de conflitos potenciais.

This paper argues that the conjugation of certain democratic, socioeconomic and political factors has led to the "nationalization" of presidential elections in Brazil since 1960. The expansion of the electoral market resulting from urbanization and the growth in the electorate, together with the progressive removal of the obstacles to voting - due to income, gender, age, and education - has democratized the electoral process by diversifying the social make-up of the electorate. Since then, the president's election has ceased to depend solely upon rural political forces and now involves multiple combinations of rural and urban political forces. Candidates can no longer count on specific and numerous social groups in order to guarantee their election, and need to widen their appeal. Based on the results of the direct presidential elections of 1960, 1989, 1994 and 1998, the author contends that the "nationalization" of the vote as described above is part of the political integration of Brazilian society and that this, together with the social complexity of the electorate, means that the candidate's appeal and political stance need to be more wide-ranging in order to prevent potential conflicts from arising during the election campaign.

Cet article veut montrer que la convergence de certains facteurs démographiques, socioéconomiques et politiques a abouti, au Brésil à partir du scrutin de 1960, à la nationalisation des élections présidentielles. L'accroissement du marché électoral, fruit de l'urbanisation et de l'augmentation de l'électorat, ainsi que la suppression progressive des barrières de vote -liées au revenu, sexe, tranche d'âge et niveau de scolarité - auraient démocratisé la participation électorale, permettant une plus grande diversification des électeurs. Dès lors, l'élection présidentielle a été moins dépendante des forces politiques rurales, requérant plutôt d'innombrables arrangements entre celles-ci et les forces urbaines. Le candidat ne pouvait plus compter exclusivement sur des segments sociaux particuliers et voyait la nécessité d'élargir ses bases sociales. Sur la base des résultats des élections présidentielles de 1960, 1989, 1994 et 1998, l'auteur montre que la nationalisation du vote, compte tenu de ces données, fait partie de l'intégration politique de la société et que celle-ci, ainsi que la complexité sociale de l'électorat, exigent que le message politique et le discours du candidat visant à récueillir le plus de voix, soient de teneur générale, de façon à éviter qu'éclate tout conflit latent pendant la campagne.

This analysis of the reform processes that have occurred in New Zealand’s, Italy’s, Japan’s, and Mexico’s electoral systems begins by examining the initial context of each reform. It then describes the changes put in place and observes subsequent effects on party systems. Reform took a common route in these four countries, that is, some type of mixed system was adopted, where each voter casts two ballots in the Lower House: the first is cast for candidates who run in single-vote districts under the plurality system; the second is cast for a proportional closed list, which may be national or state/provincial in scope. The article strongly questions the thesis that the effects of institutional changes can be controlled, since relatively similar electoral reforms have had different consequences in the cases analyzed. In New Zealand, such changes brought significant improvements to existing institutions. In Italy, the only evident effect was the breakdown of the previous party system. In Japan, one sole party no longer dominated the system, which developed instead into a moderately fragmented one. Lastly, in Mexico, one could not say the system is effectively democratic but one can detect a trend towards party diversification and less manipulation of the polls.

L’article cherche à analyser les processus de réforme du système électoral qui se sont déroulés dans quatre pays: la Nouvelle-Zélande, l’Italie, le Japon et le Mexique. On y examine les circonstances initiales de la réforme ainsi que les changements effectués, pour observer ensuite ses conséquences sur le système des partis. Les pays choisis ont un point en commun: leur réforme électorale suit un même but, c’est-à-dire l’adoption d’un système mixte où l’électeur dispose de deux voix pour la Chambre Basse. La première voix va aux candidats qui se présentent dans des districts uninominaux et au scrutin majoritaire; la seconde, à ceux qui composent des listes proportionnelles bloquées à l’échelon national ou pour un État ou province. La thèse selon laquelle il est possible de contrôler les effets des changements institutionnels est profondément mise en question, puisque les effets de ces réformes électorales assez semblables ont varié selon les cas étudiés. En Nouvelle-Zélande, le changement du système électoral s’est traduit par une nette amélioration des institutions existantes. En Italie, le seul effet visible a été le démantèlement du système de partis précédent. Au Japon, le système a cessé d’être dominé par un parti unique et a évolué vers un morcellement modéré. Enfin, au Mexique, on ne peut affirmer être désormais en présence d’un système effectivement démocratique, mais on observe une tendance à la diversification des partis et à une diminution des manipulations électorales.