RESUMO
This article analyzes party discipline in the Brazilian Senate based on a previously untapped database covering roll call votes from 1989 to 2009. As occurs in the Chamber of Deputies, political parties in the Senate display a high degree of unity that remains reasonably stable over time. Despite the similar results in the two Houses, I suggest that the explanations are not the same. As compared to the Chamber of Deputies, the Senate appears to feel less power from the Executive, the leaders are scarcely relevant, and the decision-making process is less centralized. I do not offer a solution to this paradox, but I do suggest that it involves not only the institutional rules, but also the organizational characteristics and even the political profile of the Senators themselves.
Dans cet article, on évalue la discipline des partis au Sénat fédéral, à partir de données inédites qui englobent les scrutins nominaux effectués pendant 21 ans, de 1989 à 2009. De même qu'à la chambre des députés, les partis du sénat ont manifesté une forte unité, assez stable au cours des années. Malgré les similitudes rencontrées dans les résultats pour les deux chambres législatives, les justifications ne semblent pas être les mêmes. Si on assiste aux débat des députés, la force du pouvoir exécutif au sénat semble moindre, ses leaders, moins importants et le processus décisionnel, moins centralisé. On ne propose pas de solution à ce paradoxe, mais il semble qu'elle réside non seulement dans les règles institutionnelles mais aussi dans les caractéristiques organisationnelles, voire dans le profil politique des sénateurs.
RESUMO
This article seeks to identify the determinants of the existence and powers of upper legislative chambers. By analyzing their constitutionally established powers, including all of the minimally democratic bicameral countries, it was possible to reach conclusions not previously suggested in the literature. Using a two-step model whose calculation includes the possibility of a selection bias (Heckman model), the study evaluated both the factors leading to the existence of an upper chamber and those defining its powers. Federalism proved to be an important variable for explaining the difference between bicameral and unicameral countries, but not for explaining the political strength of upper chambers. The most relevant explanatory variable in this sense was the system of government: in Parliamentary systems, upper chambers have limited powers; in Presidential systems they exercise broad powers.