Opening the black box: cabinet authorship of legislative proposals in a multiparty presidential system.

Author:Gaylord, Sylvia

Coalitions govern a large portion of the countries in the world. Even the United Kingdom, the paramount case of a majoritarian electoral system and the Westminster style of government, is presently governed by a coalition. In presidential systems, coalitions are also the norm when the president's party does not hold a majority in the legislature. In minority situations, coalitions occur 62% of the time in presidential systems and 77% in parliamentary ones (Cheibub 2010, 79). In the Western Hemisphere, the region of the world that concentrates the majority of presidential systems, 75% of all governments from 1979 to 2011 was coalition based (Figueiredo, Canelo, and Vieira 2012, 848). In other words, we cannot fully understand how countries with presidential and multiparty systems are governed without a better understanding of the policy-making dynamics of coalition governments. Yet we know little about the inner workings of the coalition, especially within the executive branch, and how it affects the role of the cabinet in the policy-making process.

Multiparty coalitions in presidential systems are not dictated by the constitution or governed by formal agreements, nor are they bound by a pact of mutual survival. The conventional view is that group decision making is not required in presidential systems: "the buck can be passed, so to speak, to the chief executive, who appoints the ministers, {...} who also allocates among them the tasks that they will have to undertake" (Blondel and Muller-Rommel 1993, 1). However, little empirical evidence has been provided to support or challenge this claim. In this article, we contribute to a greater understanding of coalition cabinets in presidential systems by exploring the role of ministers in authoring the government's agenda. In particular, we investigate how ministers' partisanship influences the government's legislative agenda.

The case of Brazil is ideal to begin exploring the inner workings of multiparty presidential cabinets. First, a body of research has already documented the formation of permanent government coalitions in Congress with representation in the cabinet and determined that the government coalition is essential for the president to govern. Second, Brazil is thought to have a dominant executive, responsible for the bulk of the legislation that is passed at the national level. Yet little is known about the process by which this policy is made and the participation, if any, of the cabinet in the production of policy. This article takes the case of Brazil to begin exploring the role of the cabinet in policy making in presidential systems.

Theories developed for parliamentary systems provide a starting point for the exploration of cabinet government in Brazil. Some theories postulate that parties, like individual politicians, are mainly motivated to win office (Downs 1957; Riker 1962). Joining the government to obtain the spoils of office is a motivation commonly attributed to politicians in Brazil (Ames 2001). It implies that cabinet ministers from parties other than that of the president do not have, or aspire to have, influence over the content of policy, or that if they do, it is secondary to office seeking. Other theories conceive government coalitions as compacts aimed at sharing policy-making responsibilities (de Swaan 1973). From this perspective, policy is an important element of governing and of government formation when single party majorities are not present (Laver and Shepsle 1996,8).

The literature on Brazil's coalition presidentialism has so far relied on the analysis of the voting record to document the role of parties in Congress supporting the government and an office-for-votes exchange between the president and its coalition partners (Ames 2001; Figueiredo and Limongi 2000a; Pereira and Mueller 2002). We know nothing about the policy aspirations of political parties joining a presidential cabinet in Brazil. Neither the parties joining the government nor the president make public statements about policy bargains or shared agendas. The executive's dominance in legislative production is explained by the extensive legislative powers granted to the executive by the Constitution, but the record of legislative approval does not tell how these prerogatives are used by the president or how the tasks of policy making are allocated within the cabinet.

Yet there is no reason to believe that parties and ministerial appointees in Brazil do not have policy agendas they wish to see implemented. We depart from the premise that, even if office seeking is the leading rationale for government formation in Brazil, policy is still important for politicians to maintain credibility with constituents (Laver and Shepsle 1996, 8) and that ministers join the cabinet in Brazil with a combination of office and policy-seeking motives. The first goal of this article is to document the participation of cabinet ministers in legislative initiatives of the executive.

The second goal is to understand the conditions that may favor policy sharing. Our hypothesis is that policy sharing between the president and the cabinet is more likely under conditions of ideological proximity. Compared to single-party governments, coalitions can be expected to experience higher levels of conflict, because parties in coalitions with multiple members are not always ideologically proximate. Additionally, parties that share a cabinet also compete against each other in elections and are under pressure to set themselves apart from government policy (Blondel and Muller-Rommel 1993, 9). Similar challenges can be expected to exist in multiparty presidential cabinets. As such, we expect cabinet participation in the executive's agenda to be more likely under conditions that facilitate internal agreement. We hypothesize that, as the ideological distance among coalition parties increases, ministerial participation in the executive's agenda will decrease and policy making will concentrate in the president's party. The empirical instrument we use is the endorsement of executive bills by cabinet ministers.

This article makes a first and original contribution to the analysis of the process of policy construction within the executive branch. Where policy is made affects the type of information that shapes policy and is revealing of the strategies used by the president in search of political success. The pattern of concentration of authorship in a few ministries that belong to the president's party that we find in our data confirms the view that coalitions in Brazil mainly serve the purpose of delivering votes in Congress. The data also show however, that the president is more willing to share policy making under conditions of ideological proximity. This suggests that the policy-making function of the cabinet is shaped by partisan conditions.

The article proceeds as follows. We begin the discussion with a review of the literature on modes of government formation in parliamentary systems, with an emphasis on policy and office-seeking motivations. We then provide a description of Brazil's system of government and a review of the literature analyzing Brazil's multiparty presidentialism, in light of what motivates parties to join the coalition. We then develop the hypotheses about patterns of legislative proposal authorship and endorsement by cabinet members. We also describe the process of policy making within the cabinet. This section is meant to illustrate the meaning of an endorsement and the relevance of ministers and ministerial staff to the content of legislative initiatives produced by the executive.

The empirical portion of the article has two parts. The first provides descriptive statistics from an original data set of ministerial endorsements of all executive bills submitted to Congress from 1995 to 2010. In addition to the partisan distribution of endorsements, we examine the frequency of endorsements by ministry and over time to

check for secular trends in executive policy making. In the second part we use the concentration of ministerial endorsements as the dependent variable to test for the effect of partisan and contextual variables on cabinet participation in executive output. We conclude with a discussion of our findings and suggestions for future research.

Office Seeking or Policy Seeking?

The literature on government formation in parliamentary systems suggests three possible motives for joining a multiparty government: appropriating the spoils of office, improving future electoral prospects, and exerting policy influence (Muller and Strom 1999a). The office-seeking motivation is about winning, and "winning means controlling the executive branch, or as much of that branch as possible" (Riker 1962, cited in Muller and Strom 1999a, 5). This motivation appears to operate in Brazil as much as in any other political system: cabinet ministers are driven to executive office because they get to command a budget, control appointments, and make decisions regarding the implementation of policy. Incumbency has electoral advantages, and parties may seek to join with an eye on future elections. Parties with a longer-term vision may seek membership in a cabinet as a way of winning a larger share of the executive at a future point in time (Muller and Strom 1999b, 6). As is often noted, getting into office is a necessary condition for the achievement of any other political or policy objective.

Another reason to join the government is to influence the content of policy. De Swaan was the first to argue that "considerations of policy are foremost in the minds of the actors and that the parliamentary game is, in fact, about the determination of major government policy" (de Swaan 1973, 88). As in the case of office seeking, policy seeking can reflect a range of motivations. It may indicate responsiveness to the party base by faithfully implementing the party platform and winning their vote in the...

To continue reading