ABSTRACT
Introduction The Participatory Budgeting Process (PBP) is an innovative experiment in direct democracy pioneered in Brazil by the Workers’ Party (Partido dos Trabalhadores, PT) and now widely implemented in hundreds of Brazilian municipalities and in cities around the world (Abers 1998; Avritzer 2002; Baiocchi 2003a, 2003b, 2005; Cabannes 2004; Santos 1998). Participatory budgeting is a form of direct democracy emphasizing deliberation among citizens as well as transparency and accountability in the budgetary process. The PBP is based on a methodology of direct popular intervention in municipalities’ investment budgets. Citizens participate in meetings at the neighborhood level, discuss budgetary priorities, and rank their choices as to where and how public funds should be spent. In these meetings, local-level delegates are elected to participate in subsequent plenary assemblies, as demands make their way to the mayor’s office. The process, therefore, establishes a clear set of universal rules for access to and influence in the budgetary process, reducing opportunities for clientelistic and personalistic forms of resource allocation (Abers 1998). In addition to its main purposes, i.e., increasing popular participation in the political system, reducing inequality by transferring resources to where citizens think they are most needed, and increasing the transparency of budgetary apportionment, some authors claim that the PBP has spillover effects. According to Avritzer (2002) and Baiocchi (2003a), the PBP creates a public space in which citizens engage in political discussion not just about the budgetary process, but also about political and social issues more broadly. Through the process of discussion and deliberation, individuals learn about the functioning of the political system, about how to negotiate, bargain and persuade, and about political competition at the local level. The public space created by the PBP plays a pedagogical role by providing new information for voters and by serving as a training ground and opportunity structure for networking and collective action. The empirical implications of these arguments are two-fold: PBP participants’ networks should be distinct from those who do not participate, and PBP activists should be better informed about politics than individuals who do not participate. Previous studies have only tested the latter conjecture through in-depth inter-
views with PBP activists or participant observation in local meetings. Hence, such studies incur a problem of selection bias; i.e., they only study activists. We are left, therefore, without a clue about how activists differ from non-activists and can come to no final conclusions about the impact of the PBP on political information. Our work contributes to this rich debate by offering a more comprehensive form of measuring the impact of the PBP process on information gain through the use of probabilistic samples and public opinion panel data allowing for comparisons of activists and non-activists. Moreover, there are no studies about how participation in the PBP affects the characteristics of activists’ social networks. Our work innovates by proposing and testing hypotheses about the structure of activists’ political discussion networks and how these networks differ from those of non-activists. Focusing on networks, as will be discussed ahead, is a way of investigating individuals’ opportunities to engage in deliberation and discussion with others, especially beyond closer circles of family and friends. In this way, we attempt to approach individuals’ opportunity structures for political discussion. Our claim is very simple: discussion and deliberation necessarily require interaction with others. Networks are a form of identifying if these interactions occur or not. Furthermore, the composition of the network, if based on strong or weak ties, for instance, makes a difference to what one learns and thinks about politics. In sum, the main goal of this chapter is to assess how the PBP influences citizens’ social networks and levels of political information about politics in the 2002 and 2006 Brazilian elections. We contrast activists and non-activists using a unique six-wave public opinion panel study in two mid-size Brazilian cities: Juiz de Fora, in the state of Minas Gerais, and Caxias do Sul, in Rio Grande do Sul. Both cities experimented with the Participatory Budget in the eight years prior to the 2004 municipal elections, albeit with very distinct levels of effectiveness. In Caxias do Sul, the PBP allocated the entire investment budget of the municipality, approaching 10 percent of the total budget. Popular participation in the PBP was very high. In Juiz de Fora, the PBP was more a political marketing strategy than an efficient mechanism of budgetary allocation. Almost nothing was distributed through the PBP, and popular participation was low. Given the differences between the two cities, are the characteristics of activists, their networks and levels of political information also distinct? That is, does the level of success of the PBP condition its influence on activists’ levels of information and political discussion networks? Moreover, in spite of its success in Caxias do Sul, in the 2004 municipal elections the PBP was discontinued in both cities by the incoming administrations.1 This offers an interesting opportunity to evaluate the effects of the demise of the PBP on former activists’ networks and on political information. Nylen (2003) has argued that long-time activists dominate the PBP and that the interruption of the process has no lasting effect on participants. We test his hypotheses using our longitudinal data, which include measures before and after the PBP. The chapter is organized as follows. The next section explores in more detail the literature on how activism creates public spaces for political deliberation and
learning. Based on this literature, we derive hypotheses that will be tested using our dataset. We then describe the data and variables. Finally, the data are analyzed and results discussed.
