Pathways to Power

< < Return to Book Page

Legislative campaigning in the United States requires that candidates raise many of their own funds, create advertising, appear on television, organize supporters, target legislation, and send mass mailings. Partisanship influences campaigns and policy positions, but legislators also take district demands into account in determining political postures and strategies. In Colombia and Brazil, legislative candidates and legislators do many of the same things, but there is also a clientelistic aspect. Partisanship is an even less important motivator for legislators in these countries than in the United States. Further, many Brazilian legislators are beholden to state-level politicians, and many more than in the United States retire in favor of state-level political or bureaucratic jobs after a stint in the legislature. Argentine legislative candidates are, by contrast, motivated by partisanship to such an extent that they almost never vote against their party in Congress. Further, their campaigns—which are dominated by newcomers, since so few Argentine legislators are reelected—are very different affairs, as the individual candidates have much less responsibility than their Brazilian counterparts in terms of raising funds or organizing supporters, though many are still involved in meeting local leaders and distributing particularistic goods. Uruguayan legislators share some similarities with the Argentinians, but that country’s defined factionalism leads legislators to consider district, factional, and partisan issues in defining their campaigns and political strategies. Some Bolivian legislators represent another distinct type, as they are clearly tied to a particular social group. As a result, these politicians’ campaigns and legislative careers are built around meeting the needs of that group, be it rural farmers or regional separatists.

Similarly, despite a commonality of basic institutional framework, presidential behavior also varies substantially across contexts. In Chile, longtime party insiders have tended to garner the presidency by campaigning along party ideological lines, and presidents have built cabinets and governed by paying careful attention to partisan and coalition dynamics. Colombia, Venezuela, and Bolivia stand in direct contrast, presenting striking cases where politicians have created their own partisan vehicles to capture the presidency. Their behavior in office, in turn, has been unrestrained by partisan considerations. Other cases lie between these extremes. In Mexico, Vicente Fox’s ambivalent yet important connections to his own party led him to emphasize business ties over partisanship in developing his cabinet and political programs. Once in office, Argentine President Raul Alfonsín, the undisputed leader of his party, distributed cabinet posts to his copartisans and obtained the support of potential internal rivals through his distribution of party and government posts. By contrast, Alfonsín’s successor, Carlos Menem, had more limited partisan ties and, depending on one’s viewpoint, worked either against the party or in an effort to remake it. Finally, Brazil’s Luiz Inácio “Lula” da Silva provides another model; perhaps because of his defined ideological position, solid position within his party, and minority status, Lula made alliances outside the left, pragmatically cultivating the support of other parties. Colombia’s Álvaro Uribe pursued a strategy like Lula’s, despite running as an independent and clearly attempting to distance himself from the Liberal Party, where he had built his career.

What explains these different types of candidates and their resultant political behavior? An important determinant is the institutional framework in which the candidates and future officials operate. Institutional theories, however, are incomplete and perhaps flawed, because their advocates too often expect uniform responses or ignore contextual variables. For example, since the time of Maurice Duverger (1951, 1954), analysts have often expected a singular response to incentives stemming from the electoral system. But such a framework, like that of a house, can influence without determining it’s inside layout and architecture. In turn, parties operating under identical institutional frameworks may take many different forms. Tastes and costs determine a house’s inside configuration; similarly, parties’ goals and opportunities lead them to react to institutional stimuli in distinct ways. Analyses of presidents, legislatures, parties, or executive legislative relations, therefore, must consider the factors that combine or interact with the institutional environment to explain political behavior.

Some recent scholarship has moved well beyond Duverger to note that although electoral systems (or other aspects of the institutional framework) operate on all parties in a country, not all parties take the same form. Rather than discarding the value of institutional variables, this recognition has led researchers to begin considering the factors that add to or interact with institutions in their search for understanding. Here we push this type of analysis one step further, discussing how party and legal variables create a framework that produces distinct types of candidates and thereby influences political behavior. By focusing on political recruitment and candidate selection—politicians’ pathways to power—this book identifies and describes this key framework that bounds political behavior. We explore this framework and its impacts with a focus on candidates for executive and legislative positions in Latin America.

The legal framework and formal institutions clearly create pressures that push political parties in particular ways, but not all parties react similarly. This differentiated response results from the variance in parties’ goals, norms, and strategies and those factors’ interactions with the institutional environment. For example, one party in a given country could use its closed list electoral rules to assure centralized control of legislative candidate nominations, whereas another could decentralize that power to provincial leaders. Here the electoral system would interact with party organizational variables—which may be dependent on the size, age, and initial development of the party—to determine the process of candidate nominations. Likewise, competitive pressures, a party’s regional strength, ideological coherence, and many other factors will influence whether national party leaders, provincial party leaders, or voters will have the greatest say in naming legislative candidates. Similarly, though party norms may dictate formalized and centralized selection procedures for presidential candidates, and the selection of candidates and presidents with a strong basis in their political parties, second-round elections may provide party dissidents with an incentive to throw their hats into the ring, obviating the importance and necessity of running with a party label. In sum, formal institutional rules (such as electoral systems and federalism) combine and interact with contextual situations (such as structures of party competition) and party-level variables (such as ideological coherence and fractionalization of the leadership) to determine particular selection methods.

Along these lines, this introductory chapter develops two key themes. First we develop an argument about the interaction of specific aspects of the electoral system with party and contextual variables in determining the recruitment and selection process and candidate type. Here we develop the idea of candidates’ loyalty to a particular “selectorate,” be it constituents or some higher party official. The direction of loyalty determines several candidate types, which in the latter part of the chapter we correlate with campaign and postelectoral behavior to show the influence of candidate type as an independent variable. We provide a parallel analysis for recruitment and selection in the legislative and executive branches, though we do acknowledge that the recruitment and selection procedures in the latter are less significant in shaping subsequent presidential behavior, for a variety of reasons. This leads to our final conclusion about the significant, but often overlooked, impact of recruitment and selection (R&S) variables on democratic process and governability.

{A}Why Pathways to Power Matter: The Study of Political Recruitment and Candidate Selection{/A}

A key goal of institutionalists has been to establish the relationship between electoral systems and party outcomes, with a particular focus on party discipline in the legislature, executive-legislative relations, and the nature of the party system (Ames 2001; Carey and Shugart 1995; Morgenstern 2004; Samuels 1999; Shugart 1995; Siavelis 2002). However, because they fail to consider the interaction of the electoral system with recruitment and selection variables, these studies cannot explain the types of candidates, parties, and interbranch relations that emerge within different countries. For example, Brazil’s open-list system, where voters choose among multiple intraparty legislative candidates, is said to work against party discipline. However, one party, the Partido dos Trabalhadores (PT), has traditionally maintained much more control over nominations than others and consequently has sustained much higher levels of party loyalty. Similarly, analyses of countries employing electoral rules that provide incentives toward party discipline, such as Argentina, fail to explain intrapartisan conflict that can arise from legislators beholden to the regional power brokers that nominated them as candidates. Systemic institutional rules also fail to explain how such disparate presidential candidates can emerge in the same country. The differences (even beyond those related to personality) among recent presidents in Argentina (for example, Raul Alfonsín, Carlos Menem, Fernando de la Rúa, and Néstor Kirchner), or between highly competitive candidates in Mexico (Felipe Calderón and Andres Manuel López Obrador) underscore this point. In short, electoral laws do not alone determine the degree to which R&S procedures are centralized; thus they fail to completely explain candidate types and the resulting behavior of politicians. That explanation, in sum, requires analysis of the R&S processes.

Scholars have generally shied from studying R&S, however, because the associated variables are only sometimes written, include party and legal statutes, and are, overall, notoriously difficult to measure (see Helmke and Levitsky 2004). However, if we are interested in analyzing the effect of political institutions, we must understand the totality of incentives operating on politicians, many of which are rooted in the R&S procedures.

The importance of R&S has not escaped scholars focusing on European parliamentary governments or the United States. Michael Gallaghar and Michael Marsh (1988) build on the early “classic” literature (Black 1972; Czudnowski 1975; Eulau and Czudnowski 1976; Marvick 1968; Prewitt 1970; Seligman 1971) and provide a comprehensive treatment of R&S issues, to which Pippa Norris (1997a) and James W. Davis (1988) have made important additional contributions. However, this literature generally ignores Latin America’s predominantly multiparty presidential systems. Literature focusing on Latin America has begun to appear, but primarily in the form of individual case studies that have not yet generated a theoretically oriented comparative study. A second limitation of the extant literature is that irrespective of the regional focus, most scholars interested in theory building tend to treat the R&S process primarily as either a dependent variable (analyzing how politicians are recruited and selected), or an independent variable (analyzing the effect of the recruitment processes on subsequent political behavior). We are aware of no study that does both simultaneously, despite the significant insights that can be gleaned by doing so. As such, we lack an overarching framework for understanding political recruitment, candidate selection, and their consequences.

Another shortcoming in the literature is the failure to adequately consider how formal institutions combine with informal institutions and processes to produce outcomes that belie predictions of much of the institutionally based literature. Following Gretchen Helmke and Steven Levitsky (2004), we distinguish between formal party (and faction and coalition) rules, and informal norms or procedures that guide the implementation of the “parchment” rules or laws. Parties develop rules—whether formally inscribed or not—with respect to seniority, incumbency, and the rights to candidacies for militants or outsiders. In some cases, party rules and processes yield centralized control over nomination decisions and tightly guarded campaign finance, whereas others allow primaries and encourage candidates to raise and spend their own funds. Most of these rules and processes are compatible with different types of electoral systems, and all can tighten or loosen the relation of candidates to party leadership. Argentina provides a good example of how the interaction of R&S processes with electoral rules affects behavior. Argentine parties fill closed lists of candidates, which gives party leaders (in this case the provincial party leaders) significant power to name and rank candidates on their lists (see chapter 2, this volume, by Mark P. Jones). Thus, in line with predictions made by the institutional literature dealing with list type, we would expect legislators who are loyal to be regularly renominated and reelected. However, the Partido Justicialista (PJ) has maintained a rule discriminating against incumbents, with the result that only about one-fifth of the legislature is filled by incumbents. In addition, the party’s organization reflects the country’s federal structure, in that the national hierarchy must contend with significant influence of provincial leaders, who wield tremendous power in the candidate selection process. Thus, Argentine deputies look to and serve the provinces and provincial party elites, a behavior that in large part is determined by recruitment variables.

< < Return to Book Page