Building Democracy in the Central Andes: Bolivia in Comparative Perspective

Miguel Centellas

Western Michigan University

 

Abstract

This study places Bolivia in comparative perspective and seeks to explain the country's unexpected two decades of democratic stability. Four potential explanations are pursued: 1) Bolivia's political institutional structure, 2) historical legacies, 3) the role of individual statesmen, and 4) the role of a patrimonial political elite. If Bolivia's democratic stability is best explained by its unique institutional arrangement of "parliamentarized presidentialism," then this would contribute to our general theoretical understanding of democracy and democratic consolidation. A variety of methods will be used to test these four possible explanations, including qualitative comparative analysis, archival research, and open-ended elite interviews.

1. Research Question and Rationale

For most of its history, Bolivia was the epitome of Latin American political instability. Thus, it is remarkable that in July 2002, Bolivian citizens will vote in their fifth regularly scheduled national election since the nation's transition to democracy in 1982. Bolivia's recent experience with democracy sharply contrasts with the experience of its Central Andean neighbors, Ecuador and Peru. All three began their democratic transition in 1978-79, though by 1992, Peru had reverted to authoritarian rule and Ecuador's democracy seems perpetually on the brink of collapse. How do we explain the comparative stability of Bolivia's democratic political system? Why has it experienced neither a democratic breakdown (as in Peru) nor chronic instability (as in Ecuador)? In the context of Latin America's troubled political history, the survival of democracy in Bolivia defied most expectations and begs for an explanation.

The dependent variable is democratic stability rather than democratic consolidation. This study does not assume that Bolivia's democracy is consolidated, only that it has been relatively stable, especially when compared to the other Central Andean democracies. The endurance of formal democratic procedures is an essential (though not a sufficient) condition for democratic consolidation. Minimally, democratic stability means that electoral calendars are respected, and that political actors do not pursue their policy preferences by extra-constitutional means.

The limited literature on Bolivia's democratization highlights a new exceptionalism. Conaghan and Malloy (1994) point out that of the three Central Andean countries, only Bolivia successfully implemented neoliberal economic reforms in the 1980s. In a cross-national study of Latin American democracies, Jones (1995) finds a significant relationship between his dummy variable for Bolivia and presidential legislative majorities, which is strongly related to the survival of presidential democracy. Mayorga (1992; 1997) suggests that Bolivia's unique institutional design accounts for its democratic stability. Nevertheless, the causal relationship between formal institutions -especially the electoral system- and Bolivia's democratic stability has not yet been systematically studied.

My goals are to: 1) explore the relationship between formal and informal political institutions or norms of behavior and democratic stability in Bolivia, 2) examine the impact of specific historical or contextual factors that may affect democratic stability, 3) isolate the key explanatory factors for stable democratic politics in Bolivia, and 4) consider the prospects for sustained democratic stability.

My study extends from the literature on modern democratic theory and political institutions. I examine post-transition politics. Rustow (1970) describes the process of democratization as involving three stages: 1) the breakdown of the nondemocratic regime, 2) the installation of a democratic regime, and 3) democratic consolidation. Here, I consider democratic consolidation as the process of deepening democracy after a democratic regime has been installed, rather than as an end-point. Because democratic consolidation requires democratic stability, this study looks to see whether Bolivia's democratic stability is the kind that can lead to democratic consolidation. The period of study begins with the installation of the first democratically elected government in 1982 and ends in 2004 (upon completion of field research). Although comparative, I adopt a case-oriented approach.

Discovering the key factors contributing to Bolivia's democratic stability may help in developing reforms meant to strengthen other new democracies. This is especially true if the key factors for Bolivia's democratic stability are institutional, rather than idiosyncratic ones. The Bolivian experience may be especially relevant for its Central Andean neighbors, but its lessons may also be applicable beyond Latin America to other countries that face the challenge of building stable democracy under historically unfavorable circumstances. Of course, it is possible that factors supporting democratic stability in early stages may hinder the prospects for long-term democratic stability. If the quality of democracy remains low for an extended period, political and social actors may no longer be willing to play the democratic game. Understanding the relationship between democratic stability and the quality of democracy help us focus on building democracy "for the long haul" (Huntington 1997).

2. Conceptual Framework

Latin America does not have a history of democratic politics. Traditional politics, described either in forms of its centralism (Vliz 1980), a caudillo tradition of personalism (Dealy 1992), or corporatism (Wiarda 1981), were the regional norm for most of the post-independence period. This was especially true in the Central Andes. By the mid-1970s, even Chile and Uruguay, two countries with long democratic traditions, were under authoritarian rule. Prospects for democracy in Latin America seemed bleak; only Venezuela, Colombia, and Costa Rica were democracies.

Beginning in the late 1970s, Latin American authoritarian regimes gave way to democratic ones. The pace of democratization and democratic consolidation, however, has differed across the region, including setbacks in Peru and Venezuela that clearly fit O'Donnell's (1994) description of delegative democracy. And while the Central Andean republics democratized first, later democratizers such as Argentina and Chile seem to be fairing better. This study aims to discover what factors account for the relative democratic stability in Bolivia.

2.1. Democracy and Democratization

In its simplest form, democracy means "a form of government in which the people rule" (Srensen 1998, 3). The so-called classical theories of democracy emphasize direct self-rule of the demos in a small community made possible by direct, active, and equal participation by members of the polity. The modern development of very large and complex polities made the application of such a definition difficult. In its place, most scholars have accepted a liberal or representative definition of democracy that rests on competitive elections. Such a political system provides the "processes by which ordinary citizens exert a relatively high degree of control over leaders" (Dahl 1956, 3).

Most of the comparative literature on democratization adopts Dahl's (1956; 1971) definition of democracy (or polyarchy) as a political system involving three dimensions: competition, participation, and civil and political liberties. These minimalist definitions of democracy focus primarily on the institutions of formal electoral democracy and are largely extensions of Schumpeter's description of democracy as "that institutional arrangement for arriving at political decisions by means of competitive struggle for the people's vote" (1943, 269).[1] Democratic self-rule was thus possible in large polities due to the development of representative, electoral institutions by which voters select legitimate policy-makers and hold them accountable.

The liberal model of democracy has not gone uncriticized. Pateman (1970) and Barber (1984) criticized the Schumpeterian model of democracy for ignoring the important role early liberals, such as J.S. Mill, had given to popular participation. For them, popular participation was an essential element in democracy and could not be reduced to elections. While recognizing that representative democracy was necessary at the national level, Pateman and Barber argued for direct, participatory democracy at the local level, such as in the workplace or in neighborhood associations. Mansbridge (1983) criticized the liberal model of democracy for its emphasis on adversarial politics. Adversary democracy cannot account for instances when disagreement is too deep for mere vote-counting; at such times, Mansbridge argued, deliberative politics are more in order. Other critics, such as Macpherson (1977), Gould (1988), and Dryzek (1996) also pointed out that liberal democracy is too closely tied to capitalism, which threatens the egalitarian premise essential to democracy.

Proponents of liberal democracy have accepted many of the arguments made by those who favor deeper forms of democracy. Dahl (1971) agreed that polyarchy was a second-best solution to the problem making democracy possible in large, pluralist polities. Attention was shifted from democracy to democratization, the constant deepening and renewal of democratic norms. Thus, like the critics of liberal democracy, Dahl (1985) argued for the extension of democratic norms into the social and economic spheres, though continuing to emphasize the procedural elements necessary for democracy.[2] To this procedural minimum have been added the importance of the accountability of rulers and civilian control over the military (Karl 1990), the rule of law (O'Donnell 1996), the existence of an effective state (Linz and Stepan 1996), and a vibrant civil society (Putnam 1993). These additions expand Dahl's formulation by focusing on democracy as a process (and the conditions necessary for that process) rather than specific policy orientations. These additional elements are essentially compatible with the conditions Dahl (1971) considered necessary for political democracy to be possible.[3]

Still, minimalist conceptions of democracy are more practical for comparative studies of democracy since "they deliberately focus on the smallest possible number of attributes that are still seen as producing a viable standard for democracy" (Collier and Levitsky 1997, 433). Thus, most comparative studies of democracy use Dahl's definition of democracy as a base reference to determine if a state is democratic. Linz and Stepan, for example, define democracy as "a free competition of power by peaceful means, free elections at regular intervals in a constitutional framework that provides conditions for such a free competition in terms of freedom of speech, of assembly, of political organization" (1978, 5-6).

Earlier scholars, while trying to explain why democracy was more prevalent in Western societies than elsewhere, had argued that certain social or economic preconditions were necessary for democracy. Lipset (1959) argued that modernization or industrialization was necessary for democracy; socioeconomic development had to come first before democracy was possible. O'Donnell (1973) presented a scathing critique of this approach, pointing out that economic development tended to produce bureaucratic-authoritarian states rather than democracies. Following the Weberian tradition, Almond and Verba (1963) argued that a certain civic culture, such as the Anglo-American one, was necessary for democracy. Huntington (1984) and Karl (1990) criticized this as an ethnocentric approach, arguing that cultures are not so readily reduced into pro-democratic vs. anti-democratic dichotomies. Moore (1966) argued that democracy required a specific type of macro-social conditions or 'social structure'. Therborn (1983) challenged Moore's proposition, that democracy is advanced by a well-developed bourgeoisie, and argued that democracy is instead produced by a struggle against the bourgeoisie.

The third wave of democracy, which began in the mid-1970s, included many countries lacking these socioeconomic preconditions and coincided with a renewed interest in states and institutional structures. Scholars such as Skocpol (1979), Stepan (1978), and Trimberger (1978) were "bringing the state back in" to the center of comparative political analysis at the same time as democratic transitions were sweeping across the globe. Similarly, many of those analyzing transitions to democracy adopted a framework of analysis that looked at the institutional framework or transition process and the strategies pursued by various elite actors. O'Donnell and Schmitter (1986) and Przeworski (1988) argued that democratization was primarily a process of negotiation between authoritarian and pro-democratic elites. Thus, recent scholars emphasize the need for institutional arrangements and agreements by which the elite crafted democracy through a series of agreements about constitutional design, timing, and trade-offs (Di Palma 1990).

2.2. Democratic Stability

Though recent literature emphasizes the concept of democratic consolidation, my study uses the more basic concept of democratic stability. Although the requirements for democratic consolidation are often debated, most scholars agree that stability is a core condition for democratic consolidation. Diamond, Hartlyn, and Linz consider a democracy stable when "the broad mass of the public and all significant actors ... believe that the democratic regime is the most right and appropriate for their society, better than any other realistic alternative they can imagine" (1999, 4). My definition of democratic stability is more modest: democracy is stable when the basic procedural democratic norms are consistently adhered to. This means that electoral calendars are institutionalized -elections go ahead as scheduled without interruptions. Elections must be free of fraud and losers must accept their outcomes. Similarly, no actors attempt to overthrow the democratic regime. In the simplest terms, there should be no coups d'tat or other attempts to use extra-constitutional power to pursue political (or personal) ambitions. I consider a democracy stable if it meets these requirements for two consecutive elections.

At its core, democratic consolidation refers to expectations that democracy will survive, that it is immune to reversal (Schedler 1998). But endurance does not by itself guarantee that a democracy is consolidated. A semi-democratic regime may also enjoy long-term stability (O'Donnell 1996). Often, definitions of democratic consolidation are closely linked to, and expand upon, procedural definitions of democracy. The concept is meant to distinguish real democracy from procedural faades or "diminished subtypes" of democracy (Collier and Levitsky 1997). Essentially, democracy is consolidated if it meets the requirements for genuine democracy as well as those necessary for its long-term sustainability.

Linz and Stepan (1996) consider democracy consolidated when it has become "the only game in town." Their commonly used definition focuses on five arenas: civil society, political society, the rule of law, a usable state bureaucracy, and economic society. For them, a consolidated democracy requires certain behavioral, attitudinal, and constitutional conditions: behaviorally, no actors try to (violently) overthrow the democratic regime; attitudinally, there is broad public support for democratic procedures and institutions; constitutionally, all actors are subject to and accept the resolution of conflict using democratic institutions.

For democracy to be the only game in town, the political and social elite must first agree to play by democratic rules. Elites and elite pacts play a critical role in the democratization process (Higley and Gunther 1992). Democracy has fared better in countries where elite pacts were prevalent than in those where they were not. The nature of elite pacts also had important consequences for democracy in Latin America (Peeler 1998). Elite consensus about the pace of democratization and the basic nature of the emerging democratic system help convince them to play the democratic game. But members of the elite may also agree to slow down the pace of democracy or to settle for a semi-democratic regime instead.

Whitehead's (2001) concept of democratic viability provides another way to assess democratization in a long-term political and historical perspective. A democracy is viable if it is capable of surviving in its environment. Related to endurance, the concept of viability points to factors that may, in the long term, threaten democracy. A democracy may be consolidated but not viable; it may simply be democracy by default if actors have accepted democracy only because non-democratic alternatives are no longer available.[4] To be viable, democracy and its institutions must enjoy widespread legitimacy. This requires effective democratic institutions, especially those that produce moderated politics and strong links between institutions and civil society. Democratic viability also requires that political institutions of liberal democracy be able to solve the critical problems of their society (Margolis 1979). Thus, democratic viability is linked to democracy's performance, its ability to resolve key social, economic, and political problems.

The concept of democratic viability distinguishes between factors that aid democratic stability from those that promote democracy for the long term. Procedural definitions of democracy often emphasize stability. But political stability and the institutionalization of political norms may, in the long term, threaten democracy. If political institutions are unable to articulate social demands or represent civil society, democracy may not be viable and may eventually break down. For democracy to endure, political democracy must be both institutionalized and flexible enough to bring in new players and deepen democracy over time (Huntington 1968).

My study, however, does not seek to predict whether Bolivia's democracy will endure nor does it argue that its democracy is consolidated. Though some argue that Bolivia's democracy is already consolidated (Whitehead 2001; Mayorga 1992; Linz 1994), what is important is that democratic procedural norms have been uninterrupted for two decades. The same cannot be said of Ecuador or Peru. This study aims to discern what factors best explain Bolivia's unexpected democratic stability.

2.3. Democracy and Political Institutions

Because elite agreements are so important, much of the recent scholarship focuses on political institutions, the commonly accepted norms and rules of behavior that frame elite (as well as mass) behavior. These include formal institutions, such as electoral rules or constitutional provisions regulating executive-legislative relations; they also include informal rules, such as those regulating bargaining and coalition building. Democracy requires institutions that encourage moderated bargaining and limited veto power, while also ensuring effective governance and the authority of the state, especially the rule of law. Much recent literature on democracy in Latin America has adopted an institutionalist perspective that sees institutions as strongly influencing political outcomes, including democratic consolidation and endurance. Thus, some literature on democratization has focused on "constitutional engineering" (Sartori 1997) or "getting the institutions right" (Diamond, et al. 1999).

A focus on political institutions is consistent with approaches to the study of democratization that adopt procedural definitions of democracy. Political institutions, the formal and informal norms of political behavior, dictate how power and authority is exercised in a democratic polity. The institutions that determine how leaders are selected and how the rights of minorities are protected are essential to the very meaning of modern representative democracy. If democracy is to be stable, of course, then the institutions that define democratic politics must also be stable and commonly accepted.

My study adopts a historical institutionalist perspective that defines institutions as "the formal or informal procedures, routines, norms and conventions embedded in the organizational structure of the polity" (Hall and Taylor 1996, 938). This approach contrasts with rational choice institutionalism, which assumes actors have predetermined interests and design institutions to maximize those interests (North 1993). Historical institutionalism focuses on both formal and informal institutions and readily admits that political outcomes are influenced by historical and cultural factors. Earlier decisions have consequences, both expected and unexpected, on the strategies and choices available in the future. Historical institutionalists accept rational choice's assumption that individuals act strategically, but argue that individual interests, choices, and strategies are also affected by their historical institutional context (March and Olsen 1989). The historical-institutionalist perspective I adopt also differs from a sociological institutionalism that more broadly analyzes social conventions and customs (Powell and DiMaggio 1991). This latter approach, however, is inherently conservative and less able to explain institutional change than either rational choice or historical institutionalism.

A historical institutionalist perspective is useful for a study of Bolivian democracy because Bolivian actors did not design their institutions. Bolivian democracy was inaugurated in 1982 under the 1967 Constitution. Thus, the Bolivian elite was constrained by an institutional framework they did not construct. Though not implemented at the time, the 1967 Constitution (drafted during the 1964-69 Barrientos military government) outlined the basic rules for democratic elections. A new constitution was not adopted until 1995, though it did not significantly alter the country's basic institutional framework.[5] Nevertheless, Bolivia's democratization was a dramatic shift from the previous political norms. Since 1982, Bolivians have created a new democratic institutional framework.

My study highlights four political institutions that may explain democratic stability: 1) the electoral system, 2) the political party system, 3) executive-legislative relations, and 4) informal coalition-building rules. Though each institution has independent effects on democratic stability, they also interact in complex ways. Of the four, only the electoral system is strictly a formal institution outlined in specific constitutional and legal provisions. The electoral system also strongly affects the other institutions, especially the political party system and the number and type of parties. The least formal of these is the set of norms governing coalition building. Though influenced by electoral outcomes, coalition building among the political elite requires trust and the willingness to cooperate and bargain with one another.

2.3.1. Electoral Systems and Electoral Laws

Elections are an essential feature of modern representative democracy. In truth, "the democratic process is indeed encapsulated in elections and electing" (Sartori 1987, 86). Of course, we must beware of the electoralist fallacy: while elections are a necessary condition for democracy, they are not a sufficient condition (Linz and Stepan 1996). Nevertheless, free and fair elections allow citizens to choose between competing political elites and different policy options. Electoral systems make voting possible by stipulating, among other things, the number and types of offices contested, how votes are cast, and the counting rules used to determine winners and losers. Essentially, democracy becomes the only game in town when all actors agree to use competitive elections as the mechanism to decide who wields power in the polity. Implied, of course, is the stipulation that political actors agree to the specific electoral rules and that losers agree to respect electoral outcomes.

The procedural model of democracy relies on elections to make popular self-government possible in large political systems. While citizens of large, modern polities are no longer able to directly decide political issues, they can freely select their own representatives. Through competitive elections, citizens are able to influence public policy, articulate their interests, and hold government officials accountable (Manin 1997). Although democracies may also use referenda, ballot initiatives, or other electoral mechanisms, this study focuses on national-level elections for the executive and legislature.

The kind of electoral system used reflects the elementary foundations of the political system. Each counting rule aims to build a different type of majority or popular consensus. Different types of counting rules, however, vary significantly in their approaches. While single-member district systems are often meant to build elective majorities, proportional representation (PR) systems are usually designed to increase minority representation. By dictating how votes are translated into seats, different electoral systems affect citizen and elite behavior by providing different strategic choices (Lijphart 1994).

Electoral systems also strongly affect other political institutions, especially the party system. Duverger (1954) was one of the first to outline the relationship between electoral systems and party systems. PR systems tend to be associated with multiparty systems; simple majority systems tend to coincide with two-party systems. Majoritarian systems have a constraining effect on voters and a reductive effect on the number of parties (Sartori 1997). By limiting the possibilities that smaller parties can win, majoritarian electoral systems encourage voters and elites to limit their ballot choices. In contrast, PR systems, especially those with larger district magnitudes and lower thresholds, encourage a greater number of parties. Voters are more likely to expect their party to win some representation; consequently, minority parties are more likely to campaign rather than support a larger party.

2.3.2. Political Parties and Party Systems

Modern representative democracy is impossible without political parties and an institutionalized party system. Political parties link elites to voters, organize and articulate public political discourse, help make representatives accountable, and allow for challenges to political authority. Political parties are naturally consistent with democracy -even direct democracy. Finley (1985) and Manin (1997) point out that the Athenian social elite played an important role.[6] Although any Athenian citizen was able to voice a proposal in the Assembly, specially trained orators often served as de facto representatives. Of course, modern political parties are more highly organized and differentiated than simple political factions. In contrast to other political arrangements, electoral democracy allows citizens to choose the elites who will govern on their behalf. Political parties allow voters to organize behind elites they believe will best represent them.

Because an institutionalized party system is indispensable for democracy, significant attention should be given to the development of stable party systems in new democracies (Lipset 2000). A party system is institutionalized if parties are more than temporary or personal electoral vehicles. Institutionalized parties are linked to, and legitimately represent, important social groups and constituencies; they should coincide with the significant social cleavages and retain relatively stable bases of electoral support. Party systems also give voters intellectual shortcuts; voters should be able to identify basic policy tendencies of political parties or, at the very least, who their core leaders are and what they stand for. If parties are not institutionalized, if they are merely empty labels used haphazardly during elections, then voters are essentially voting randomly and without clearly articulated preferences.

Political parties also train and prepare potential government teams. Unlike other civic associations or interest groups, parties deliberately seek to place their members into government positions. When in power, parties are expected to translate the party's program into government policies. Thus, the recruitment and training of political elites is an important function of political parties. Because voters tend to hold parties accountable for their management of government, parties have incentives to nominate capable and responsible candidates for election.

Though electoral systems influence the party system's character, political parties and party systems also have independent social and historical roots. Lipset and Rokkan (1967) argued that Western European party systems froze in the 1920s before the full impact of electoral democracy. Because party systems in large part reflect and articulate a society's cleavages, the underlying social structure also influences the number and type of political parties in the party system. These historical legacies were especially felt in new democracies. Political parties were common in Latin America, even in countries with little or no history of democracy and democratic elections. The ability of these parties to learn to play the electoral game proved crucial.

Finally, political parties and party systems have an impact on the political process between elections. Competitive elections produce both a government (the winners) and an opposition (the losers). When they agree to play by the electoral rules, political parties must concede the right of the winners to exercise political power. Thus, the way parties work together after an election is important, not only for day-to-day government, but also for the continued survival of democratic politics. In large part, how political parties interact is shaped by the constitutional structure, especially those regulating executive-legislative relations.

2.3.3. Executive-Legislative Relations

Modern democratic systems make clear distinctions between the executive and legislative powers. Even in parliamentary systems, where the prime minister is technically a member of parliament, voters recognize that the prime minister and his or her cabinet wields the executive power, that is, the execution of government policy. The relationship between the executive and legislature can vary significantly, both between presidential and parliamentary systems and within them. These differences are often stipulated by constitutional structures; they are also, however, affected by the electoral system, the party system, and informal coalition-building norms. Like electoral systems, the norms regulating executive-legislative relations reflect underlying assumptions of the political system. The two basic constitutional types stem from different views of democracy.

Parliamentary systems closely bind the executive and legislature and reflect a populist theory of democracy. The populist theory identifies democracy with popular sovereignty and majority rule (Dahl 1956). Popular sovereignty is reflected in the election of a representative assembly. Executives in parliamentary systems are not elected by direct popular vote, but rather by parliament; the ability of parliament to call for a vote of confidence also makes the executive dependent on legislative support. Although parliamentary systems do tend to focus executive power in the cabinet, rather than the legislature as a whole, parliamentary cabinets are more collegial and spread decision-making beyond the prime minister (Lijphart 1999).

In contrast, presidential systems keep executive and legislative powers separate and reflect a Madisonian theory democracy. The Madisonian theory reflects an effort to restrain majority (and minority) tyranny by building compromise between competing interests (Dahl 1956). Presidential systems hold separate elections for the executive and legislature, who may represent different and competing social groups or interests. Thus, unlike in parliamentary systems, divided government is a very real possibility in presidential systems. Although executive power is centralized within the chief executive, the legislature retains its independent base of support and can check the president. Similarly, presidents can exercise veto power and restrain legislative power. The ability of different political parties to work together is therefore critical in presidential systems.

Linz (1990; 1994) argued that presidential systems are inherently unstable and less likely to lead to democratic stability than are parliamentary systems. Linz's critique focused on the two most prominent features of presidential systems: dual legitimacy and temporal rigidity. Separate elections for the executive and legislature give each a competing claim to legitimacy; since each is popularly elected, "no democratic principle can decide who represents the will of the people" (Linz 1994, 7). Similarly, because presidential systems do not allow for votes of confidence and tend to limit terms of office, they are less flexible than parliamentary systems; popular and effective governments cannot extend their mandate while voters are stuck with unpopular and ineffective governments until the next election. When presidents and legislatures disagree, dual legitimacy and temporal rigidity can collide dangerously.

Stepan and Skach argue that the conflict between executive and legislative powers inherent in presidentialism "systematically contributes to impasses and democratic breakdowns" (1993, 19). Like Linz, they argue that the failure of presidential democracy explains why democracy has not succeeded in Latin America, where most countries adopted presidential systems of government. Historically, conflicts between executives and legislatures were often solved by the military, acting as the poder moderador. More recently, presidential democracy is criticized for producing executives with authoritarian tendencies who claim to rule in the name of the people against the legislature (O'Donnell 1994).

Nevertheless, most new democracies have adopted some type of presidential system. This is especially true in Latin America, where no country has yet adopted a parliamentary system. A further problem is that many new democracies have adopted PR electoral systems for their legislature, which tend to increase the number of political parties and make presidents less effective. Again, Latin American electoral systems have followed this pattern closely. Still, subtle differences in the design and operation of presidential systems are significant. Differences in types of executive decree powers, for example, affect how legislatures and presidents interact.

The relationship between executives and legislatures is deeply affected by other institutional factors. Jones (1995) demonstrates that presidential democracy fares better when electoral laws can provide executives with majorities or near-majorities. Similarly, Shugart and Carey (1992) point out that different presidential systems balance executive and legislative power differently. Electoral systems intervene into the working of executive-legislative relations and affect them as much as do formal proscriptions separating their powers. Electoral systems also offer different incentives for building inter-party coalitions, which affect the relationship between executives and legislatures.

2.3.4. Coalition-Building Rules

Coalition-building rules are informal, commonly accepted norms of behavior that specify how different political parties can collaborate. Although liberal democracy relies on political parties that vie for power in competitive elections, coalitions allow rival parties to reduce some of the antagonism of electoral politics by coming together to build policy consensus. A key element of democracy is the principle of majority rule. Governing majorities can be built through coalition-building processes that allow for different political parties to come together through deliberation and agreement.

While formal institutions cannot make coalitions inevitable, they can make them more likely. Institutions offer incentive structures that can either encourage or discourage cooperation between rival elites. Electoral systems that promote antagonistic campaign strategies limit the possibility that political elites will work cooperatively after the election. 

Coalition-building rules may be highly institutionalized and broadly based, such as in consociational norms that rely on elite cooperation (Lijphart 1984; 1999). Consociational power-sharing agreements between elite cartels are credited with stabilizing liberal politics in societies with deep social cleavages, such as Austria, Belgium, and The Netherlands. In pluralist societies, the principle of majority rule effectively becomes majority tyranny if the alternation of power is not possible. One danger with consociationalism, however, is that it can lock power-sharing agreements into place for too long. This is especially true if cleavage structures change and new groups do not have access to political power. In Latin America, consociational agreements in Colombia and Venezuela were credited with preventing authoritarianism. Still, bipartisan agreements in both countries excluded political movements that emerged in the 1970s; the democratic crises in both countries have been in great part blamed on their consociational elite agreements (McCoy 1999; Gonzlez and Crdenas 1998).

Other types of coalition-building rules may be narrower and less static, such as the ad hoc governing coalitions common in parliamentary systems. Because executives are elected by the legislature in parliamentary democracy, multi-party coalitions are necessary whenever no single party wins a simple majority. Of course, different parties only need to agree to vote together to elect a prime minister; there is no reason why parties cannot subsequently return to the role of opposition. Governing coalitions, in which two or more different political parties agree to share and exercise power together, are nevertheless the norm in parliamentary systems. In such coalitions, the various member parties agree to share cabinet and ministerial positions between them. In many cases, coalition partners tend to be fairly stable and predictable, with some parties commonly joining together, often due to ideological closeness. Coalition governments have been relatively common in Latin America, though their character and frequency across countries is heavily affected by electoral and party systems (Deheza 1998).

Differences in coalition norms influence executive-legislative relations. In government coalitions of the type common in parliamentary systems, legislation and government policy is routinely hammered out between the coalition parties before going to the legislature for approval. Similarly, majority coalition governments may decide to grant the executive discretionary decree powers over wide policy areas.

3. Bolivia's Democratic Experience

Assessments of Bolivian democracy are mixed, despite its widely recognized stability. The normalization of politics around consensual practices and moderated elite bargaining point to the possibility that Bolivia's democracy was already consolidated by the late 1980s (Mayorga 1992). In contrast, these same elite agreements may merely be a continuation of patrimonial and dominated-dominated political discourse (Gamarra 1996). Although this study does not directly consider whether Bolivia's democracy is consolidated, it explores the relationship that political institutions have on democratic stability. This study aims to test whether Bolivia's democratic stability was the result of its unique political institutions (as Mayorga argues) or if it was caused by other factors, such as elite consensus (as Gamarra proposes). I also consider two other variables that may account for political stability: 1) the historical legacies of the 1952 National Revolution and 2) the role of particular individuals in building democracy.

3.1. The Transition to Democracy

Bolivia's democracy was inaugurated in 1982 after a long and difficult transition. Although several civilian governments ruled throughout its history, none of these qualify as democratic. The transition to democracy began in 1978 when then-dictator Hugo Bnzer Surez stepped down in favor of elections. The Democratic Popular Union (UDP), led by Hernn Siles Zuazo, won the 1978 election but was prevented from holding power when Bnzer's chosen successor, Juan Pereda Asbum, launched a military coup and declared the elections invalid. After another military coup, by David Padilla, elections were also held in 1979 and 1980, though no presidential candidate won a majority of the popular vote; parliament was unable to elect a president in either election and chose instead to elect a legislator as interim president until new elections could be held within a year.[7] This introduced a period of extreme political crisis as one military junta replaced another until the last junta stepped down in 1982.[8]

After civilian rule was restored in 1982, the parliament elected in 1980 chose Siles Zuazo, the UDP presidential candidate and plurality winner in all three elections. Siles Zuazo's UDP, a broad coalition of left-of-center parties, came to power at the beginning of the nation's worst economic crisis. As the debt crisis of the mid-1980s spiraled out of control, the UDP fell apart. By 1985, Siles Zuazo's vice-president, Jaime Paz Zamora, had effectively abandoned the president and taken his party, the Movement of the Revolutionary Left (MIR), into the opposition. Only the Bolivian Communist Party (PCB) remained steadfast; but its ideological commitments -along with its ability to mobilize the Bolivian workers, especially the strategically important miners- pressured Siles Zuazo against the difficult economic policy choices the situation required. Unable to solve the key problem of hyperinflation, Siles Zuazo stepped down and called for elections a year ahead of schedule.

The Bolivian left was soundly defeated in the 1985 elections, with most votes split between Bnzer's center-right Democratic National Action (ADN) and the centrist National Revolutionary Movement (MNR).[9] The effective collapse of the Bolivian revolutionary left after 1985 is significant. Bolivia's left had deep historical roots and was highly militant; the Trotskyite Revolutionary Workers Party (POR) and the Moscow-line Revolutionary Party of the Left (PIR) traced their roots back to the 1930s (Roln 1999). The Bolivian Labor Confederation (COB), the federation of Bolivian worker's unions, had been a backbone of the 1952-68 Revolutionary governments and the most powerful force (outside of the military) in Bolivian politics for most of the twentieth century. In post-1985 Bolivian politics, only MIR has managed to survive.[10] Similarly, though the COB is still able to mobilize significant support, it has not come close to dominating political discourse as it once did.

The 1985 election gave Bnzer a plurality over the MNR candidate, Paz Estenssoro, but it did not give him a clear majority. Under Bolivia's electoral system, presidential candidates must win by a majority. The potential stalemate was ended after parliament implemented Article 90 of the Bolivian constitution, which calls for the legislature to select the president in the event that no candidate wins a simple majority of the popular vote. Because many Bolivians were uneasy about giving the former military ruler control of the government, especially so soon after the return to democracy, the MNR was able to convince legislators to vote for Paz Estenssoro.

Barely two months after parliament elected Paz Estenssoro, in October 1985, the MNR founder entered into a political alliance with Bnzer. The "Pact for Democracy" ensured the new president a legislative supermajority. This was critical in order to pass urgently needed emergency measures to deal with hyperinflation nearing 25,000 percent and the general economic crisis. In exchange, the Paz Estenssoro government implemented many of the neoliberal economic reforms favored by ADN.[11] This New Economic Policy (NEP) reduced inflation to just eleven percent in one year and stabilized the economy while also fundamentally dismantling much of the previous state-centered development model (Sachs and Morales 1988). Ironically, it was Paz Estenssoro and the MNR that had originally created this statist economic model after coming to power in the 1952 National Revolution. Subsequent governments have also continued neoliberal economic policies.

3.2. Building Democratic Stability, 1985-present

The 1985 election established the pattern for democratic politics in Bolivia. Since 1985, no president has been elected directly; parliament has elected the president in each election. Parliamentary election of the president has also made coalition government unavoidable. To win support for their presidential candidate, parties have built governing coalitions under complex political pacts. In intense inter-party negotiations during the brief month between the election and the swearing in of the new president on 6 August, parties agree to share cabinet and other ministerial posts as well as to adopt policy packages in exchange for parliamentary support for their candidates.  For the most part, these pacts have held surprisingly well, making Bolivian governments since 1985 true coalition governments. Stongly disciplined political parties have guaranteed every Bolivian president since 1985 a working legislative majority.

The MNR-ADN alliance, however, did not last into the 1989 elections. Though MNR and ADN had signed a secret addendum to the Pact for Democracy in May 1988 under which the MNR agreed to back Bnzer in the 1989 election, the new leader of the MNR, Gonzalo Snchez de Lozada, chose to run for president. As Paz Estenssoro's Planning Minister, and responsible for most of the economic stabilization reforms, Snchez de Lozada hoped that his popularity would propel him to an electoral victory. Bnzer took Snchez de Lozada's decision to run for the presidency personally, sparking a bitterly contested election campaign.

The MNR won a narrow plurality in 1989 with 25.6 percent of the vote. Bnzer, who placed second with 25.2 percent, was unwilling to cede the presidency a second time to the MNR. Meanwhile, MIR had doubled its vote share from 1985 and placed third with 21.8 percent. Under Article 90, parliament was able to select from among the top three candidates. Consequently, all three parties pursued the presidency. Opposed to the neoliberal economic policies of Snchez de Lozada, the left-of-center MIR was not prepared to back MNR. Paz Zamora and MIR were also unwilling to back ADN since Bnzer's military regime had brutally repressed the left during his military regime. The stalemate was ended when Bnzer ordered his party to vote for the third place MIR-candidate, Paz Zamora. The result was another political pact, the "Patriotic Accord" (AP), an unanticipated alliance of the traditional right and left against the center.

Although Paz Zamora took the presidency in 1989, Bnzer and ADN held considerable power. Under the AP agreement, half of the cabinet and ministerial appointments were given over to ADN. Similarly, most government policy was initiated in the Committee of the Patriotic Accord, chaired by Bnzer. The heavy reliance on ADN for a legislative majority partially explained why, despite having campaigned against the neoliberal economic model imposed under Paz Estenssoro, Paz Zamora continued the same neoliberal policy orientation initiated by the 1985-89 MNR-ADN government. Although originally an uneasy alliance, Paz Zamora and Bnzer were able to reign in their parties and hold the coalition together with considerable discipline. Like the MNR-ADN agreement, the AP accord called for MIR to support Bnzer in the next elections.

In 1993, Bnzer again ran for the presidency, this time under a single AP electoral list binding ADN and MIR. Accusations of corruption within the Paz Zamora government, however, dogged Bnzer's presidential campaign. At the same time, two new populist parties, Conscience of the Fatherland (CONDEPA) and Solidarity Civic Union (UCS), cut deeper into the anti-incumbent vote. Popular resentment against the neoliberal economic policies was especially heavy among the urban poor. Led by the charismatic media mogul Carlos Palenque, CONDEPA captured most of the Aymara vote in La Paz and El Alto, the large slum city attached to the capital. Max Fernndez, owner of the largest national brewery, propelled UCS by building schools, hospitals, and even distributing food and consumer goods throughout the countryside.

Snchez de Lozada won an overwhelming plurality in the 1993 elections with 35.6 percent of the vote. Bnzer's combined ADN-MIR list, in contrast, took a mere 21.1 percent -less even than the third-place MIR had won on its own in 1989. Shortly after the election, Bnzer conceded defeat and the presidency was securely in the hands of Snchez de Lozada and the MNR. Although the 1993 election gave the MNR a comfortable majority in the Senate with 17 of the 30 seats, it had won only 51 seats in the 130-seat House of Deputies. To secure a legislative majority, Snchez de Lozada pieced together a coalition an agreement known as the "Pact for Governability" with the small leftist Free Bolivia Movement (MBL) and the populist UCS.

Part of Snchez de Lozada's electoral appeal stemmed from his alliance with one of the most significant campesino (indigenous peasant) parties, the Tupaj Katari Revolutionary Movement of Liberation (MRTKL). Although the kataristas never won more than three percent of the vote on their own, the inclusion of Vctor Hugo Crdenas, the MRTKL leader, as the vice-presidential candidate for the MNR-MRTKL electoral front appealed to millions of campesino voters. The MNR-MRTKL victory made Crdenas the first indigenous vice-president in Bolivia's history. Under their "Plan for Everyone" platform, Snchez de Lozada and Crdenas campaigned with the promise of social and institutional reforms to improve the condition for Bolivia's historically neglected indigenous majority.

The Snchez de Lozada government implemented numerous second-generation reforms meant to modernize the economy and decentralize Bolivia's political system. One of these, the Law of Popular Participation (LPP), divided the country into 311 popularly elected municipal governments. The law had important implications. First, it introduced democratic politics to the local level; previously only the major cities held mayoral elections. Second, the LPP also guaranteed state resources -20 percent of the national budget- to be allocated on a per capita basis to each municipal government. This brought badly needed economic resources to long-neglected rural communities.

 

Table 1. Presidents and government coalitions, 1982 to present

 

Year

President

Party

Presidential votes

Seats in lower house

1982-85

Hernn Siles Zuazo

UDP

38.7%

43

 

 

 

 

 

1985-89

Vctor Paz Estenssoro

MNR

30.4%

43

 

Supporting parties

ADN

 

41

 

 

 

 

 

1989-93

Jaime Paz Zamora

MIR

21.8%

33

 

Supporting parties

ADN

 

38

 

 

 

 

 

1993-97

Gonzalo Snchez de Lozada

MNR-MRTKL

35.6%

52

 

Supporting parties

MBL

 

7

 

 

UCS

 

21

 

 

 

 

 

1997-

Hugo Bnzer Surez

ADN-NFR-PDC

22.3%

32

 

Supporting parties

MIR

 

23

 

 

UCS

 

21

 

 

CONDEPA1

 

19

Sources: Based on data provided by Corte Nacional Electoral.

Notes: 1CONDEPA was expelled from Bnzer's government coalition on 6 August 1998.

 

Other programs initiated by the Snchez de Lozada regime included a variety of socioeconomic reforms. The package of amendments to the constitution included inclusive language that identified Bolivia as a "pluricultural and multiethnic" republic, broad recognition of indigenous rights, and lowered the voting age from 21 to 18. Like the LPP, the reform of the agrarian law (INRA) and the Law of Administrative Decentralization meant to streamline and modernize Bolivia's political system while making local resources more readily available to citizens. Finally, the Capitalization Law further liberalized the national economy by allowing foreign investors to capitalize half the stock of state owned enterprises. In contrast to simple privatization, foreign investors were to invest in equal proportion to the corporation's market value, essentially doubling the value of Bolivian corporations. The capitalization of strategic public enterprises was unpopular among many Bolivians who saw the MNR regime as giving away much of the state sector to foreign capital.

Bnzer, again ADN's presidential candidate, won a narrow plurality in the 1997 elections over the MNR's Juan Carlos Durn. With the support of MIR, CONDEPA, and UCS, Bnzer assumed the presidency on 6 August 1997. The centerpiece of Bnzer's foreign policy (in terms of US-Bolivian relations) was an escalation of the US-backed war on drugs in Bolivia. The campaign was successful, practically eradicating for-export coca crops by 2001, despite several violent confrontations with organized cocaleros, the coca-growing peasants of the Chapare region. Due to health reasons, Bnzer left the presidency in 2001 and turned power over to his vice president, Jorge Quiroga Ramirez.

3.3. Explaining Democratic Stability

One possible explanation for Bolivia's democratic stability might be its unique political institutions. Since 1985, Bolivia's political institutions have merged into a system of "parliamentarized presidentialism" (Mayorga 1997). The system is characterized by: 1) a fused-ballot PR electoral system; 2) congressional election of the executive; and 3) informal coalition rules that provide presidents with legislative majorities. Bolivian voters choose from party lists headed by presidential candidates (fused-ballot PR). Presidential candidates win directly only if their list wins a majority. If not, the new legislature elects a president from among the front-runners. To gain support for candidates, parties join coalitions in exchange for shares of state patronage. Because no party list has yet won a simple majority, parliamentarized presidentialism has become institutionalized.

Three other possible explanations must also be considered: 1) the role of the political class, 2) the impact of historical legacies, and 3) the role of key individual statesmen. A key moment in Bolivia's history is the National Revolution. Launched by the MNR on 9 April 1952, it fundamentally altered Bolivia's political landscape (Malloy and Thorn 1971). If Bolivia's revolution is consolidated, democratic stability may be facilitated by broad social and elite consensus around the political discourse of revolutionary nationalism (as in Mexico). Bolivia's democratic stability might also be the product of the role of elites. Determined elites could, possibly, make democracy work under any circumstances and institutional arrangements. The actions of such elite personalities as Siles Zuazo, Bnzer, or Paz Estenssoro, may better explain democratic stability. Finally, political stability in Bolivia might be a function of patrimonialism among the political class, as Gamarra argues. If a small political class benefiting from government privilege maintains elite stability, then Bolivian democracy may be stable, but it is not representative and faces limited prospects for long-term durability.

3.3.1. Parliamentarized Presidentialism

The key element of parliamentarized presidentialism is the electoral system. The fused ballot is, essentially, a parliamentary ballot; it closely binds presidents to the legislature and eliminates the problem of dual legitimacy plaguing many presidential systems (Linz 1994). The simple structure of the Bolivian ballot fuses the election of the executive and legislature into one singular vote choice (resembling ballots in parliamentary systems). When voting, Bolivian citizens receive a simple multi-color, multi-sign ballot that has the name of each presidential candidate along with the name, signs, and colors of their party. Voters are then given a pencil and simply asked to mark the box for their presidential candidate. Seats in the lower and upper chambers of the legislature are then given out in proportion to vote shares. If a candidate wins a majority of the popular vote, he or she is automatically chosen president.

The 1994 constitutional reforms, which went into effect in the 1997 election, did not significant alter Bolivia's political system. Parliament is now limited to selecting from among the top two (rather than three) presidential candidates.[12] The practice of building coalitions has gone largely unaffected and the change may instead have helped to streamline the process. The introduction of a German-style multi-member proportional (MMP) electoral system also does not seem to have significantly altered the political system. Although MMP allows for half the lower house to be elected in single-member districts, the system is still a PR system since the other half of the seats are apportioned in compensatory fashion. The upper house is still elected under the old PR formula.

Bolivia's electoral system also calls for the legislature to elect the president if no candidate wins a majority, rather than a second round popular election. No presidential candidate has yet won a popular majority and it seems highly unlikely that any will in the near future. The legislature has chosen the president in every election since 1985 after intense inter-party bargaining and negotiation. Here, informal coalition building rules play an important role. The legislature could simply choose to elect a president without building a formal governing coalition. Instead, every government since 1985 has involved a formal power-sharing agreement between coalition partners. These coalitions have provided each president with a sustained and disciplined legislative majority.

3.3.2. The Role of the Political Class

Gamarra and Mayorga both agree that the normalization of political pacts is a key factor in Bolivia's political stability. They disagree, however, on the origins and long-term implications of these pacts. Where Mayorga (1997) sees parliamentarized presidentialism as evolving from a combination of institutional incentives and political learning, Gamarra (1994) suggests that elite cooperation was made possible by the collapse of organized labor, elite convergence around neoliberal economic politics, and patrimonialism. While consociational practices "reveal a degree of political maturing of Bolivia's political class," they also tend to exclude the opposition (Gamarra 1997).

The political pacts characteristic of Bolivian politics since 1985 may, as Gamarra suggests, be more the product of elite patrimonial bargaining than on the incentive structures of political institutions. If so, stable politics in Bolivia may simply revolve around an actively engaged political class made up professional politicians and the small middle class elite that lives off of politics -this includes high and mid-level professional bureaucrats and party leaders- whose careers depend on the success of their political parties.

Bolivia's political parties pose a key question. Several Bolivian social scientists have called for institutional reforms to improve the weak links between political parties and civil society (Rojas and Zuazo 1996; Tornado and Exeni 1994; F. Mayorga and Paz 1999). If parties have weak connections to civil society, the elite bargaining arrangements that supported governmental stability since 1985 have little to do with institutionalized political parties and more to do with strong caudillos who command their supporters' obedience. If so, a lack of representation could erode support for democracy in the long term. Survey evidence suggests Bolivian voters support democracy in principle, though they have negative evaluations about Bolivia's democracy in practice (Lazarte 1993; Rojas and Verdesoto 1997). Especially negative are assessments of the formal institutions of democracy and their representative character.

 

Table 2. Current vote and seat shares for Bolivian political parties

 

Party

Percent vote share

Seats in lower house

Seats in upper house

ADN-NFR-PDC

22.3

32

11

CONDEPA

17.2

19

3

IU

3.7

4

0

MBL

3.1

5

0

MIR

16.8

23

7

MNR-MRTKL

18.2

26

4

UCS

16.1

21

2

Sources: Based on data provided by Corte Nacional Electoral.

 

The growth of two populist parties, UCS and CONDEPA, in the 1990s is significant. Both parties have mounted effective electoral challenges to the traditional parties. Led by highly charismatic populist leaders, both seemed to be anti-system alternatives. Neither party has played a destabilizing role, however, and they have instead been brought into the normal political discourse by the political system (Mayorga 1995). Bolivia's political institutions provide opportunities for the accommodation of such new political forces. UCS and CONDEPA have joined MNR, MIR, and ADN as the major political parties. They have also emerged as potential kingmaker parties able to shift the balance of power from the two political poles, one centered on the MNR and the other on the ADN and MIR.

3.3.3. Historical Legacies

A historical institutionalist perspective also considers the effect historical legacies have on contemporary politics. These legacies form part of the context within which political actors interact. Actors are constrained by institutions (especially those they inherit rather than those they create or redesign) as well as the historical lenses through which they view the costs and incentives of institutional strategies. This study distinguishes between political institutions and broader historical or contextual institutions. Unlike political institutions (such as electoral systems) that constrain actors' preferences in strategic ways, historical legacies (such as a social revolution) alter actor's preferences indirectly by shaping the political culture within which they interact.

The 1952 National Revolution was one of only four successful social revolutions in modern Latin America. The other three revolutions occurred in Mexico (1910), Cuba (1959), and Nicaragua (1979). Most accounts of the 1952 Bolivian Revolution describe it as either "incomplete" or "unconsolidated" (Selbin 1999; Malloy 1970). Nevertheless, there is reason to suspect that the core revolutionary project may have been accepted by a majority of Bolivia's current political elites.

Prior to 1952, the Bolivian state was dominated by a small oligarchy composed of wealthy mining barons and owners of large landed estates (the latifundias). The nation's majority indigenous population was excluded from political, social, and economic life. The small middle class and the growing working class (especially miners) were similarly excluded from politics. Through its control over the military, the Bolivian oligarchy was able to suppress the increasingly militant labor movements that emerged in the 1920s.

The Chaco War (1932-35) dramatically altered Bolivia's political developments and was the major contributing factor to the 1952 Revolution. After a series of clashes along the disputed Bolivia-Paraguay border, full war broke out in 1932 after oil deposits were discovered in the region. The war was an unmitigated disaster with more than 65,000 Bolivian casualties (one quarter of combatants) and the loss of significant territory (doubling the size of Paraguay). The end of the war led to a nation-wide rejection of the white oligarchy, which was blamed for mismanaging war. Similarly, accusations that the war had been fought on behalf of Standard Oil led to sweeping critiques of the role of international capital in Bolivian national politics and helped radicalize the small urban middle class. Immediately after the Chaco War, disaffected mid-ranking officers usurped the civilian government.

The MNR emerged in the aftermath of the Chaco War, along with other radical and revolutionary movements. Led by young veterans and intellectuals, most of these movements sought to create a new more progressive and modern Bolivian state and society. In 1943, the MNR participated in a civil-military coup, though it was soon eclipsed within the governing coalition. Only a few years later, however, the MNR became the nation's most significant political party. The MNR, campaigning behind an exiled Paz Estenssoro, won a clear majority in the 1951 elections. After the military prevented the MNR from assuming power, the MNR began preparing to take power by force.

Bolivia's social revolution completely altered the country's power structure. Backed by highly politicized and militant miners' militias, the middle-class MNR came to power after a brief, three-day civil war that shattered the armed forces and swept away the traditional landed oligarchy. On 9 April 1952, the MNR civilian militias led by Hernan Siles Zuazo took to the streets of La Paz. The brief civil war was tipped in favor of the MNR after organized miners militias joined the revolution, attacking loyalist military forces throughout the country and marching on the capital. Only three days after it began, the revolution was over and Paz Estenssoro returned from exile to assume the presidency.

Among its reforms, the MNR introduced universal adult suffrage, nationalization of the mining industry and the creation of the COB, sweeping land redistribution to the peasants, and a purge of the armed forces. Even after the military seized power in 1964, it did not attempt to reverse or challenge the central gains of the Bolivian Revolution.[13] Essentially, many of the military rulers were members of the new Bolivian military and represented the military cell of the MNR.[14]

The return to democracy also returned many of the revolution's key actors (Siles Zuazo and Paz Estenssoro). Some even portray democratization as a new "democratic revolution" (Bedregal 1996). None of the post-1982 political parties could be classified as counter-revolutionary parties; most embrace its myths and symbols and trace their lineage to the MNR and the revolution. UDP was headed by Siles Zuazo, one of the founders of the MNR and leader of the movement's center-left faction. MIR was founded by MNR's leftist student wing in 1971 and headed by Paz Zamora (Paz Estenssoro's nephew). Even Bnzer's ADN is rooted in the Revolution's political discourse of revolutionary nationalism and represents some of the MNR's rightist factions.[15] The absence in Bolivia of a genuine political right, coupled by the collapse of the Marxist left in the 1980s, may help explain Bolivia's democratic stability. The centrist character of Bolivian politics, with its agreement on basic neoliberal policies, may be the result of the historical development of Bolivia's party system, rather than due to the molding influence of its electoral system.

3.3.4. The Role of Individual Statesmen

There is no denying that the actions of individual political actors have profound implications. Much of the democratic transition literature points out the important role that key individuals had in promoting democracy (Linz and Stepan 1996; Di Palma 1990). Bolivia's democratic stability might be the product of powerful statesmen who may have made democracy work under even the worst institutional conditions. Bolivian democracy may have been saved on different occasions by the actions of key individuals. At least three individuals stand out as important in Bolivia's democratic transition: Paz Estenssoro, Siles Zuazo, and Bnzer.

Paz Estenssoro was, arguably, the most important figure in twentieth-century Bolivian politics. As a principal founder of the MNR in 1941-2 (along with Siles Zuazo), Paz Estenssoro was one of the chief architects of the post-1952 Bolivian state. As leader of the MNR, Paz Estenssoro was also among the most visible and important civilian elites involved in the democratic transition process. His ability to discipline the MNR as it adopted neoliberal economic reforms to solve the economic crisis in 1985 are also important factors. Siles Zuazo's resignation one year ahead of schedule in 1985 in favor of early elections saved Bolivia from enduring one more year with an ineffective government during the nation's worst economic crisis. Bnzer's support for the democratic process since 1978 may also have been crucial (Klein 1992). The former military dictator's willingness to concede the presidency to Paz Estenssoro in 1985 (despite having won a plurality) and his dramatic overture in 1989 towards MIR were critical junctures in Bolivia's democratic development.

Finally, younger, second-generation political elites may also play key roles in supporting democratic stability. These may include politicians such as Snchez de Lozada, Paz Zamora, and Max Fernandez (founder of UCS) or important social, economic, or military players.

4. Hypotheses

The following is my general hypothesis: Twenty years after Bolivia's transition to democracy, the political system of parliamentarized presidentialism appears institutionalized. Since 1985, every president has been elected by a parliamentary coalition. These coalitions brought previously antagonistic groups of political elites together into cooperative power-sharing arrangements. The literature suggests at least four different explanations for Bolivia's democratic stability. Thus, I specifically test the following hypotheses:

Hypothesis 1.         Bolivian democracy has been stable because the behavior of political elites is strongly influenced by parliamentarized presidentialism's incentive structures.

Hypothesis 1a.   Bolivian democracy has been stable because of its electoral system.

Hypothesis 1b.   Bolivian democracy has been stable because of its party system.

Hypothesis 1c.   Bolivian democracy has been stable because of its executive-legislative relations.

Hypothesis 1d.   Bolivian democracy has been stable because of its coalition-building rules.

Hypothesis 2.         Bolivian democracy has been stable because of elite patrimonialism.

Hypothesis 3.         Bolivian democracy has been stable because political elites share a common political discourse inherited from the 1952 Revolution.

Hypothesis 4.         Bolivian democracy has been stable because of the determination or willingness of key political elites.

I do not ignore the possibility that Bolivia's democratic stability may stem from a combination of factors; nor do I dismiss the possibility that the strength of the influence of explanatory variables may change over time. For example, democratic stability may have, in the earliest stages, been more a factor of the role of strong individuals (such as Paz Estenssoro and Bnzer) to support democracy. Over time, however, the influence of individuals may have waned, giving way to some other factor, such as patrimonial politics. Thus, my research aims to uncover: 1) which of these factors were most influential in different periods in time and 2) which factor is most important at this time.

It is important to note the implications of each of the four central hypotheses. If democratic stability is the product of Bolivia's unique political institutions, it is possible that these institutions may similarly support democratic stability in other similar cases. If stability is merely the product of patrimonial politics, then Bolivia's democracy may not be viable. The current status quo could easily be unsettled if any of the groups of political elites decide it no longer wants to play the democratic game and decides to mobilize either the military or the popular masses. If Bolivia's democratic stability is a legacy of the 1952 Revolution, then it is possible that Bolivia's democracy may endure but the lessons of the Bolivian case are only generalizable to other countries that experienced social revolutions. If Bolivia's democracy is stable because of the determination of key elites, then Bolivian democracy is fragile at best (it may not survive the death of these individuals) and the case is not generalizable (other than to call for powerful pro-democratic elites).

5. Research Design

This study is essentially a case study of Bolivia's democratic system since 1982 based on fieldwork in that country. Nevertheless, I employ an embedded case study approach, placing the Bolivian case within a comparative perspective involving two similar countries, Ecuador and Peru. As a comparative study, this study adopts a "most similar with different outcomes" (MSDO) research design (Przeworski 1987). This approach assumes that the differences in outcomes between cases should be explained by their differences, rather than by their similarities. Placing a study of Bolivia's democratic stability within the context of its Central Andean neighbors allows the research to narrow the number of possible explanatory variables.

5.1. Case Selection

The three Central Andean republics -Bolivia, Ecuador, and Peru- were among the first countries in the third wave of democracy. Nevertheless, their political outcomes have been dramatically different. While Bolivia's democracy has been stable and uninterrupted for two decades, democracy has not been stable in Ecuador or Peru. Peru's experiment with democracy ended abruptly in 1992 after president Alberto Fujimori's military-backed autogolpe (self-coup) disbanded the legislature, purged the judiciary, and suspended the constitution. Ecuadorian democracy has also had a troubled road. Since 1979, the forced removal of two presidents from office, the brief kidnapping of another by the military, and several military uprisings and coups have kept the future of Ecuador's democracy uncertain. Comparing such different outcomes along the dependent variable (democratic stability) is possible because the cases are substantially similar.

The Central Andean republics are classified as "developing" countries by the United Nations Development Program (1999). They rank the lowest in South America (except for Paraguay, which ranks among them) along socioeconomic indicators such as annual per capita GDP, human development index, life expectancy, and adult literacy rates. Along these indicators, Ecuador and Peru are roughly comparable, while Bolivia is the least developed country in the region. The overall failure of democracy in the Central Andes could be explained by modernization theory to be a product of the region's low levels of socioeconomic development. Here, the stability of Bolivia's democracy is exceptional. Bolivia is the poorest and least developed of the Central Andean republics, suggesting that socioeconomic variables alone cannot account for the stability of democracy.

The Central Andes is also culturally homogenous. Beyond a similar Spanish cultural legacy, the three are the only South American countries to have indigenous majorities. These Quechua- and Aymara-speaking populations have been historically excluded from national political, economic, and cultural life.

All three countries also face similar obstacles to democracy. Neither had any previous historical experience with democracy. They share a long history of authoritarian rule. Although civilian governments have been common at different times in all three countries, none of these would meet the criteria for polyarchy. None of the three countries had an institutionalized party system.[16]

The three also share other similarities that may or may not have positive effects on democratic stability. US and Western pressure in favor of democracy has been strongly felt in the region. International pressure, including from the Organization of American States, was powerful enough to convince Fujimori to hold elections (though far from under ideal circumstances) shortly after he seized power. The demonstration effects of other successful democratizations in the region (Argentina, Brazil, Uruguay) were also significant across the region. All three cases shared similar transition processes; pacted transitions were the norm. Bolivia, Ecuador, and Peru have also been involved in the US-led drug war, receiving significant military and technical support. Finally, all three have embraced neoliberal economic policies.

Comparing Bolivia with Ecuador and Peru allows for control of an important variable: level of social unrest. While popular protest and unrest has been common in all three countries, only Peru experienced a guerilla war during the democratization process. No doubt the Shining Path guerrilla movement played some role in the breakdown of democracy in Peru.

5.2. Data and Method

My study aims to more rigorously tests the rival hypotheses presented in the literature by Mayorga (stability as product of institutional design) and Gamarra (stability as product of patrimonialism). In addition, I will test two other possible explanations for Bolivia's democratic stability. The data I will collect includes an extensive survey of political elites, content analysis of the written historical record, and comparative analysis.

A focus on political elites is consistent with the procedural definitions of democracy. The electoral process puts politicians and their political parties at the center of politics. How politicians behave -the strategies they employ-significantly affects the democratic process. I will consider elites including politicians, bureaucrats, business leaders, public intellectuals, and military leaders.

5.2.1. Content Analysis

To explore changes in elite behavior and attitude between 1982 and 2002, I will use qualitative and quantitative content analysis of published statements. Content analysis allows me to determine if elite attitudes (individual or collective) shifted over time in favor of moderated politics and coalition-building norms. I do not expect elite statements to be true representations of their attitudes. Instead, I will use content analysis to interpret their statements.

While I will focus on statements made in national newspapers, I will also seek statements made in other sources, such as in party archives (when made available), the Library of Parliament debate and voting records, and the public intellectual literature. Newspaper and print media archives are readily available at the Center for Documentation and Information (CEDOIN) in La Paz. CEDOIN is an independent archive of print media; it maintains indexed copies of articles from Bolivian newspapers and news magazines going back through the 1980s.

Those relevant materials for analysis include public statements made by politicians, party leaders, public intellectuals, and leaders of social movements. Such statements are included in direct quotations of such individuals in the press or statements made by them in the media meant for public consumption.[17] Public academic or intellectual literature is also important, since it provides a written record of the public intellectual discourse over time.[18]

The specific content categories I will measure include: 1) assessments of the quality of democracy, 2) evaluation of individual political elites, 3) evaluation of democratic institutions (e.g. the presidency, the judiciary, the National Electoral Court), 4) evaluation of the electoral system, 5) evaluation of political parties, 6) evaluation of the party system, 7) evaluation of social movements, and 8) evaluation of the 1952 Revolution. Each statement about any of the eight categorized will be described along a five-point scale from "negative" (0) to "positive" (4). Statements referring to individuals will also be separated to distinguish the target of the statement (the individual the statement refers to). The same procedure will be made for similar statements (references to specific social movements, political parties, ministries, or institutions).

The recording unit for analysis will be the entire statement, regardless of its length, made by one individual or group. Since I am interested in analyzing how elite opinion changed over time, recording the number of negative or positive sentences or paragraphs in a single statement is not as important as noting whether the statement as a whole negatively or positively assesses one of the content categories. I am primarily interested in the distinguishing the change in opinion for individual elites over time. However, some public statements may be made by a group (such as a political party) to reflect its official position as a group. If the author or spokesperson is a single individual, I will treat the statement as coming for a single individual, even if the person claims to speak for the whole party. If the statement has numerous authors (as in a public manifesto or party position paper), I will treat the statement as being made by the group.

The time period for each section of analysis will be the month. This provides 240 possible observations.[19] While scores on the negative-positive evaluation scale will not be additive, I will compile a cumulative index of the number of statements in each category with the same score.

I will seek to establish intercoder reliability by providing a sample of documents to other researchers in Bolivia to code independently. If our coding is comparable, I will continue independent coding. I may, however, ask Bolivian colleagues to independently code especially difficult documents to ensure that I my coding is reliable.

Analysis of the written record has several advantages: 1) easy access to the data, 2) the data is non-reactive, 3) allows for study across time, 4) allows for a large sample size, and 5) the cost of collecting the data is born primarily by archivers. There are, however, potential problems with using written record. Access to documents may be limited, leading to sample bias. Such problems are most limited, though, in the CEDOIN archives. Thus, I will rely primarily on their newspaper and news magazine archives.

5.2.2. Open-Ended Interviews with Political Elites

I will also interview political elites (in Spanish) using an open-ended survey format. The interviews will cover the same categories used in the content analysis of the written record. Additionally, I will ask them questions relating to: 1) the factors they consider most important for Bolivia's political development since 1982, 2) their assessment of the quality of Bolivia's democracy, 3) what changes are necessary to improve Bolivia's democracy, and 4) their assessment of the 1952 Revolution and its impact on contemporary politics.

Elite interviews are useful because of their special knowledge of and personal involvement with the political process. While mass opinion surveys are useful, they do not allow me to study political elites motivations and attitudes, as well as their interpretation of events. Elite interviews also supplement the written record; I can ask elites to elaborate on or respond to statements they made previously, evaluate events of which they have personal first-hand knowledge, or reflect on changes in attitudes over time.

Open-ended surveys are more useful when interviewing elites. Because of their special knowledge and involvement in the political process itself, highly structured, close-ended surveys lack the level of informative power that open-ended surveys provide. An open-ended survey format will allow me the flexibility to pursue different lines of inquiry. It also gives me the flexibility to specifically tailor surveys to the respondent, allowing me to ask only the most relevant questions to the knowledge of that individual. Nevertheless, for standardization purposes, I will ask some standard opening questions. These will be designed both to set the respondent at ease as well as to establish myself as a qualified researcher.

Like all surveys, open-ended surveys are subject to potential question order effects. To some extent, I can control for this by letting the respondent guide the interview in certain respects. While still making sure that each interview covers all relevant material, an open-ended format allows me to let the respondent choose the most salient issues first and ask him or her to elaborate on them. Further questions are, in some regard, responses to elite statements during the interview.

I will try to limit each interview to one hour in length. This limit will help to prevent redundant or repetitive questions as well as to prevent fatigue (both for the researcher and, more importantly, the respondent). When permitted, I will record the interview on cassette. I will provide respondents with full confidentiality, though I will keep track of the identity of each interviewee. I plan to interview about 60-80 elites during my stay in Bolivia.

5.3.3. Comparative Methods

I will also use comparative methods, including Qualitative Comparative Analysis (QCA), to test relationships between different political institutions and democratic stability. Cross-national comparisons will include a variety of institutional variables. Some of these will include differences in electoral rules, coalition-building norms, and number and types of executive decrees. Designed for research where the number of variables exceeds the number of cases (Ragin 1987; 1994), QCA enables cross-national comparisons between Bolivia, Ecuador, and Peru. The method makes it possible for comparative analysis to retain the complexity of each case. Each case is considered as a constellation of possible dichotomous (or multichotomous) variables. Nevertheless, Boolean procedures allow for the reduction in the number of possible explanatory variables. QCA also allows for multiple and combinatorial causal relationships.

Many of the institutional variables I consider using QCA are stipulated in constitutional frameworks and other laws. These include, among other things: whether elections use PR systems, whether executive and legislative elections are linked, whether executive and legislative elections are synchronic, whether second round elections are used, the number of effective political parties, levels of electoral disproportionality (the proportion of votes to seats), and changes in electoral volatility (party vote share changes between elections). The unit of observation for this part of the study is the electoral cycle.

Qualitative Comparative Analysis between Bolivia, Ecuador, and Peru does not require fieldwork in the control cases and can be finished before fieldwork in Bolivia begins. QCA does not, however, make the fieldwork portion of the study (content analysis of the written record and elite interviews) unnecessary. Even if QCA establishes a possible causal relationship between Bolivia's specific constellation of institutions and democratic stability, the institutional hypothesis is not validated. Political institutions offer political actors incentive structures, but political actors might act in ways that produce the same outcome (democratic stability) for other reasons. Fieldwork is essential to test and refute the alternate hypotheses.

6. Necessity of Field Research and Previous Research

Further research for this project requires fieldwork. Archival data is available only in Bolivia. Elite interviews are essential to examine elite attitudes twenty years after democratization. I conducted two preliminary research trips: In 1998 I studied the Law of Popular Participation and spent three months in Bolivia consulting with experts and government agencies, gathering archival data, and familiarizing myself with Bolivia's democratization literature. This trip persuaded me to study the role of institutions in promoting democratic stability. I returned to Bolivia in the summer of 2001 to reestablish and broaden my research contacts in anticipation of this longer research trip.

I presented previous research at various conferences. A paper on the Law of Popular Participation was presented at the March 2000 congress of the Latin American Studies Association (LASA). Papers on Bolivia's democratization process were presented at annual meetings of the Midwest Political Science Association (MPSA) in 1999 and 2001, as well as at the September 2001 LASA congress. A paper comparing the quality of democracy in Bolivia, Colombia, Ecuador, Peru, and Venezuela was presented at the 2000 MPSA annual meeting. These papers were well received and I was highly encouraged at these conferences to pursue research on Bolivia's democratic stability.

7. Affiliation Sites, On-Site Activities, and Timeframe

During my fieldwork, I will work closely with Bolivian social scientists. I have been extended personal invitations from Carlos Toranzo, co-director of the Latin American Institute for Social Research (ILDIS), and Ren Antonio Mayorga, director of the Bolivian Center for Multidisciplinary Studies (CEBEM). I will also continue expanding relationships with researchers at the National Electoral Court (CNE) and at the USAID program for Democratic Deepening and Citizen Participation (USAID-DDPC). Throughout my stay in Bolivia, I will participate in academic exchanges (conferences, seminars, round tables) and continue to familiarize myself with recent political literature.

I plan to spend the first month gathering archival data, focusing specifically on newspaper archives at CEDOIN. After organizing the archival sources, I plan to spend the second month refining my interviewee list in consultation with Bolivian colleagues and after reviewing archival data. By the end of the third month, I hope to have enough interviews completed to allow for preliminary content analysis of both archival data and elite interviews, reviewing with Bolivian colleagues to test for intercoder reliability. After the prelimary content analysis, I will refine my content analysis as needed. I plan to spend the remainder of my time (at least four months) continuing to gather elite interviews.

 

8. Preliminary Outline

Chapter 1 will discuss modern democratic theory, especially as it relates to Latin America and the third wave of democracy. It will outline the definition of democracy and democratic stability used in this study, linking them to the liberal theory of representative democracy and the theory of political institutions.

Chapter 2 will test the relationship between political institutions and democratic stability in Bolivia, Ecuador, and Peru. This analysis will rely primarily on electoral data from all three countries from 1979-2004. I will also test for relationships between institutional variables to test for interactive effects.

Chapters 3-7 will provide a detailed history of Bolivia's democratization process, presenting the four possible explanations for Bolivia's democratic stability presented in this study and liked to my research hypotheses. Chapter 3 will give a history of the 1952 National Revolution, focusing on its legacies and implications for the current democratic system. Chapter 4 will give biographical histories for the three most important politicians of the democratization period-Paz Estenssoro, Siles Zuazo, and Bnzer. Chapter 5 will present the history of Bolivian democracy since 1978, the year of the first transitional election. Chapter 6 will give a history of Bolivia's political class and its relationship to the democratization process. Chapter 7 will detail the development of Bolivia's parliamentarized presidential system and how its institutions interacted with the democratization process.

Chapters 8-9 will present findings from the content analysis of the written record and from the open-ended surveys, respectively. Finally, Chapter 10 will evaluate the evidence from the previous chapters and determine which of the hypotheses seems to be the most plausible explanation for democratic stability. I will conclude by assessing the future of Bolivia's democracy.


 List of Abbreviations

ADN

Democratic National Action

AP

Patriotic Accord

CEBEM

Bolivian Center for Multidisciplinary Research

CEDOIN

Center for Documentation and Information

COB

Bolivian Labor Confederation

CONDEPA

Conscience of the Fatherland

FRI

Revolutionary Leftist Front

ILDIS

Latin American Institute for Social Research

INRA

National Institution for Agrarian Reform

LPP

Law of Popular Participation

MBL

Free Bolivia Movement

MIR

Movement of the Revolutionary Left

MNR

National Revolutionary Movement

MRTKL

Tupaj Katari Revolutionary Movement of Liberation

MSM

Movement Without Fear

NEP

New Economic Policy

PCB

Bolivian Communist Party

PIR

Revolutionary Party of the Left

POR

Revolutionary Workers Party

UCS

Solidarity Civic Union

UDP

Democratic Popular Union


References

Almond, G. and S. Verba. 1963. The Civic Culture: Political Attitudes and Democracy in Five Nations (Princeton: Princeton University Press).

Barber, B. 1984. Strong Democracy: Participatory Politics for a New Age (Berkeley: University of California Press).

Bedregal, G. 1996. Bolivia; Gobernabilidad y Cambio Revolucionario (La Paz: Amigos del Libro).

Collier, D. and S. Levitsky. 1997. "Democracy with Adjectives: Conceptual Innovation in Comparative Research," World Politics 49 (2): 430-51.

Conaghan, C. and J. Malloy. 1994. Unsettling Statecraft: Democracy and Neoliberalism in the Central Andes (Pittsburgh: University of Pittsburgh Press).

Dahl, R. 1956. A Preface to Democratic Theory (Chicago: University of Chicago Press).

----. 1971. Polyarchy: Participation and Opposition (New Haven: Yale University Press).

----. 1985. A Preface to Economic Democracy (New Haven: Yale University Press).

----. 1990. After the Revolution? Authority in a Good Society, Rev. ed. (New Haven: Yale University Press).

----. 1998. On Democracy (New Haven: Yale University Press).

Dealy, G. C. 1992. The Latin Americans, Spirit and Ethos (Boulder: Westview University Press).

Deheza, G. 1998. "Gobiernos de coalicin en el sistema presidencial: Amrica del Sur," in Nohlen and Fernndez, eds., El pesidencialismo renovado (Caracas: Nueva Sociedad), 151-169.

Diamond, L., et al. 1999. Democracy in Developing Countries: Latin America, 2nd ed. (Boulder: Lynne Rienner).

Diamond, L., J. Hartlyn, and J. Linz. 1999. "Introduction: Politics, Society, and Democracy in Latin America," in Diamond, et al., Democracy in Developing Countries: Latin America, 2nd ed. (Boulder: Lynne Rienner), 1-70.

Di Palma, G. 1990. To Craft Democracies: An Essay on Democratic Transitions (Berkeley: University of California Press).

Dominguez, J. and A. Lowenthal, eds. 1996. Constructing Democratic Governance: South America in the 1990s (Baltimore: Johns Hopkins University Press).

Duverger, M. 1954. Political Parties: Their Organization and Activity in the Modern State, trans. B. and R. North (London: Wiley).

Dryzek, J. 1996. Democracy in Capitalist Times: Ideals, Limits, and Struggles (Cambridge: Cambridge University Press).

Finley, M. I. 1985. Democracy Ancient and Modern (New Brunswick: Rutgers University Press).

Gamarra, E. 1994. "Crafting Political Support for Stabilization: Political Pacts and the New Economic Policy in Bolivia," in Smith, et al., Democracy, Markets, and Structural Reform in Latin America (Boulder: Lynne Rienner), 105-27.

----. 1996. "Bolivia: Managing Democracy in the 1990s," in Domnguez and Lowenthal, Constructing Democratic Governance: South America in the 1990s (Baltimore: Johns Hopkins University Press), 72-98.

----. 1997. "Hybrid Presidentialism in Bolivia," in von Mettenheim, ed., Presidential Institutions and Democratic Politics (Baltimore: Johns Hopkins University Press), 109-35.

Gonzlez, F. and M. Crdenas. 1998. "Partidos y crisis del sistema politico en Colombia," in T. Manz and M. Zuazo, eds., Partidos politicos y representacin en Amrica Latina (Caracas: Nueva Sociedad).

Gould, C. 1988. Rethinking Democracy: Freedom and Social Cooperation in Politics, Economy, and Society (Cambridge: Cambridge University Press).

Hall, P. and R. Taylor. 1996. "Political Science and the Three New Institutionalisms," Political Studies 44: 936-57.

Higley, J. and R. Gunther, eds., 1992. Elites and Democratic Consolidation in Latin America and Southern Europe (Cambridge: Cambridge University Press).

Huntington, S. 1968. Political Order in Changing Societies (New Haven: Yale University Press).

----. 1991. The Third Wave: Democratization in the Late Twentieth Century (Norman: University of Oklahoma Press).

----. 1997. "Democracy for the Long Haul," in Diamond, et al., eds., Consolidating the Third Wave Democracies: Themes and Perspectives (Baltimore: Johns Hopkins University Press), 3-13.

Jones, M. 1995. Electoral Laws and the Survival of Presidential Democracy (Notre Dame: Notre Dame University Press).

Karl, T. L. 1990. "Dilemmas of Democratization in Latin America," Comparative Politics 23 (1): 1-21.

Klein, H. 1992. Bolivia: The Evolution of a Multi-Ethnic Society, 2nd ed. (New York: Oxford University Press).

Lazarte, J. 1993. Bolivia: Certezas e Incertidumbres de la Democracia. Problemas de Representacin y Reforma Poltica (La Paz: ILDIS).

Levitsky, S. 1999. "Fujimori and Post-Party Politics in Peru," Journal of Democracy 10 (3): 78-92.

Lijphart, A. 1984. Democracies: Patterns of Majoritarian and Consensus Government in Twenty-One Countries (New Haven: Yale University Press).

----. 1994. Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies, 1945-1990 (Oxford: Oxford University Press).

----. 1999. Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries (New Haven: Yale University Press).

Linz, J. 1990. "The Perils of Presidentialism," Journal of Democracy 1 (1): 51-69.

----. 1994. "Presidential or Parliamentary Democracy: Does It Make a Difference?" in J. Linz and A. Valenzuela, eds., The Failure of Presidential Democracy, Vol 1 (Baltimore: Johns Hopkins University Press), 3-87.

Linz, J. and A. Stepan. 1978. The Breakdown of Democratic Regimes: Crisis, Breakdown, and Reequilibrium (Baltimore: Johns Hopkins University Press).

----. 1996. Problems of Democratic Transition and Consolidation (Baltimore: Johns Hopkins University Press).

Lipset, S. M. 1959. Political Man: The Social Basis of Politics (Baltimore: Johns Hopkins University Press).

----. 2000. "The Indispensability of Political Parties," Journal of Democracy 11 (1): 48-55.

Lipset, S. M. and S. Rokkan, eds. 1967. Party Systems and Voter Alignments: Cross-National Perspectives (New York: Free Press).

McClintock, C. 1999. "Peru: Precarious Regimes, Authoritarian and Democratic," in Diamond, et al. Democracy in Developing Countries: Latin America, 2nd ed. (Baltimore: Johns Hopkins University Press), 309-65.

McCoy, J. 1999. "Chvez and the End of 'Partyarchy' in Venezuela," Journal of Democracy 8 (1): 64-77.

Macpherson, C. B. 1977. The Life and Times of Liberal Democracy (Oxford: Oxford University Press).

Malloy, J. 1970. Bolivia: The Uncompleted Revolution (Pittsburgh: University of Pittsburgh Press).

Malloy, J. and R. Thorn, eds. 1971. Beyond the Revolution: Bolivia Since 1952 (Pittsburgh: University of Pittsburgh Press).

Margolis, M. 1979. Viable Democracy (London: Penguin).

Manin, B. 1997. The Principles of Representative Government (Cambridge: Cambridge University Press).

Mansbridge, J. 1983. Beyond Adversary Democracy (Chicago: University of Chicago Press).

March, J. and J. Olsen. 1989. Rediscovering Institutions: The Organizational Basis of Politics (New York: Free Press).

Mayorga, F. 1997. Ejemonias? Democracia representativa y liderazgos locales (La Paz: Plural).

Mayorga, F. and R. Paz. 1999. Sociedad civil y democracia participativa (La Paz: ILDIS).

Mayorga, R. A. 1992. "La democracia en Bolivia: el rol de las elecciones en las fases de transicin y consolidacin," in Cerdas-Cruz, Rial, and Zovatto, eds., Elecciones y democracia en Amrica Latina (1988-91) (San Jos: IIDH-CAPEL).

----. 1995. Antipolitica y Neopopulismo (La Paz: CEBEM).

----. 1997. "Bolivia's Silent Revolution," Journal of Democracy 8 (1): 142-56.

Moore, B. 1966. Social Origins of Dictatorship and Democracy (Boston: Beacon Press).

North, D. 1993. Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge University Press).

O'Donnell, G. 1973. Modernization and Bureaucratic Authoritarianism (Berkeley: Institute of International Studies).

----. 1994. "Delegative Democracy," Journal of Democracy 5 (1): 55-69.

----. 1996. "Illusions About Consolidation," Journal of Democracy 7 (4): 151-68.

O'Donnell, G. and P. Schmitter, eds. 1986. Transitions from Authoritarian Rule: Tentative Conclusions about Uncertain Democracies (Baltimore: Johns Hopkins University Press).

Pateman, C. 1970. Participation and Democratic Theory (Cambridge: Cambridge University Press).

Peeler, J. 1998. Building Democracy in Latin America (Boulder: Lynne Rienner).

Powell, W. and P. DiMaggio, eds. 1991. The New Institutionalism in Organizational Analysis (Chicago: University of Chicago Press).

Przeworski, A. 1987. "Methods of Cross-National Research, 1970-1983," in M. Dierkes, H. N. Weiler, and A. B. Antal, eds., Comparative Policy Research: Learning from Experience (Aldershot: Gower), 31-49.

----. 1988. "Democracy as a Contingent Outcome of Conflicts," in J. Elster and R. Slagstad, eds., Constitutionalism and Democracy (Cambridge: Cambridge University Press), 59-81.

Putnam, R. 1993. Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press).

Ragin, C. 1987. The Comparative Method: Moving Beyond Qualitative and Quantitative Stategies (Berkeley: University of California Press).

----. 1994. "Introduction to Qualitative Comparative Analysis," in T. Janoski and A. Hicks, eds., The Comparative Political Economy of the Welfare States (New York: Cambridge University Press), 299-319.

Rojas, C. and L. Verdesoto. 1997. La Participacion Popular como reforma de la Politica. Evidencias de una cultura democratica boliviana (La Paz: UIA-SNPP).

Rojas, C. and M. Zuazo. 1996. Los problemas de resentatividad del sistema democrtico boliviano. Bajo el signo de la reforma del Estado (La Paz: ILDIS).

Roln Anaya, M. 1999. Poltica y Partidos en Bolivia, 3rd ed. La Paz: Juventud.

Rustow, D. A. 1970. "Transitions to Democracy: Toward a Dynamic Model," Comparative Politics 2 (3): 337-65.

Sachs, J. and J. A. Morales. 1988. Bolivia, 1952-1986. San Francisco: ICEG.

Sartori, G. 1987. The Theory of Democracy Revisited, 2 vols. (Chatham: Chatham House).

----. 1997. Comparative Constitutional Engineering: An Inquiry into Structures, Incentives, and Outcomes (New York: New York University Press).

Schedler, A. 1998. "What Is Democratic Consolidation?" Journal of Democracy 9 (2): 91-107.

Schumpeter, J. 1943. Capitalism, Socialism and Democracy (New York: Harper & Row).

Shugart, M. S. and J. Carey. 1992. Presidents and Assemblies: Constitutional Design and Electoral Dynamics (Cambridge: Cambridge University Press).

Selbin, E. 1999. Modern Latin American Revolutions, 2nd ed. (Boulder: Westview Press).

Skocpol, T. 1979. States and Social Revolutions: A Comparative Analysis of France, Russia, and China (Cambridge: Cambridge University Press).

Srensen, G. 1998. Democracy and Democratization: Processes and Prospects in a Changing World, 2nd ed. (Boulder: Westview Press).

Stepan, A. 1978. State and Society: Peru in Comparative Perspective (Princeton: Princeton University Press).

Stepan, A. and C. Skach. 1993. "Constitutional Frameworks and Democratic Consolidation: Parliamentarism versus Presidentialism," World Politics 46 (4): 1-22.

Therborn, G. 1983. "The Role of Capital and the Rise of Democracy," in D. Held, et al., States and Societies (Oxford: Martin Robertson).

Toranzo, C. and J. L. Exeni, eds. 1994. Relexiones sobre la Ley de Necesidad de Reforma de la Constitucin Politica del Estado (La Paz: ILDIS).

Trimberger, E. K. 1978. Revolution from Above: Military Bureaucrats and Development in Japan, Turkey, Egypt and Peru (New Brunswick: Transaction Press).

United Nations Development Programme. 1999. Human Development Report 1999 (http://www.undp.org/hdro).

Vliz, C. 1980. The Centralist Tradition in Latin America (Princeton: Princeton University Press).

Whitehead, L. 2001. "The Viability of Democracy," in Crabtree and Whitehead, eds., Towards Democratic Viability: The Bolivian Experience (Oxford: St. Martin's Press), 3-20.

Wiarda, H. 1981. Corporatism and National Development in Latin America (Boulder: Westview Press).



[1]   While Schumpeter does not stipulate that democracies necessarily require civil liberties, Dahl's (1971) extension of the Shumpeterian logic explicitly makes them necessary.

[2]   These are: freedom to form and join organizations, freedom of expression, the right to vote, the right of political leaders to compete for support, alternative sources of information, free and fair elections, and institutions for making government policies accountable to voters. See Dahl (1971, 7).

[3]   Polyarchy is not limited to elections; elections must be meaningful. Dahl's later works (1990; 1998) include many of these "additions" as procedural minimums.

[4]   Whitehead distinguishes viability from both consolidation and institutionalization. Like Selbin (1999), Whitehead points out that most of the literature on democratic consolidation emphasizes the institutionalization of democratic procedures. A democratic system may be institutionalized, but a lacks popular support for the democratic project, underlying socioeconomic inequalities, or other contextual variables may lead to a breakdown of democracy.

[5]   The Snchez de Lozada government oversaw a package of constitutional amendments (popularly referred to as the "new" constitution) with the consensus of government and opposition parties. Despite changes such as the introduction of a multi-member proportional (MMP) electoral system and lowering the voting age from 21 to 18, the basic framework of parliamentarized presidentialism was not significantly altered. One important exception is Article 90: parliament now must select the president from among the top two (rather than three) placing candidates.

[6]   Demagogues (such as Pericles) played an important role in Athenian politics. Often trained by Sophists, they frequently spoke for some particular faction of supporters in the assembly.

[7]   In 1979 parliament chose the MNR's Walter Guevara Arce; in 1980 parliament chose Lydia Gueiler Tejada, also of the MNR. Both were overthrown by hard-line military coups (Guevara by Alberto Natusch Busch and Gueiler by Lus Garca Mesa).

[8]   This was perhaps the most chaotic period in Bolivian politics. Celso Torrelio overthrew Garca Mesa in August 1981. Following an abortive coup by Garca Mesa, Guido Vildoso overthrew Torrelio in July 1982. In October 1982, Vildoso stepped down in favor of the Congress of 1980.

[9]   ADN and MNR won 33 and 30 percent of the vote respectively. The only UDP member to do well in 1985 was MIR, which placed third after ADN and MNR with a mere 10 percent of the vote.

[10]  Traditional left-wing political parties do, of course, still exist. Some, such as the Revolutionary Leftist Front (FRI) continue to play important roles in regional politics (FRI is still powerful in Tarija). Nevertheless, their impact is primarily limited to local politics, except when included in the MIR candidate lists. Some parties of the new left, such as MBL and the Movement Without Fear (MSM) emerged in the 1990s. These parties, however, are also mostly effective in local elections or as part of larger alliances, such as MBL's close ties to MNR since 1993.

[11] Support for neoliberal reforms was not limited to ADN. The MNR was specifically vague about its economic program during the 1985 electoral campaign. Some elements within the MNR, such as Paz Estenssoro's planning minister, Snchez de Lozada, supported neoliberal reforms and were deeply involved in crafting the NEP.

[12] The change was due, in part, to public outcry that a distant third place winner could win the presidency, such as Paz Zamorra's election in 1989.

[13] Even the hard-line military dictatorship of Garca Mesa did not alter the statist economic model established by the 1952 Revolution. Similarly, military dictatorships often rested on a campesino-military alliance that retained the peasants' social gains of the revolution, especially agrarian reform. Most of the military regimes were essentially bureaucratic-authoritarian regimes that aimed to limit the ideological debates that emerged within the MNR-dominated regimes by 1964.

[14] Bnzer himself was a member of the MNR's military cell. The Paz Estenssoro wing of the MNR also participated in the early period of the Bnzer regime from 1972 to 1974.

[15] Self-defined as a "centrist" party, ADN's manifesto outlines its continuation with the national revolutionary project: the party advocates an "integrative democratic nationalism" along with an "agrarian revolution" and a "social market economy" (Rlon Anaya 1999, 355-57).

[16]  Peru and Bolivia did, however, have institutionalized parties. Peru's APRA and Popular Action and Bolivia's MNR were firmly established parties dating back through most of the twentieth century. The absence of any significant period of democratic competition means that neither party was institutionalized as part of a democratic political party system.

[17]  Bolivian political parties (and individual politicians) commonly publish statements or manifestos in the popular media; these are not limited to electoral cycles. Unlike typical political advertising, these are essays of substantial length meant to define the position a party, faction, or individual holds on a specific issue.

[18]  This public academic-intellectual discourse is not limited to social scientists or other traditional intellectuals. Politicians and leaders of popular social movements also participate in the numerous conferences and seminars; these debates are often published for public consumption.

[19]  In case a smaller time period is more appropriate, I will keep track of the data to enable the use of weekly time series. This would provide a possible 1040 observations.