Radical Change at the Ballot Box:
Causes and Consequences of Electoral Behavior in Venezuela's 2000
Elections
José E. Molina V.
Carmen Pérez B.
Abstract
This study aims to explain the victory of Hugo Chávez and his party in the 2000 Venezuelan elections, to analyze the factors that made this victory possible, and to examine the consequences for future developments in the Venezuelan political system. The decay of traditional party loyalties without the emergence of new parties deeply rooted in society (dealignment without realignment); underdevelopment; and an institutional setting dominated by a president elected by a plurality electoral system have opened the door to personality-centered politics and weak parties, which are the main features of the current political situation. Compared to the 1993 and 1998 elections, the 2000 elections once again confirm an increase in personality politics and the decay of parties as instruments for articulating interests, representation, and governance. As a consequence, this article argues, instability is likely to remain a feature of Venezuela's party system for some time.
The national elections in the year 2000 gave Venezuela the opportunity to ratify or reverse the choice of radical change taken in 1998. A majority of voters chose to confirm the government of Hugo Chávez. This article attempts to explain why, and to assess the political meaning of that decision. The analysis is made from the perspective of electoral behavior theory and previous explanations of political attitudes in Venezuela.
Electoral behavior analysis has moved from the search for a dominant single factor to the search for multicausal explanations that accommodate the factors stressed by various historical approaches: social cleavage voting (the sociological approach), party identification (the psychological approach), and government evaluation and issue preferences (the rational choice approach) (Dalton 2002, 145-214; Dalton and Wattenberg 1993; Miller and Niemi 2002).
The search for a multicausal explanation appears clearly in Dalton's work. According to Dalton (2002), in advanced industrial democracies, modernization has brought forth a sizable sector of politically interested, highly sophisticated, independent voters; and this has meant an important decrease in the influence of social cleavages and party identification in the voting decision. Because of this process, which Dalton calls [End Page 103] "cognitive mobilization," voting is increasingly a matter of government evaluation and issue voting; yet social cleavages and party identification still have importance, even if very reduced (Dalton 2002, 145-214). Acknowledging the decline in party and cleavage-oriented voting, Miller and Niemi (2002) stress the importance of including in the analysis the constraints on the voter from institutions (party and electoral systems) and the conditioning that accrues from the media and the sociogeographic context. In short, in most advanced industrial societies, voting can be explained as the result of long-term factors that have tended to lose influence (social cleavages, values, party identification) and short-term factors (issues and candidates), within the constraints posed by institutions and the conditioning of the media.
In the case of Latin America, Venezuela included, it has been pointed out by Dix (1989), Mainwaring (1999), and Coppedge (2001) that the region's political instability, the late appearance of the socioeconomic cleavages, and the weakness of the political influence of other cleavages, such as ethnicity and race, have meant that social cleavages and party identification have had a weaker influence on voting than has been found in advanced industrial democracies. So, although party loyalties have developed and, in some cases, have been the main influence on electoral behavior, they have tended to be unstable and more prone to a process of dealignment (Dalton and Wattenberg 1993) than in advanced industrial democracies. Indeed, dealignment, in which the percentage of citizens holding stable party identification decreases significantly, seems to be occurring in several countries of Latin America: Peru, Venezuela, Colombia, Ecuador, Bolivia, Argentina, and Costa Rica. This process, as predicted by Dalton (2002), has opened the door to personality politics and issue voting, particularly in an institutional context dominated by presidentialism.
The main cause of party dealigment in Latin America, however, does not seem to be modernization or the rise of the cognitive, mobilized, highly educated, individualistic independent, although this might be a minor factor. The main cause seems to be dissatisfaction with the performance of the established parties, in a context in which party identification has had neither the anchor of social cleavages (Dix 1989) nor enough time to consolidate, because of political discontinuity (Mainwaring 1999, 54-62). It is a context also characterized by underdevelopment, which means that the electorate is much more prone to discontent than is usual in advanced industrial democracies, as Molina (2001) shows. The population's expectations of any elected government tend to be very high; at the same time, the government's available resources to meet those expectations in nonindustrialized countries are low. As a result, incumbents in Latin America tend to lose support more intensely and to lose elections more frequently than in industrialized [End Page 104] countries. Molina (2001) calls this tendency "endemic discontent." With political discontinuity, weak links between parties, and social cleavages (Mainwaring 1999, 21-62; Dix 1989), and endemic discontent (Molina, 2001), in the case of Latin America, we should expect dealignment to be more frequent and intense than in industrialized countries. Also, we should expect electoral behavior to be more "volatile"--as Roberts and Wibbels (1999) have shown it to be--and based more on government performance and personalities than it is in industrialized democracies. This should be more the case when the party system is less institutionalized (Mainwaring 1999; Mainwaring and Scully 1995).
These effects are present particularly when the party system has undergone an intense process of dealignment and deinstitutionalization, as it has in Venezuela. In Latin America, therefore, we must be aware not only that several factors concur to explain the electoral results, but that their weight is likely to have changed in the past and probably will do so again in the future. Because of these changes, a flexible multicausal approach seems to be appropriate; multicausal because it allows for the concurrent influence of several factors, and flexible because no one factor is necessarily expected always to be or remain predominant.
In Venezuela, the parties that developed in the nineteenth century were linked to church-state and center-periphery cleavages. They were never more than elite parties, and they disappeared altogether because of political discontinuity and the Juan Vicente Gómez dictatorship. The party system that developed in the struggle against Gómez and his political heirs was linked to the socioeconomic cleavage and the church-state cleavage. Acción Democrática (AD), a Social Democratic party, developed a close link with the peasant and trade union movements (which where largely built by the party) and took an anticlerical stand during its 1945-48 government. COPEI, a Christian Democratic party, was originally linked to the middle classes and the Catholic Church (Levine 1973). The rise of a new dictatorship from 1948 to 1958 interrupted the development of the party system, however, and softened the links between the parties and their original constituencies. After democracy was reestablished in 1958, AD and COPEI built their loyalties as catch-all parties and became the axis of an attenuated two-party system that remained in place from 1973 to 1993. During that period, partisanship was the main causal factor in Venezuelan electoral behavior (Baloyra and Martz 1979).
The economic crisis of the 1980s, a growing discontent with how the governments of the two main parties handled the crisis, a significant split in COPEI in 1993, and an institutional setting that, through decentralization and direct elections for governors and mayors, promoted the personalization of politics were factors that came together to "thaw" the party system (using the classic terminology of Lipset and Rokkan 1967). They opened the system to personalities and electoral behavior based [End Page 105] increasingly on the evaluation of government performance, issues, and personalities. The institutional setting for these changes was characterized by a plurality electoral system for presidential elections, which stimulates personality voting and vote concentration in favor of the main parties when legislative elections are concurrent (Shugart and Carey 1992; Mainwaring and Shugart 1997)--as was the case in 1993 and 2000.
Thus, from 1993 on, the Venezuelan party system became deinstitutionalized because of weakened party loyalties, the decline of party organizations, the rise of autonomous interest groups and further weakening of the social penetration of Venezuelan parties, and the lost legitimacy of the party as an institution (Mainwaring 1999). This created the right environment for the weakening of party-oriented voting and the rise of issue and personality voting, in a context of increasing fractionalization and polarization of the party system. It is in this context that the 2000 elections occurred.
According to this analysis, we should expect issues, particularly government evaluation, and personalities to be the main factors behind the reelection of Chávez. Parties and cleavages should be of secondary importance. This hypothesis also implies that the supposed class character of the support for Chávez and the Bolivarian Revolution was not true, at least until this election (Canache 2002a, 148-50). It would also explain the nosedive in support for the Bolivarian Revolution that most opinion polls reported in 2002 across all sectors of society.
To test this hypothesis, this study considers the background to the 2000 elections, the institutional framework of the election, the campaign, and the turnout. Building on this analysis, the study then examines the factors that explain the electoral result, using a multivariate statistical analysis on the Chávez vote based on a survey carried out during the month before the election. After the National Assembly results are also considered, conclusions are presented.
Events That Set the Stage for the 2000 Elections
The so-called mega-elections were initially scheduled for May 28, 2000, but could not be carried out. As a result, the elections were split, so that the elections for president, National Assembly, governors, state legislatures, and mayors were held on July 30, 2000. City councils and ward councils were elected on December 3, 2000. Most of the attention focused on the vote for president, but 165 representatives were also elected to the first National Assembly, a unicameral body that replaced the Congress of the Republic. Its members represent 24 federal entities (23 states and the Capital District), plus 3 single-member districts representing the indigenous population. [End Page 106]
The people who registered to vote for the presidential elections numbered 11,720,971. According to figures released by the National Statistical Institute, there were approximately 13,650,000 citizens over the age of 18 and eligible to vote on July 30, 2000. This means that approximately 1.9 million eligible voters failed to register.
The 2000 elections came at the end of a chain of events set in motion by Hugo Chávez's victory in 1998 (Njaim 1999; Vaivads 1999). In that election, Chávez led a leftist coalition called the Patriotic Pole (Polo Patriótico) and embodied the hope for change shared by many Venezuelans. The 1998 coalition consisted of the president's party (Movimiento V República), Movimiento al Socialismo, Patria para Todos, and several smaller leftist parties.
Chávez's 1998 campaign platform promised to convene a constitutional assembly, which he did shortly after winning the election. He began the process by organizing a series of electoral events in 1999, beginning with a referendum on April 25 to convene the constitutional assembly, followed by the election of the assembly delegates on June 25, and culminating on December 15 with a referendum that approved the new constitution of Venezuela. This process also functioned as a mechanism for consolidating Chávez's popularity. Both referendums approved the president's proposals by huge margins. Candidates from the Patriotic Pole also won a huge majority in the election for representatives to the constitutional assembly, taking 66 percent of the vote and securing 121 of the 128 available seats. 1
The overwhelming victories of Chavismo, the disintegration of the opposition, and the president's energetic and dominant style gave the new National Constitutional Assembly enormous powers. The opposition-controlled congress elected in 1998 could not hold its ground. The Constitutional Assembly stripped Congress of any real power in August 1999 and eventually dissolved it in December without serious resistance, which is a clear sign of how weak the traditional parties were at that moment. The Constitutional Assembly replaced Congress with a National Legislative Council composed of members appointed by the Constitutional Assembly instead of popularly elected. The Constitutional Assembly also replaced the members of the Supreme Tribunal of Justice, the National Electoral Council, the Attorney General, and the General Comptroller of the Republic with government supporters and passed a new electoral statute for the first elections of the new regime.
After the constitution was approved and took effect, the last step in this process of "reconstruction' was the election of new public authorities. With the sole exception of local authorities whose term had ended, the "mega-elections" were not legally necessary. They were not mandated by the new constitution. However, the motivation behind the decision to call these elections was clearly political: to capitalize on the [End Page 107] enormous popularity Chávez enjoyed in order to secure a solid majority in the National Assembly and in the state legislatures, whose representatives were elected as recently as 1998.
With regard to the economy, high oil prices in 1999 helped sustain a minor recovery in some macroeconomic indicators. The recession, however, continued to have deeply felt social consequences. The lack of national and foreign investment that resulted from the political conflicts, the insecurity about the rules of the game that would prevail, and the uncertainty regarding the government's economic policy prevented a more significant economic recovery from occurring. As a consequence, unemployment levels soared from 11 percent at the end of 1998 to 15.3 percent by the first quarter of 2000. Although inflation stayed under control, several economic analysts argued that this was the result of a decline in consumer spending, including spending on basic foodstuffs. The economic recession and the severe social consequences it generated neutralized the positive impact of higher oil prices so that when the 2000 elections rolled around, the economic crisis was far from over.
Electoral Organization and Electoral System
Venezuela's electoral processes are coordinated by the National Electoral Council (CNE), an independent body to which the 1999 Constitution gives the rank and autonomy of a separate branch of government. Under the pressure of public opinion, civil society organizations, and some political parties, after 1998 the council's members ceased to be representatives of political parties and instead were selected from among independent citizens.
The new constitution confirmed the independence and impartiality of the CNE's members. In December 1999, however, the Constitutional Assembly removed the CNE members and replaced them with officials it designated with the support of the ruling coalition. As soon as this move occurred, it became a centerpiece of the electoral campaign. Under pressure from the government, the newly selected Electoral Council agreed to organize the elections in a very short time, by May 28, 2000. Amid a swirling controversy over the council's partial nature and lack of efficacy, however, the council members were forced to admit to the Supreme Tribunal of Justice on May 25 that they could not carry out the election on time (Valery Gil and Ramírez Prado 2001).
The court therefore decided to postpone the elections. The entire board of the CNE resigned in response. The government then agreed to choose the new representatives to the CNE in negotiations with the nongovernmental organizations that had spearheaded the opposition against the council's previous members. Constitutional procedures still were not [End Page 108] followed, however, and the ultimate decision was laid in the hands of the National Legislative Council. Although civic associations had some say in the matter, in the end the CNE was composed primarily of government supporters, with only a few well-known public figures. The elections were now divided to take place on the two different dates.
The National Constitutional Assembly passed the so-called Electoral Statute of Public Power (Estatuto Electoral del Poder Público), published in the Gaceta Oficial on February 3, 2000. This was a special statute created for the 2000 elections only. The electoral statute retained the electoral system used during the 1998 election, with a few important modifications. The president, governors, and mayors were still elected by a plurality system in a single-round election. The legislative bodies (the National Assembly, the state legislatures, city and ward councils) were elected using a system of personalized proportional representation (also known as the additional member system). This system has been in place in Venezuela since 1989, and is based on the German electoral system. The most important alteration introduced for the 2000 election was to raise the proportion of seats elected in the National Assembly and the state legislatures using a plurality system in single-member districts, from 50 percent to 60 percent. This made it much more likely that the largest political force, if its support were distributed homogeneously across the states, could win a share of seats significantly higher than its proportion of votes, and made it highly plausible that it could win a majority of the seats without receiving a majority of the votes.
It must be pointed out that single-member seats had been in the legislation since 1989. There was no public outcry for further personalization of the system that supported the move to increase the proportion of single-member seats and to reduce proportionality. On the contrary, the experience with an all-plurality system in the National Assembly elections of 1999 made its staunchest supporters realize that in the context of Venezuelan politics, proportionality was necessary to guarantee political pluralism and a National Assembly that would not exclude any meaningful opposition. The system was skewed further to the disadvantage of minority parties with the elimination of the rule that distributed additional compensatory seats according to the percentage of the national vote each party received (Molina 1991, 39).
The total number of seats in the parliament was also reduced. Until 1998, the Venezuelan parliament consisted of two chambers. The Chamber of Deputies elected in 1998 had 207 representatives, and the Senate had 54 elected senators. The new national legislature has only 165 representatives.
These three measures had, and have, important political implications. They tend to reduce pluralism in the legislative branch; they skew the distribution of seats, reducing the proportionality of the system; and [End Page 109] they make it easier for the largest party or coalition to gain majority control over the National Assembly.
Parties and Candidates
In his highly influential work on party system theory, Sartori (1976) shows how two dimensions are necessary to understand the relationship between party systems and democratic politics: the relevant number of parties and the ideological distance between them (ideological polarization). According to Sartori, a political system with a large number of relevant parties (more than five) and intense ideological polarization tends to become unstable, ineffective, and prone to breakdown.
For Latin American countries, Mainwaring and Scully (1995) and Mainwaring (1999) have shown that a third dimension should also be considered: the level of institutionalization. Institutionalization depends on the electoral stability of the parties in the system, the roots these parties have in the social fabric, the legitimacy of parties as the main vehicle for political representation, and the strength of their organizations. The more institutionalized the party system, the greater the chances for democratic consolidation, stability, and responsiveness and the lower the possibility of personalistic politics (Mainwaring 1999).
Because the extent of party identification is one of the elements associated with the depth of the party system's social roots, it follows that the higher the level of institutionalization, the more likely that voting will be explained by party loyalties and social cleavages. Conversely, the weaker the level of party system institutionalization, the more likely that the short-term factors stressed by the rational choice approach to voting analysis (evaluation of government performance, issues, and personalities) would account for electoral behavior. Mair (2002) classifies party systems in two types on the basis of the structure of party competition: open or closed. He argues that this is similar to the classification of party systems on the basis of the level of institutionalization (Mair 2002, 100). Given the more elaborate specification of the consequences of institutionalization for democracy in Latin America, Mainwaring and Scully's typology will be used here.
The Venezuelan party system was among the most institutionalized in Latin America until the 1988 elections (Kornblith 1998; Alvarez 1996; Hidalgo 1998; Levine 1998). It met the characteristics that, according to Mainwaring and Scully (1995), make for a highly institutionalized party system. Popular support for the major parties tended to be stable over time; the parties were deeply rooted in society; they were the central actors of the electoral process; and they were highly organized. As regards the number of relevant parties (Sartori 1976), from 1973 until just before the 1993 elections, Venezuela's party system could be characterized [End Page 110] as an attenuated two-party system (Molina and Pérez 1998) in which two parties, AD and COPEI, dominated political life. A much smaller third party, Movimiento al Socialismo (Movement Toward Socialism, MAS) was a distant third competitor, gaining between 5 and 10 percent of the votes between 1973 and 1988.
The economic crisis that began in the mid-1980s, the repeated corruption scandals, and the government's inability to deal with the fundamental problems that concerned most social classes undermined traditional party loyalties and ushered in, first, a new political climate characterized by a majority of voters' desire for change; and second, a weakly institutionalized and polarized multiparty system since the 1993 elections. In this new political climate, parties were overshadowed by charismatic personalities, to which they lost the central stage in the elections; they also were progressively weakened in their organizational capacity and their societal roots (Alvarez 1996).
This process, which has not yet been reversed, was evident during the 1998 and 2000 elections. These elections were primarily competitions focused on government evaluation, issues such as "change," and personalities. This was true not just of the presidential elections but also the elections for state governors and mayors. Whereas until 1988, electoral victories were primarily a function of traditional party loyalties and candidates contributed a smaller part of the equation, today the importance of these two factors is reversed. Political parties today are almost exclusively organizational tools designed to assist candidates (Pereira 1998).
President Hugo Chávez and the evaluation of his government were the protagonists of the 2000 elections. The president launched his campaign backed by Polo Patriótico, the same coalition of parties that supported him in 1998. During the negotiations over how to divide among themselves the nominations for the National Assembly and for local offices, it became evident that the president's party, the MVR, would attempt to capitalize on the president's popularity to reduce the participation of its allies. MVR's goal was to secure the greatest number of possible positions, relegating the other coalition partners to a subordinate position. The result was that Patria para Todos left the alliance, denying Chávez its support, and that in some cases, candidates from the remaining coalition competed against each other.
Three opposition members of the Constitutional Assembly created the organization Encuentro Nacional (National Encounter) in January 2000. Their leader was Claudio Fermín, a dissident from AD. The organization was an attempt to create a pole for the opposition at a time when it seemed that no one could contend against the president and his party. With speeches that had clear neoliberal overtones, they nominated Fermín as their candidate. But the group's neoliberal leanings played against it and [End Page 111] in favor of Chávez. Already in 1993, the electorate had soundly rejected the candidates who promoted policies that could be identfied with the International Monetary Fund-sponsored program of the ill-regarded government of Carlos Andrés Pérez; and Fermín had been one of the rejected candidates. The emergence of Francisco Arias Cárdenas as a candidate relegated the Encuentro Nacional to a distant third place; most of its potential backers and some of its leaders broke ranks and decided to back the opposition candidate with the greatest chance for victory.
On February 4, 2000, the anniversary of Chávez's 1992 coup attempt, the other three surviving leaders of that failed insurrection (Francisco Arias Cárdenas, Joel Acosta Chirinos, and Jesús Urdaneta) published a slightly critical statement in which they called on the government to change direction. They gave a press conference at which they accused important officials, among them Luis Miquilena (second in command in the MVR and then-president of the National Legislative Council) of corruption. This unleashed a ferocious public opinion campaign, which made clear that two radically opposed groups coexisted within the government. On the one hand were the three former commanders of the coup, along with some civilians; and on the other, the politicians of the left who had founded the MVR and formed Chávez's organizational base. The president sided with the latter group.
Shortly before the deadline for registering candidates, Arias Cárdenas agreed to run for president with the support of La Causa R (LCR), Izquierda Democrática (ID), Movimiento por la Democracia Directa (MDD), and Bandera Roja (BR). LCR is a center-left movement whose principal leader is Andrés Velásquez, who ran for president in 1993 and received 22 percent of the vote. ID was made up of dissidents from MAS who opposed the Chávez candidacy in 1998. ID would be dissolved in 2001; its leaders would create a new political organization called Unión with Arias Cárdenas as its main leader. MDD is a dissident group from the MVR. BR is a party that supported subversive movements during the 1970s and today is a small opposition leftist party.
Arias Cárdenas's campaign was designed to show willingness to generate confidence among business groups and to slow the flight of capital allegedly sparked by the president Chávez's combative language and actions. Arias Cárdenas also emphasized support for decentralization as a way to achieve a more effective public administration, in contrast with Chávez's opposition to greater autonomy for mayors and governors. Arias Cárdenas opposed the use of the military as a political arm of the government and proposed limiting the armed forces to their traditional role of securing national defense. Likewise, he underscored his respect for democratic values and procedures, using as proof his work as governor of Zulia. His platform was on the center-left of the Venezuelan political spectrum. [End Page 112]
AD, COPEI, and Proyecto Venezuela (PV) did not run presidential candidates, but they did run a slate for the National Assembly and for local and regional elections. Just as all the others did, the fortunes of these parties depended on the popularity of their local and state candidates. Some of their best-known leaders, however, decided to run for office using their own organizations without the symbols of their parties. In general, these campaigns featured individual personalities more prominently than parties, in order to avoid falling victim to the low popular support of traditional parties in Venezuela, which Chávez constantly vilified. PV, which had won 40 percent of the vote in the 1998 elections, was limited to its home state of Carabobo, where the son of the party leader ran for reelection as governor of the state.
Convergencia, the party founded by former president Rafael Caldera, managed to win only 2.5 percent of the vote in the 1998 parliamentary elections. In 2000 it was competitive only in Yaracuy (Caldera's home state), where the party was overshadowed by its regional leader, governor Eduardo Lapi, who was running for reelection. Convergencia did not run a presidential candidate.
Primero Justicia (PJ) is an organization that arose among the middle-class-dominated municipalities in the capital city. The 2000 elections were its first ever. It ran candidates for the National Assembly and local offices; its relative success and the continuous decline of the traditional parties placed it as the first option for replacing the latter as the main center-right opposition to Chávez. Its platform emphasized respect for the rule of law, institutionalism, democracy, and the market system. It is a good example of the new type of political party that emerged during the 2000 elections, which has a local base and draws strength from individual leaders.
The Campaign
The official start of the campaign period was May 1, 2000, but practically speaking, it began much sooner. President Chávez claimed as his principal achievement the transformation of the national political panorama, marked by the near-annihilation of AD and COPEI, the discrediting of 40 years of "false democracy," and the approval of a new constitution. On social and economic issues, he argued that the government needed more time for its policies to bear fruit. The opposition countered by pointing to the government's failed economic and social policies, characterized by high unemployment and lack of private investment. According to the opposition, one year was enough time for the government's policies to promote some improvement, but all they had produced was a worsening of the economic situation.
The opposition also accused the government of creating an authoritarian regime characterized by a greater concentration of power in the [End Page 113] hands of the president, caudillismo, an increased political role for the military, and increasing pressure on the media to fire reporters and editors who were critical of the government. 2 The appearance at the beginning was that of a campaign in which the opposition was content to win a moderate number of seats in the National Assembly and a few victories in state and local elections.
The electoral outlook changed radically when Francisco Arias Cárdenas decided to enter the fray as a presidential candidate and when the other ex-leaders of the 1992 coup attempt distanced themselves from the government. One reason is that Arias Cárdenas was initially seen as someone who could pose a real challenge to the president. Another is that new issues and attitudes were put on the electoral agenda. Arias Cárdenas and his team, including Urdaneta, who had served as director of Chávez's political police, accused the administration's civilian leadership of corruption, claiming that Chávez had not taken any measures to investigate or stop corruption. This was a powerful criticism because it came from the heart of the movement that had swept Chávez into power in 1998 and because it was the first attempt to undermine the image the government had promoted of itself as the antithesis of the corrupt administrations that had preceded it. These charges gained credibility when the attorney general, Javier Elechiguerra, who had been appointed by the governing coalition, requested that the Supreme Tribunal of Justice allow a criminal trial of Luis Miquilena. The Tribunal decided not to do so.
The administration responded quickly to the launching of the Arias Cárdenas campaign and to the corruption allegations. Using an aggressive rhetoric that described Arias Cárdenas as a Judas and a traitor, the government accused him of attempting to restore the past and of being a tool to halt the revolutionary changes that were designed for the benefit of the lower classes. This campaign succeeded. Arias Cárdenas was unable to respond effectively, and the rhetoric benefited from the enormous reserve of hope that voters had deposited in the administration. This reserve of hope remained intact in the belief that economic and social improvements would follow shortly after the regime's political consolidation.
The president's aggressive verbal assault incited some of his supporters to use physical violence against members of the opposition. They tried to block Arias Cárdenas from visiting lower-class neighborhoods, and on one occasion they threw stones at him. Fortunately, the overwhelming response from public opinion and the mass media put a stop to the violence, although the verbal assaults continued. The campaign was charged nonetheless with a level of aggressiveness unusually high for Venezuela.
The belligerent tone of the president's speech did not mean that his campaign lacked a theme, or themes. The main themes could be summarized as follows: [End Page 114]
- The government had succeeded in pushing through a political transformation. Economic and social improvements were under way, promoted by the ever closer ties between the military and the masses.
- The absence of private investment during the first phase was the result of the political conflict that was necessary for the political transformation. But the economic climate would improve after the elections.
The president's aggressiveness led him to engage in bitter confrontations with anyone who dared to criticize him or failed to support him unconditionally. When the Catholic Church suggested the need to renovate the membership of the National Electoral Council in order to guarantee its credibility and the legitimacy of the elections, the president accused the bishops who had signed the letter of being reactionaries, supporters of AD with robes, hypocrites, and instruments of the devil. He treated publishers and journalists who were critical of his administration in the same way. Napoleon Bravo, the anchor of the most-watched morning television show, was taken off the air overnight after a particularly critical program on May 4.
Arias Cárdenas attempted to convince voters that the government had strayed from its original electoral platform and that he, if elected, would once again direct the country down the path of change. Without renouncing his support for the process of political change begun with the 1998 campaign, he emphasized the corrupt practices that surrounded the president and the failure of his economic policies. He attributed the economic collapse to the president's pro-Castro whims; to his authoritarian style, which had alienated investors; and to his attempts to undermine efforts to decentralize power. Arias Cárdenas attempted, without much success, to appear as aggressive as his opponent. His strategy was to portray himself as a strong candidate, capable, unlike Chávez, of carrying out the changes that Venezuelan citizens desired.
The president, meanwhile, maintained a commanding lead from the beginning of the campaign until polling day. According to a national survey with 1,500 respondents carried out by Consultores 21 in March 2000 (El Nacional 2000), out of those who declared the intention to vote, 62 percent favored Chávez. This supports the hypothesis that citizens' preestablished attitudes toward the government were the deciding factor in the election. Most people held a positive evaluation of the government. President Chávez was originally elected in 1998 to serve as an instrument with which to forge a radical political transformation; most Venezuelans looked favorably on the political changes produced by the constitutional process and were willing to give the government more time in office so that these political changes would have the chance to affect the social and economic situation. Also, according to the July 2000 [End Page 115] poll by Consultores 21 on which our statistical analysis is based, 64.1 percent of respondents had a favorable opinion of the government. 3
Turnout and Invalid Votes
Turnout in Venezuela's presidential elections used to be around 90 percent, but in 1988 it decreased to 82 percent. It diminished further in 1993, when it reached 60.2 percent. Turnout has remained low since then. It was 63.8 percent in 1998 and 56.5 percent in 2000. Comparative political research has pointed out how the level of turnout in a country is, to a large extent, the result of institutional factors, such as compulsory voting, the electoral system, the system of voter registration, the legal voting age, decisiveness and importance of the election, competitiveness of the party system, the strength of the link between social sectors and parties, party identification, and the mobilization capabilities of parties (the level of institutionalization of the party system). It also is influenced by factors related to the socioeconomic environment, such as the level of development; and to a lesser extent, individual attitudinal factors, such as political interest and satisfaction with the political system (Blais and Dobrzynska 1998; Crewe 1981; Verba et al. 1978; Pérez-Liñán 2001; Franklin 2002; Molina 1991).
The decrease in turnout that occurred in Venezuela from 1988 on fits nicely within the boundaries of that literature. It can be explained partly by institutional factors, such as the weakening and final elimination of compulsory voting and the deinstitutionalization of the party system, which severely reduced the parties' capability to mobilize the electorate. Individual-level factors, especially an increase in negative attitudes toward the political system, also contributed. Previous studies provide evidence that one-third of the people who abstained from voting cited dissatisfaction with the political system as the reason (Molina 1991; Molina and Pérez 1995; Pérez 2000). By 1998, that dissatisfaction was considerably channeled to support for Chávez, reducing that factor as a cause for abstentions. This explains why the rate of nonvoting was lower in this presidential election than in 1993. The new rise of nonvoting in the 2000 elections is probably not the result of fresh dissatisfaction with the regime, but of other factors.
Despite the importance the government placed on the 2000 elections, 43.5 percent of registered voters failed to go to the polls, exceeding the 1998 elections figure, 36.2 percent, by seven percentage points. A low level of turnout corresponds to a party system that has lost its mobilization capabilities due to deinstitutionalization, and, as noted, a noncompulsory voting setting. There are several possible explanations for this increase in abstentions. First, Venezuelan citizens were called to participate in numerous elections in 1998 and 1999, which may have [End Page 116] proved too much for some voters, especially those who were the least motivated or the least interested in the first place (Lijphart 2000). Second, the debacle associated with the postponement of the elections from the original May date and the resignation of the CNE members made apparent the deficiencies in the organization of the elections and eroded citizens' confidence in the CNE and in the electoral process itself. Third, the absence of a strong and well-organized opposition made Chávez seem invincible, as the perception appeared from opinion polls. An election's degree of competitiveness is one of the factors that influence turnout rates (Lijphart 2000, 317; Powell 1986; Blais and Dobrzynska 1998, 251; Molina 1991, 143). The less competitive the elections, the less motivated people are to participate. Thus, the perception that Chávez would run away with the election in 2000 can plausibly be regarded as one of the factors that caused people to stay away from the polls. To the extent that Arias Cárdenas could not attract the votes of people disillusioned with traditional political parties but who did not support Chávez, abstention became their best option.
The predominance of strong personalities during the campaign made voting for executive offices the primary motivation for going to the polls, thereby reducing the importance of the legislative elections. One result of this was the large number of spoiled ballots cast in the election for the National Assembly (31.9 percent in the list vote and 46.3 percent in the nominal vote) and for the state legislatures (34.1 percent in the list vote and 38.5 percent in the nominal vote), compared to the elections for executive offices (5.3 percent in the vote for president, 6.8 percent in the vote for governors, 6.4 percent in the vote for mayors). 4 Complexity is not an explanation here because the list vote for legislative offices is as easy to cast as the vote for an executive office. The elections for representatives of indigenous groups, which were held for the first time, had the highest number of void ballots (72.4 percent). This is a sign that the general public was not particularly interested in these elections, thinking perhaps that they were relevant only for indigenous groups.
The lack of information about candidates and procedures created difficulties for many voters. Even this was one of the reasons for postponing the elections in the first place; the CNE was unable to solve this problem effectively by July.
Electoral Results: President Chavez
Voters overwhelmingly preferred President Chávez, who
received 3,757,773 votes, or 59.8 percent of the votes cast (see table 1). In this election he managed to raise both
the absolute number and the percentage of votes over what he received
in the 1998 elections (3,673,685 votes, or 56.2 percent of the valid
votes cast). One way to compare the
[End Page 117]
 Click for larger view |
Table 1
Results for the Presidential Elections, 2000 |
results from these two elections is to examine the percentage of votes Chávez received from registered voters. In this way, we take into account the increase in the number of registered voters as well as the number who abstained. This comparison shows that in 1998 Chávez received 33.5 percent of the votes from registered voters, while in 2000 he [End Page 118] received 32.2 percent. We can conclude that the president managed to maintain his level of support; he lost a small fraction of the share of registered voters, not to opposition candidates but to the rise in nonvoting.
What explains Chávez's victory, having similar support levels as those he achieved in 1998? Why did being the incumbent not hurt Chávez, as we might expect from our theory about the electoral effect of underdevelopment and "endemic discontent"? For the 2000 presidential election, this erosion of electoral support did not occur in Venezuela, partly because of the short time span that separated the elections. Indeed, the government was still operating within the so-called honeymoon period of electoral support when it called for new elections. This effect of the timing of the election has been supported by the findings of the neoinstitutionalist approach to electoral behavior. As Shugart and Carey have shown (1992, 242), during this honeymoon period, citizens are less willing to demand results from the government, and tend to maintain positive expectations about the administration.
Chávez, moreover, was not an instrument through which voters sought a normal change of government but a political weapon with which to punish the traditional parties. Those parties suffered from such low esteem among the electorate in 1998 that they were unlikely to recover their popularity by 2000. Indeed, they were unable to run their own presidential candidates. This caused voters to feel more benevolence and optimism toward Chávez, and made it harder for the other actors to mount a strong challenge. The opposition that did emerge—the Arias Cárdenas candidacy—was unable to convince voters that the government had lost its way and could not bring about the desired changes. Arias failed to do this because Chávez actually did succeed in bringing about significant institutional changes during his first year-and-a-half in office; and because citizens were willing to believe that for the economic and social situation to improve, the government needed to consolidate more institutional power in the 2000 elections. People perceived that not enough time had elapsed to pass judgment on socioeconomic policy, and therefore tended to be satisfied with the government so far.
Chávez's success cannot be explained by "class-cleavage voting" or by popular support for either his economic policy or his verbal rejection of neoliberalism. Indeed, statistical analysis of the different variables associated with electoral support for Chávez in the 2000 elections reveals that the two most influential factors were the public's evaluation of the government's performance, and negative party identification (Rose and Mishler 1998), which is a measure of the electorate's rejection of the two main parties, AD and COPEI.
From the 1998 election on, the prevailing assumption was that most of President Chávez's support came from the lower classes, while the [End Page 119] middle and upper classes rejected him (Canache 2002a, 148-50). This conjecture was based on the high expectations he generated among the lower class and on the nature of his political base, which comprises left-of-center parties, some of which wanted to polarize the social classes as part of their revolutionary aims. Nevertheless, both the July 2000 preelection survey conducted by Consultores 21 and the World Value Survey carried out after the election suggest that Chávez won with the support of all social classes, including the top 10 percent of income earners. 5
Lower-income groups did tend to support the president more than the middle class did, but the difference was moderate. According to data from the 2000 World Value Survey, there was no substantial difference in the support for Chávez among the poorest 10 percent of the population, whose monthly income is below the minimum wage, and the second-richest tenth of the population; 71 percent of the former group reported voting for the president, compared to 76.1 percent of the latter. 6 A sizable difference in support for Chávez appears between the lowest and highest tenths of the population according to income; but even here, the majority (55 percent) reported having voted for Chávez.
Statistical analyses show that there is actually a negative correlation between social class, measured both by income levels and by respondents' self-identification, and the propensity to vote for Chávez. Given the available information, however, the difference in support among the middle and lower classes was not huge at the time of the 2000 elections. The level of association between social class and voting for the president (Chávez vs. others), measured using Somers'd asymmetric, was -0.06 according to data from the July 2000 survey by Consultores 21, and -0.10 according to the World Value Survey (dividing respondents into lower working class, lower middle class, higher middle class, and upper class). The statistical association between income levels and support for the president in the former survey was -0.11 and in the latter, -0.10.
This level of association is quite low, and does not support the hypothesis of a polarization of support among the social classes. According to the data, support for Chávez was not rooted in class cleavage or class consciousness; neither was it based on enduring party loyalties. This explains the volatility of that support. What seems to be true, according to opinion polls taken during 2001 and 2002, is that Chávez's support tended to fade more quickly among the middle class than in the lower-wage sectors; but it was unstable and personality-based in both. By the second semester of 2002, according to available opinion polls, Chávez's support had declined into a minority in all sectors of the population (El Universal 2002; El Nacional 2002). Chávez and the MVR may have wanted to root their support in the class cleavage, but they did not succeed. Their failed attempt to win the elections of the Workers' Union Confederation (Confederación Venezolana del Trabajo, CTV) in 2001 [End Page 120] using blatant government intervention, moreover, was a watershed in this matter.
Analysis of Venezuela's previous electoral evolution (Molina and Pérez 1998) shows, as Dalton argues in dealignment theory (2002, 186-93), that as party identifications have declined, the decision to vote for or against one candidate has increasingly depended on government evaluation and short-term factors, such as candidate personality and electoral campaign issues. These the public assesses, to some extent, under the influence of longstanding political predispositions, such as ideology (left-right self-placement). 7 Accordingly, we expected the variable "evaluation of the government" to have had a strong effect on voting. 8 We also expected the rejection of the traditional government parties (AD and COPEI), as indicated by the variable "negative party identification," to show a significant influence, as it did in 1998. 9 "Negative party identification" refers both to a longlasting political predisposition toward the two traditional government parties and to the main issue in the 1998 and 2000 elections, change. We expected the influence of government evaluation, negative party identification, and issues such as change to be stronger than social cleavages and longstanding predispositions, such as political ideology.
Our first procedure was to find out the level of statistical association between the relevant variables available in the July 2000 Consultores 21 survey and the dependent variable "vote for Chávez." (This variable is dichotomous: vote for Chávez = 1; vote for Arias or Fermín = 0.) The independent variables included in the analysis were
Sociodemographic variables:
- Sex (0 male, 1 female)
- Age (the respondents were classified in five groups: 1) 18-24; 2) 25-34; 3) 35-44; 4) 45-54; 5) 55 and older)
- Income10
- Social Stratum11
- Education12
Longstanding political predisposition:
- Ideology
- Attitude toward democracy 13
- Negative party identification
Issues and short-term factors:
- Evaluation of the government
These independent variables allow us to contrast the weight of the
factors our hypothesis assumes most relevant with the other variables
that contemporary theories of voting consider influential, as was
indicated,
[End Page 121]
in the introduction (Dalton and Wattenberg 1993; Manza and
Brooks 1999
9-19; Niemi and Weisberg 2001, 14).
 Click for larger view |
Table 2
Presidential Election, 2000: Logistic Regression with Vote for Chacez as Dependent Variable |
The variables that individually were found to have a significant association with the decision to vote for or against Chávez in 2000 were the following (in order of the strength of the association measured by the statistic Somers'd asymmetric, as seen in table 2):
- Evaluation of the government (the more positive the evaluation of Chávez's administration the greater the likelihood of voting for him)
- Negative party identification with AD and COPEI (those who said they would never vote for AD or COPEI tended to vote in greater proportions for Chávez)
- Income (the lower the income the higher the probability of voting for Chávez)
- Education (the lower the level of educational attainment the greater the probability respondents voted for Chávez)
- Gender (women voted less often for Chávez than men did) [End Page 122]
- Social stratum (the lower the social position the higher the probability of voting for Chávez)
- Ideology (the more to the left respondents placed themselves on a ten-point left-right ideological scale, the greater the propensity to vote for Chávez)
In order to identify the importance of each variable in relation to the others and to tell which ones remain significant in the multivariate model, we carried out a logistic regression on "Vote for Chávez" as the dependent variable. After this procedure, the only statistically significant explanatory variables that remained in our model were "evaluation of the government," "negative party identification with AD or COPEI," and income. These three independent variables explain 66 percent of the variation in the vote (Pseudo R 2 Nagelkerke = 0.66), an indication of the importance of these variables to the likelihood an individual voted for Chávez. Of these three variables, "evaluation of the government" had the strongest impact on the decisions to vote for or against Chávez, "negative party identification" was the second most important, and income showed the third-strongest effect (see table 2).
We can conclude from this analysis that Venezuelan citizens' positive evaluation of the Chávez government was the most important factor leading to his electoral victory in 2000. Voters' desire for change, especially the desire to keep the traditional parties out of power, was also an important factor, as is evidenced by the influence of the variable "negative party identification" in the statistical analysis. Income, which can be taken as an indicator of social cleavage, also had a significant impact, even though its influence on the election results was not as strong as that of the other two variables. This finding gives support to our thesis that after the decline of party loyalties in Venezuela, short-term factors, such as government evaluation and personalities, are the main influence on the vote. The socioeconomic cleavage was not the paramount factor in the 2000 election, neither in the poor-rich format nor as a class cleavage. Therefore, our data do not give support to the idea that Chávez succeeded in dividing Venezuelan politics along class lines--at least not up to the 2000 elections.
Factors such as people's attitudes toward democracy or their ideological position on a left-right scale, which were important in the 1998 elections, were not significant in the 2000 elections (Canache 2002b). For attitudes toward democracy, the statistical analysis shows no significant difference between democrats and no democrats regarding voting for Chávez. This is in contrast to the 1998 elections, when the "no democrats" were significantly more likely to vote for Chávez than the "democrats" (Pereira 2001a).
In the 1998 elections, left-of-center citizens voted for Chávez in much greater proportions than right-of-center voters. To a large extent, [End Page 123] this was because leftist organizations, excepting only LCR, supported Chávez. In the 2000 elections, the principal opposition candidate, Arias Cárdenas, also identified himself with groups of the left and, during his campaign, emphasized that he came from the same military movement as Chávez. He also received the support of several leftist parties (La Causa R, Izquierda Democrática, Democracia Directa). This led voters who placed themselves on the left of the ideological spectrum to divide their votes more or less equally among the candidates and in similar proportions to voters on the right of the spectrum. When included in a statistical model of support for Chávez as the only independent variable, ideology shows a small but statistically significant effect. This effect disappears, however, when "evaluations of the government," "negative party identification," and "income" are included. This shows that the effect of the left-right dimension on Venezuelan elections, like that of values and issues (Miller and Niemi 2002; Dalton 2002, 195-215), is contingent on the options put forward to the electorate.
Placing this analysis in the theoretical context of recent research on electoral behavior and adopting a flexible multicausal perspective as outlined in the introduction, we believe that besides the underlying and structural effect of underdevelopment (endemic discontent), short-term variables have become the ones with the greatest influence on voting decisions--especially those associated with rational choice theory, such as the evaluation of government performance (Downs 1957) and "economic voting" (Lewis-Beck and Paldam 2000; Remmer 1991; Robberts and Wibbels 1999; Molina 2001). It appears that the Venezuelan voter weighs preferences for the government or the opposition in each election, with a tendency to pass a negative judgment unless the election is held in the honeymoon period, as in 2000. This explains why the issue of political change has carried so much weight since 1993. In 1993 and 1998, the desire for change led first to the election of Rafael Caldera and then to the election of Hugo Chávez, each at the expense of AD and COPEI, which were seen as agents of the past. In the 2000 elections, the period between 1958 and 1993 continued to be the reference point for people's desire for change, with the traditional government parties seen as the villains and President Chávez as the agent for change. Party identification had been the most influential factor on the vote until the 1988 elections, but it became much less powerful as a result of the process of party dealignment, in which individual personalities replaced political parties at the center of electoral contests (Dalton 2000).
Nevertheless, psychological models of electoral behavior, especially the concept of party identification (Miller and Shanks 1996), are still useful for understanding certain aspects of Venezuelan electoral behavior. First, the process through which the affective ties between citizens and the traditional parties were broken, known as partisan dealignment [End Page 124] (Dalton 2000), is one of the main variables that explain the personalization of politics, the volatility of electoral behavior, and the importance that short-term factors have acquired. Second, negative party identification, understood as negative feelings toward AD and COPEI, is one of the variables with the most influence, making it difficult to return to "the way things were." Both the partisan dealignment and the negative party identification were caused by the failure of those parties to manage effectively the economic crisis that struck Venezuela after oil prices plummeted during the 1980s, with the resulting increase in poverty and income inequality. This failure to manage made the corrupt and clientelistic practices of the traditional parties intolerable and eroded any remaining goodwill voters might have had toward them.
Our analysis also takes account of institutional variables (March and Olsen 1989). We have emphasized, for example, that because the 2000 elections were scheduled so soon after the first elections that saw Chávez victorious, he was able to benefit from an extended honeymoon period in which voters' opinions of his administration where still overwhelmingly positive (Shugart and Carey 1992). The 1999 referendum in which the new constitution was approved reinforced this effect because it was seen as a personal victory for President Chávez. The proximity of the two presidential elections (December 1998 and July 2000) helps explain why the Chávez administration was able to maintain high levels of support among the public, whereas most governments in nonindustrialized democracies tend to lose their electoral support by the end of their mandate because they are unable to meet the high expectations generated by the electoral victory (Molina 2001).
The electoral system, in which a presidential candidate needs only to win a plurality of the votes, also affected the outcome by creating incentives for parties to coalesce around two strong candidates, making it more difficult for smaller parties and enhancing the parliamentary representation of the president's party in the concurrent legislative elections, by the coattail effect of the presidential vote on the legislative vote (Shugart and Carey 1992, 229). Other studies of institutional factors have argued that when one of the candidates is an incumbent running for reelection, the party in power tends to maintain greater electoral support than when reelection is not on the agenda. The president has more incentives to promote popular policies during his administration, knowing that he will be judged on his performance while in power; and also has more resources than nonincumbents for making his case (Dix 1994; Molina 2001). Chávez was the first incumbent presidential candidate in Venezuela's modern history, and it is likely that he benefited from the effects of this status. [End Page 125]
Electoral Results: National Assembly
The increasing importance of individual personalities in Venezuelan elections has also had a significant impact on the party system. Beginning approximately a decade ago, a progressive weakening has afflicted the traditional political parties in Venezuela (namely AD and COPEI), evidenced by the erosion of voter loyalty toward these parties and by their worsening electoral fortunes, which have reduced their capacity to win seats in the legislature. The new political forces also have not been able to consolidate stable party loyalties; they depend heavily on the appeal of individual candidates and leaders. This is true both for national elections and for elections at the state and local level.
Both these developments, the personalization of campaigns and the decay of traditional party loyalties, have had an impact on the party system, which can be described as a deinstitutionalized and ideologically polarized multiparty system (Mainwaring 1999, 22-39; Sartori 1976) on the basis of the results of the 2000 elections for the National Assembly. It is this kind of system that the literature considers least prone to democratic stability and governmental efficacy (Mainwaring 1999; Sartori 1976).
AD and COPEI have seen the number of seats they control in the legislature plummet in only two years. In 1998, they held 87 seats in the Chamber of Deputies, or 42 percent of the total seats. But after the 2000 elections, they controlled only 38 seats, or 23 percent of the seats in the new National Assembly (see table 3). At the same time, the MVR and MAS, the core parties in the governing coalition, won a comfortable majority in the National Assembly, assisted no doubt by President Chávez's landslide victory (see table 3). Of the parties in the coalition, only the MVR (80 seats) and MAS (21 seats) succeeded in winning seats in the Assembly. Three of MVR's seats were won by CONIVE, an indigenous movement with ties to the MVR. In other words, the MVR won 77 of the 162 seats corresponding to the national electoral districts, in addition to the three seats from indigenous districts.
Some of the winning candidates nominated by the MVR were independents
or members of small regional organizations, such as the two
seats of Organización Fuerza en Movimiento (OFM) in the
State of Lara. After the elections, the PPT (1 seat) rejoined the
governing coalition. As is apparent in table 3,
the coalition between the Movimiento V Republica and the Movement
Toward Socialism was helped by the diminished proportionality of the
electoral system. They received a total of 49.4 percent of the votes
for the National Assembly but were awarded 101 of the 165 seats (61.2
percent). This overrepresentation has proven to be politically very
important. Because of the diminished proportionality, the government
has been able to maintain control of the Assembly after
[End Page 126]
the split in the
MAS and the defection of four MVR parliamentarians to a new opposition
center-left party (Solidaridad).
 Click for larger view |
Table 3
Logistic Regression with Vote for Chávez as Dependent Variable |
In contrast, Francisco Arias Cárdenas, Chávez's main competitor, was not able to translate his performance in the presidential elections into electoral victories in legislative contests. Of all the parties that supported him, [End Page 127] only Causa R managed to win three seats. During the entire campaign, Arias cultivated an image of a candidate without any commitments to the parties and political movements that backed his candidacy. Not only did he not campaign for any party, he also avoided being identified with any of them, especially with the discredited traditional parties. The distance he cultivated from political parties explains why his campaign failed to yield electoral victories for his supporters in the legislative elections.
The other parties that ran presidential candidates in the 1998 elections, such as PV and IRENE, which had also managed to win some seats in the national legislature, lost their support in the 2000 elections and in some cases were unable to win any seats at all in the National Assembly. 14 IRENE disappeared, and PV became a regional party in the state of Carabobo. In general, candidates with regional bases of support filled the vacuum left by these parties.
The 2000 election results confirm the existence of a deinstitutionalized and polarized multiparty system. The Chávez effect, which improved the electoral fortunes of the MVR, reduced the "effective number of parties" (Laakso and Taagepera 1979) in Venezuela from 7.6 to 4.3, and reduced the total number of parties holding seats in the national legislature from 21 to 13. 15 Still, this reduction in the number of parties does not mean that the party system became more institutionalized. Electoral volatility remained high, new parties appeared, and previously relevant parties disappeared as quickly as they came. The linkage between the party system and social groups remained extremely feeble; the electoral process still had personalities as its main actors; and party organizations for all the parties were very weak, without exceptions.
The Venezuelan party system continues to suffer from a process of deinstitutionalization; that is, it lacks strong, stable parties with strong ties to citizens. Instead, the new parties that have emerged since 1993 are dominated and controlled by a few leaders, and they operate more like electoral machines. The MVR, which received the largest number of votes, is a typical example; it has yet to consolidate its organizational structure and secure a stable support base. Chávez's leadership has made the MVR successful in the electoral arena, but its reliance on a charismatic leader is also its biggest weakness, one that threatens to jeopardize its organizational and ideological consolidation (Pereira 2001b).
Conclusions
The 2000 elections ratified the desire of Venezuelan voters to change the political system that had been in place since democracy was established in 1958. Opinion polls indicate that voters' desire for reforms is aimed at creating a better government without necessarily moving away from democratic rule (Pereira 2001a). This was the first time that a president [End Page 128] was reelected in Venezuela, and according to our analysis, his reelection was the result both of voters' desire for change and their positive evaluation of Chávez's administration, in a political context characterized by party system deinstitutionalization and personality-centered politics. This suggests that most Venezuelan voters no longer use political parties to guide their voting decisions. Instead, they tend to base their decisions on their opinion of individual candidates, the campaign issues, and their evaluation of the government.
While some analysts did not believe that the two-party system based on the political domination of AD and COPEI had come to an end by 1993, its demise is beyond doubt today. Disciplined organizations with clear ideological positions based on the conflict between labor and capital have disappeared. Along with them, party loyalties cultivated throughout the country and across generations, which had penetrated to all corners of society to the point of strangling organizations in civil society, have eroded. Today, political organizations are a different species. Their keystone is one or more national or regional personalities, and their electoral fortunes depend on the popularity of those individuals. They have limited penetration into social groups and flexible political platforms with no ideological anchors. In some cases, they lack ideological coherence altogether and do not enjoy stable popular loyalties. If the MVR owed its electoral success almost exclusively to the popularity of Hugo Chávez, the recent success by the opposition is not at all rooted in support for organized parties. Rather than describe a party system, it is tempting to speak of a system without real parties, although the concept of deinstitutionalization of parties to which we have referred makes greater theoretical sense.
This new political system has had important effects on the way the Chávez government has been run. The government has sought both to govern and to build a party structure grounded in the working class and the poor, one that can mediate between citizens and the state and can organize campaigns against the opposition. This was a fight against the clock, because the public did not have any loyalty toward the government programs beyond their expectations placed on Chávez as the leader of the so-called Bolivarian Revolution. By the end of 2003, the government seems to have lost this battle for party building. Popular support has decreased from 60 percent at the time of the 2000 elections to about 30 percent. The weight of Chávez's personalistic leadership has not allowed the MVR to consolidate an autonomous leadership, and the party's attempt to steer Venezuelan politics on a class-cleavage path (poor vs. rich) seems to be fading as quickly as the support for Chávez among the poor and the working class.
On the other hand, the likelihood that competition among well-organized
parties with stable and coherent ideological positions will
[End Page 129]
return
to Venezuela is hindered by the negative attitude toward parties
as institutions that is a legacy of the traditional parties'
failure. Another factor
working in the same direction is that past party loyalties were, to a
certain extent, originally built on contrasting ideologies, or at least
political programs, whose differences have become clouded since the fall
of the Berlin Wall. It seems that building up party identification is,
in Venezuela and elsewhere, a more difficult task than it was in the
twentieth century. These are not obstacles that the current Venezuelan
political organizations can easily overcome, so it seems that the end
to political instability in Venezuela is not something that can be
foreseen in the short run, let alone a phenomenon that could disappear
when Chávez and Chavismo leave government.
José E. Molina V. is a professor at the Instituto de Estudios Políticos y Derecho Público, Universidad del Zulia, Venezuela. From 2000 through 2003 he was a visiting professor at the University of Michigan. Electoral systems and electoral behavior in Venezuela and Latin America are the focus of his research. Some recent publications include "The Rise and Decline of COPEI in Venezuela" (with Brian Crisp and Daniel Levine), in Christian Democracy in Latin America: Electoral Competition and Regime Conflict, ed. Scott Mainwaring and Timothy Scully (2003); and "The Electoral Effect of Underdevelopment: Government Turnover and Its Causes in Latin American, Caribbean, and Industrialized Countries," Electoral Studies 20 (2001).
Carmen Pérez-Baralt is a professor at the Instituto de Estudios Políticos y Derecho Público, Universidad del Zulia. Her research specializes in electoral behavior in Venezuela. She has collaborated with José Molina on the recent articles "Los procesos electorales de 1999 en Venezuela: la nueva institucionalidad política," Ciencias de Gobierno (2000), and "Participación política y derechos humanos," Revista IIDH (2002).
Endnotes
The research for this paper was made possible by support from the Consejo de Desarrollo Científico y Humanístico of the Universidad del Zulia and Fondo Nacional para la Ciencia y la Tecnología de Venezuela. We are grateful to Consultores 21 for making survey data available to us. We would like to thank five anonymous referees for their very helpful comments on a previous version of the manuscript.
1. Three additional seats were reserved for representatives of indigenous groups. They were elected by a convention of indigenous associations.
2. This was the case for Teodoro Petkoff, director of the newspaper El Mundo, who was fired even though his administration of the newspaper was generally regarded as successful.
3. National survey with 1,500 respondents carried out between the June 26 and the July 7, 2000. We are grateful to Consultores 21 for making the survey available to us.
4. The CNE does not distinguish between annulled ballots and blank ballots that were not used. It combines both into a single category called void ballots.
5. The World Value Survey for Venezuela was coordinated by Red Universitaria de Cultura Política (REDPOL) and administered by DATOS using a sample of 1,200 respondents in December 2000 (1,120 valid cases). The World Value Survey is a worldwide study coordinated by Ronald Inglehart at the University of Michigan.
6. The minimum wage in Venezuela at the time was Bs144,000 a month, approximately US$210.
7. The variable "ideology" presents the respondent's self-placement on a ten-point left-right scale. In the July 2000 Consultores 21 survey, a total of 1,231 (82 percent) of those surveyed placed themselves on the scale. Recoding the results as left (1-4), center (5-6), and right (7-10), we find that 21.2 percent placed themselves on the left, 26.4 percent in the center, and 52.4 percent on the right.
8. The question was, "In general terms, how would you classify the Chávez government?" (1) very bad, 9 percent; (2) bad, 15 percent; (3) regular to bad, 11.9 percent; (4) regular to good, 25.2 percent; (5) good, 31.3 percent; (6) very good, 7.6 percent. [End Page 130]
9. The question was, "For which party would you never vote?" This variable classifies the respondents between (1) those who said they would never vote for AD or COPEI and are not voters of the other traditional government party (54.3 percent), and (0) those who did not reject AD or COPEI, or reject one of them but vote for the other (45.7). Valid cases 1,317; total cases 1,500.
10. The respondents were divided into those with (1) income up to Bs700,000 monthly, approximately US$1,030 a month at the time of the survey (88.4 percent); and (2) income over Bs700,000 monthly (11.6 percent). Valid cases 1,364; total cases 1,500. Any other recoding of the variable yielded insignificant results.
11. The Consultores 21 2000 survey included several social stratum variables, but only one of them has a significant association with voting intention: the variable that classified type of barrio or urban area in which respondents live. This variable includes 9 categories which we codified from (1) marginal to (9) high and upper middle class.
12. This variable presents the strongest association with voting intention when it is recoded in the following manner: (1) from no education to complete secondary education; (2) more than secondary education. This is the classification used in the statistical analysis.
13. The question was, "Which would you prefer, a democracy like the one we have or a dictatorship?" Accordingly the respondents were classified in two categories: (1) Democrats: those who answered that they preferred a democracy like they one we have (94.5 percent percent); and (0) No democrats: those who responded "dictatorship" (5.5 percent). Valid cases 1,452; total cases 1,500.
14. IRENE was the personalistic party set up by Irene Sáez, former mayor (1992-98) of Chacao, a municipality of Caracas, and a presidential candidate in 1998.
15. In this paper, the "effective number of parties" (N) is calculated from the parties' share of the vote.
References
Alvarez, Angel. 1996. La crisis de hegemonía de los partidos políticos venezolanos. In El sistema político venezolano. Crisis y transformaciones, ed. Alvarez. Caracas: Universidad Central de Venezuela. 31-154.
Baloyra, Enrique, and John Martz. 1979. Political Attitudes in Venezuela: Societal Cleavages and Political Opinion. Austin: University of Texas Press.
Blais, André, and Agnieszka Dobrzynska. 1998. Turnout in Electoral Democracies. European Journal of Political Research 33: 239-61.
Canache, Damarys. 2002a. Venezuela: Public Opinion and Protest in a Fragile Democracy. Coral Gables: North-South Center Press.
----. 2002b. From Bullets to Ballots: The Emergence of Popular Support for Hugo Chávez. Latin American Politics and Society 44, 1 (Spring): 69-90.
Consejo Nacional Electoral (CNE). 1998. Resultados electorados 1998. CD-Rom. Caracas: CNE.
----. 2000. Elecciones 30 de julio de 2000. Votos diputados listas a Asamblea Nacional Venezuela. <www.cne.gov.ve/estadisticas/e018.pdf> [End Page 131]
Consejo Nacional Electoral-INDRA. 2000. República Bolivariana de Venezuela Elecciones 2000. CD-ROM (Bulletin of August 11, 2000). Caracas: CNE.
Consultores 21. 2000. National survey of 1,500 respondents carried out June 26-July 7.
Coppedge, Michael. 2001. Political Darwinism in Latin America's Lost Decade. In Political Parties and Democracy, ed. Larry Diamond and Richard Gunther. Baltimore: Johns Hopkins University Press. 173-205.
Crewe, Ivor. 1981. Electoral Participation. In Democracy at the Polls, ed. David Butler, Howard Penniman, and Austin Ranney. Washington, DC: American Enterprise Institute. 216-43.
Dalton, Russell. 2000. The Decline of Party Identification. In Parties Without Partisans: Political Change in Advanced Industrial Democracies, ed. Dalton and Martin Wattenberg. Oxford: Oxford University Press. 19-37.
----. 2002. Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies. 3rd ed. New York: Chatham House.
Dalton, Russell, and Martin Wattenberg. 1993. The Not So Simple Act of Voting. In Political Science: The State of the Discipline II, ed. Ada Finifter. Washington, DC: American Political Science Association. 193-218.
Dix, Robert H. 1984. Incumbency and Electoral Turnover in Latin America. Journal of Interamerican Studies and World Affairs 26, 4 (November): 435-48.
----. 1989. Cleavage Structures and Party Systems in Latin America. Comparative Politics 22: 23-37.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper.
Franklin, Mark. 2002. The Dynamics of Electoral Participation. In Le Duc et al. 2002. 148-68.
Hidalgo, Manuel. 1998. Consolidación, crisis y cambio del sistema venezolano de partidos. Politeia 21: 63-104.
Kornblith, Miriam. 1998. Venezuela en los 90. Las crisis de la democracia. Caracas: Ediciones IESA.
Laakso, Markku, and Rein Taagepera. 1979. Effective Number of Parties: A Measure with Application to West Europe. Comparative Political Studies 12: 3-27.
Le Duc, Lawrence, Richard Niemi, and Pippa Norris, eds. 2002. Comparing Democracies 2: New Challenges in the Study of Elections and Voting. 2nd ed. Thousand Oaks, CA: Sage. 148-68.
Levine, Daniel. 1973. Conflict and Political Cleavage in Venezuela. Princeton: Princeton University Press.
----. 1998. Beyond the Exhaustion of the Model: Survival and Transformation of Democracy in Venezuela. In Reinventing Legitimacy: Democracy and Political Change in Venezuela, ed. Damarys Canache and Michael Kulischeck. Westport: Greenwood Press. 187-214.
Lewis-Beck, Michael, and Martin Paldam. 2000. Economic Voting: An Introduction. Electoral Studies 19 (June-September): 113-21.
Lijphart, Arend. 2000. Turnout. In International Encyclopedia of Elections, ed. Richard Rose. Washington, DC: CQ Press. 314-22.
Lipset, Seymour Martin, and Stein Rokkan. 1967. Cleavage Structures, Party Systems, and Voter Alignments: An Introduction. In Party Systems and Voter Alignments: Cross-National Perspectives, ed. Lipset and Rokkan. New York: Free Press. 1-64. [End Page 132]
Mainwaring, Scott. 1999. Rethinking Party Systems in the Third Wave of Democratization. Stanford: Stanford University Press.
Mainwaring, Scott, and Matthhew Shugart. 1997. Conclusion: Presidentialism and the Party System. In Presidentialism and Democracy in Latin America, ed. Scott Mainwaring and Matthew Shugart. Cambridge: Cambridge University Press. 394-440.
Mainwaring, Scott, and Timothy Scully. 1995. Introduction. In Building Democratic Institutions, ed. Mainwaring and Scully. Stanford: Stanford University Press. 1-34.
Mair, Peter. 2002. Comparing Party Systems. In Le Duc et al. 2002. 89-107.
Manza, Jeff, and Clem Brooks. 1999. Social Cleavages and Political Change: Voter Alignments and U.S. Party Coalitions. New York: Oxford University Press.
March, James, and Johan Olsen. 1989. Rediscovering Institutions: The Organizational Basis of Politics. New York: Free Press.
Miller, Warren, and J. Merrill Shanks. 1996. The New American Voter. Cambridge: Harvard University Press.
Miller, William, and Richard Niemi. 2002. Voting: Choice, Conditioning, and Constraint. In Le Duc et al. 2002. 169-88.
Molina, José. 1991. El sistema electoral venezolano y sus consecuencias políticas. Valencia, Venezuela: Vadell Hermanos/Instituto Interamericano de Derechos Humanos.
----. 2001. The Electoral Effect of Underdevelopment: Government Turnover and Its Causes in Latin American, Caribbean, and Industrialized Countries. Electoral Studies 20, 3: 427-46.
Molina, José, and Carmen Pérez. 1995. Participación y abstención electoral. In El proceso electoral de 1993, ed. CENDES. Caracas: CENDES. 29-42.
----. 1998. Evolution of the Party System in Venezuela, 1946-1993. Journal of Interamerican Studies and World Affairs 40, 2 (Summer): 1-26.
El Nacional (Caracas). 2000. Ventaja de Chávez sobre Arias es de 28 puntos. April 4: D-1.
----. 2002. Sesenta y seis por ciento de los venezolanos no quiere regresar al pasado. Investigación ordenada por la Universidad de Harvard. October 26: D7.
Niemi, Richard, and Herbert Weisberg. 2001. Introduction. In Controversies in Voting Behavior, 4th ed., ed. Niemi and Weisberg. Washington, DC: CQ Press. 1-21.
Njaim, Humberto. 1999. La campaña electoral venezolana de 1998 como proceso de comunicación. In Campañas electorales y medios de comunicación en América Latina, vol. 2, ed. Frank Priess and Fernando Tuesta. Buenos Aires: Konrad Adenauer Stiftung, Centro Interdisciplinario de Estudios sobre el Desarrollo Latinoamericano. 629-710.
Pedersen, Mogens. 1983. Changing Patterns of Electoral Volatility in European Party Systems, 1948-1977: Explorations in Explanations. In Western European Party Systems: Continuity and Change, ed. Hans Daalder and Peter Mair. Thousand Oaks, CA: Sage. 29-66.
Pereira, Valia. 1998. La personalización de la política y estabilidad democrática en América Latina. Boletín Electoral Latinoamericano 18 (July-December): 321-53. [End Page 133]
----. 2001a. Cambio político radical y actitud hacia la democracia en Venezuela. In Venezuela en transición: elecciones y democracia 1998-2000, ed. José Vicente Carrasquero, Thais Maingón, and Friedrich Welsch. Caracas: REDPOL-CDB. 52-68.
----. 2001b. Movimiento V República. In Partidos políticos de América Latina. Países andinos, ed. Manuel Alcántara and Flavia Freidenberg. Salamanca: Ediciones Universidad de Salamanca. 585-99.
Pérez, Carmen. 2000. Cambios en la participación electoral venezolana: 1998-2000. Cuestiones Políticas 25 (July-December): 11-26.
Pérez-Liñán, Aníbal. 2001. Neoinstitutional Accounts of Voter Turnout: Moving Beyond Industrial Democracies. Electoral Studies 20: 281-97.
Powell, G. Bingham, Jr. 1986. American Voter Turnout in Comparative Perspective. American Political Science Review 80 (March): 17-44.
Remmer, Karen. 1991. The Political Impact of Economic Crisis in Latin America in the 1980s. American Political Science Review 85 (September): 777-800.
Roberts, Kenneth, and Eric Wibbels. 1999. Party Systems and Electoral Volatility in Latin America: A Test of Economic, Institutional, and Structural Explanations. American Political Science Review 93: 575-90.
Rose, Richard, and W. Mishler. 1998. Negative and Positive Party Identification in Post-Communist Countries. Electoral Studies 18: 305-22.
Sartori, Giovanni. 1976. Parties and Party Systems: A Framework for Analysis. Cambridge: Cambridge University Press.
Shugart, Matthew Soberg, and John Carey. 1992. Presidents and Assemblies: Constitutional Design and Electoral Dynamics. Cambridge: Cambridge University Press.
El Universal (Caracas). 1999. Cincuenta y tres por ciento de los congresistas apoya reforma constitucional. Survey conducted among congressional representatives, published in three installments. January 18-20.
----. 2000. La cuarta parte de la Asamblea piensa en moderar a Chávez. August 14.
----. 2002. Chávez se debilita por la base. Datanalisis survey. November 15: A1.
Vaivads, Henry. 1999. La teoría de realineamiento partidista. Una aproximación explicativa para el caso venezolano. Cuestiones Políticas 22: 133-46.
Valery Gil, Yolanda, and Roselena Ramírez Prado. 2001. Megafracaso 2000. La conspiración de los sordos. Caracas: Fundación Andrés Mata.
Verba, Sidney, Norman Nie, and Jae-on Kim. 1978. Participation and
Political Equality. Cambridge: Cambridge University Press.
