SciELO - Scientific Electronic Library Online

vol.13 issue25Los riesgos de cruzar: La migración centroamericana en la frontera México-GuatemalaFuentes de crecimiento del empleo en el sector agrícola del norte de México author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • Have no similar articlesSimilars in SciELO


Frontera norte

On-line version ISSN 2594-0260Print version ISSN 0187-7372

Frontera norte vol.13 n.25 México Jan./Jun. 2001




Implications of the PRI's First Presidential Primary. A View from Ciudad Júarez


Steven Barracca*


* Assistant Professor, Department of Political Science-University of Texas at El Paso. Email;


Artículo recibido el 14 de julio del 2000.



This article explores voters' perceptions of the PRI's first-ever presidential primary held on November 7, 1999. An exit poll conducted on election day in Ciudad Júarez, Chihuahua, indicates that the majority of voters viewed the elections as well organized, transparent, and honest, although these perceptions varied significantly according to partisan and candidate sympathies. Responses from the exit poll suggest that the PRI will benefit from the primary by strengthening its electoral support among its base and among independent voters. This is crucial for the PRI in Chihuahua, given the competitive and fluid two-party electoral environment in the state. At the national level, by helping to solidify its base, the primary was a factor that likely contributed to the PRI obtaining pluralities in the Senate and the Chamber of Deputies in the July 2000 election.



Este artículo investiga las percepciones de los votantes en la primera elección del PRI para elegir candidato a la presidencia, llevada a cabo el 7 de noviembre de 1999. Una encuesta de salida realizada el día de la elección en Ciudad Juárez, Chihuahua, indica que la mayoría de los votantes percibían las elecciones bien organizadas, transparentes y honestas, aunque estas percepciones variaron significativamente de acuerdo a las simpatías hacia el candidato y al partido. Los resultados de la encuesta de salida sugieren que el PRI se beneficiaría con la elección al reforzar el apoyo electoral de su base y entre los votantes independientes. Esta situación era crucial para el PRI en Chihuahua dada la competitividad y el fluido ambiente bipartidista en el estado. A nivel nacional, al ayudar a solidificar su base, la elección fue un factor que posiblemente contribuyó a que el PRI obtuviera pluralidad en el Senado y en la Cámara de Diputados en las elecciones de julio del 2000.



On November 7, 1999 Mexico's Institutional Revolutionary Party (PRI) held its first-ever presidential primary to select a candidate for the July 2, 2000 general election. This event marked a historic moment in Mexico's transition toward democracy, and a watershed in the process of democratic reform within the PRI, the party that had ruled Mexico uninterruptedly at the federal level for the past seven decades. With the primary, the ruling party left behind the 70-year old tradition of the dedazo—literally the finger tap—by which the sitting Mexican president handpicked his successor. In its place, the PRI allowed the Mexican electorate to choose the party's presidential candidate in an election process that was open to all registered voters. An exit poll conducted in Ciudad Juárez, Chihuahua, indicates that the majority of voters viewed the elections as well organized, transparent, and honest, although these perceptions varied significantly according to partisan and candidate sympathies. The majority of respondents also believed that the primary would strengthen the PRI, making it a more competitive regional and national political force.

The winner of the primary came as no surprise to pollsters and political analysts. In a race among four candidates, Francisco Labastida Ochoa, the former Interior Minister and presumed favorite of incumbent President Ernesto Zedillo, won the election in a landslide, winning 59 per cent of the vote and taking 272 of 300 electoral districts.2 His closest competitor, Roberto Madrazo Pintado, the controversial Governor-on-leave from Tabasco, won 30 per cent of the vote coming out ahead in only 21 electoral districts —of which all but one were located in Southeastern Mexican states, reflecting Madrazo's regional support. The two other competitors finished far behind. Manuel Bartlett Diaz, the former Governor of Puebla, won 6 per cent of the vote and 7 electoral districts, all in his home state. Humberto Roque Villanueva, an ex-party president of the PRI, received 4 per cent of the vote but no electoral districts.3

While the outcome of the election was not surprising, the election itself raises many issues that are of great interest to observers of Chihuahuan politics, Mexican politics, and more broadly to students of democratization. For Mexicanists, the election merits special attention because it is an important measure of the progress of democratic reform within the PRI. From the inception of its forerunner in 1929, the PRI has been known for its top-down process of candidate selection, whereby nominations are controlled by incumbents and party leaders with only token input from the party's rank-and-file. Its authoritarian internal practices, coupled with the public's perception that the party has been responsible for the nation's persistent economic crises over the past three decades, had led to a steady erosion of electoral support for the PRI over the 1980's and 1990's.4 Whereas prior to the 1980's the party could count on receiving 70 to 80 per cent of the vote for its presidential candidates, in 1988 and 1994 the PRI candidates won barely 50 per cent of the official vote nationally.5 In 1994, when Ernesto Zedillo received 46.4 per cent of the vote in Chihuahua, it marked the first time that a PRI presidential candidate failed to get a majority of vote in that state. As a result of electoral reforms, the growing strength of opposition parties, and greater political pluralism in Mexico, the PRI has found itself in an increasingly competitive electoral environment over the past decade. Increased party competition has led to greater demands within the party for democratic reforms. Members of the parry's progressive wing have seen reform as necessary in order to placate party activists that have grown restless with their lack of voice in internal affairs. Moreover, party reformers see democratizing the candidate selection process as essential for choosing candidates with greater grass-roots appeal, a prerequisite for winning elections that are increasingly clean and competitive.6 Party reform has become an even greater imperative in the wake of the July 2000 Presidential and Congressional election. In the most dramatic demonstration of the PRl's declining electoral support, the PRI lost its 71-year hold on the presidency, with its candidate Francisco Labastida gamering 36.1 per cent of the vote, compared to 42.5 per cent for the victor, Vicente Fox and his Alliance for Change.7 In Chihuahua, Fox carried the state with 48.7% of the vote, compared to 40.9% for Labastida and 6.8% for Cárdenas. Thus, not only does the PRI have to survive in a more competitive electoral environment, now for the first time, but also the PRI no longer has the electoral advantages associated with controlling the resource of the federal government, a perk on which depended upon heavily in the past.

Against this backdrop, the PRl's first-ever presidential primary offers a window on the party's efforts to reform and gives rise to a number of questions. Most basic, was the primary a truly democratic process heralding the advent of a new PRI, or was it—as critics from the opposition claimed it would be—a "simulation" of democracy? That is, an event that gave the appearance of giving voters a choice, but whose outcome was really predetermined by the party leadership? More specific, was the campaign process characterized by equity for all candidates, or did the party and government favor President Zedillo's choice to such an extent that the primary election was merely "window dressing" for a decision that had been made beforehand by the President in the traditional autocratic manner? Was the ballot count free from fraud, or did the PRI resort to its old tricks of manipulating the results? Did the candidates and the voters perceive the process to be fair, and if so, to what extent did the primary improve the PRl's image among the electorate as a genuine democratic alternative? Did the primary have a unifying effect on the PRI or did it exacerbate factionalism in the party? These are one set of questions addressed by this research.

Not only are public perceptions of the primary a useful tool for assessing the PRl's future as a national political actor, but they also have important implications for politics in Ciudad Juárez and the whole state of Chihuahua. As is well known, the past two decades have witnessed the steady erosion of electoral support for the PRI in this region, and a parallel rise in the electoral fortunes of the PAN (see Tables 1-3)(2).8 This trend has been especially strong in Ciudad Juárez, Chihuahua's largest municipality, accounting for almost 40 per cent of the state's population.9 Since 1992, both electoral markets appear to have settled into a tight two-party competition between the PRI and the PAN. Given that a major factor underlying the erosion of the PRl's electoral support has been its lack of democratic credibility, the primary election, as an indicator of party reform, could be a significant variable in halting or reversing its decline in Chihuahua.10

This, in turn, would help strengthen the party's competitive position vis-à-vis the PAN in close state and local elections. This article will examine to what extent public opinion, as expressed in the exit poll, suggests a strengthening or weakening of the PRI in this closely contested local and regional electoral market.

Perhaps the primary's broadest significance stems from the fact that it would likely be an important factor, among several, affecting voters' preferences in the July 2000 Presidential and Congressional elections. In turn, the outcome of this election would have profound ramifications for the trajectory of Mexico's democratic transition. Therefore, this research explores a third set of questions; Namely, what effect did the primary have for the PRl's electoral support in the 2000 general election, and what implications do the results of the general election have for Mexico's transition? While it is difficult to determine with certainty what impact the primary had on the outcome of the national election, prior to the July 2 contest one could speculate about three general scenarios: unified PRI government, unified opposition government, and divided government.11 In the first scenario, if party activists and the electorate viewed the primary as a clear advance in the PRl's efforts to democratize, this would likely help reverse the trend of diminishing electoral support for the party. All things being equal, this would improve the chances for the PRI to win the presidency, regain its absolute majority in the lower house of the Chamber of Deputies, and maintain its majority in the Senate.12 Under this scenario of a unified PRI government, the ruling party would have the upper hand in shaping electoral reforms and other reforms of the state during the 2000-2006 sexenio, allowing it to control the pace and scope of the transition process for the immediate future. I emphasize, "all things being equal", because of course, voters' perceptions on the primary would not be the only factor affecting their preferences in the general election. Other factors, such as the state of the economy, candidate image, and party loyalty also shape the choice of voters. More will be said on this point below.

A second scenario would result if party activists and the general electorate did not view the primary as credible. Further disillusionment with the PRI stemming from a seriously flawed primary would contribute to the continuing downward trend in electoral support for the ruling party leading to greater support for its two main opposition rivals, the National Action Party (PAN) and the Party of the Democratic Revolution (PRD). If disgust over the primary reached a significant level, this could be one factor in paving the way for an opposition candidate to win the presidency and for the opposition to achieve absolute majorities in both houses of the legislature. At the time of the PRI primary in fall of 1999 this outcome appeared highly improbable, as polls were showing Labastida with a commanding lead over the other major candidates.13 However, opinion polls conducted ten days before the July 2 vote showed Labastida in a virtual tie with the PAN's candidate, Vicente Fox.14 Historically, the opposition had seldom appeared to have as great a possibility of winning the presidency as it did in the eve of the 2000 election.15 If the election were to result in unified opposition government—an unlikely event—this would, for many scholars, signal the end of the first stage of the transition process: the installation of a democratic regime. Moreover, a unified opposition government would produce the most auspicious conditions for the further extension and deepening of democratic reforms in Mexico.

As noted above, a united opposition government appeared an unlikely scenario prior to the election. Even if an opposition candidate did win the presidency, the probability was low that he would have strong enough "coattails" to help his party capture an absolute majority of seats in both chambers of the Congress. This is due to the fact that none of Mexico's three major parties control an absolute majority of the Mexican electoral registered voters. A much more likely outcome that presented itself on the eve of the election was the third scenario: some form of divided government. A divided government could be led either by a president of the PRI or the opposition. In either case, the president's party would not likely receive an absolute majority in the two chambers of the Congress, making it necessary for the executive to engage in coalition building with other parties to pass democratic reforms. This situation would create less favorable conditions for the extension and deepening of democratic reforms in Mexico than in the case of unified opposition government, but greater chances than under unified PRI government.

Above it was emphasized that voters' perceptions of the primary would only be one among many factors shaping the electoral choices of individual Mexicans in the 2000 election. Nevertheless, an important analysis of Mexican voting behavior does suggest, indirectly, that voters' perceptions of the primary could be a highly salient issue. Domínguez and McCann, in their seminal study of voting behavior in the 1988 and 1991 national elections, have highlighted the primary importance of "regime issues" in explaining voter choice.16 By "regime issues", the authors mean "Voters ask themselves, above all, whether they continue to support the ruling party" or not.17 In other words, the key variable determining electoral choices is voters' evaluations of whether they continue to support the one-party regime, or whether they favor opening up power to an opposition party. In fact, Domínguez and McCann find that "voter behavior is not well explained by (1) attachments to social cleavages, (2) attitudes on policy issues, or (3) general assessments about the present circumstances and the prospects for the nation's economy or personal finances".18 While subsequent studies have modified these findings, arguing that retrospective and prospective assessments about the economy have become strong determinants of voting behavior in more recent elections,19 these same studies suggest that regime issues remain critical variables. Assuming that the voters' evaluations of the PRI would be strongly influenced by their perceptions of the party's primary, it seems clear that it would be an important factor shaping voting behavior in Mexico.

With these questions in mind we undertook an effort to study the PRI presidential primary.20 The main component of our evaluation took the form of an exit poll consisting of the survey responses of 309 voters to 28 questions from 14 randomly selected mesas receptoras (polling booths) in Ciudad Júarez, Chihuahua.21 While our questions touched on a number of themes, the majority of our queries was designed to reveal voters' perceptions about the fairness, honesty, and transparency of the primary, and what implications the primary would have for the future strength of the PRI and its prospects for the 2000 election. The results of this survey and an analysis of these results are presented below following an overview of the electoral environment in Ciudad Juárez and Chihuahua, and the recent developments in the PRl's methods of candidate selection.



The significance of the PRI primary to politics in Chihuahua centers on the fact that state and local elections occur in a party system that is fluid and characterized by close two-party competition.22 In this environment, developments that have even a modest impact on voters' perceptions of the two parties can end up making a significant difference in election results. The state did not always have a bipartisan system. Prior to 1983, Chihuahua, like the rest of Mexico, could be classified as having a hegemonic party system with the PRI garnering 70 to 80 per cent of the vote in state and local elections.23 Typical of this period was the 1980 state elections, in which official statistics gave the PRI 70.44 per cent of the total vote in the gubernatorial race, a margin of 56.04 points over the PAN (see Table 1).24 In this same election, the PRl's margin of victory over the PAN for local deputies was even greater at 59.11 percentage points (see Table 2). As Table 3 indicates, ruling party hegemony extended to Juárez, where the PRI was given 80 per cent of the vote in municipal elections in 1977 and 70 per cent in 1980.25 This situation changed dramatically, however, in the 1983 state and local elections. In this contest the PAN won 11 elections in the state's largest urban municipalities, including Chihuahua City, Juárez, Delicias, Camargo, and Parral. This victory meant that the PAN was governing 65 per cent of the state's population and the PRI just over 30 per cent.26 In addition, the PAN won 5 of the 14 seats in the local congress.27 This election marked the beginning of the two-party system in the state.

The bipartisan electoral scenario continued into the state and local elections of 1986 and 1989. However, the precise contours of voter support for the two parties in these elections were clouded by electoral fraud and high rates of abstention. In 1986, the PRI resorted to blatant manipulation of the vote and reversed all the gams the PAN had made three years earlier.28 The PRI swept all 14 local congressional districts, with official statistics indicating a difference between the PRI and PAN of 22.09 percentage points (see Table 2). Moreover, according to official numbers, the PRI made a dramatic comeback in municipal elections in Juárez, winning by a margin of 17.25 percent (see Tables 3). However, the real figures were no doubt much closer.29 The PRI resorted to fraud once again in 1989—although to a lesser degree—enabling it to maintain the gams it made in 1986. The PRI took Juárez once again, by a margin of 13.61 points, and the ruling party swept all 18 congressional distncts.30 However, the 1989 election is a particularly poor measure of voter preferences since the rate of abstention reached an all time high for the decade of almost 70 per cent.31

The 1992 state election was a watershed, bringing the PAN to victory for the first time in a gubernatorial race in Chihuahua. The PAN also won 10 of 18 single-member district seats in the State Congress, giving it a majority in the legislature for the first time. In local races, the PAN captured most of the large urban municipalities in the state, including Juárez, Parral, Cuauhtemoc, Nuevo Casas Grande, and Camargo. The key point for this discussion is that with the 1992 election, the state entered a period of even closer two-party competition. The 1992 and 1998 gubernatorial races have been won by narrow margins and have shown volatility in voter support (see Table 1). In 1992 Francisco Barrio's margin of victory was a slim 6.44 per cent. In 1998 the PRI was able to regain the governorship, running as its candidate the progressive former mayor of Chihuahua City, Patricio Martinez. Like Barrio, Martinez won that race by a similarly narrow margin of 7.89 per cent. Close competition and fluidity also characterize the elections for the local Congress. In 1992 the PAN won 10 of 18 majority seats with a very slim overall margin of just 2.12 per cent of the vote. In the 1995 mid-term election, the PAN suffered a great setback, losing its majority in Congress by a margin of 7.26 per cent. In this contest the PRI won 15 of the 18 majority seats. In the most recent congressional election of 1998, the PRI was able to maintain its majority in the legislature, yet its margin of victory (5.31 per cent) was smaller than three years earlier.32 In sum, these statistics draw a picture of state elections that are closely contested and fluid.

During the 1990's, municipal elections in Juárez have been even more competitive than gubernatorial and congressional elections, yet they have been more stable in terms of outcomes. The PAN has won every municipal election in Juárez since 1992, butas Table 3 indicates, the PAN's margin of victory has grown extremely narrow in the two most recent contests. In 1995 the PAN won by a very slim majority of 0.74 per cent and in 1998 by slightly less than 1 per cent. These highly competitive elections in Juárez and at the state level, combined with a large number of independent voters, make for a volatile electoral market. In 1995 Aziz conducted a survey of voters' party preferences for local elections in 13 of the state's most populous municipalities. He found that 33 per cent of all the survey respondents classified themselves as independents, not having any partisan sympathy. In Juárez, this figure was 24 per cent.33 More recently, in an exit survey conduct by the author in Juárez during the July 2000 election, some 5.7 per cent of the voters identified themselves as independent when asked to indicate which party they were most sympathetic with.34 While these survey results are not directly comparable, given the different time frame, election, and questions asked, they nevertheless indicate that a fairly sizable percentage of voters in the region do not have a strong attachment to any party. This number can be estimated to be anywhere between 5 and 33 per cent, depending on the specific location and election. The PRI presidential primary gains importance in light of this significant number of independent voters. Obviously, partisans of the PAN and PRD are not likely to shift their support on the basis of the PRI running a fair and honest primary. However, if independents perceived the PRI primary as credible, this could be a significant factor persuading them to vote for the PRI. Given their numbers in the electorate, this shift in voting could be one factor helping the PRI to win elections in the competitive elections in Chihuahua and Juárez.


The PRl's first-ever presidential primary is also significant because it marks a major reform in the party's process of candidate selection. Every Mexican president since 1940—ten in all—were chosen by the sitting president toward the end of their single six-year term, in a process known as the dedazo. This top-down process of candidate selection within the PRI was also extended to the state and local level. Under informal party norms, presidents would be given the leading role in selecting national legislators, state governors, and even the municipal presidents of the country's most important urban areas. In turn, governors and national legislators were given a leading role in selecting PRI candidates for state legislatures and the remaining municipal presidencies. The evolution and practice of these conducts have been well documented and analyzed elsewhere.35 For the purposes of this discussion, it is important to note how these entrenched norms began to change in recent years, ultimately resulting in the PRI moving fairly rapidly from a very closed system of candidate selection to a very open one.

Only recently did the PRl's highly centralized system of candidate selection begin to open up. The impetus for change grew in the face of an increasing number of losses to opposition parties in state and municipal races over the past decade. Prior to 1989, the PRI never conceded the loss of a state government, but by January 2001, the opposition controlled 13 of 31 states. Moreover, in 1994, the opposition controlled 222 municipal governments, or some 9 per cent of the total in the country. However, by the start of 2001, this number increased to 687 municipalities, or 28 per cent of the total.36 In addition to this general trend, recent incidents that were particularly troubling to the PRI were the defeat of its candidates to its left-wing foe the PRD in two gubernatorial elections in 1998 in the states of Zacatecas and Tlaxcala. In both elections, unsuccessful aspirants for the party's nomination became disgruntled with the PRl's tightly controlled process of candidate selection and ended up bolting from the party to run successful campaigns under the banner of the PRD.

In this context, the PRI began to experiment in 1998 with primaries for gubernatorial elections in five states. In the state of Chihuahua this change seemed to payoff. The internal election was won by Patricio Martínez García, a telegenic former mayor of Chihuahua City. Martínez's successful campaign against a veteran party official helped him galvanize PRI sympathizers behind his candidacy, giving him the needed momentum to defeat the PAN's candidate in the July 5 general election.37 This victory marked the first time that the PRI won back a state controlled by the opposition. However, in that same state, a primary to choose the mayoral candidate for Ciudad Juarez demonstrated to party leaders the risks involved in opening up the selection process. In the May 9 primary, the supporters of one faction of the PRI occupied the party headquarters in Juárez where they interrupted the ballot count in order to protest what they said was an attempt by party leaders to steal the election.38 The PRI went on to loose the general election to the PAN.

Based on the success of the gubernatorial primary in Chihuahua, the PRI continued to use gubernatorial primaries for six of the seven state elections scheduled in 1999. In these contests the result were mostly positive for the PRI. Despite primary campaigns that were often acrimonious, the ruling party was able to unite behind its candidates for governor and win decisive victories in four of the seven states: Quintana Roo, Hidalgo, Mexico, and Coahuila. Official result also gave the PRI candidate the victory in the closely contested gubernatorial election in Guerrero, despite the efforts of the PRD to have electoral authorities overturn the results based on allegations of electoral fraud. The PRI fared less well, however, in the states of Baja California Sur and Nayarit. The former case was evidence to the PRI that primaries were not a panacea for preventing splits in the party. In the party's gubernatorial primary in Baja California Sur the losing candidate, former mayor of La Paz, Leonel Cota Montano, switched his allegiance to the PRD, and won the February general election as the standard bearer of this opposition party.

The gubernatorial election in Nayarit was the one state where the PRI did not use a primary, but stayed with the old formula of arriving at a candidate through the negotiations of party leaders. Resorting to the old system may have been a factor accounting for the PRI candidate's subsequent loss to an unusual alliance between the PAN, PRD, and two smaller opposition parties. Certainly, the loss provided further evidence to the national PRI of the utility of primaries.

In view of the three successful gubernatorial primaries held in February m the states of Guerrero, Quintana Roo, and Hidalgo, the 331 members of the PRl's National Political Council met on May 17, 1999 to decide what mechanism the party would use to choose its presidential candidate for the July 2000 election. Comments made by party leaders in the months preceding this meeting gave the impression that the party would be adopting a more democratic method. From the very start of his term in office, President Zedillo clearly indicated that he would break with the tradition of the dedazo and not choose his successor. In a speech to a New York audience in July 1998 Zedillo made his views more concrete when he proposed the idea of a U.S. style primary in Mexico. Moreover, at the PRl's 70th anniversary party in early March 1999, several party leaders, including President Zedillo, endorsed the idea of a primary, or a series of state primaries culminating in a national convention.

Despite these clues as to the thinking of party elites, the announcement by the National Political Council that the PRI would use an open primary was view by political analysts with some surprise and great interest. After all, an open primary seemed to carry considerable risk for the party A seemingly safer strategy would have been to elect the candidate through a party convention in which the votes of delegates were respected. This method would have given party leaders greater control over the outcome, and at the same time it would have been sufficient to increase the party's democratic credibility, since this was the method of candidate selection being used by its main rivals, the PAN and PRD. Instead, the ruling party was gambling with the more adventurous step of conducting an open primary. Specifically, the risk in this method was that an acrimonious internal campaign would lead to a split in the party's ranks going into the 2000 general election. Early public opinion surveys were showing that the PRI was favored to win the election in 2000, but this was based on the assumptions that the party would remain unified and the opposition parties would fail to form an alliance.39 In view of the fact that the PRI, as the top vote-getting party, only received 38 per cent of the national vote in the 1997 midterm elections, with 59 per cent of the vote divided among seven opposition parties, most analysts thought that a division in the PRI would be the one event that could seriously jeopardize its chances of victory in the 2000 presidential election. Yet, while these risks were real, the PRI had a lot to gain if it could hold a primary that was viewed as free and fair and avoided divisions within the party. Such an outcome would increase its democratic credibility among the voters and give it momentum going into the general election. Other analysts argued that the PRl's national leadership saw an open primary as preferable to a party convention due to their assessment that a convention would favor the party's hard-line candidates, an outcome that the technocratic wing of the party headed by President Zedillo wanted to avoid.40 The question of how voters actually viewed the primary is taken up below following an overview of the candidates and the campaign.



Very early in 1999 it became clear that the process of selection for the PRl's presidential candidate would be very different from the past. According to party tradition, the handful of presidential aspirants that emerge each sexenio—usually members of the president's cabinet—would keep their political ambitions hidden from the media and the public. This party norm against would-be candidates mounting public campaigns developed in order to preserve the party's image of unity before the public. Meanwhile, behind the scenes, these presidential hopefuls worked hard trying to strengthen their alliances with various political cliques within the party and curry the favor of the president in order to win his all-important endorsement. The presidential campaign would only become public after the destape (the unveiling) of the candidate by the president some six or seven months prior to the general election. Once the president had made his choice known, party activists and leaders of the PRI-affiliated labor, peasant, and popular sectors would throw their support behind the "official" candidate in a show of unity.41

However, the 2000 presidential succession marked a radical break from this tradition. By the end of January 1999, a full year and a half before the general election, three candidates had already mounted very public campaigns to obtain the party's nomination. These three candidates, Manuel Bartlett Diaz, Roberto Madrazo Pintado, and Humberto Roque Villanueva were considered old style, hard-line politicians, or "dinosaurs". Months before President Zedillo indicated his preference for a candidate, Bartlett and Madrazo were airing television commercials that criticized the government and the national PRI for having moved away from its populist roots. These ads were a source of mild embarrassment for President Zedillo and other members of the technocratic wing of the party, who were the architects of the neoliberal reforms that sharply curtailed many of the regime's populist policies. Also, as candidates with questionable democratic credentials, the campaigns of Bartlett and Madrazo irritated the reform-minded wing of the PRI, which had been working hard to give the ruling party a more democratic public image.

In addition to their reputations as hardliners, Bartlett and Madrazo were controversial candidates due to their previous records in office. Bartlett, who was interior minister in the administration of President Miguel de la Madrid (1982-1988), is considered to be the mastermind behind the electoral fraud in the 1988 presidential election that many Mexicans believe deprived the center-left coalition's candidate, Cuauhtémoc Cárdenas, of victory. Bartlett also has a tarnished reputation in Washington DC. among advocates of the war on drugs. It was during Bartlett's watch as interior minister in the mid-1980s that drug traffickers began to establish operations in Mexico. At this same time Bartlett was in charge of overseeing the Federal Security Department (DFS), an agency that earned a reputation for protecting drug traffickers. Also during Bartlett's term as interior minister, U.S. DEA agent, Enrique Camarena, was kidnapped, tortured, and killed in 1985. More recently, Bartlett revived the perception that he is "soft" on drug trafficking, when he defended former governor of Quintana Roo, Mario Villanueva, who was accused in 1999 by Mexico's Attorney General's office of being involved in the illicit drug trade. Despite these black marks, Bartlett was a popular governor in the state of Puebla, the position he held just prior to launching his bid for the nomination. Moreover, Bartlett was encouraged by early surveys showing he had considerable support among the party's rank-and-file.42

Like Bartlett, Roberto Madrazo was launching his bid for the PRl's presidential nomination at the tail end of a six-year term as Governor. Madrazo gained notoriety in the wake of the November 1994 elections that brought him to power in the state of Tabasco. The PRD claimed that the PRI resorted to widespread fraud to win these elections. The anomaly in the electoral process drawing the most criticism from the opposition was what appeared to be the blatant disregard for campaign finance laws by Madrazo's team. A 1996 inquiry conducted by Mexico's Attorney General determined that Madrazo spent $38.6 million on his campaign, about thirty times the legal limits. While this incident hurt his standing among PRI reformers, Madrazo's handling of the post-electoral conflict made him very popular among the PRI in Tabasco and with other party hardliners around the country. In this dispute over the election results, Madrazo resisted efforts on the part of the interior minister and President Zedillo to negotiate a solution to the electoral conflict by making concessions to the PRD, including Madrazo's stepping down from office.43 By standing his ground, Madrazo became the champion of the many old guard priístas that had come to resent what they saw as the technocratic wings' over eagerness to make concessions to the opposition to resolve electoral disputes in exchange for the opposition's cooperation in approving the president's legislative initiatives in the Congress. With these credentials, Madrazo gained the support of PRI activists and sympathizers who were more resistant to opening up political space to the opposition and who supported a return of the party to its more populist roots. In addition, and rather ironically, both Bartlett and Madrazo also gained a following by criticizing government corruption and the failure of the Zedillo administration to deal with crime. Nevertheless, voter support for Bartlett and Madrazo remained limited to regional strongholds in Puebla and the southeastern states, and neither candidate attracted much of a following m Chihuahua or the other northern states.

Francisco Labastida Ochoa, the man considered to be President Zedillo's choice, did not announce his candidacy until May 18, some four months after his opponents had begun their campaigns. Before entering the campaign, Labastida was the interior minister in the Zedillo administration, and from 1986 to 1992 he was the governor of the state of Sinaloa. In all, Labastida was a 37-year veteran of government posts. With these credentials, Labastida tried to portray himself in the campaign as the more steady and experienced candidate who was seeking gradual change. In doing so, his campaign was trying to draw a sharp contrast between himself and his closest rival, Madrazo, who was casting himself as a party-rebel and an outsider who wanted to wrest control of the PRI away from the elites in Mexico City and return it to the people. Like the other three candidates, Labastida's campaign mainly focused on three themes: combating crime, which has become rampant in Mexico in recent years; fighting government corruption, which swelled to new heights m the Salinas administration; and increasing social spending in a country were living standards for the majority of citizens were devastated by the severe economic recession set off by the peso devaluation in late 1994.

When Labastida announced his candidacy, the key concern for the other three contenders was that the playing field for the campaign and the election would remain level. Traditionally, Mexico's presidents would choose the candidates to succeed them from among the members of their cabinet. Although President Zedillo consistently asserted his neutrality m the primary, as the only cabinet member running for the party's nomination, it was clear that Labastida was Zedillo's choice, and the "official" candidate. In this first-ever primary would the party organization and the party in government follow tradition and throw their support behind the president's choice, making it impossible for the other candidates to compete fairly? In an effort to provide for an equitable process, the PRI established rules for the primary including, campaign spending limits, rules against government officials campaigning for a candidate, and prohibitions against the use of public resources for campaigns. Despite these precautions, Bartlett, Madrazo, and Roque knew well that these practices were standard procedure for the PRI in past elections. Therefore, these candidates, their supporters, and members of opposition parties were very much on guard to denounce any perceived violations of these rules and evidence of favoritism toward Labastida.



What did the results of our survey reveal about voters' perceptions of the fairness of this historic primary? An overarching conclusion that can be drawn from the data is that the opinions of Juárez voters about the contest varied markedly according to candidate and partisan sympathies. To explore these views, it is useful to divide up the analysis of voters' perceptions into three segments: views pertaining to the fairness of the campaign, the transparency of the ballot count, and the implications of the primary for the PRI and its future prospects. Two questions were asked in the survey to assess voters' perceptions about the fairness of the campaign: "Do you believe the elections were well organized?" and "Do you think that there was transparency and honesty in the election?" The responses to these questions are presented in Table 4. When asked if the primary election had been well organized, 86 per cent of supporters of the PRI and 80 per cent of supporters of the PRD said yes. In contrast, only 48 per cent of PAN supporters characterized the election this way. To the question of whether they thought the process was honest, 81 per cent of priístas and 80 per cent of perredisas —porters of the PRl's bitter center-left rival—responded yes. Reflecting a dramatically different view, only 29 per cent of panistas responded in the affirmative. Given the historic and fierce competition between the PRI and PAN in Juárez, the responses of PAN supporters are no surprise. However, the high level of confidence expressed by perredistas is less easy to explain. Before an explanation is offered it needs to be pointed out that only five of the 309 respondents identified themselves as sympathizers of the PRD. Of course, one cannot make any generalizations based on such a small sample. Yet having made this qualifying statement, it might be posited that these perredistas were recent converts from the PRI, thus retaining less antagonism for the ruling party than longtime panistas.

More interesting than the differences that fell along party lines are the variations in opinion expressed by PRI supporters who backed different candidates (see Table 5). In response to the question of whether the primaries had been well organized, Almost 89 per cent of Labastida supporters and some 85 per cent of Madrazo supporters answered in the affirmative, whereas supporters of the other contenders were less laudatory. Some 20 per cent of Roque supporters and 50 per cent of Bartlett supporters described the election as poorly organized.44 These critical views are no doubt a response, in part, to the fact that the campaign organization of Bartlett only succeeded in placing representatives at approximately 50 per cent of the polling stations in Juarez and the organization of Roque did not have any representatives in the city. In contrast, the campaign of Labastida and Madrazo demonstrated their greater organizational capacity by placing representatives at 100 per cent of the polling stations in the city. A similar trend can be seen in the response to the question of whether the process was characterized by transparency and honesty. Only 13 per cent of Labastida supporters responded that the process was not transparent and honest. In contrast, 29 per cent of Madrazo supporters and 75 per cent of Bartlett supporters responded this way.

What is one to make of the fact that priístas who supported Labastida expressed a great deal more confidence in the electoral process than priístas who supported loosing candidates? Most likely, the negative views of the supporters of the underdog candidates were influenced by numerous incidents from the campaign indicating that elements in the government and the ruling party were showing favoritism toward Labastida. These incidents included reports from half a dozen states that governors and municipal presidents had been distributing building materials, food, money, and school scholarships to voters in exchange for supporting Labastida.45 For example, state government officials in Colima dispensed roofing materials and cement; in Michoacán, a PRI state legislator distributed chickens; and in Oaxaca a government social assistance agency paid indigenous citizens 60 pesos to attend a Labastida campaign rally.46 One highly publicized example of campaigning by an incumbent PRI official was illustrated by the response of Governor of Chiapas, Roberto Albores, to a group of state legislators when they questioned the Governor's instructions to work in favor of Labastida. Albores is reported as answering, "Don't act like a bunch of idiots, we have to guarantee the victory of Labastida".47 Another example of partiality on the part of actors in the government included an effort by priístas in the municipality of San Andrés Tuxtla, Veracruz, to prevent citizens from attending a political rally for Madrazo by blocking a road with heavy machinery. Yet another widely publicized episode was the accusation that high-level officials in the Interior Ministry had the host of a radio program in Sinaloa removed from the air for broadcasting an interview with Madrazo. Cases were even reported in Chihuahua of municipal presidents threatening municipal employees with being fired if they attended Madrazo campaign rallies. Perhaps more significant than these glaring yet episodic cases are studies of media coverage of the campaign indicating that Labastida received substantially more news coverage on T.V. and radio than the other three candidates.48 Faced with these incidents, the campaign organizations of Madrazo, Bartlett and Roque all complained to the National PRI that the rules established by the party were being violated. The Labastida campaign acknowledged that some of these acts of favoritism were going on, but said that the national party organization was not able to control the actions of all its members at the state and local levels, nor could it control media coverage. Perhaps the illegal activities of party members could have been curtailed had the party's National Executive Committee made a serious attempt to punish these violations, however, they appeared to lack the will to do so. Clearly, the fears of official favoritism and the use of public resources in support of Labastida were realized. These cases illustrate the difficulty that national-level party reformers face in trying to remove long-entrenched electoral practices where state and local party organizations are often less committed to reform.



In terms of the integrity of the vote count—as distinct from the campaign and election process as a whole—voters' describing themselves as supporters of the PRI expressed a great deal of confidence in the process. To the question of whether they thought their vote would be respected, 92 per cent of the supporters of the PRI and 75 per cent of the supporters of the PRD responded yes. In contrast, only 47.5 per cent of the supporters of the PAN responded yes, with over half believing that their vote would not be respected. The lack of faith expressed by panistas in the process is not surprising given the PRl's legacy of resorting to electoral fraud, especially in the 1985,1986, and 1989 elections in Chihuahua. However, according to our survey, it appears that the expectations of PAN supporters were not realized. Official nationwide results gave Labastida 59 per cent of the vote, Madrazo 30 per cent, Bartlett 6 per cent, and Roque 4 per cent. This vote distribution was very close to our survey results that predicted Labastida would receive 56 per cent of the vote, Madrazo 29 per cent, Roque 3 per cent, and Bartlett 2 per cent. The fact that the official results correspond closely with the survey results suggests that the ballot count was conducted fairly. The legitimacy of the election outcome was also supported by nationwide exit polls. Although at least two organizations conducting exit polls argued that the PRI inflated the number of total votes cast to give the results more legitimacy, no one disputed the accuracy of the percentage of the vote won by each candidate.49 In sum, the transparency of the vote count should improve the PRl's image among the electorate as a genuine democratic alternative.

While the survey results clearly indicated that a majority of voters believed the PRI would respect their vote, looking at the responses of PRI identifiers who supported different candidates suggests some trouble for the party. To the question of whether their vote would be respected, almost 95 per cent of priístas casting a ballot for Labastida and 84 per cent of those voting for Madrazo answered yes. In contrast, the yes responses of the few priístas voting for Roque, and Bartlett were 66.7 per cent and 50 per cent respectively. This demonstrates a significant lack of confidence among PRI sympathizers in their own party's commitment to the integrity of the electoral process. Clearly the ruling party's reputation for tampering with electoral results, built up over seventy years, will not change overnight, even among those sympathetic to the party.



What conclusions can be drawn about the primary and its implications for the party's future as a national and regional political force? Fortunately for the PRI, the party leadership was able to maintain the loyalty of the losing candidates in the aftermath of the primary and enter the 2000 general election unified. Nevertheless, party unity was not enough to prevent the loss of the presidency. Here Labastida lost to Fox by a margin of 6.4 per cent. Yet, as was suggested above in the third scenario for the general election, Fox's solid margin of victory did not translate into a majority of seats for the PAN-PVEM alliance. Instead, the outcome of the election was a divided government. Specifically, in the 500-seat Chamber of Deputies the PRI was given 211 seats, the PAN 207, the PRD 51, the PVEM 16, the PT 8, and the remaining seven were split among smaller parties. In the 128-seat Senate, the PRI was also able to win a plurality of seats with 60. Of the remaining seats, the PAN was given 46, the PRD 15, the PVEM 5, the PT 1, and 1 to another party.

Was the primary a factor in helping the PRI win the plurality of seats in the Chamber and the Senate? Despite newspaper reports of scattered anomalies such as ballot box stuffing in states like Querétaro,50 surveys suggest that the actual ballot count was by-and-large free and fair. Running a clean primary election probably helped the PRI boost its credibility among the majority of its party activists, thus solidifying its base of support. That this may be the case is indicated by the responses to our survey question, "Will the internal competition between the candidates help the PRI in the July 2, 2000 elections?" Almost 83 per cent of priístas and 76 per cent of all respondents answered yes to this question. An improved image for the ruling party resulting from the primary may have even swayed some independent voters to support the PRI in the general election. This possibility finds partial confirmation in the survey responses in Table 4 indicating that 50 per cent of independent voters believed that the internal competition within the PRI would help the party in the year 2000. Maybe the primary convinced independents to vote for PRI congressional candidates in 2000, helping PRI in capturing a plurality of seats in the Congress, but such a conclusion must be tentative, however, since one cannot predict national voting behavior from a survey of voters in one city.

The survey results also suggest that the primary will strengthen the PRI as a political force in Juárez. To explore voter opinion on this issue, the survey asked the question, "Do you believe the elections will strengthen the PRI here on the border". Again, it was not surprising that opposition party supporters were skeptical that the primary would mark a real change for the PRI and its public image. Some 80 per cent of PRD supporters and 70.7 per cent of panistas held the view that the primary would not strengthen the PRI locally. Nevertheless, priístas and the important independent voters had a more optimistic view. Of the PRI supporters, 89.3 per cent said the primary would strengthen the party on the border, suggesting that the primary helped to solidify the party's base. Perhaps most significant, 75 per cent of independents were convinced that the primary would help the PRI become a more competitive regional political actor. This is particularly important in Juárez, were independent voters make up at least 5 per cent of the electoral force and the two most recent municipal elections have been decided by a margin of less than 1 per cent of the vote.



In sum, by holding a primary that most voters perceived as fair and honest, the survey suggests that the PRI was able to strengthen its base and improve its image among independent voters in Juarez. One can speculate that the primary had a similar affect on voters in other parts of the country. However, in considering the future of the PRI both nationally and regionally, the primary will only be one among many factors shaping voter's perceptions of the party. More important than what the PRI has done in the past, will be what it does in the future. Moreover, in Chihuahua, what the state and local PRI organizations do is just as important in influencing voters' opinions of the party as what the national PRI does. As the PRI prepares for Chihuahua's state and local elections in July 2001, state party officials will need to use processes of candidate selection that are perceived to be open and honest. On this point, the party's past record is uneven. At times candidates for state and municipal offices have been chosen in processes that are open and transparent, and at other times the party has relied on the traditional, closed, top-down method. The PRI will only be able to strengthen its position in the border region to the extent that it consistently and uniformly adopts methods of candidate selection that are perceived as democratic by party members and voters at large.

Outside the state, the PRI has had mixed fortunes since the July 2000 election. On one hand, the party has not disintegrated as a result of loosing the presidency. Members of the PRl's national legislative group have not bolted to other parties, and have even shown signs of party unity in opposing presidential initiatives, such as Fox's efforts to attract foreign investment to the electricity industry and extending the value added tax to most types of food and medicine. In addition to opposing tax increases, the PRI legislative block also led the charge in two other populist measures, securing a larger portion of the federal budget for pensions and legalizing hundreds of thousand of pickup trucks ("chuecos") illegally imported from the United States. If PRI legislators can continue to walk a line between supporting populist measures, and at the same time not appear like they are obstructing the popular President's efforts at reform, the party may improve its image among the sizable segment of the electorate without firm partisan attachments. This, in turn, could allow the PRI to gain more seats in the 2003 mid-term elections.

On the other hand, the PRI has shown signs of struggling since the 2000 Presidential election. First, while the party has not disintegrated, the national PRI is plagued by internal factionalism over who will take over the presidency of the party and how the party will elect a new National Executive Committee. These questions will likely be decided at the February meeting of the PRl's National Political Council, in which a hard-line faction led by the governors and ex-governors of Mexico's Southeastern states are likely to play an influential role. If these PRI hardliners dominate the meeting, it would likely mean a slowing of internal party reforms. Tensions within the national leadership have been exacerbated by a second problem: unfavorable results in the three state elections held since July's national election. The biggest shock for the party was the loss of the August 2000 gubernatorial election in Chiapas to an alliance of parties headed by the PRD. Until recently, this state was considered a PRI stronghold. Two months latter, in October, the PRI suffered another setback when the nation's highest electoral court annulled the apparent electoral victory of the PRI in the gubernatorial race in Tabasco, another one-time PRI stronghold.51 Most recently the PRI lost the very closely contested election in November for the governorship of Jalisco to the incumbent PAN.52 These problems for the PRI illustrate that reforms within the party will have to go beyond holding presidential primaries. In order for the PRI to continue to be a serious electoral force in the future, efforts to democratize the party's methods of selecting its leaders and candidates will need to extend throughout all levels of the political system.



1 This article would not have been possible without the assistance of Gregory Rocha, Irasema Coronado, and Thomas Longoria. These colleagues from the Department of Political Science at the University of Texas at El Paso did the bulk of the work in organizing the exit poll that this research is based on. I am also indebted to Thomas Lengona and the anonymous reviewers for comments on an earlier version of this article. Finally, I wish to thank our student-pollsters from UTEP who actually conducted the surveys. This research was funded by a grant from the Center of Inter-American and Border Studies of the University of Texas at El Paso.

2 According to the rules established by the party for the primary, the winner would be the candidate receiving a simple majority of the votes in a majority of the 300 electoral districts in the country.

3 For complete results of the primary see

4 See Beatriz Magaloni, "Is the PRI Fading? Economic Performance, Electoral Accountability, and Voting Behavior in the 1994 and 1997 Elections", in Toward Mexico's Democratization: Parties, Campaigns, Elections, and Public Opinion, Jorge I. Domínguez and Alejandro Poiré (eds.), New York, Routledge, 1999, pp. 209-210.         [ Links ]

5 Joseph L. Klesner, "The Mexican Midterm Congressional and gubernatorial Elections of 1997: End of the Hegemonic Party System", in Electoral Studies, Vol. 16, No. 4, December 1997, pp. 567-575.         [ Links ] Also by Klesner, see "The 1994 Mexican Elections: Manifestation of a Divided Society?", in Mexican Studies/Estudios Mexicanos, Vol. 11, No. 1, Winter 1995, pp. 137-149.         [ Links ]

6 For a discussion of the challenges presented by internal reforms in the PRI see Lorenzo Meyer, "Democratization of the PRI: Mission Impossible?", in Mexico's Alternative Political Futures, Wayne A. Cornelius et al (eds.), La Jolla, CA, Center for US-Mexican Studies, University of California, San Diego, 1989, pp. 325-348.         [ Links ] See also Victoria E. Rodriguez and Peter M. Ward, "The New PRI: Recasting Its Identity", in Dismantling the Mexican State, Rob Aitken et al. (eds.), London, MacMillan, 1995, pp. 92-112.         [ Links ] Also see Rogelio Hernández Rodriguez, "The Partido Revolucionario Institucional", in Governing Mexico: Political Parties and Elections, Mónica Serrano (ed.), London, Institute of Latin American Studies, 1998, pp. 71-94.         [ Links ]

7 For complete election results see

8 On this subject see Tonatiuh Guillén López, "Political Parties and Political Attitudes in Chihuahua", in Electoral Patterns and Perspectives in Mexico, Arturo Alvarado (ed.), Center for US-Mexican Studies, University of California, San Diego, 1987, pp. 225-252;         [ Links ] Alberto Aziz Nassif, "Chihuahua: de la euforia a la indiferencia", in Frontera Norte: Una década de política electoral, Tonatiuh Guillén López (ed.), El Colegio de la Frontera Norte, 1992, pp. 69-95;         [ Links ] Alberto Aziz Nassif, Territorios de alternancia: El primer gobierno de oposición en Chihuahua, México, Triana Editores, 1996;         [ Links ] and Yemile Mizrahi, "Las Elecciones de 1995 en Chihuahua", in Las elecciones y partidos políticos en México 1994-1995, Manuel Larrosa and Leonardo Valdés (eds.), México, Universidad Autónoma Metropolitana-Iztapalapa, 1997.         [ Links ]

9 XII Censo General de Población y Vivienda 2000: Resultados Preliminares. See         [ Links ]

10 Aziz argues that the PRI's credibility crisis, stemming from its corruption and authoritarian ways, was an even greater factor accounting for its declining electoral support in Chihuahua than the economic crisis of the 1980s. See Alberto Aziz Nassif, "Electoral Practices and Democracy in Chihuahua, 1985", in Electoral Patterns and Perspectives in Mexico, Arturo Alvarado (ed.), Center for U.S.-Mexican Studies, University of California, San Diego, 1987, pp. 181-205.         [ Links ]

11 These three scenarios are suggested by Joseph L. Klesner, The 2000 Mexican Presidential and Congressional Elections: Pre-Election Report, Western Hemisphere Election Studies Series, Vol. 18, No. 1, CSIS Americas Program, June 15, 2000.         [ Links ]

12 The PRI lost its majority m the Chamber of Deputies for the first time ever in the midterm elections of 1997. It did, however, win 239 of the 500 seats in the Chamber, giving it a plurality.

13 An example of such a poll was conducted by the newspaper El Diario (Júarez, México) and published by that paper on October 31, 1999.         [ Links ] The survey asked 1,000 registered voters in Chihuahua the following question: "If the election where held today, of the following candidates mentioned, which would you vote for?" The results were: 45 per cent for Labastida (PRI), 33 per cent for Fox (PAN), and 7 per cent for Cardenas (PRD).

14 The New York Times reports a poll published by Reforma (Mexico City) on June 22, 2000 showing Labastida favored by 34 per cent.         [ Links ] Fox with 32, and the PRD's candidate, Cuauhtémoc Cárdenas, with 13. The margin of error was plus or minus 3 percentage points. See Sam Dillon, "Polls Show a Virtual Tie in the Crucial Presidential Race in Mexico", New York Times, June, 24, 2000.         [ Links ]

15 Other strong showings for opposition presidential candidates were Juan Almazán in 1940, Miguel Henriquez Guzmán in 1952, and Cuauhtémoc Cárdenas in 1988. It is widely held that Cárdenas actually won the 1988 election, but was denied victory through widespread electoral fraud.

16 Jorge I. Domínguez and James A. McCann, "Shaping Mexico's Electoral Arena: The Construction of Partisan Cleavages in the 1988 and 1991 National Elections", in American Political Science Review, Vol. 89, No. 1, 1995, pp. 34-48.         [ Links ]

17 Domínguez and McCann, 1995, p. 34.

18 Dominguez and McCann, 1995, p. 34.

19 For an argument about the salience of retrospective voting behavior see Alejandro Poiré, "Retrospective Voting, Partisanship, and Loyalty in Mexican Elections: 1994", in Toward Mexico's Democratization: Parties, Campaigns, Elections, and Public Opinion, Jorge I. Domínguez and Alejandro Poiré (eds.), New York, Routledge, 1999, p. 45.         [ Links ] For a study confirming the importance of prospective voting see Magalom, 1999, p. 233.

20 See footnote number 1.

21 The exit poll was conducted on the day of the primary, November 7, 1999. The survey consisted of 28 questions. Thirty bilingual students from University of Texas at El Paso conducted the interviews. These students broke up into teams of two or more, who were assigned to one of 14 randomly selected mesas receptoras (polling booths). It was our initial goal to survey 600 voters, however, due to an unexpectedly low turnout, our student-pollsters were only able to obtain the responses of 309 persons. Nationwide, turnout for the primary was only 16.9 per cent, dramatically lower than the almost 78 per cent turnout in the 1994 presidential election and 58 per cent in the 1997 midterm elections. In the state of Chihuahua, a stronghold for the opposition pan, turnout for the primary was even lower at 10.8 per cent.

Juárez is divided up into three federal congressional districts: numbers II, III, and IV. The 14 mesas receptoras where we conducted surveys were fairly evenly distributed among the three districts, such that, of the 309 responses, 103 (33.3%) were from District II, 127 (41.1%) were from District III, and 79 (25.6%) were from District IV. The complete data set for the survey can be obtained by requesting it from the author. The poll has a margin of error of ± 6 per cent.

22 Prior to the 1994 presidential elections, the electoral dynamics in Chihuahua for national elections has been distinct from that of state and local elections. The two-party system emerged for state and local elections in 1983, but did not come to characterize national elections in Chihuahua until 1994.

23 Here I follow the party system typology set out by Cornelius. His scheme has five categories: (1) PRI monopoly = PRI vote > 95 per cent; (2) strong PRI hegemony = PRI vote < 95 per cent but > 70 per cent (3) weak PRI hegemony = PRI vote < 70 per cent, but the difference between the PRI and the second party is > 40 percentage points; (4) two-party competition = PRI vote < 70 per cent, difference between PRI and second party is < 40 percentage points, second party vote > 25 per cent, and third party vote < 10 percent (5) multiparty competition = PRI vote < 70, difference between PRI and second party is < 40 percentage points, and second party < 25 per cent or third-party vote > 10 per cent. See Wayne A. Cornelius, Mexican Politics in Transition: The Breakdonm of a One-Party Dominant Regime, Center for U.S.-Mexican Studies, University of California, San Diego, 1996.

24 See Alberto Aziz Nassif, "Chihuahua: de la euforia a la indiferencia", in Frontera norte: una década de política electoral, Tonatiuh Guillén López (ed.), México, El Colegio de México and El Colegio de la Frontera Norte, 1992.         [ Links ]

25 Due to electoral fraud, voting statistics prior to 1992 are not really an accurate indicator of voter preferences.

26 Aziz, 1992, p. 74.

27 For a discussion of the reasons for the rise of the PAN in Chihuahua see Alberto Aziz Nassif, "Elections in Chihuahua, 1985", in Electoral Patterns and Perspectives in Mexico, Arturo Alvarado (ed.), Center for U.S.-Mexican Studies, University of California, San Diego, 1987.         [ Links ] Also on this topic see Rubén Lau et al., Sistema político y democracia en Chihuahua, Instituto de Investigaciones Sociales, Universidad Autónoma de Ciudad Juárez, 1995.         [ Links ]

28 See Juan Molinar Horcasitas, "Regreso a Chihuahua", in Nexos, No. 111, March, 1987.         [ Links ]

29 In contrast to the official statistics, a June 1986 survey of Juárez voters by Guillén found that 41.9 per cent preferred the PAN and 31.7 per cent preferred the PRI. See Tonatiuh Guillén López, "Political Parties and Political Attitudes in Chihuahua", in Electoral Patterns and Perspectives in Mexico, ed. Arturo Alvarado. Center for U.S.-Mexican Studies, University of California, San Diego, 1987, p. 232.         [ Links ]

30 Until 1988 the Chihuahua legislature consisted of 14 single-member district seats (aka. majority seats). In 1989 the number of majority seats was increased to 18, and 10 proportional representation seats (aka., plurinominal seats) were added.

31 Aziz, 1992, p. 75.

32 For an analysis of the results in the 1995 election see Aziz, 1996. Also see Yemile Mizrahi, "Dilemmas of the Opposition in Government Chihuahua and Baja California", Mexican Studies/Estudios Mexicanos, Vol. 14, No. 1, pp. 151-189.         [ Links ]

33 Aziz, 1996, p. 177.

34 This exit poll was conducted by the Political Science Department at the University of Texas at El Paso at randomly selected casillas in Juárez on July 2, 2000. There were 631 respondents and the poll has a margin of error of ± 4 per cent.

35 On the dedazo see Jorge G. Castañeda, La herencia: Arqueología de la sucesión presidencial en México, México, Extra Alfaguara, 1999.         [ Links ] On the historical evolution of the PRl's methods of candidate selection see Jeffrey Weldon, "The Political Sources of Presidencialismo in Mexico", in Presidentialism and Democracy in Latin America, Scott Mainwaring and Matthew Soberg Shugart (eds.), Cambridge, Cambridge University Press, 1997, p. 225-258.         [ Links ]

36 687 municipalities represent 28.3 per cent of the 2,427 in the country. Not counted in the total of opposition-controlled municipalities were the 418 municipalities governed by usos y costumbres and the 10 municipalities governed by municipal councils. The data on opposition controlled municipalities comes from the Centro Nacional de Desarrollo Municipal (Cedemun), "Radiografía de filiación política de los municipios de México: Cuadro histórico 1994-Enero 2001". Retreived from the World Wide Web at         [ Links ]

37 Sam Dillon, "Mexico opens up elections", Austin American-Statesman (Austin, TX), 26 July 1998, pp. A15 and A19.         [ Links ]

38 Dillon, 1998, p. A19.

39 Opinion polls showing a significant advantage for Labastida in the 2000 election led many to believe that the only way the opposition could win the presidency would be for the PAN and PRD to form an alliance. These two parties met for months trying to forge a unified front, however the effort failed in the end due to an inability to agree on a method for choosing the coalitions presidential candidate.

40 Sam Dillon, "Zedillo Suggests US-Style System to Pick Mexico's Presidential Nominees", New York Times, 5 March 1999, A1 and A6.         [ Links ] Since the late 1970s, the major rift in the PRI has been between the so-called politicos and técnicos. Políticos tend to be the old-guard career politicians who have moved up in the party by holding elective office; they typically have received their university educations in Mexico; they tend to be critical of the governments neoliberal economic policies, favoring a return to the more populist approaches of the past; and they tend to take a more hard-line position about opening up the political system to the opposition. In contrast, técnicos is a term that refers to the mostly US-educated bureaucrats with advanced degrees whose careers have been made exclusively within the bureaucracy. The técnicos are the group that initiated and continue to support Mexico's shift to a neoliberal economic strategy, and they tend to be more willing to concede space to the opposition.

41 For a description of this process see Robert E. Scott, Mexican Government in Transition, rev. ed. Urbana, IL, University of Illinois Press, 1964, pp. 197-223.         [ Links ] Beginning with the 1988 presidential succession, the president's ability to unity the various wings of the party behind his candidate became much more difficult. On this topic see Peter H. Smith, "The 1988 Presidential Succession in Historical Perspective", in Mexico's Alternative Political Futures, Wayne A. Cornelius et al. (eds.), La Jolla, CA, Center for U.S.-Mexican Studies, University of California, San Diego, 1989, pp. 391-415.         [ Links ]

42 Susan Ferriss, "PRI 'dinosaur' is early favorite in Mexico race", Austin American-Statesman (Austin, TX), 14 March 1999, pp. A21;A31.         [ Links ]

43 Todd A. Eisenstadt, "Electoral Federalism or Abdication of Presidential Authority? Gubernatorial Elections in Tabasco", in Subnational Politics and Democratization in Mexico, Wayne A. Cornelius et al (eds.), La Jolla, CA, Center for U.S.-Mexican Studies, University of California, San Diego, 1999, pp. 269-293.         [ Links ]

44 Conclusions regarding the opinions of Bartlett and Roque supporters cannot be made with any confidence given the small number of their supporters in the sample.

45 Julia Preston, "La política antigua sigue viva". El Diario (Juárez, México), 8 November 1999, p. 3A.         [ Links ]

46 Preston, 1999, p. 3A.

47 Preston, 1999, p. 3A.

48 Preston, 1999, p. 3A.

49 Sam Dillon, "Mexican Pollsters Challenge Size of Turnout m the Primary", New York Times, 17 November 1999, A12.         [ Links ]

50 SUN, "Surgen inconformidades entre 'madracistas'", El Diario (Juárez, México), 9 November 1999, p. 4A.         [ Links ]

51 Negotiations among the parties resulted in a solution, whereby the PRI candidate, Enrique Priego Oropeza will serve as interim-governor until new elections are held in November 2001.

52 The PRI is currendy challenging this election in the federal electoral courts.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License