The Collapse of PRI Strongholds
On July 1, Mexico’s political map was redrawn. As Andrés Manuel López Obrador won a resounding victory in the presidential election, candidates for his Morena party triumphed in state and local races across the country. At the same time, the incumbent PRI, which governed just under half of all states, suffered a historic defeat.
The PRI’s collapse (further discussed here) represents one of the most important outcomes of the 2018 election. Facing a sudden struggle for survival, how will the former ruling party respond?
Examining patterns in states considered PRI strongholds where the party had won gubernatorial elections in 2016 and 2017 perhaps offers some insights on what happened to the PRI on the ground, and what the future might hold. The maps and analysis below look at four states—Coahuila, Zacatecas, Estado de México, and Hidalgo—and highlight a revealing region in each.
Three conclusions emerge from this perspective:
First, the PRI’s resilience in marginal urban neighborhoods appears to be tenuous: it retained influence in Torreón, Coahuila’s Zaragoza Sur neighborhood, where it could rely on the Antorcha Campesina, but was trounced in Tizayuca, Hidalgo, where the party’s internal unity faltered. These traditional urban and suburban bastions have long been slipping from the PRI's grasp, but are now likely gone for good. Somewhat paradoxically, contestation of these areas could increase incentives for corruption and clientelist practices.
Second, the safest PRI strongholds seem to be more isolated rural areas, such as Pinos, Zacatecas, but the traditional clientelist ties that often underpinned this control have weakened substantially in places such as the Mazahua Region in Mexico State. It is in these areas that the PRI will likely fight hardest to prevent incursions by rival parties, and deploy a full range of persuasive and coercive tactics.
Third, the magnitude of the PRI's defeat in Hidalgo does not necessarily suggest a long-term catastrophe. There, Morena had failed to challenge in the 2016 gubernatorial race and finished fourth, but the total anti-PRI vote surpassed 50%. In 2018, the lack of a compelling PAN-PRD coalition candidate allowed López Obrador to capture that protest vote and expand the coalition, leading to staggering gains for his party in the state. If the PAN is able to rebuild in Hidalgo, the PRI will likely find survival on a local level easier.
Interestingly, the state-level results were unrelated to approval levels of the PRI governors: Coahuila and Mexico State had nearly identical approval/disapproval ratings in 2017, but had vastly different patterns in 2018. Hidalgo, where the PRI fared worst, had the highest approval rating for its governor of the four states.
This discussion, focusing on the PRI’s strongholds, should not disguise the magnitude of its defeat in 2018, nor does it seek to. Understanding where the PRI remains strong is less about envisioning a return to dominance than it is about seeing where and how the party will cling to power as it struggles to survive. Identifying where the PRI felt the 2018 defeat most keenly allows us to watch the party's tactics as it attempts to rebuild there. And knowing where it successfully held off its opponents gives clues as to how it might do so.
While the analysis here begins on the micro level, building upwards from individual precincts, the implications go beyond the results of local races for mayor or state congress. In attempting to retain control of the few states the party still rules, PRI governors might funnel significant resources to areas where local control remains strongest, or where rival parties are not yet fully established. Local politicians, in turn, will face pressure to defend those strongholds vigorously, with journalists and activists potentially facing increased risks there. The PRI will not go quietly: though wounded, it cannot be ignored.
Notes on methodology and approach: I have also not included information on votes for either the PAN-PRD candidate or the independent candidate, though some is available here. There are two reasons for this decision. First, there is a perception that the PRI's losses are Morena's gains, and that clientelist networks may have migrated from the PRI to Morena. The data presented here suggests that pattern may be more complicated than often assumed, and vary from region to region. Second, I am primarily interested in the future dynamics between Mexico's ascendant political force and a decaying rival that has the most extensive political machinery - both the PAN and PRD are also greatly weakened after a crushing defeat in 2018.
A "bastion" was defined as an electoral section (subsequently described as a precinct) where the PRI won more than 50% of the vote. If Morena won more than 50% of the vote in 2018, it was deemed "flipped." Given that Mexico is a multiparty system, where four candidates contested the presidency in 2018, achieving 50% of the vote is a significant result.
If the PRI won less than 25% of the vote in a precinct it had previously won by more than 50%, it was deemed "lost." I have also mapped precincts that were previously bastions (but where the PRI did not win more than 50% of the vote in 2018), and precincts that were bastions in both electoral cycles. The underlying cloropleth maps are coded by the difference between vote percentage for the Morena candidate in 2016/17 and López Obrador in 2018.
Data for some sections was unavailable or could not be rendered due to changes in the electoral map. Sections where zero total votes were recorded in either 2016/7 or 2018 are empty and do not react to user input.
Coahuila represented the PRI’s strongest showing in 2018. Despite a history of scandalous misrule and corruption during the governorships of the Moreira brothers (who continue to occupy prominent roles in the national PRI organization), bastion precincts were relatively safe in 2018. Only one of the 195 precincts where the PRI won more than 50% of the vote in 2016 flipped for Morena in 2018 (and represented fewer than 80 votes). Of those 195, nearly 39% (76) of those remained bastion precincts and in none did the PRI win less than 25% of the vote. In those bastions, Morena was only able to garner 15% of the vote, suggesting that local machinery continues to exercise effective control.
Of the PRI’s 358,000 votes, a third came from the agricultural southeast region of the state where 39 of the PRI’s remaining bastion precincts are located. More intriguingly, perhaps, the PRI also remained relatively durable in marginal neighborhoods on the southern edge of Torreón.
In the city of Torreón, it is clear that some of the classic PRI machinery has persisted. The party’s bastions are largely located in peripheral areas of the city, where residents of informal settlements are organized by the Antorcha Campesina group. The Antorcha, one of the few remaining urban clientelist organizations that are solidly loyal to the PRI, guaranteed two million votes for the party in 2018. It has survived by controlling political access for those living in areas without municipal services or legal titles, and has a reputation for corruption and violence.
Zaragoza Sur is one such neighborhood. The product of a land invasion and subsequent regularization between 2000 and 2010, it has changed from a neighborhood of cardboard shacks to one of blockwork houses and paved streets but remains one of the poorest colonias in Torreón. It is solidly Antorcha territory, both due to the group’s role in its history and because of its ongoing role in organizing residents—often to protest for infrastructure and services. In 2014, the group celebrated its 40th anniversary with a 6,000-person rally in the neighborhood. The Antorcha also maintains a steady presence in the community, running, for example, a community center, and may have supported recent nearby land invasions.
The PRI’s influence in neighborhoods like Zaragoza Sur could prove sturdy: Coahuila does not hold gubernatorial elections until 2023, and the state government will continue to funnel resources to and through Antorcha leaders. While federal government policies could possibly flip the loyalties of community organizers, the dynamics of local politics (and Torreón is governed by the PAN) will complicate that process.
Where Coahuila is a lingering PRI stronghold, Zacatecas, just to its south, represents a more complicated picture. After a narrow defeat in the 2016 gubernatorial election, Morena gained ground across much of the state and won a resounding victory in 2018. This is not totally surprising, given that Morena’s regional coordinator was former state governor Ricardo Monreal. Indeed, the PRI retained only 33 of its 186 bastions from 2016, and in 39 of those former bastions it received less than 25% of the vote. Statewide, the party took a beating.
On a local level, however, the PRI in Zacatecas appears to have successfully dug in. In many precincts Morena did significantly worse in 2018 than it did in 2016, while the PRI held steady. As in Coahuila, only one bastion precinct flipped, and it was similarly a smaller precinct of only 130 total votes. In precincts that had been 2016 bastions, Morena only averaged 23.6% of the vote—11% worse than it did overall in the state—and the PRI averaged nearly 37% of the vote (a figure that only impresses by comparison to national rates.)
Perhaps the most striking example of PRI entrenchment is in the rural Pinos municipality on the state’s eastern border with San Luis Potosí. There, the PRI won 41% of the vote, down 7% from 2016, while Morena received 27%, down 10%. More than half (17) of the PRI’s bastion precincts are in Pinos.
The PRI’s longstanding control of Pinos is characteristic of the party’s survival in poor rural areas. The municipality has never been governed by an opposition party, and has long been a stronghold for priistas in Zacatecas. An arid agricultural area, residents cultivate cactus fruit, beans, and maguey for mezcal production, as well as raising some livestock. It is very poor, with 73% of the population living in poverty per the 2010 census, and like most of Zacatecas it has experienced significant migration to the United States.
This poverty helps sustain the PRI’s influence, as municipal officials have engaged in classic paternalism, handing out cash to supplicants at town hall, as well as ensuring the flow of resources from social programs. The Antorcha Campesina is also active in the municipality, though its influence appears to be less than in Zaragoza Sur.
Recent elections have been marred by allegations of votebuying, coercion, intimidation, and fraud. Following the 2016 municipal election, Morena supporters protested by occupying town hall during a tense standoff. Though Morena had hoped to gain ground in the municipality on the strength of López Obrador’s support for the development of the agricultural countryside, the influence and control of regional priistas appears to be solid. If Pinos is emblematic of classic PRI fiefdoms, such bastions have become rare, and the party is likely to fight to preserve them. Given the relative remoteness and isolation of such places, eradicating corrupt practices will also likely prove more difficult than in urban areas.
Rural bastions have been central to the PRI’s control of Mexico State, the party’s historic heartland. There, however, the party failed to rebound after a razor-thin victory over a Morena challenger in the 2017 gubernatorial election. While it was expected that the PRI would fare poorly there in 2018, but behind an uninspiring candidate with little connection to the party’s base, Mexico State turned solidly for López Obrador.
Of its 441 bastion precincts in 2017, the PRI only retained 25, and won less than 25% of the vote in 102, and on average Morena won more than 30% of the vote in former bastions. Neither was the party able to mount any effective rearguard actions: though Morena only succeeded in flipping 11 bastions, in 1,842 precincts where Morena won 30% or more of the vote in 2017, López Obrador’s party increased its vote share by more than 15% in 2018.
Much of this collapse was concentrated in rural areas where the data suggest that, unlike Zacatecas, the PRI’s coercive control seems to be faltering. Of particular interest is a region in the western part of the state, where the party has long maintained power through the support of the Mazahua indigenous group. This zone includes Atlacomulco, the birthplace of a legendarily powerful PRI political clique. Indeed, the Mazahua region contains 17 of the PRI’s 25 remaining bastion precincts—but also 44% of the “lost” precincts and 8 of the 11 precincts that flipped for Morena.
The Mazahua’s traditional linkages to the PRI are bolstered by classic machine politics. In the region, effective clientelist use of social programs and exploitation of poverty have contributed to the power of PRI bosses in the region. One such figure, Mario Santana, has served as municipal president of Villa Victoria on three separate occasions and won election to a fourth in 2018.
Nevertheless, Morena found growing support in the region during the campaign, and won 37% of the vote, 18% more than in 2017. The PRI, meanwhile, lost nearly 24% of its 2017 vote in the Mazahua region, compared to only 12% statewide.
Whether López Obrador’s party will be able to build on that trend, however, is unclear. In a number of Mazahua municipalities Morena’s 2018 increases were dramatic, such as improving from 9 to 25% in Villa Victoria, which could be attributed to López Obrador’s personal appeal. Long-term change in the region, then, will likely depend on Morena’s ability—independent of the new president’s popularity—to build political networks in the region and recruit local leaders.
Of the four states discussed here, Hidalgo represents the most striking and shocking defeat. Of the four states, Hidalgo's PRI governor had the highest approval ratings in 2017. Yet despite the efforts of PRI campaign coordinator and former Interior Minister Miguel Osorio Chong, Morena improved by 41% over its 2016 showing and flipped 61 of the PRI’s 197 bastions, leaving the party with only a fraction of its former strength. On average, Morena won 40% of the vote in precincts where the PRI had formerly been dominant.
What happened in Hidalgo? For one, the state saw a rash of PRI defections that prompted the governor to complain about traitors, and internal divisions contributed to apparent widespread discontent with the party. This opened the door for Morena to successfully recruit influential local politicos, and reward former priistas with candidacies.
Statewide patterns repeated in microcosm in the municipality of Tizayuca, a small city bordering on Mexico State. There, despite having elected a PRI mayor by a large margin in 2016, voters migrated to Morena massively in 2018. Of the PRI’s 14 bastion precincts in Tizayuca, 9 flipped and the other 5 were lost.
This reversal is intriguing because there is no clear explanation. The current PRI administration has failed to address growing insecurity in the area, blaming the problems on spillover from Mexico City. Rapid urban growth—often via shoddy government backed construction—has resulted in semi-abandoned neighborhoods that have encouraged criminal activity as well. Some of these recently developed neighborhoods, such as Pedregal and Unidad Deportiva, have histories similar to that of Zaragoza Sur in Coahuila (they only received electric service in 2011), but both communities flipped for Morena in 2018.
Another factor may have been divisions within the local PRI, where long-running factionalism may have pitted the current mayor against the municipal party president. It was also rumored that the PRI’s selection of a state congressional candidate from an established political family may have alienated younger party activists in the area.
A third element was the decision by local PAN-PRD candidates to withdraw from their races and back López Obrador and Morena, as the prevailing winds became clear.
All told, the PRI’s defeat in Tizayuca suggests not only that it will be extremely difficult for the party to retain urban strongholds, but that the biggest threat to survival in these areas might be internal.