The hegemony of PRI: absence of alternance, PRI alliances and margin of victory

The absence of alternance in government is a constant in the case study. The state of Mexico and Coahuila are the only two states that up to February 2023 have not experienced a political alternance in the state government. Part of the explanation offered has to do with the interaction between electoral alliance and opposition candidate, who in the case of being external (former PRI militant, entrepreneur or citizen) is propitious for alternance ^{(Espinosa, 2015)} . In this state, where this combination has not occurred, it would seem impossible to change officialism, though Coahuila experienced a PAN-PRD alliance in 1999, which in any case did not produce alternance. The state of Mexico is the only state which has neither had a PAN-PRD antihegemonic alliance nor experienced alternance in state government.⁷

Previous studies have identified this “anomaly” in the state of Mexico, given the segmentation of its local and federal legislative elections, in which there are usually high electoral competence levels and alternance, however, not in the state government ^{(Arzuaga et al., 2007}; ^{Cedillo, 2012)} .

Other indicators of the officialist force are individual participation in elections and favorable conditions for adding political partners in the alliance when more competitive elections take place. The supposition is that forging alliances intends to increase their vote turnout. As noticed in Chart 2, with the exception of 1999, PRI has forged “PRI” electoral alliances, adding partners every electoral cycle.

Chart 3 shows increasing margins of victory for PRI, above 30% in the analyzed period. Between 1999 and 2005, it barely increases six percentage points, but has an increase of 14 points in the 2011 elections; these two figures show the consolidation of its electoral hegemony. It grows slightly more than 15 percentage points between 1999 and 2005, and the difference with its closest competitor reaches 15 points in 2011. In that year, electoral results and margin of victory exhibited the slim chances of PAN and PRD to compete on their own, or as an alliance with PAN or PRD ideology; only an antihegemonic alliance might work in such asymmetric electoral scenario.

It is not idle to bring to mind that with the ascension to power of MORENA, the 2017 election shows a clear breakage in the partisan dynamic in force for at least three decades, and was evinced in electoral registrations. Between 2011 and 2017, PRI lost 27.24 percentage points, and 38.14 points of margin of victory. The election was decided with a margin of victory of merely 2.87%.

The electoral records of the potential PAN-PRD alliance reveal that the hegemonic PRI in the state of Mexico would have been defeated in 1999 and 2005; however, data show it was not enough to win the state government in 2011 and 2017. Even if in the elections to renew governor it was never possible to forge an antihegemonic alliance opposing to PRI, the successful ascension of MORENA at national and local level triggered the forging of an officialist alliance composed of PRI, PAN and PRD to retain-defend the state government in the hands of PRI. Over time, only the common enemy to defeat changed: PRI up to 2017, and now MORENA in the 2023 elections.

Electoral cycle and concurrence with presidential elections

The electoral cycle is expected to foster antihegemonic alliances whey they are nonconcurrent with presidential elections and when many elections take place the same day ^{(Reynoso and Espinosa, 2017: 414-415).} The state of Mexico meets this postulate for its elections to renew the local Executive took place in 1999, 2005, 2011 and 2017, that is to say, a year before the presidential elections in 2000, 2006, 2012 and 2018; although there is no full vertical concurrence, there is clear “closeness” between the election of state governor and president.

To enquire whether this distorts the hypothetic relationship, all the cases with such characteristic were examined. In 1999, the review of electoral calendars for governors produces seven states with this closeness to the presidential election: Baja California Sur, Guerrero, Quintana Roo, Hidalgo, state of Mexico, Nayarit and Coahuila. All lack vertical-presidential concurrence, but the thorough review shows the presence of local concurrence,⁸ an aspect omitted in previous studies.

In the group three patterns are distinguishable: states with local concurrence (Baja California Sur, Quintana Roo, Nayarit, Coahuila); states with differed local concurrence⁹ (Hidalgo and Guerrero); and, states with no local concurrence (state of Mexico).¹⁰

Out of those with local concurrence, Nayarit had an antihegemonic alliance and accomplished the first change in power in 1999, and in 2021, once again, there was an antihegemonic alliance;¹¹ Coahuila also tried a PAN-PRD alliance in 1999, and lost the election; Quintana Roo had a PAN-PRD alliance in 2016 —two years before the presidential in 2018— and accomplished the first alternance in state government; Hidalgo, one of the two states with differed concurrence, presented a PAN-PRD alliance in 2010 —two years before the 2012 presidential election— and lost. That is to say, out of the seven states with no presidential concurrence, four that held simultaneous or differed elections at local level presented PAN-PRD alliances; this is a finding compatible with the hypothesis of no vertical concurrence and many elections on the voting day.

Baja California Sur and Guerrero did not have any PAN-PRD alliance in spite of sharing the same conditions of the previous four, though since they modified the electoral calendar to match it with the federal lost their inclusion in the analysis group after 2011.¹²

By 2017, only three states had governor elections a year before the presidential: state of Mexico, Coahuila and Nayarit. As noticed in Chart 4, the first is the only without vertical and local concurrence. In 1999, there was at least seven states with governor elections in a single year, a condition deemed a key factor for the concretion of PAN-PRD alliances, in this way it made sense to register them in Coahuila and Nayarit, as in addition to the vertical non-concurrence there were many elections (governor, local representatives and town halls¹³). By contrast, the state of Mexico has two conditions that hinder antihegemonic alliances: absence of local concurrence, and after 2011, few governor elections the same day as in other states.

Although closeness to presidential election is not exclusive of this state because it was shared it with six states in 1999 and with two in 2017, these had at least an antihegemonic alliance in the analyzed period. What is distinctive in the election of the state of Mexico is that it systematically combines the absence of vertical concurrence and the absence of local concurrence. This makes it difficult to negotiate an eventual opposition alliance in this state, since there is only the election of governor; none can offer the other payments or benefits attractive to join.

Available elite and PRI alliances¹⁴

In the renewal of governor in 1999 and 2005, the state of Mexico registered the most competence inside officialism. It has been documented that selection processes of candidates are moments of division because the governor in office and his candidate compete with advantages against other candidates, who ask for impartial mechanisms to define the candidacies. By and large, it is almost impossible to counteract the candidate of the governor in office ^{(Hernández, 2008: 66-67)} . When the contest closes and this candidate is imposed, some candidates left PRI and looked for a candidacy in the opposition, causing a triple effect: divide the local PRI, offer competitive candidacies for the opposition, and make the formation of PAN-PRD alliances easy, since it is an ideological neutral in the face of left and right stances of traditional opposers ^{(Salazar, 2017)} .

In 1999, Arturo Montiel’s candidacy provoked the renounce of Humberto Lira Mora ^{(Hernández, 2005: 187-188)} , who joined the ranks of PRD in the state of Mexico, but was not the opposition candidate; this was also the case of Isidro Pastor —president of PRI in the state of Mexico— in 2005, who moved away from politics, but did not emigrate in the electoral conjuncture to the opposition to head an alliance. The main grudge in both cases came from the favoring of a certain candidate by the governor in office ^{(Niño, 2017)} . This practice was repeated again in 2011, when the candidacy had to be chosen and Eruviel Ávila strongly intended to be, though he was not part of the group of by-then Governor Peña Nieto. Unlike Lira and Pastor, the governor gave up on his preferences in the face of the risk of division amidst a competitive campaign and Eruviel Ávila’s extensive electoral base in Ecatepec, one of the most populated municipalities in the state ^{(Vivero, 2017)} . By 2017, there is no record of renounces in the officialist campaign of Alfredo del Mazo Maza.

Alliance policy in the state of Mexico, 1997-2023

Electoral alliances have been intensively used in elections in the state of Mexico. Out of 2133 registrations, the most have been ideological alliances (PRI, PAN and PRD and MORENA) with 84.32% (1,799), and to a lesser extent, antihegemonic alliances (PAN-PRD, PRI-PAN-PRD) with 15.65% (334); out of this group, PAN-PRD alliances account for 9.79% (209), a marginal option (see Chart 5). The party with the most alliances has been PRI with 48.61%; MORENA, 18.09%; PRD, 13.78%; and, PAN, barely 3.84%. The last two figures are key, they show the traditional opposition far from PAN-PRD alliances as instruments to defeat PRI. PAN militants in the state of Mexico have privileged their participation on their own and to a lesser extent, ideological alliances before joining with PRD. Partly, this makes sense due to the increase complexity of elections in the state of Mexico as of the 1990’s and the facture of PAN between those dogmatic and the pragmatic in the party.

Three alliance tendencies are identified over the period under study. The first, the solitary participation of PRI, and ideological alliances of PAN and PRD (1997-2000). The second is PRI’s exclusively alliance participation, PRD’s ideological alliance and the poor forging of PAN ideological alliances (from 2000-2017). The third is a change stage: PRI presents its last ideological alliances, late PAN-PRD alliances in concurrent elections —though not for governor—, while MORENA alliances appear and for the first time an alliance between PAN, PRI and PRD (from 2018 — to present).

For as long as PRI was in federal and local alliances between 2003 and 2018, political opposition displayed differenced patterns: PAN forged alliances for governor in 1999 and 2005, for the federal election in 2000, and for municipal presidencies in 2009 and 2015, though in the rest of elections competed on its own. For its part, PRD forged alliances in federal elections in 2000, but not in local elections in 2000 and 2003, and between 2005 and 2012 forged more federal than local alliances. PAN and PRD competed on their own between 2015 and 2017. The late forging of PAN-PRD antihegemonic opposition alliances took place in the concurrent 2018 elections, while antihegemonic PAN-PRI-PRD officialist alliances occurred in the concurrent in 2021.¹⁵

Electoral cycle, nationalization and candidacies

Even if it is expected that non-vertical concurrence and holding elections on the same day incentivize the formation of PAN-PRD alliances, in the thorough revision of the case, the state of Mexico verifies nonvertical concurrence, though unusually it records nonlocal concurrence and scarce elections on the same day. This seriously complicates local politicians because they do not have positions in return for an antihegemonic opposition alliance. The elections of local and federal representatives, municipal presidencies and senators are always held a year later. There were many elections on the same day in other states in 1999, nevertheless, as noticed over time, it reduces to three in 2017 and two by 2023. In this year, only Coahuila will be available for a possible national negotiation, in which PAN and PRD are opposition for in this state and those spaces serve as negotiation resources for local politicians. Opposition in the state of Mexico depends to a large extent on access to resources in other spheres, where they would have to negotiate and receive other sorts of interchange resources at the local level, interesting for national politicians. As displayed below, the state of Mexico has visible characteristics that promote sufficient incentives for multilevel negotiations.

Owing to its dimensions, the state of Mexico not only holds the largest population¹⁶ and nominal listing,¹⁷ accounts for the largest legislative contingent in the federal congress by state (64 legislators representing 12.8% out of 500 representatives)¹⁸, and because of its closeness to Mexico City, the local politicians reach national noticeability, and usually venture as presidential candidates or members of the federal cabinet (^{Hernández, 2005 and 2015)}. In this way, the electoral cycle ends up nationalizing significant resources that may be controlled by controlling the government in the state of Mexico, with the clear goal of increasing their odds of succession in government and in the presidential election. In this context, by nationalization, we understand the centralization of decisions made by the dominant coalitions of political parties at national level regarding party and electoral strategies at various levels.

However, nationalization does not mean the arbitrary meddling of national elites, it refers the decisive influence of national partisan elites when political processes concur with presidential elections —called first-order— ^{(Carreras, 2018: 543)} and/or when highly-relevant particular spaces for national politics are detected ^{(Battle and Puyana, 2011: 53)}.

The opportunity window for the local-national interchange of the opposition is provided by the national electoral calendar when presidential candidates are under definition. As we will expose following, evidence reveals that when the local opposing party was the same as the officialist president’s party, the heavy influence of presidency was strengthened in the state; when the local opposition party was also national opposition, the influence of the national dominant coalition of the party on the local sphere increased.

PRI state officialism does not experience this situation because it never has lost its position of state incumbent¹⁹ and has resources so that the governor and his party ^{(Arzuaga et al., 2007)} may “add votes allotting seats” ^{(Reynoso, 2010)} , hence, fulfilling agreements with other parties when candidacies are defined, or even conceding positions in public state administration.²⁰

By contrast, for the opposition this cycle acts against it for there are neither conditions for local and national negotiations nor interchanges with credible promises ^{(Reynoso and Espinosa, 2017: 417)}. As previously observed, many elections increase the resources to be allotted to candidates and opposition parties to forge the alliance; in the absence of groups for interchange, the option is to reach out for national resources.

Since PAN was in power between 2000 and 2012, the elections to renew state governments may be grouped into those with no federal government influence (1999 and 2017), and those that experienced it (2005 and 2011).

In 1999, PAN’s candidate José Luis Durán was a local politician with a traditional dogmatic profile, where the governmental experience was incipient, mainly as regards local representations and town halls —out of the influence from PAN federal government, with increased competitivity— and PAN’s antiregime struggle took place at national level (Chart 6). For her part, 2017 Presidential Candidate Josefina Vázquez showed the dynamic and wear of PAN after being in power for two presidential terms; she did not have supporters in the state, neither had she developed her political or administrivia career there, nor had ever been a local representative.²¹ Therefore, her candidacy for state governor made it clear the depletion of local PAN bases, the division of PAN in the state of Mexico (traditional vs. neo-PAN militants), ideological reinforcing, and mainly, the higher centralism of PAN national leaders. These two candidates, Durán Reveles and Vázquez Mota, evince the influence of PAN dynamic at two contrasting moments: the first, when the party was the main national and local opposition, and the second, national political electoral wear with no local competitive groups.

The influence of PAN federal government was noticed in the electoral processes in the state in 2005 and 2011. After an accidented process Rubén Mendoza Ayala won the state of Mexico PAN’s internal elections, however, owing to an applicable inconformity by the losing candidate, José Luis Durán Reveles, the candidate was appointed by the National Executive Committee with a divided result: 17 in favor, 14 against, and 3 abstentions ^{(Román, 2005)} .

The presidential and party insistence on ratifying Mendoza Ayala in spite of raising tensions and creating division between state party leaders could not be understood without the division that was taking place in the succession of PAN national leaders, headed by presidential couple Vicente Fox-Martha Sahagún (2000-2006); Santiago Creel and Manuel Espino (PAN’s Secretary General, 2002-2004) on one side; while on the other, Luis Felipe Bravo Mena (president of PAN’s National Executive Committee from 1999 to 2005), Carlos Medina Plascencia, and Felipe Calderón. Rubén Mendoza Ayala was appointed as a candidate in November 2004, when Luis Felipe Bravo Mena was about to leave the direction of the national executive committee.²² With the triumph of the national leaders, a PAN-PRD alliance was unthinkable due to the confrontation between Fox and López Obrador, the most likely PRD candidate.

Owing to the forging and cancelation of a national PAN-PRD anti-alliance agreement, which included the state of Mexico²³, the legislation that disincentivized the union of these parties was reformed and Luis Felipe Bravo Mena, former particular secretary of President Calderón²⁴— was registered as a candidate for governor. The 2011 election was somehow influenced by PAN federal government against the candidates and local leaders.

In their way, presidents from PAN subjected the party in the selection of candidates in the state of Mexico, and the governor elections were used as hard currency in national strategies. That is to say, despite the lack of vertical concurrence, inductively, history has taught us that presidential succession has indeed influenced the likely integration of an opposing alliance in the state of Mexico.

PRD in the state of Mexico experiences the same pressures, though extremely, because if has systematically been federal and local opposition. It wholly depends on the national party, particularly, in its dominant coalition at national level. Higinio Martínez postulated when Andrés Manuel López Obrador was PRD president while its founder, Cuauhtémoc Cárdenas, was governor of Federal District, which for his part, was the presidential candidate of PRD. As PAN over the 1990’s, PRD also resorted to the few local politicians trained as local representatives, municipal presidents, or local state leaders (Chart 8) .

The information reveals a clear closeness between López Obrador and Yeidckol Polevnsky, candidate to governor in the state of Mexico in 2005. She had a notable entrepreneurial career, though she only started in politics hand in hand with the favorite in the surveys in 2005. Though López Obrador, as a PRD leader, promoted alliances in 1999, in 2005, his clash with Vicente Fox due to the impeachment process cancelled every approach to forge a PAN-PRD alliance.

The influence of the presidential candidate, López Obrador, is verified in the 2011 state elections, when a close collaborator in his government, Alejandro Encinas, was appointed as candidate to governor for the state of Mexico, who as a native of this state had made a career in Mexico City’s government, where he was a direct subordinate of the head of government, López Obrador. Despite entering into antinatural agreements in a number of states in 2010, a year later, the national PRD leaders and López Obrador rejected this sort of coalition in the state of Mexico. Though PRD in the state of Mexico never witnessed the meddling of a president from PRD, it indeed experienced the influence of their main presidential candidates, Cárdenas in 2000, and López Obrador in 2005 and 2011.

In 2017, López Obrador was no longer the main character in PRD, its governor candidate Juan Zepeda —with a short political career — expressed his disapproval for the forging of a PAN-PRD alliance, albeit in 2018, he was the only competitor on behalf of his party. If in previous years, the addition of these oppositions was able to defeat the local PRI, by 2017, the uneven presence in relation with PRI or MORENA had been a sufficient reason to forge an alliance as a sort of antihegemonic contention.