Introduction
Recent scholarship has begun to shed light on the role of religious institutions in the process of autocratizationFootnote 1 during the post-Third Wave period (Leininger Reference Leininger, Croissant and Tomini2024; Lotfy Reference Lotfy2024). This research builds on earlier work that highlighted religious institutions’ significant political influence during the Third Wave of democratization (Haynes, Reference Haynes2009; Huntington Reference Huntington1991; Mantilla Reference Mantilla2010; Resende Reference Resende2018; Toft, Philpott, and Shah Reference Toft, Philpott and Samuel Shah2011). Given that a considerable amount of democratic backsliding follows from elections (Bermeo Reference Bermeo2016), it is crucial to examine the growing role of religious institutions during electoral manipulation in the post-Third Wave period.
According to the Varieties of Democracy (V-Dem) 2023 Report “Defiance in the Face of Autocratization” (Papada et al. Reference Papada, David Altman, Lisa Gastaldi and Martin Lundstedt2023), elections are increasingly being manipulated across the world as the quality of elections is worsening in 30 countries. This trend reverses 35 years of global democratic gains, fueling the so-called “Autocratization Wave” (Lührmann and Lindberg Reference Lührmann and Lindberg2019). Aside from the traditional tactic of election-day vote fraud in polling stations, manipulating elections strategically is one of the contemporary forms of democratic backsliding (Bermeo Reference Bermeo2016). Electoral manipulation is defined as a variety of tactics used to sway elections in the incumbents’ favor prior to election day (Bermeo Reference Bermeo2016, 13), including constraining media access, supporting incumbent candidates with government funds, intimidating and preventing opposition candidates from running for office, impeding voter registration, hampering electoral commissions, and manipulating election laws to the advantage of incumbents.
Due to the co-optation of media, political institutions, and electoral laws, recent scholarship has increasingly emphasized the importance of studying resistance to autocratization (Gamboa Reference Gamboa2022; Tomini, Gibril, and Bochev Reference Tomini, Gibril and Bochev2023). In this article, I focus on Catholic churches, as they are not only recognized as one of the primary dominant religious institutions that adapted to democracy in Western Europe (Warner Reference Warner2000), but they also emerged as key actors playing significant roles in democratic transitions during the Third Wave (Huntington Reference Huntington1991).
Scholars framed the Third Wave as a “Catholic wave,” as most of the countries that witnessed democratic transitions were predominantly Catholic (Huntington Reference Huntington1991; Philpott Reference Philpott2004, Reference Philpott2007). But Catholic churches were not unanimous in their response—scholars argued that Catholic churches helped bring about democratization in some countries, while other Catholic churches hindered democratization or played no part (Edmonds Reference Edmonds2013; Mantilla Reference Mantilla2010; Philpott Reference Philpott2004; Troy Reference Troy2009). Hence, it is important to study the public role of Catholic churches in the current wave of backsliding and explore the reasons behind their varying positions.
I collected data on the stances of national Catholic churches toward presidential electoral manipulation post-Third Wave (1990s–2023), studying 59 cases across Latin America, Africa, Eastern Europe, and Asia. Despite these geographical differences, these regions experienced the Third Wave of democratization from the 1970s to the 1990s (Huntington Reference Huntington1991), during which national Catholic churches played significant and varied roles (Toft, Philpott, and Shah Reference Toft, Philpott and Samuel Shah2011, 82-120). It is thus important to consider the role of these national Catholic Churches in the current democratic setback, as the level of democracy in these regions has since deteriorated (Papada, et al. Reference Papada, David Altman, Lisa Gastaldi and Martin Lundstedt2023).Footnote 2 I found that 32% of national Catholic churches resisted the electoral manipulation that happened in their countries, 34% called for peace, and 34% took no stances. This begs the question: Why do Catholic churches adopt different stances toward electoral manipulation in the post-Third Wave period?
I approach this question from the perspective of religious economies, wherein the responses of religious institutions to democratic backsliding depend on their cost-benefit calculations regarding their freedom from regulation, competition, and historical commitments. I argue that Catholic churches resist electoral manipulation when the state’s official support for Catholicism is too low, even when they control a considerable proportion of the population. Meanwhile, Catholic churches call for peace when government favoritismFootnote 3 toward Catholicism is high. This can be termed “religiously friendly autocratization,” as opposed to Driessen’s “religiously friendly democratization” (Driessen Reference Driessen2014), where incumbent regimes favor certain religions or religious institutions in exchange for legitimacy or to prevent opposition while consolidating power.
Additionally, the historical pro-democratizing role of Catholic churches may have a positive impact on their decision to resist electoral manipulation in the post-Third Wave period, and especially for those who are facing high competition in the religious market. Hence, this article contributes to the political economy of religion approach by arguing that the cost-benefit calculations of religious institutions are shaped not only by state-religion regulation and the dynamics of the religious market but also by their historical involvement in the Third Wave of democratization. These arguments are tested using logistic and multinomial regression models.
The article proceeds as follows: the first section summarizes the literature on religion and elections. The second section describes the different ways a Church can engage in the electoral process. The third section theorizes the incentives behind Catholic churches’ different stances toward electoral manipulation. The fourth section lays out the data and the methodological tools. The fifth section tests the hypotheses before the discussion and conclusion in the final sections.
Religion and electoral politics: a review of the literature
Existing work on religious actors’ engagement in elections tends to study public and political speeches of clergy (e.g., Alavaa and Ssentongo Reference Alavaa and Spire Ssentongo2016; Djupe and Gilbert Reference Djupe and Gilbert2002) and their partisanship during elections (e.g., Guth and Smidt Reference Guth and Smidt2022), the clergy’s influence on congregants’ vote choices, and the citizens’ response to such influence (e.g., Djupe and Gilbert Reference Djupe and Gilbert2008; Olson Reference Olson, James, Lyman and Corwin2009; Roso and Chaves Reference Roso and Chaves2023; Smith Reference Smith2019; Smith and Boas Reference Smith and Boas2024). According to this body of literature, there are motivational reasons behind these religious actors’ engagement in elections, such as religious competition (e.g., Smith Reference Smith2016), secularism and religious pluralism (e.g., Rosenberg and Smith Reference Rosenberg and Erica Smith2021), clerical conservatism (e.g., Glazier Reference Glazier2018), and theological considerations (e.g., Frahm-Arp Reference Frahm-Arp2019; Holman and Shockley Reference Holman and Shockley2017).
Recent scholarship has also started to recognize the growing salience of religion in the electoral manifestos of political parties, driven by the rise of radical right-wing parties (e.g., Layton, et al. Reference Layton, Erica Smith, Moseley and Cohen2021; Molle Reference Molle2018; Schwörer and Fernández-García Reference Schwörer and Fernández-García2021; Schwörer and Romero-Vidal Reference Schwörer and Romero-Vidal2020). Additionally, recent studies have highlighted how clergy, particularly Evangelical clergy, are not only supporting certain political candidates but are also engaging in elections as political candidates themselves (e.g., Boas, Reference Boas2020, Reference Boas2023; Hurst Reference Hurst2023).
Yet much of this literature is missing important issues. First, much of this research focuses on the individual level (clergy) and not on the institutional level (religious institutions). Only a few studies have tackled the phenomenon at an institutional level (e.g., Gherghina and Mișcoiu Reference Gherghina and Mișcoiu2022; Grzymała-Busse Reference Grzymała-Busse2015; Holman and Shockley Reference Holman and Shockley2017). Second, existing research focuses more on Protestants’ role in elections than on the Catholics’, except for a few studies (e.g., Byrnes Reference Byrnes1991; Castle Reference Castle2018; Holman and Shockley Reference Holman and Shockley2017; Jelen Reference Jelen and Corwin2004; Mantilla Reference Mantilla2021; Smith Reference Smith2016). Third, existing research does not examine the role of religious institutions and their clergy in election-related autocratic practices. Finally, most, if not all, of the aforementioned studies focus on a single case or region, rather than addressing the issue across multiple regions. Thus, I cover these shortcomings in this article by examining the stances of Catholic churches toward presidential electoral manipulation in the post-Third Wave period across regions.
Religious engagement in elections
There are many ways the Church can engage in the election process, whether before, during, or after elections. The religious engagements that precede election day can take the form of electioneering and engaging in political campaigns by forming alliances with politicians for future policy concessions. Such alliances risk politicizing the Catholic Church and weakening its moral authority (Grzymała-Busse Reference Grzymała-Busse2015, 10). The Church can also engage in electioneering without overtly interfering in politics, as clergy can individually support certain politicians, whereas the Church as an institution does not take a side. In Kenya’s 2007 elections, the Catholic Church remained silent, while some leaders supported President Kibaki due to ethnic ties. Maupeu (Reference Maupeu2008) argues that this silence damaged the Church’s moral authority, undermining its image as a non-ethnic national institution.
Religious engagements that precede election day can also take the form of mediation efforts between the incumbent ruler and the political opposition, especially before potentially divisive elections. For example, the Episcopal Conference of the Democratic Republic of the Congo (CENCO) mediated between the government and opposition before the 2018 elections, originally scheduled for 2016 but delayed due to the president’s unconstitutional bid to extend his term in office (Africanews 2015). Churches may also play a role in educating citizens about elections, electoral laws, and the voting process (Buckley, Mohan, and Bledsoe Reference Buckley, Mohan and Bledsoe2023). Churches also play a role in warning citizens against political inducements, bribery, vote-buying, and being used by politicians to cause violence.
Another important role can take the form of encouraging voter participation. For instance, the Tanzania Episcopal Conference president urged Catholics not to boycott the 2019 elections (Karashani and Kidanka Reference Karashani and Kidanka2019). Churches may also advocate for electoral reforms. For example, ahead of Togo’s 2020 presidential elections, the Episcopal Conference issued a pastoral letter urging the government to revise the Electoral Code and restructure the Electoral Commission and the Constitutional Council to ensure transparency (Atemanke Reference Atemanke2019).
Religious engagement may also be witnessed before and during the elections through election monitoring. Churches train observers to ensure fair elections; if violations occur, they may withdraw, delegitimizing the results. This happened in Burundi less than a week prior to the 2015 elections (La Croix Reference Croix2015). This may explain why governments sometimes ban Church election observers, as in the Republic of Congo, where the Catholic Church was prevented from deploying 4,000 Justice and Peace Commission observers for the 2021 presidential elections (La Croix Africa 2021).
Finally, religious engagement can take place after the elections. Churches can pressure electoral commissions to release delayed results. For instance, the Honduran Bishops’ Conference expressed concern over the 2017 election result delay, arguing it raised suspicions and could lead to instability (Agren Reference Agren2017). Churches may also bless electoral results and urge losing candidates to accept them or await a court ruling. For example, the Episcopal Conference of Guatemala supported the 2023 election results amidst attempts by the public ministry to delegitimize them (AICA 2023; Roman Reference Roman2023). Meanwhile, churches may denounce election results. For example, a Catholic Church representative in Venezuela condemned the 2018 election manipulation and called for the presidential elections to be re-held (El Nacional Web 2018). In the Democratic Republic of Congo’s 2018 elections, the Catholic Church deployed 40,000 observers, urged the electoral commission to publish results (Maclean Reference Maclean2019), and challenged the official outcomes, claiming they differed from its data (Powell Reference Powell2019).
Protests usually take place because of these manipulated elections. In such contexts, churches may act as national representatives and security preservers (Grzymała-Busse Reference Grzymała-Busse2015, 12). In other situations, churches may engage in protests and demonstrations, as happened in the Democratic Republic of Congo after the 2011 elections (BBC News 2012). Electoral violence can also be witnessed before, during, and after the elections. Churches typically condemn violence, call for prayers and dialogue, and play the role of the mediator. For instance, the Conference of Catholic Bishops of Burundi denounced violence during the 2010 elections and recommended a dialogue between the ruling regime and the opposition (Cath.ch 2011).
The above reveals that Catholic churches, whether before, during, or after elections, may resist electoral manipulation, call for peace and dialogue, or remain silent without taking a clear stance. This article examines Catholic churches’ post-election stances when electoral manipulation is evident, focusing on observed—not just anticipated—democratic backsliding. Some Catholic churches criticized elections beforehand but stayed silent or only called for peace afterward, as seen during Uganda’s 2021 elections (The Independent 2021). Consequently, this article does not categorize their stances as resistant. The focus thus remains on the actual actions taken by these churches after the elections. In the following section, I examine why Catholic churches take different stances, using insights from the political economy of religion perspective.
Theoretical insights on the incentives underlying Catholic churches’ stances toward controversial elections
Religious institutions decide whether to resist autocratizing moves by incumbent regimes through a cost-benefit analysis. The potential benefits of each option should reduce the threat of losing followers and maintain moral authority. The state regulation of religion, along with the level of religious competition, is regarded as the primary factor influencing the political offerings of the religious institution. There is a strong body of religious economic theories that tackles the influence of state-religion regulation on religious freedom (e.g., Finke and Martin Reference Finke and Martin2014; Gill Reference Gill2008; Fox Reference Fox2008) and regime legitimacy (e.g., Fox and Breslawski Reference Fox and Breslawski2023; Gill Reference Gill2005; Sarkissian Reference Sarkissian2015; Schleutker Reference Schleutker2021). Other studies have tackled the state-religion regulation’s effect on religious institutions’ political stances, especially toward democratic transitions (e.g., Driessen Reference Driessen2010; Gill Reference Gill1994; Philpott Reference Philpott2007).
Given that this article focuses on a contemporary form of autocratization—electoral manipulation—a religious economy approach proves useful in outlining the crucial variables, namely religious regulation and competition, that influence whether a Church would push back against or accept electoral manipulation. Therefore, this article follows the literature that looks at the motivations of religious institutions rather than focusing on the motivations of the government when testing the effect of state-religion regulation.
From the government’s motivation perspective, governments that favor one religion and regulate others can expect a return on their investment: the favored religious body will, in turn, legitimize the authority of the ruling regime (Fox and Breslawski Reference Fox and Breslawski2023; Gill Reference Gill2005, 20; Toft, Philpott, and Shah Reference Toft, Philpott and Samuel Shah2011, 41). Therefore, the motive for the government is to ensure regime survival by neutralizing any potential religious threats and reducing the threat of religious mobilization against the ruling regime (Grim and Finke Reference Grim and Finke2011, 50–51). The effect of these transactional dynamics will materialize in critical times like elections, especially those that are manipulated by the incumbent regime. The regime that favors the Catholic Church and regulates others will then expect the Catholic Church not to criticize the elections, welcome the results, and congratulate the leader for their new term. This is what Daniel Philpott (Reference Philpott2007) calls “a consensual arrangement,” where both parties are satisfied with the status quo. This can also be described as “religiously friendly autocratization,” where regimes favor certain religions to secure legitimacy or suppress potential opposition during power consolidation.
Just as the government’s motivation for supporting religion is to enhance and maintain its legitimacy, Catholic churches also need government support in order to maintain their position in the religious market. The state subsidies can take the form of funding religious buildings and activities, financing clerical salaries, or supporting religious education, etc. (Sarkissian Reference Sarkissian2015, 27). From the Catholic churches’ perspective, Catholic churches desire to have friendly relations with the regime even when there is a separation of Church and state so as to influence policy decisions and get the maximum benefits of state subsidies (Gill Reference Gill1994, 405). Churches with such allies are significantly more effective in shaping policy and benefit from financial and legal privileges, which often dissuade them from opposing the incumbent regime (Grzymała-Busse Reference Grzymała-Busse2015; Philpott Reference Philpott2007).
Therefore, churches seek to maintain friendly relations with the state in order to lower the costs of fierce competition and to use their access to state resources to retain their current members and to attract new ones, which means suppressing competition (Gill Reference Gill1994, 412). When this kind of state support is not available, which means being threatened by competition, membership loss, and not being the only religious authority that can speak for the nation, Catholic churches then try to retain their position in the market by adopting appealing political platforms, challenging authoritarian regimes and regimes with autocratic intent, and promoting pro-poor movements (Gill Reference Gill1994; Trejo Reference Trejo2009). Therefore, autonomy from the state enables religious institutions to engage in democratic politics (Philpott Reference Philpott2007; Toft, Philpott, and Shah Reference Toft, Philpott and Samuel Shah2011).
A good example of the freedom and incentives to engage in democratic politics is the Catholic Church in the Democratic Republic of the Congo (DRC). Since the Third Wave of democratization, the Catholic Church has been high on autonomy and has had strong leadership. Catholics comprise almost half of the population. Independents and Protestants are the main competitors in the religious market, as they attract the adherence of the other half of the population (Johnson and Grim Reference Johnson and Grim2022). Government favoritism toward Catholicism is almost similar to other religions in the country (Brown Reference Brown2020). Moreover, the Catholic Church manages large public sector entities in the field of education and health—more or less 50% of the schools and 40% of health structures (Vatican News 2022). As the Catholic Church is fulfilling the Congolese state’s responsibilities and being threatened in the religious market, it has the incentives and capacity to call on the people to mobilize in the streets, demanding political reforms (Africa Center for Strategic Studies 2018). It has also advocated against alleged electoral manipulations in 2011, 2018, and 2023.
Hypothesis 1(A): When there is low government favoritism toward Catholicism, the national Catholic Church is more likely to resist electoral manipulation than to call for peace or take no stance.
Additionally, Catholic churches that control a large portion of the religious market but enjoy low government favoritism may also engage in democratic politics. An example of this is the Venezuelan Catholic Church, which comprised 80.5% of the population by 2020, according to the Association of Religion Data Archives (ARDA) (Johnson and Grim Reference Johnson and Grim2022). Despite its significant share of the population, the Venezuelan Catholic Church faces challenges from growing Evangelical support, backed by the administrations of Hugo Chavez and Nicolas Maduro, weakening its traditional dominance (Smilde Reference Smilde2018; Smilde and Pagan Reference Smilde, Pagan, Smilde and Hellinger2011). Catholicism receives minimal government favoritism, even lower than its Protestant counterparts (Brown Reference Brown2020). For instance, according to the US Department of State’s International Religious Freedom Report (2022), the Venezuelan government halted funding for some Catholic schools despite the 1964 concordat and restricted Church activities, including denying religious visas.
As a result, the Catholic Church has emerged as a significant political player in the Venezuelan humanitarian and political crisis since President Hugo Chavez took power (Smilde Reference Smilde2018). For instance, it criticized Nicolas Maduro’s unconstitutional push for a constituent assembly in 2018 and condemned the government’s electoral manipulation attempts in the same year, advocating for the repetition of elections (Smilde Reference Smilde2018).
Hypothesis 1(B): When there is no high government favoritism toward Catholicism, the national Catholic Church is more likely to resist electoral manipulation, even if it controls a considerable proportion of the population.
Meanwhile, if there is no competition and the national Catholic Church is highly favored by the government, it is in the interest of the Church to maintain this kind of friendly relationship with the current regime, regardless of any legal separation between Church and state, since the government controls resources. Therefore, there is a constant incentive for the Church to maintain amicable relations with the incumbent regime (Gill Reference Gill1994, 412). However, a Church with high moral authority can be highly criticized for its stances overtly supportive of the regime. This Church, which is highly favored by the government, may instead act as a preserver of the nation by promoting patience and nonviolence. This could mean that the Church is keen not only on avoiding bloodshed but also on preserving a favorable regime (Grzymała-Busse Reference Grzymała-Busse2015, 41).
A good example of this is the Polish Catholic Church’s stance in the 2020 elections, which are believed to have been plagued with irregularities and fraud (Santora Reference Santora2020). The Catholic Church in Poland claims the adherence of 90% of the population (Johnson and Grim Reference Johnson and Grim2022). It is believed to be closely aligned with the Law and Justice Party (PiS), which came to power in 2015 and which promotes policies that align with traditional Catholic values, like endorsing a total abortion ban (Cienski Reference Cienski2016). Faced with low threats in the religious market and favored by the government, the Catholic Church was not motivated to take a resistant stance against the alleged electoral manipulation in 2020. The Polish Episcopal Conference took no stance after the elections, but, prior to the elections, it called for dialogue between the government and the opposition. It played a role by calling for the love of the homeland, appealing for peace, and asserting that it is not taking part in the political disputes (Rada Stała Konferencji Episkopatu Polski 2020).
Hypothesis 2: As government favoritism toward Catholicism increases, the national Catholic Church is less likely to overtly resist electoral manipulation and instead call for peace or take no stance.
The previous theory of churches’ incentives to support or resist the regime during critical moments in times of controversial elections follows the supply-side theoryFootnote 4 of the religious economy approach. Meanwhile, Lotfy (Reference Lotfy2024) argues that the religious economy’s expectations should not end with the type of state regulation specified. It’s vital to consider their historical pro-democratizing role in the Third Wave when examining their support for regime survival in today’s autocratic wave. Following Warner (Reference Warner2000, 15), the historical legacies of Catholic churches heavily influence their strategic considerations of costs and benefits.
The national Catholic Church may take the credit or the blame for its historical record during foundational moments: after a regime collapse, upon gaining state independence, and when building a brand-new democracy (Grzymała-Busse Reference Grzymała-Busse2015, 51 & 61). The historical record of the democratic role is basically dependent on the political theologies that Catholic churches adopted following the Second Vatican Council.Footnote 5 According to Philpott (Reference Philpott2007), the Catholic churches that came to oppose authoritarian regimes were those that adopted a liberal political theology, advocating human rights and democracy—themes significantly supported by Pope John Paul II. There were also churches that were neutral in their political theology, as they were willing to support any type of regime, failing to advocate democracy. Therefore, Catholic churches’ beliefs about legitimate authority and justice were key in driving opposition to authoritarianism, while those lacking such views stayed sidelined during the Third Wave (Philpott Reference Philpott2007, 512).
Thus, in the post-Third Wave, I argue that Catholic churches’ stable commitment toward their own political theology that was adopted during the Third Wave is important to consider while examining their stances toward the current backsliding wave. The guilt for endorsing authoritarian regimes and the credit for promoting democracy may both fall on the shoulders of the Catholic churches. Therefore, it is expected that Catholic churches with a historical pro-democratizing roleFootnote 6 in the Third Wave during the era of Jaun Paul II will be more likely to resist autocratic practices in the post-Third Wave during the eras of Benedict XVI and Francis.
Hypothesis 3(A): The national Catholic Church, with a historical pro-democratizing role in the Third Wave, is more likely to resist electoral manipulation than those that did not have a historical pro-democratizing role.
Nevertheless, this kind of stable commitment is more likely to exist when religious institutions are threatened in the religious market in order to maintain their adherents and attract new ones (Lotfy Reference Lotfy2024). The national Catholic Church faces a high cost, particularly in the face of competition from other religious groups, if it deviates from its historical pro-democratic role in the Third Wave. In such a scenario, the national Catholic Church risks losing its moral authority among its adherents, who may opt to freely convert to other competitive religions. Therefore, it is expected that Catholic churches, with a lower share of the religious market and a historical pro-democratizing role in the Third Wave, will be more likely to resist electoral manipulation than those that also have a historical pro-democratizing role but contrarily monopolize the market.
On one hand, the Polish Catholic Church played a vital role in the Polish struggle against communism and in advocating for human rights and democracy in the Third Wave (Eberts Reference Eberts1998). Nevertheless, the PiS’s electoral success in 2015 made it easier for the Church to gain political clout. In exchange, the PiS used this alliance as a legitimation of its illiberal and anti-democratic policies (Resende and Hennig Reference Resende and Hennig2021, 10). As a result, the Catholic Church became embroiled in Poland’s democratic crisis, losing its standing as the highest moral authority. Additionally, there is no other religious institution that has the capacity to both challenge the Catholic Church’s dominance in the religious market and act as a champion of democracy. The Church’s historical commitment to liberal democracy was then eroded by the absence of threats from alternate religions and the threat of communism.
On the other hand, the DRC Catholic Church, with its threatened position in the market, as previously illustrated, has always played vital roles in promoting democracy since the Third Wave to preserve its position and retain its followers. It is believed that the “Democracy in the DRC owes a lot to the Church” (Châtelot Reference Châtelot2023). Therefore, the stable commitment of the DRC Catholic Church toward democracy is stronger than that of the Polish Catholic Church.
Hypothesis 3(B): The national Catholic Church, with a historical pro-democratizing role in the Third Wave and a low religious market share, is more likely to resist electoral manipulation than those that have a historical pro-democratizing role but a high religious market share.
Proceeding from this, I augment the political economy of religion approach by asserting that the cost-benefit calculations of religious institutions are influenced not only by the state-religion regulation and the dynamics of the religious market but also by their historical role in the Third Wave of democratization.
Data and method
I constructed a dataset of the stances of Catholic churches toward allegedly manipulated presidential elections across regions in the post-Third Wave period (1990s–2023). The unit of analysis is national Catholic Churches (whether they are a majority, a second majority, or a minority). It is worth noting that I do not consider Catholic churches as monolithic entities. I acknowledge the potential divisions that may arise within the Catholic Church regarding electoral controversies. However, the focus of this article is on the official stance of the national Catholic Church as an institution. The official stance refers to the public position of the national Catholic Church regarding the electoral manipulation that occurred in their respective countries. This stance can manifest as a resistant stance, a call for peace stance, or a decision not to take any stance at all.Footnote 7
I gathered information about the Catholic churches’ stances from their pastoral letters, episcopal conference’s statements, and news agencies. See Table A1 in the Appendix for further elaboration on the Catholic churches under investigation in this article and their stances. Furthermore, I adopt Nancy Bermeo’s definition of strategic electoral manipulation as a range of tactics used to influence elections in favor of incumbents (Bermeo Reference Bermeo2016, 13). The data of alleged electoral manipulation come from the National Elections across Democracy and Autocracy (NELDA) data setFootnote 8 (Hyde and Marinov Reference Hyde and Marinov2012) and different media sources.
Dependent variable
The dependent variable is the national Catholic Church’s stance toward presidential electoral manipulation. I employ two forms of dependent variables. The first form is a binary variable (DV1) whereby the Catholic Church takes 1 if it resisted the electoral manipulation and 0 if not. The second form is a nominal categorical variable (DV2) that includes the three following stances:
-
- The stance of the Catholic Church is coded as “resist” when the Church heavily criticized the elections and human rights violations, organized protests, refused to attend the presidential inauguration, or called for new elections.
-
- The stance of the Catholic Church is coded as “Peace” when the church condemned violence, called for dialogue, prayers, and peace, and engaged in mediation efforts.
-
- The stance of the Catholic Church is coded as “No Stance” when there is no evidence that the Catholic Church took any stance.
Independent variables
The first key independent variable is the Catholic population size. The data come from ARDA’s national profile dataset (Johnson and Grim Reference Johnson and Grim2022). The variable is a proportion that ranges from 0 (low) to 1 (high). The second key independent variable is government favoritism toward Catholicism. The data are based on Government Preference Religion 2.0 (Brown Reference Brown2020) and are recoded with a range from 0 (low) to 1 (high). This variable includes five state-religion policy components (official status, religious education, financial support, regulatory burdens, and free exercise).
The third key independent variable is the historical pro-democratizing role of the Catholic Church in the Third Wave. The Church takes the value 1 if it had a pro-democratizing role in the Third Wave and 0 if not. The data are taken from Toft, Philpott, and Shah (Reference Toft, Philpott and Samuel Shah2011).
Control variables
Electoral violence is included, as it is more likely that the Catholic Church will call for peace when high levels of violence occurred prior to, during, or after the elections. The data come from the V-Dem dataset (Coppedge et al. Reference Coppedge, John Gerring, Lindberg, Jan Teorell and Bernhard2023; Pemstein et al. Reference Pemstein, Marquardt, Eitan Tzelgov, Juraj Medzihorsky and Miri2023), specifically from “Election other electoral violence,” which focuses on the violence related to the conduct of the election and the campaigns, but not to the government and its agents. The democracy score of the country is also controlled. The data come from Freedom House (Freedom House 2022). Whether the political ruler is Catholic or not is also controlled. The data are gathered from different media sources and from the Religious Characteristics of States Data Project (Brown Reference Brown2021).
It is also important to account for the poverty and inequality level of the country. As examined by Olson (Reference Olson2000) at both clerical and neighborhood levels, churches are expected to be more politically active in communities with high levels of poverty and inequality. The data are the Gini index (lagged one year) (World Bank 2023).Footnote 9 Furthermore, in the post-Third Wave period, many regimes tried to bypass term limits (Lotfy Reference Lotfy2022), sparking strong opposition from national Catholic churches, especially dominant ones, who largely opposed such amendments (Lotfy Reference Lotfy2024). Hence, cumulative autocratic practices, such as amending term limits, are controlled for, as they may incentivize Catholic churches to adopt a more resistant stance. The variable takes 1 if the term limits got amended, whether immediately before elections or not, and 0 if otherwise. The data are collected by the author from different media sources and term limit amendment literature (e.g., Lotfy Reference Lotfy2022).
Descriptive statistics are presented in Table 1. In the following section, I test my theory using two types of statistical analyses. The first is logistic regression, which is appropriate for the binary form of the dependent variable. The second is multinomial regression, used to compare the categories of the dependent variable.
Table 1. Descriptive statistics
Results
To begin, I test the three hypotheses in relation to the binary form of the dependent variable—resisting electoral manipulation—as shown in Table 2. Four logistic regression models were constructed. The first model tests the main explanatory variables (hypotheses 1(A), 2, & 3(A)). The second model controls additional variables, including amended term limits, electoral violence, democracy score, being a Catholic ruler, and Gini score. The third model tests hypothesis 1(B) and includes the interaction term between the “Catholic population size” and “government favoritism toward Catholicism” variables. The fourth model tests hypothesis 3(B) and includes the interaction term between the “catholic population size” and “historical pro-democratizing role” variables. Models 3 and 4 control for the “amended term limits” variable, as it is the only control variable that has a significant effect.Footnote 10
Table 2. Logit models for resisting electoral manipulation
Standard errors are in parentheses
***p < .01, **p < .05, *p < .1
The results of Models (1 & 4) suggest a negative relationship between government favoritism toward Catholicism and the national Catholic churches’ inclination to resist electoral manipulation. It is revealed that as government favoritism toward Catholicism increases, the national Catholic Church is less likely to resist manipulation. This result confirms hypotheses (1(A) & 2). Moreover, Models 1, 2, & 3 confirm hypothesis 3(A), showing that the national Catholic Church, with a historical pro-democratizing role in the Third Wave, is more likely to resist electoral manipulation than those that did not have a historical pro-democratizing role.
The third model confirms hypothesis 1(B), which indicates that when the government’s favoritism toward Catholicism is not high, the Catholic Church is more likely to resist electoral manipulation, even if it has a considerable share in the market. Figure 1 shows the substantive significance of the interaction term between the “catholic population size” and “government favoritism toward Catholicism” variables in Model 3.
Figure 1. The substantive significance of the interaction term in Model 3, 90% Cis.
The plot, using 90% confidence intervals (CIs), generally shows that with increasing governmental favoritism toward Catholicism (along the x-axis), the effect of the coefficient of the Catholic population size on resisting electoral manipulation decreases (along the y-axis). This effect becomes significant at lower levels of government favoritism (less than 0.42 score). The plot clearly shows that the effect of the Catholic population size on the Catholic Church’s decision to resist electoral manipulation is significantly positive if government favoritism toward Catholicism is low. Meanwhile, if the government favoritism is too high, the Catholic population size has no significant effect on resisting electoral manipulation.
The fourth model confirms hypothesis 3(B), which predicts that Catholic churches, with a historical pro-democratizing role in the Third Wave and a low religious market share, will be more likely to resist electoral manipulation than those with a historical pro-democratizing role and a high religious market share. Figure 2 shows the significance of the interaction term between the “catholic population size” and “historical pro-democratizing role” variables in Model 4.
Figure 2. The substantive significance of the interaction term in Model 4, 95% Cis.
The plot, using 95% CIs, clearly shows that with increasing Catholic denomination size (along the x-axis), the effect of the coefficient of having a historical pro-democratizing role on resisting electoral manipulation decreases (along the y-axis). This effect becomes significant when the Catholic population size is less than 50%. This means that, whether Catholicism is a majority or minority religion, the historical pro-democratizing role of the Catholic Church with a low market share has a positive effect on the Church’s decision to resist electoral manipulation. Meanwhile, the historical pro-democratizing role of the Catholic Church has no significant effect on the Church’s decision to resist electoral manipulation if it holds a high share of the religious market.
Additionally, Models 2, 3, & 4 show that national Catholic churches are more likely to resist if the country has witnessed term limit amendments aimed at prolonging the political leader’s stay in power prior to controversial elections. The other control variables in the second model had no significant effect. Additional control variables such as government regulation of religion,Footnote 11 the predominant religions competing with Catholicism in each country (whether Protestantism or Islam),Footnote 12 and the religious affiliation of political leaders (whether Protestant or Muslim)Footnote 13 are also accounted for; see Table A2 in the Appendix. They had no significant effect.
For further testing of hypotheses (1(A), 2, & 3(A)), as well as for robustness checks, I employed multinomial logistic regression, with “No stance” serving as the base outcome, as shown in Tables A3-A5 in the Appendix. For a better understanding of the stance categories, Figure 3 shows the significant and non-significant relations among the categories.
Figure 3. The (non) significant relations among the categories of Catholic churches’ stances, 95% Cis.
Note: The lines indicate non-significant relations. ATL = Amended Term Limits; GFC = Government Favoritism towards Catholicism; CPS = Catholic Population Size; HPD = Historical Pro-Democratizing Role
Figure 3 confirms hypotheses (1(A) & 2), as it reveals that a standard deviation increase in government favoritism toward Catholicism increases the odds of calling for peace compared to resisting electoral manipulation and taking no stance. However, it shows that government favoritism toward Catholicism does not distinguish between resisting electoral manipulation and taking no stances. Furthermore, the results confirm hypothesis 3(A) by revealing that Catholic churches with historical pro-democratizing roles have a higher impact on the odds ratio of resisting electoral manipulation relative to taking no stances and calling for peace. However, the historical pro-democratizing role of Catholic churches in the Third Wave does not distinguish between taking no stances and calling for peace.
Figure 3 also shows that term limit amendments have a greater impact on the odds ratio of resisting electoral manipulation compared to calling for peace. However, the term limit amendment does not distinguish between taking no stances and calling for peace. The term limit amendment also does not distinguish between resisting electoral manipulation and taking no stances (p-value = 0.065, see Table A3 in the Appendix). Figure 3 also shows that the Catholic population size does not distinguish between the different stances.
I further check for discrete changes by calculating the marginal effect of each main explanatory variable while holding other variables at their means. Figure 4 shows amended term limits increase the likelihood of Catholic churches resisting electoral manipulation by about 0.45, compared to respected term limits. However, they also show a decrease of 0.3 in the predicted probability of adopting a call for peace stance and a decrease of 0.14 in the probability of not taking any stances, compared to respected term limits. Additionally, a standard deviation increase in government favoritism toward Catholicism has a high effect on predicted probabilities, increasing the probability of adopting a call for peace stance by 0.27 and decreasing the probability of resisting electoral manipulation by about 0.23.
Figure 4. Marginal effects of main variables on outcome probability.
Note: ATL = Amended Term Limits; GFC = Government Favoritism towards Catholicism; CPS = Catholic Population Size; HPD = Historical Pro-Democratizing Role
Moreover, national Catholic churches with a historical pro-democratizing role demonstrate a higher predicted probability of resisting electoral manipulation compared to those without such a historical role (approximately 0.52). Additionally, those with a historical pro-democratizing role exhibit predicted probabilities 0.28 and 0.24 lower for taking no stances and calling for peace, respectively, than their counterparts lacking such a historical role. Nevertheless, the Catholic population size has a small effect on predicted probabilities, decreasing the probability of taking no stances by 0.07 and increasing the probability of taking stances (resist or call for peace) by 0.03 and 0.04, respectively. See Table A6 in the Appendix for further details.
Discussion
This article has set out to examine the stances of Catholic churches toward presidential electoral manipulation in the post-Third Wave period. The findings reveal that low government favoritism motivates religious institutions to oppose incumbent regimes’ autocratization efforts through election manipulation. This holds true even when the religious institution holds a large share of the religious market. Meanwhile, being highly favored by the government secures, at the very least, a “call for peace” stance rather than resistance to electoral manipulation. This frames the autocratization process as a religiously friendly one.
This finding challenges Fox & Breslawski’s (Reference Fox and Breslawski2023) claim that state support for religion in Christian-majority countries reduces government legitimacy. While their study focused on citizens’ perceptions of state legitimacy, this research examines the impact of state support on religious institutions’ views of state legitimacy after manipulated elections. It emphasizes the need to consider the effects on institutions, not just individuals. Furthermore, this article contributes to the political economy of religion approach by suggesting that the political engagements of religious institutions extend beyond the confines of the religious market context. Historical context is vital for understanding religious institutions’ cost-benefit calculations regarding autocratic practices amid the current wave of backsliding. A stable commitment to democracy is pivotal to understanding the public roles of religious institutions in the post-Third Wave period. The findings reveal that such a stable commitment is stronger when the religious market is highly competitive.
Based on this, religious institutions’ self-interests are recognized as the main drivers behind resisting autocratization, maintaining peace, or remaining silent. Moreover, it is possible to simultaneously achieve both goals—preserving stability and opposing autocratization. Meanwhile, certain religious institutions may prefer the former to the latter. This preference stems from the fact that resisting autocratization doesn’t ensure success and may provoke retaliation from regimes with authoritarian intentions. As argued by Tomini, Gibril, and Bochev (Reference Tomini, Gibril and Bochev2023), there have been numerous resistance attempts by opposition actors, but only a few have been successful in halting autocratization.
The cost-benefit calculus of the religious institution takes into consideration the impact of its stance on its moral authority. When a religious institution is highly favored by the government and/or has no historical democratic burdens, it is less likely to engage in risky behavior. Instead, it may adopt a neutral position by calling for peace to avoid aligning itself with any particular side. This publicly displayed neutrality or peace advocacy may conceal underlying private resistance or complicity. Hence, this peace-prayer stance may undermine the moral authority of the religious institution, as it fails to support democracy and instead contributes to autocratic transitions or the deepening of authoritarianism.
Conclusion
In this article, I examine why Catholic churches adopted varying stances toward presidential electoral manipulation through the lens of religious economies. I argue that the responses of religious institutions to democratic backsliding are influenced by their cost-benefit assessments concerning freedom from regulation, competition, and historical commitments. Future research could test the hypotheses of this study regarding other religious groups, such as Protestants and Muslims.
Further research should explore internal divisions among religious actors within the same institution, as this article focuses only on national Catholic churches’ public actions, which have a significant political and social impact. Nevertheless, behind-the-scenes actions could also be considered. Future studies could address theories that better capture the complexity of religious institutions’ political behavior to interrogate and expose their private affairs. Qualitative methods, like interviews with key religious leaders, could support this line of research.
Beyond the political economy approach, other explanations, such as the internal organization of the Catholic Church, should also be examined. As Warner (Reference Warner2000, 8) notes, national Catholic churches differ in history, structure, and ideology, with varying doctrinal and pastoral emphases. These differences call for qualitative case studies. Future research could build on the comparative cases used in this study to explore how internal church organization shapes its stance on electoral manipulation and other autocratic practices. Future research could also explore how foreign religious engagements affect local religious leaders’ stances on domestic crises. For example, Henne (Reference Henne2023, 78 and 79) points out the role of the U.S. Department of State’s Office of Religion and Global Affairs in engaging with Nigerian religious leaders to prevent violence, support peace during Nigerian elections, and combat corruption.
In sum, this article furthers the debate on resisting backsliding in general and on the specific role of religious institutions in autocratization. The development of the analysis regarding religious institutions’ stances urges further studies to address the gaps suggested by this article. Future research can also investigate the choice between maintaining stability and fighting for democracy by civil society actors other than religious institutions.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1755048325100114
Acknowledgements
I would like to express my sincere gratitude to the leaders and discussants of the “Religion and Democracy: New Research Frontiers” Workshop at the 2024 APSA Virtual Meeting. Special thanks are due to Professors Paul Djupe, Ramazan Kiliniç, David Buckley, and Günes Tezcür for their invaluable feedback on the first draft of this paper. I am also grateful to Prof. Michael Driessen for his insightful comments during the 2024 Rome Summer Seminars on Religion and Global Politics and to Prof. Madalena Meyer Resende for her helpful feedback at IPSA’s 75th anniversary conference. Additionally, my heartfelt thanks go to Professors Mohamed Safy El-Din Kharboush and Mazen Hassan for supervising this work as part of my Ph.D. dissertation.
Competing interests
The author declares none.
Dina Osama Lotfy is an assistant professor in the Department of Political Science at the Faculty of Economics and Political Science, Cairo University, Egypt. Her research interests include autocratic transitions, religion and politics, and terrorism. Her publications have appeared in Democratization, and Government and Opposition.
Open access
