Introduction
Throughout his papacy, Pope Francis has been very active in expressing his opinions on several social and political issues, such as immigration, climate change, and social justice. His sermons, discourses, apostolic exhortations, and encyclicals have often included critiques of social and economic injustice and climate change denial. Therefore, his role goes beyond the religious and moral sphere, and his messages have been addressed to a larger community than the Catholic one. Although the content of his appeals does not radically diverge from his predecessors' social teachings (Federico, Reference Federico2021), given the emphasis on certain topics some pundits have depicted Pope Francis as liberal and leftist (Neumayr, Reference Neumayr2017). Moreover, in the media representation he gained the status of modernizer (de Rooij, Reference De Rooij2019). Overall, his figure appears influential in the public debate. However, when studying the impact of Pope Francis' positions on public opinion, extant research has focused almost exclusively on environmental issues, especially climate change (Li et al., Reference Li, Hilgard, Scheufele, Winneg and Jamieson2016). In other words, so far there is little evidence toward the existence of a Francis effect in the public opinion on other topics of interest. Moreover, extant research has mostly focused on the US context (Landrum and Vasquez, Reference Landrum and Vasquez2020).
By acknowledging that the Pope's message could have an influence on a broader spectrum of topics in several contexts, we aim at testing whether Pope Francis has played a role in affecting support on measures for poverty alleviation among the Italian public. We focus on the Pope's proposal of introducing measures of poverty alleviation in a letter sent to popular movements during Easter 2020, amid the COVID-19 emergency, when the debate over poverty and fragility of marginal groups was particularly vivid in Italian politics and society (Vicentini and Galanti, Reference Vicentini and Galanti2022). In the Italian context Pope Francis is still highly regarded as both religious leader and political actor, potentially capable of influencing the public debate (Genovese, Reference Genovese2015; Garelli, Reference Garelli2020). By means of a survey experiment carried out in Italy that manipulates the source (actor) who proposed the introduction of a universal basic income for all workers (Pope Francis versus some people), we test whether activating the papal cue affected public support for the anti-poverty measure. Moreover, by employing simple linear regression models we analyze whether the Francis effect varied depending on trust in the Pope himself, religiosity, and political ideology. By using original survey data (ResPOnsE COVID-19 project) collected from April 17 to May 15, 2020, during the first wave of the COVID-19 pandemic in Italy, we contribute to expanding the knowledge on the Francis effect outside the US context and on a topic different from climate change.
Our contribution speaks to the theoretical perspective of heuristic processing in opinion formation (Mondak, Reference Mondak1993), according to which external cues—such as the source of a statement—guide people's opinions toward debated social issues. Here, we analyze whether the reference to a prominent figure such as Pope Francis influences people's support to anti-poverty measures. Considering the role of religious authorities in the public sphere and their influential role in times of crisis and international instability, we aim at further assessing the existence of a Francis effect and its societal consequences (Genovese, Reference Genovese2015; Marchetti et al., Reference Marchetti, Mazzoni, Stanziano and Pagiotti2019).
The experimental results indicate the existence of an overall Francis effect among the Italian public. In addition, they also revealed that having trust in the figure of the Pope and leftist political orientation actually increased this effect, while religious denomination and adherence to institutional religion did not lead to a substantial change. Results provide evidence toward the political influence of the Pope on public opinion. Also, our analysis suggests that the Pope is more influential among some social groups who do not constitute the main target of the Catholic Church. In the concluding section, we will offer some reflections regarding our findings, in light of the debate on secularization.
Background: religious authorities in the public sphere
Several studies have provided empirical evidence on the ongoing process of secularization in Western societies, by showing a decline in several indicators of individual religiosity (Chaves, Reference Chaves1994; Bruce, Reference Bruce2002; Voas, Reference Voas2009; Voas and Chaves, Reference Voas and Chaves2016; Molteni and Biolcati, Reference Molteni and Biolcati2023). It has been argued that the process of secularization consists, among other things, of a loss of religious authority and a decline of the social significance of religion in a society (Wilson, Reference Wilson1966; Chaves, Reference Chaves1994). Nonetheless, since the 1990s some scholars have started to highlight the contemporary public and political role religion still has (Casanova, Reference Casanova1994; Habermas, Reference Habermas2006). Following Habermas' argument, religious language and actors continue to bring a remarkable contribution to the public sphere, influencing people's opinions, attitudes, and civic and political behaviors (Portier, Reference Portier2011). Within such a framework, the role of religion in affecting the public sphere was labeled as “public religion” (Casanova, Reference Casanova1994). Regardless of the argumentation one supports in contemporary societies, churches and their authorities still play a role that is not confined to the religious sphere.
Focusing on the Catholic Church, on account of its universalistic claims and being one of the big five global confessions (together with Islam, Judaism, Buddhism, and Hinduism), it can potentially communicate to people from every corner of the world and influence large masses. The Church can exert influence even beyond religious and moral issues, as it engages with multiple socio-economic themes, that are relevant for people irrespective of their religious beliefs (Marchetti et al., Reference Marchetti, Mazzoni, Stanziano and Pagiotti2019; Marchetti and Pagiotti, Reference Marchetti and Pagiotti2023). By behaving as an actor in civil society with a language that can reach both adherents and non-adherents, the Church often sustains, spreads, and argues in favor of social issues that are also closely related to its mission (Marchetti et al., Reference Marchetti, Mazzoni, Stanziano and Pagiotti2019; Marchetti and Pagiotti, Reference Marchetti and Pagiotti2023).
As the Catholic Church can have a potential impact on the public sphere, by behaving as a political actor addressing various issues connected to its agenda, a crucial role is played by its main representative, the Pope. Indeed, he possesses critical characteristics to assert power over political decisions, such as the capacity to mobilize resources and masses and the legitimacy that can lead to obedience and conformity (Genovese, Reference Genovese2015). Voicing the worries and passions of large groups of people when there is an open political debate, the Pope aims to exercise his influence by gaining support, strengthening his reputation among non-Catholics, and building public opinion (Genovese, Reference Genovese2015).
Moreover, several contextual (social, economic, and political) factors play a role in creating the conditions in the public arena for the Church to fit in (e.g., processes and threats of globalization, global economic instability in light of the financial crisis, periods of political instability). In this regard, Genovese's (Reference Genovese2015) analysis of the content of papal encyclicals shows that the Church vocalizes its opinion and concern on topics of social and political relevance to the general public especially when there is a political void to fulfil, a relevant crisis or a situation of international instability. Especially in those cases, the Church can undertake the role of political actor and decide to engage in the public sphere in troubled times as topics become more controversial and people are divided on them, sometimes because the political actors fail to address those issues (such as immigration, poverty, social justice, labor market, environment, and terrorism: Marchetti et al., Reference Marchetti, Mazzoni, Stanziano and Pagiotti2019). Furthermore, when the moral component of an issue prevails, religious authorities are perceived as more credible (Tavits, Reference Tavits2007; Genovese, Reference Genovese2015).
Furthermore, the current pontiff—Pope Francis—has a remarkable status as modernizer on account of his communication strategy that is innovative for the Church (Gandolfi, Reference Gandolfi2016; de Rooij, Reference De Rooij2019). He takes advantage of social media like Twitter while using a more informal approach than his predecessors to express the concerns of most people (not only Catholics) mainly about the environment and social justice (de Rooji, Reference De Rooij2019), although maintaining a top-down approach in the use of those new media (Gandolfi, Reference Gandolfi2016). These characteristics contribute to making him a deeply charismatic figure, capable of influencing the public sphere on debated topics with his legitimacy and communication power (de Rooji, Reference De Rooij2019). As a result of the proliferation of media sources and the speeding up of information, the Church's potential audience has become increasingly broader (Gili and Nardella, Reference Gili and Nardella2019). Not only does news travel faster, but now they are more able to cross borders and sway people from the most different backgrounds and social categories. In light of the mediatization of religion (Hoover, Reference Hoover2006; Hjarvard, Reference Hjarvard2011), the Church aims to take advantage of the various media sources in its public discourse, carving its space in the shared arena of debate and acting as an agent of civil society.
Although the active role of Pope Francis in the public sphere has been widely highlighted, extant research on the Francis effect on public opinion has mostly focused on attitudes toward climate change in the US context. The influence of Pope Francis on public opinion is supposed to go beyond the climate issue, given the variety of stances addressed in his discourse, from modernization to social justice (De Rooij, Reference De Rooij2019). As argued before, this is especially true in periods characterized by enduring crises (Genovese, Reference Genovese2015). In this respect, in light of the COVID-19 pandemic and its economic consequences, the issue of providing a feasible strategy for economic relief for low-income social groups has emerged. The debate remains open on whether to introduce special policies to address people's basic needs, targeting people with lower incomes who tend to have more unstable, underpaid jobs. By recognizing the importance of the issue related to both the labor market and poverty, the economic and the social sphere, Pope Francis seized the opportunity and during the Sunday of Easter (April 12, 2020) expressed his opinion via a letter to civic movements and organizations (Schneck, Reference Schneck2020). The addressees were defined by the Pope himself as those left behind, those “excluded by the benefits of globalization” but facing the worst downsides of it, in other words an ensemble of minorities, among which poor people, migrants, women and youth (Pope Francis, 2020; Schneck, Reference Schneck2020). This letter appeared on many prominent Italian newspapers and information channels (like Corriere della Sera, La Repubblica, RaiNews, Avvenire) and international ones (like CBC, CNBC, Forbes), highlighting its broad resonance. Using his authority and legitimacy, Pope Francis expressed his support toward measures of poverty alleviation with the aim to address a problem appealing to everyone troubled by the precarious situation, while also dealing with social injustice, as follows:
“Many of you live from day to day, without any type of legal guarantee to protect you. […] you who are informal, working on your own or in the grassroots economy, you have no steady income to get you through this hard time. This may be the time to consider a universal basic wage which would acknowledge and dignify the noble, essential tasks you carry out […]–” (Pope Francis, 2020)
The Pope's words contained in this letter could be relevant in shaping the public sphere, considering not only the aforementioned influential role of the source, but also the strategic timing (at the height of the religious celebrations), the highly debated topic (under the circumstances of a global crisis such as the COVID-19 pandemic) and the vast audience addressed. Within such a framework, it becomes consequential to understand whether and to what extent Pope Francis influenced public opinion on poverty alleviation measures.
Context of the study
Whereas the influence of religious authorities on public opinion remains an important topic to be investigated worldwide, it assumes particular relevance when looking at the Italian context. First of all, Italy hosts the Vatican State, the seat of the Catholic Church in the world, and the large majority of the Italian population identifies as Catholic (in 2019, 72%, according to European Social Survey Round 9 Data, 2018). Hence, the Pope still retains a considerable potential impact on the public both as a religious leader and a political actor in the public discourse (Genovese, Reference Genovese2015; Garelli, Reference Garelli2020). Second, overall Italians have highly positive attitudes toward Pope Francis, among the most favorable in the world (91% of favorable views, compared to 84% in Europe, 78% in the US, and 60% around the world, Pew Research Center, 2014), creating advantageous preconditions for the Pope to influence people's opinions in situations where politics fails. Lastly, the Italian political scenario has been dramatically shaken by the outbreak of COVID-19 and, at the same time, the economy has been greatly impacted by the restrictions, causing to the population a deep sense of destabilization and uncertainty (Vicentini and Galanti, Reference Vicentini and Galanti2022). Because of the unprecedented circumstances created by COVID-19 and the extreme difficulty of Italian politics to cater urgent needs of its impoverishing population, Pope Francis could potentially fulfil this vacant role in the social sphere. In addition, the COVID-19 crisis has made re-emerge the debate on income support and welfare measures to tackle poverty (Natili et al., Reference Natili, Jessoula and Caizzi2021). Around the world, the two main strategies used to address the issue are the universal basic income (a certain amount of money given to every adult in a country) and the guaranteed minimum income (stipulated minimum income granted to every worker) (Bryan, Reference Bryan2021). However, in Italy only in 2018 an income support measure has been developed firstly as a Reddito di inclusione (Inclusion Income) and then widened as Reddito di Cittadinanza (Citizenship Income), which gives monetary help to people after passing a means test (Jessoula and Natili, Reference Jessoula and Natili2020; Maino and De Tommaso, Reference Maino and De Tommaso2022). As a matter of fact, this was the only anti-poverty welfare measure (apart from a special COVID-19 emergency income); also, there is no minimum income threshold for workers and the public debate is still heated as the Italian population is very divided on poverty alleviation policies (Jessoula and Natili, Reference Jessoula and Natili2020).
Considering this peculiar context, Pope Francis decided to publicly express his opinion in the letter sent to popular movements in 2020 during Easter, at the peak of the very first COVID-19 wave and of the Catholic holy days. Mentioning in generic terms a “universal basic wage” Pope Francis aimed at raising attention to strengthening welfare measures for poverty alleviation during the time of crisis. While there are different approaches to address the issue of poverty alleviation, following the Pope's words our analysis focuses on the basic income.
Francis effect on public opinion: literature review and theoretical expectations
This study aims to test the effect of Pope Francis' endorsement of poverty alleviation measures on the public opinion, and whether such effect varies depending on various individual characteristics. According to the perspective of heuristics processing, people often employ external cues when expressing their opinions toward a certain issue. In this regard, the source cues—such as the individual exposure to the opinion of a political leader—are expected to strengthen the opinion holding and guide the opinion direction (Mondak, Reference Mondak1993).
So far, almost all the existing studies have analyzed Pope Francis' influence on attitudes toward climate change considering the publication of the encyclical Laudato Sì in 2015 and focused mostly on the US context (for a review, see Landrum and Vasquez, Reference Landrum and Vasquez2020). By analyzing both cross-sectional survey data collecting information on the exposure to the papal encyclical (Li et al., Reference Li, Hilgard, Scheufele, Winneg and Jamieson2016), panel data collected before and after the publication of the encyclical and the papal visit in the US (Maibach et al., Reference Maibach, Leiserowitz, Roser-Renouf, Myers, Rosenthal and Feinberg2015; Landrum et al., Reference Landrum, Lull, Akin, Hasell and Jamieson2017), and experimental survey data where a random group of respondents was exposed to information about the Pope's messages (Schuldt et al., Reference Schuldt, Pearson, Romero-Canyas and Larson-Konar2017; Myrick and Evans Comfort, Reference Myrick and Evans Comfort2019, Reference Myrick and Evans Comfort2020; Buckley, Reference Buckley2022), those studies provided some evidence toward a Francis effect on the overall US population. Between March and October 2015 (before and after the release of the encyclical), Maibach et al. (Reference Maibach, Leiserowitz, Roser-Renouf, Myers, Rosenthal and Feinberg2015) showed a slight increase in the climate change concern and belief that climate change is happening, while they reported no shift in climate change attribution. On a nationally representative sample of the US population, Schuldt et al.s' (Reference Schuldt, Pearson, Romero-Canyas and Larson-Konar2017) survey experiment showed that the exposure to Pope Francis' message was—overall—associated with an increase in the perception of climate change as a moral issue (also in Buckley, Reference Buckley2022) and in personal responsibility for contributing to it, but not with the personal responsibility for helping to mitigate climate change. Indeed, not all studies agree on a significant presence of a Francis effect. For instance, Li et al. (Reference Li, Hilgard, Scheufele, Winneg and Jamieson2016) found no direct effect between encyclical awareness and agreement with the Pope's positions on climate change, while Buckley (Reference Buckley2022) reported—overall—a non-significant impact of the exposure to Pope Francis' message on opinions on the causes of climate change and the government action for mitigation. Also, the only study carried out in the European context showed that, when primed with a moral message from Pope Francis' encyclical, on average Italian people did not substantially change their willingness to support a petition for a meat tax (Mrchkovska et al., Reference Mrchkovska, Dolšak and Prakash2023).
Nonetheless, given several characteristics of the Italian context (the strong religious tradition, the high percentage of Catholics, and the overall positive evaluation of the figure of Pope Francis among the public), we should expect that the papal cue can positively affect the overall support for the proposed measure of poverty alleviation. Thus, our first hypothesis stands as follows:
– H1: Overall, Pope Francis' endorsement increases support for anti-poverty measures.
All in all, previous studies detected some heterogeneity in the Francis effect on public opinion, depending on individual attitudes toward the figure of Pope Francis himself and political views (Landrum and Vasquez, Reference Landrum and Vasquez2020).
Previous research has shown that the figure of the pontiff goes beyond the mere religious role-model and has overcome traditional barriers (Bohigues and Rivas, Reference Bohigues and Rivas2021). Indeed, the same authors show that in Latin America Pope Francis' evaluation is higher among Catholics, but the level of religiosity proves not to be associated with the Pope's evaluation. Therefore, we cannot consider support for Pope Francis as a mere indicator of religiosity.
Notably, again according to the heuristic processing perspective, individuals tend to extend their evaluations of a political figure to the policies he/she supports (Mondak, Reference Mondak1993). In other words, the impact of a source cue is stronger when the individual evaluation of a political figure is more intense. In line with this expectation and the extant empirical research, those individuals who hold positive attitudes toward the Pope should be more likely to employ the papal cue when expressing their opinion toward the poverty alleviation measure. In other words, it is likely that people trusting the pontiff will be more subjected to the Francis effect, increasing their support toward the measures the Pope supports. In this regard, Buckley (Reference Buckley2022) showed that among people sustaining Pope Francis the exposure to his message on climate change substantially increased the attribution of human activity as the main cause of climate change, the moral salience of the issue and the support toward a government action for its mitigation. Therefore, we can hypothesize that:
– H2: Pope Francis' endorsement increases support for anti-poverty measures especially among people trusting him.
Moreover, one could expect that the Francis effect on public opinion can depend on religious denomination and religious involvement. As Pope Francis is the leader of the Catholic Church, Catholic people are supposed to be more inclined to modify their opinions to follow the Pope's message when compared to non-Catholic ones. Extant studies from the US context provided mixed evidence in this regard. Li et al. (Reference Li, Hilgard, Scheufele, Winneg and Jamieson2016) do not report substantially different relationships between encyclical awareness and respectively climate change concern and perceived credibility of Pope Francis on the climate issue when comparing Catholics and non-Catholics. Instead, Maibach et al. (Reference Maibach, Leiserowitz, Roser-Renouf, Myers, Rosenthal and Feinberg2015) show that the change in attitudes toward climate change after the release of the encyclical was more remarkable among Catholics than other confessions and that Catholics were more inclined to self-report being influenced by Pope Francis' stances. Moreover, people highly involved in institutional religion are supposed to be more prone to follow religious teachings and, accordingly, the Pope's teachings. Nonetheless, previous research on the Francis effect on public opinion has generally neglected the moderating role of religious involvement. A rare exception is represented by Mrchkovska et al.s' (Reference Mrchkovska, Dolšak and Prakash2023) work, which surprisingly shows that in Italy Pope Francis' priming message had a negative effect on the support for a meat tax among highly religious people.
Previous literature showed mixed evidence toward the moderating role of religiosity, but it is reasonable to think that the Francis effect should be higher among those Catholics who are more involved in institutional religion. Since attendance to religious services “exposes the laity to the messages from the clergy” (Vezzoni and Biolcati-Rinaldi, Reference Vezzoni and Biolcati-Rinaldi2015: 104), practicing Catholics should be more inclined to follow the Pope's indications and teachings, Although previous studies have shown that sometimes bishops and local clergy deviate from Pope's public messages (Calfano, Reference Calfano2009; Holman and Shockley, Reference Holman and Shockley2017), individuals who invest time and resources in practicing religion are more likely to be influenced by religious messages (Iyer, Reference Iyer2016; Curtis and Olson, Reference Curtis and Olson2019). Thus, our third hypothesis stands as follows:
– H3: Pope Francis' endorsement increases support for anti-poverty measures especially among practicing Catholics.
Finally, Pope Francis' influence on public opinion is supposed to be moderated by political ideology. In the media representation, Pope Francis has often been depicted as leftist because of his several warnings about inequality, social injustice, and environmental degradation (Neumayr, Reference Neumayr2017). This is reflected also among public opinion, despite his conservative positions on cultural issues, rather in line with his predecessors. In the US leftist people are indeed more likely to positively evaluate the Pope than rightist ones, even though among Catholics the role of ideology in predicting the Pope's evaluation tends to disappear (Federico, Reference Federico2021). Other survey data shows, however, that Pope Francis is more appreciated among Catholics who identify as Democrats than among Catholics who identify as Republicans (Pew Research Center, 2021). Instead, in Latin America left-right ideology was found not as a significant predictor of Pope's evaluation. For what concerns political ideology as a moderator of Pope Francis' effect on public opinion, extant research provides us heterogeneous results. Li et al. (Reference Li, Hilgard, Scheufele, Winneg and Jamieson2016) show that in the US democrats grant more credibility to Pope Francis on the topic of climate change when compared to republicans, who tended to reject the encyclical message even when they identify as Catholics. Schuldt et al. (Reference Schuldt, Pearson, Romero-Canyas and Larson-Konar2017) show that, on the one hand, brief exposure to Pope Francis increased the perception of climate change as a moral issue especially among Republicans, and on the other hand enhanced the feeling of personal responsibility for contributing to climate change and its mitigation only among Democrats.
Since on the one side the themes of social justice and reduction of inequality traditionally pertain to the left, and on the other side the Pope has been often depicted as leftist in the public debate, we should expect that the Pope's voice on poverty alleviation measures is heard more loudly by leftist people. Following some existing evidence on the Francis effect on climate change, we hypothesize that:
– H4: Pope Francis' endorsement increases support for anti-poverty measures especially among leftist people.
Data, methods and measures
Data
We test our hypotheses by employing survey data coming from the first wave of the ResPOnsE COVID-19 project carried out by the Sps Trend Lab of the University of Milan (Vezzoni et al., Reference Vezzoni, Chiesi, Biolcati, Dotti-Sani, Guglielmi, Ladini, Maggini, Maraffi, Molteni, Pedrazzani, Moroni, Piacentini, Sarti and Segatti2022; for further information on the ResPOnsE COVID-19 project see Vezzoni et al., Reference Vezzoni, Ladini, Molteni, Dotti Sani, Biolcati, Chiesi, Guglielmi, Maraffi, Pedrazzani and Segatti2020). The survey aimed to monitor the dynamics of Italian public opinion during the various phases of the COVID-19 pandemic. Besides the core questionnaire—fielded between April 6 and July 8, 2020—covering behavior compliance, well-being, risk perception, attitudes toward several issues and socio-demographics, some questions were asked only during specific periods of the fieldwork. This was the case of the survey experiment and the questions on religiosity. Survey data analyzed in this paper were indeed collected between April 17 and May 15, 2020 (n = 4,601), during the first wave of the COVID-19 pandemic. The sample was drawn from an opt-in panel of an Italian survey research institute (Swg S.P.A.) and reproduces Italian population distributions for sex, geographical area of residence, and age class (see the socio-demographic distribution of the sample in Table S1 in the Online Supplementary Material).
The experimental design
The aim of the study is to test whether Pope Francis' endorsement on the introduction of measures of poverty alleviation influences citizens' support for that measure, by employing a randomized experimental design. We focus on those measures proposed by Pope Francis in the letter sent to popular movements and organizations during Easter 2020, namely, a universal basic wage for all working people. Although the non-probabilistic nature of the sample is a potential threat to the external validity of the findings, the use of a randomized experimental design has the potential to test a specific mechanism on public opinion on anti-poverty measures. While non-experimental studies have some limits in investigating the effect of the exposure to Pope Francis' public statements and their influence on the public sphere because of the interplay of a variety of external factors, an experimental approach allows isolating the effect of papal cues on public opinion. Here, as a result of the randomization of the experimental condition we can assume that differences in the distributions of the dependent variable depending on the experimental condition are explained by the experimental condition itself. Also, previous research showed large correspondence in experimental results between probabilistic and non-probabilistic samples, (Coppock et al., Reference Coppock, Leeper and Mullinix2018).
The sample was randomly divided using a split-half design into one treatment group (N = 2,254) and one control group (N = 2,347). Respondents expressed their support for the introduction of a universal basic wage for all workers, including temporary and self-employed ones, either proposed by Pope Francis (treatment group) or by some people (control group). The wording of the two experimental conditions stood as follows:
− Control group: “Some people proposed the introduction of a universal basic income for every worker including temporary and self-employed workers. Are you in favor or against?”
− Treatment group: “Pope Francis proposed the introduction of a universal basic income for every worker including temporary and self-employed workers. Are you in favor or against?”
The only difference between the experimental conditions consists in the source (actor(s)) proposing the introduction of the measure.
Respondents were asked to give an answer to the question on a four-point Likert scale, where 1 corresponds to the least supportive position (strongly against) and 4 the most supportive one (strongly in favor), while 99 corresponds to the “don't know” option. Therefore, higher values of the dependent variable indicate higher support for the introduction of the anti-poverty measure.
Measures
As we aim to analyze whether Pope Francis‘ effect on support for anti-poverty measures varies depending on certain individual characteristics, we consider three main moderating variables measured in the survey.
First, we include as moderating variable the trust in Pope Francis, measured on a 0–10 scale where 0 corresponds to “None at all” and 10 to “A great deal” (mean = 6.59, S.D. = 3.05). Second, to simultaneously account for both religious denomination and involvement in institutional religion we employ a typology of individual religiosity as the combination of religious affiliation and attendance to religious services (Jagodzinski and Dobbelaere, Reference Jagodzinski, Dobbelaere, van Deth and Scarbrough1995; Ladini et al., Reference Ladini, Biolcati, Molteni, Pedrazzani and Vezzoni2021). For what concerns the religious denomination, we distinguish between Catholics (n = 2,922, 63.6% of the sample), and non-affiliated people (n = 1,202, 26.2% of the sample), by excluding from the analysis respondents belonging to other denominations as they are a very small and heterogeneous minority in the Italian context (in our sample, 6.0% of the respondents, n = 276) and respondents providing a don't know answer (n = 193, 4.2% of the sample). As a measure of institutional religiosity, we consider the attendance to religious services (see Nicolet and Tresch, Reference Nicolet and Tresch2009). Nonetheless, the survey was carried out during the first wave of the pandemic, when churches were closed, and mass celebration forbidden because of the lockdown. Therefore, the measure refers to the frequency of attendance to religious services via web, radio, and tv in the week before the interview (employed also in Molteni et al., Reference Molteni, Ladini, Biolcati, Chiesi, Dotti Sani, Guglielmi, Maraffi, Pedrazzani and Vezzoni2021). The original variable included the answer options “every day”, “more than once a week”, “once a week”, and “never”. Considering the precept of the Catholic Church of attending religious service every Sunday, we aimed at simply distinguishing between those adhering to institutional religion, thus attending at least once a week (n = 1,491, 32.4% of the sample), and those not adhering (n = 2,894, 63.0% of the sample), while excluding the missing cases (211 respondents, 4.6% of the sample). Thence, we construct a religious typology by combining the outcomes of the two variables, with the resulting three categories: “practicing Catholics” (n = 1,303) define themselves as Catholic and attended religious services at least once in the week before the interview; “nominal Catholics” (n = 1,512) define themselves as Catholic but did not attend religious services in the week before the interview; “non-religious” (n = 1,174) are those who declare not to belong to any religion. Finally, we consider as third moderator political ideology, measured in terms of left–right self-placement on a 0 (left) −10 (right) scale, with the “don't know” and “do not locate” options (mean = 4.56, S.D. = 2.93).
While the question on political ideology was asked before the survey experiment, religious affiliation, attendance, and trust in Pope Francis were measured after the survey experiment. Nonetheless, the distribution of the three variables does not differ across the two experimental conditions (see Table S2 in the Online Supplementary Material). In the concluding section, we will discuss the implications of the placement of the moderating variables in the analysis of this survey experiment.
Empirical strategy
To test our research hypotheses, we run linear regression models where the dependent variable is the individual support for the measure of poverty alleviation. Model 1 allows testing H1 by including as main independent variable the experimental condition (‘Pope Francis” versus “Others” priming). Each of the models 2, 3, and 4 include a single interaction term between the experimental condition and respectively trust in Pope Francis, religious typology, and left–right self-placement. Those models allow testing H2, H3, and H4.
Results
We first analyze whether the distribution of support for anti-poverty measure varies by experimental condition.
Table 1 shows that the majority of respondents were favorable toward the proposal of introducing a guaranteed basic wage for all workers, irrespective of the experimental condition. Moreover, it provides evidence toward H1: although the difference between the two distributions is not large, the percentage of total support for the measure of poverty alleviation is substantially higher among those respondents exposed to the papal endorsement (26.0% of respondents are totally favorable, versus 20.9% in the control group). While the percentages of respondents who are totally against the measure and do not answer are analogous between the two experimental groups, in the control group there is a slightly higher percentage of people declaring to be favorable and against it. All in all, the Pope's endorsement significantly enhances the support for the measure of poverty alleviation (χ2 p-value < 0.001).
Table 1. Percentage distribution of support for the anti-poverty measure by experimental condition (n = 4,601)
Pearson χ2(4) = 23.55; p-value = 0.000.
Regression analysis allows formal testing of whether the effect of the papal cues on support for anti-poverty measures is moderated by trust in Pope Francis, religiosity, and left–right position (in Figure S1 and S2 in the Online Supplementary Material we also reported the raw means of the dependent variable by experimental condition and each moderating variables). In regression analysis, we excluded those people who answered don't know to the item on support for anti-poverty measures (similar share of respondents across the two groups: 15.2% in the control group, and 15.9% in the treatment group). In the following analysis, the dependent variable is treated as cardinal.
In Table 2, Model 1 confirms the empirical evidence for H1 coming from Table 1. Being exposed to Pope Francis's endorsement led to a significantly higher support for the measure of poverty alleviation. The mean difference with people not exposed to the papal endorsement was equal to 0.09 (p-value < 0.01). The result is the same even after controlling for socio-demographics (Model 1 in Table A1 in the Appendix).
Table 2. Regression models with support for the anti-poverty measure (1–4 scale) as dependent variable
Note: Standard errors in parentheses. ***p < 0.01, **p < 0.05, *p < 0.1.
In Table 2, Model 2 allows testing whether such an effect increases as long as the trust in Pope Francis increases. The interaction term is positive and statistically significant, therefore results support H2: Pope Francis' endorsement increases the support for measures of poverty alleviation especially among people trusting him. Figure 1 (Panel A) allows for better visualizing the moderating effect of trust in Pope Francis, by reporting the predicted means of support for the measure of poverty alleviation by experimental condition and trust in the Pope. For people with high trust in the Pope, the level of support for the anti-poverty measure is substantially higher when the Pope endorsed the measure (estimated mean difference equal to 0.21 among people with the highest—equal to 10 on a 0–10 scale—trust in the Pope), while the papal endorsement had a backlash effect on support for the measure among people who do not trust him (estimated mean difference equal to −0.17 among people with the lowest—equal to 0 on a 0–10 scale—trust in the Pope). Moreover, the figure shows that higher trust in Pope Francis is associated with higher support for the measure of poverty alleviation in both the control and the treatment group. In the Appendix, the regression model with socio-demographic controls (Model 2 in Table A1) reports the same interaction coefficient of Model 2 in Table 2.
Model 3 allows testing whether the effect of the papal cue varies according to the individual religious typology. In the regression models, the coefficient of the interaction term between the experimental condition and the dummy “nominal Catholics” (with “practicing Catholics” as reference category) is negative (−0.11) but statistically significant only at the 0.10 level, while the coefficient of the interaction term between the experimental condition and the dummy “non-religious” is negative (−0.10) but not statistically significant. Figure 1 (Panel B) shows that among people not exposed to the papal endorsement, the mean support for the measures of poverty alleviation is similar across the different categories of religious typology. As suggested in the regression analysis and shown in Figure 1 (Panel B), the mean difference in support for the measure of poverty alleviation between the “Pope Francis” and the “some people” condition is slightly higher among practicing Catholics, but such difference is not statistically different—at the 0.05 level—than the ones related to nominal Catholics and non-religious, respectively. In other words, there is weak evidence to affirm that the effect of the papal cue on support for the measure is stronger among practicing Catholics than among nominal Catholics and non-religious people. Therefore, although the pattern is in line with our theoretical expectations, the findings offer little evidence for supporting H3. Moreover, when controlling for socio-demographics (see Model 3 in Table A1 in the Appendix) both the interaction terms between the experimental condition and the dummies of religious typology become non-statistically different from 0. To avoid idiosyncrasies related to the self-reported measure of church attendance referred to the pandemic period, characterized by a slight religious revival and impossibility of physically attending religious services (Molteni et al., Reference Molteni, Ladini, Biolcati, Chiesi, Dotti Sani, Guglielmi, Maraffi, Pedrazzani and Vezzoni2021), we estimated the same regression model in which the religious typology is constructed by employing a measure of church attendance referred to the pre-pandemic period (see Model 3bis in Table S3 in the Online Supplementary Material).Footnote 1 Nonetheless, the results are consistent with the ones presented in Model 3 in Table 2 (and in Model 3 in Table A1 in the Appendix).
Finally, Model 4 in Table 2 tests H4, namely, whether the effect of the Pope's endorsement is stronger among leftist people. The interaction term (equal to −0.02) is negative and statistically significant at the 0.05 level, namely, the Francis effect decreases as long as moving from the left to the right on the ideological scale. Similar to Figure 1, Figure 2 allows for a better interpretation of the moderating effect of the political ideology. Among right-wing people, the exposure to the Pope Francis' endorsement does not have any effect on support for the measure of poverty alleviation. Instead, among left-wing people the reference to Pope Francis substantially enhances the support for the measure (among people scoring 0 on the 0–10 left–right scale, estimated mean difference between the two experimental groups equal to 0.20 on a 1–4 scale). Therefore, empirical evidence supports H4, and is further reinforced by the regression model which controls for the main socio-demographic variables (Model 4 in Table A1 in the Appendix).Footnote 2 However, we should consider that the three moderating variables are not independent. Indeed, trust in Pope Francis is substantially higher among practicing Catholics (mean = 8.1) when compared to nominal Catholics (mean = 6.8), who show in turn higher trust in the Pope when compared with non-religious (mean = 5.3). Furthermore, trust in Pope Francis was shown to be substantially higher among leftist people (mean = 7.3 among people scoring 0–3 on the left–right scale) when compared to rightist ones (mean = 5.6 among people scoring 7–10, while the mean is equal to 7.0 among people scoring 4–6 on the left–right scale). To exclude the presence in the results of composition effects due to the interconnections between the three variables we have estimated in Table A2 in the Appendix the regression models (Model 2, 3, and 4) where controlling for the other two variables not considered as moderators and the other socio-demographics. As shown by the interaction terms, results are consistent with the ones shown in Table 2 (and in Figures 1 and 2), by offering empirical evidence toward H2 and H4 (interaction terms statistically different than 0 at the 0.05 level) and only weak evidence toward H3. We should also stress that these latter analyses were performed on samples substantially smaller when compared to the analyses shown in Table 2, because of the significant number of cases with a missing value for at least one of the independent variables.
Figure 2. Predicted means of support for the measure of poverty alleviation (1–4 scale) by experimental condition and left–right ideology (estimated by Model 4 in Table 2. N = 3,193).
Discussion and conclusions
Since he was elected Pope in 2013, Jorge Bergoglio's statements on several social and political issues have been pivotal in the public debate. Nonetheless, so far there is limited evidence of the impact that Pope Francis has had on public opinion, especially on issues not related to climate change. Our contribution has thus aimed to answer two main research questions: Is Pope Francis influential on public opinion, and to what extent? And which are the social groups whose opinions can be more affected by the Pope's? Focusing on the issue of social justice and the proposal for introducing poverty alleviation measures during the emergency period of the COVID-19 pandemic and implementing a priming survey experiment on a sample of Italian respondents, our study shows that the exposure to Pope Francis' endorsement of the anti-poverty measure slightly increased the overall support for it. Moreover, the results show that the exposure to the Pope's endorsement had a larger impact among people with high trust in the Pope and left-wing political orientation, while individual religiosity proved not to be a significant moderator. Applying an experimental design which allows isolating the effect of a source cue on public opinion, our results can be interpreted in causal terms. In other words, we can talk about the existence of a Francis effect in Italian public opinion, differentiated by some individual characteristics. Although the effects are not huge, our findings can be suggestively read in light of the debate on the secularization of Western societies. On the one hand, in line with Casanova's (Reference Casanova1994) argument, the high activity of Pope Francis in expressing his opinions about the social and political sphere and the consequential effect detected on public opinion confirms the persistence of public religion in contemporary societies. According to our findings, the Pope is still an influential actor who can impact the public. In other words, the Catholic Church is still a voice that can matter in the public debate, even beyond the strictly speaking religious sphere. On the other hand, in line with the secularization thesis, the Catholic Church and the Pope do not have a particularly strong influence on their adherents' opinions. Practicing catholic people, indeed, are not highly affected by the Pope's statement on an issue related to the Catholic social teachings, dealing with social justice and solidarity. All in all, the influence of the Pope on public opinion is mostly explained by his trustworthiness (since the Francis effect is substantially higher among people trusting him, even net of individual religiosity and political orientation) rather than his role as a religious authority.
Pundits have often stressed the media representation of Pope Francis as liberal and leftist, and survey data showed a higher preference toward his figure among left-wing people, at least in the US. Besides showing that this finding applies also to the Italian context, our study has further shown that the Pope's statement supporting the introduction of poverty alleviation measures affected mostly the leftist public opinion. According to such empirical evidence, the Pope was shown as an influential leader especially for the left-wing people. This is peculiar for a country in which the right-wing political elites are not in opposition with the Catholic Church and, in particular, Catholics have still a stronger preference for right-wing parties when compared to non-religious people (Barisione et al., Reference Barisione, Keeling, Ladini and Maraffi2023). We can therefore argue that the reception of Pope Francis' message should be better interpreted through a political rather than religious lens.
Furthermore, these results have some implications for the political communication strategy. As the Pope was shown to influence public opinion on issues related to left-wing stances—even beyond the climate change issue—left-wing political leaders could have benefits in exploiting the Pope's endorsement of policies they support in their political communication. In this regard, future research is invited to analyze to what extent the references to the Pope are present in the discourse of political leaders of different orientations, around which issues, and finally how are they effective in influencing public opinion.
Although our analysis contributes to expanding the knowledge of the Francis effect on public opinion, it does not come without limitations. First, we analyzed public support for poverty alleviation measures which has several traits in common with both guaranteed minimum income and universal basic wage, but it does not exactly coincide with any of those measures. For the sake of external validity, indeed, we focused on the measure endorsed by the Pope in the letter sent to popular movements (a universal basic wage for all workers) and analyzed people's support for that measure. Nonetheless, given the common leftist ideological root of anti-poverty measures, we guess that our conclusions could be extended to other anti-poverty measures. Our conclusions, however, cannot be generalized to public opinion support for policies framed into the rightist ideology, and less-ideological policies. Only future experimental research will allow comparing the Francis effect on support for rightist and leftist policies, as well as concerning different topics. Second, the survey did not include any manipulation check aimed at directly assessing individual-level attentiveness in answering the questionnaire (Kane and Barabas, Reference Kane and Barabas2019). Therefore, we cannot detect whether the respondents have paid sufficient attention to the source of the message (Pope Francis versus some people) and, accordingly, we can interpret the experimental effects only in terms of intent-to-treat effects. Thus, the reported estimates of the regression models should be intended as conservative. Third, some of the moderating variables (religious denomination, attendance to religious services, trust in Pope Francis) did not precede the survey experiment in the questionnaire. In this regard, previous research has outlined the issue of post-treatment bias in experimental research (Montgomery et al., Reference Montgomery, Nyhan and Torres2018), while other scholars argue that measuring moderators before the treatment variable can prime respondents and change the treatment effect (Klar et al., Reference Klar, Leeper and Robison2020). A recent study, however, shows little evidence toward the hypothesis of differences in treatment effects based on moderator placement (Albertson and Jessee, Reference Albertson and Jessee2023). Notwithstanding, our moderating variables measure rather stable constructs, thus we tend to minimize the presence of an ex-post rationalization in respondents' answers. Moreover, we have shown that the distribution of the moderating variables did not vary according to the experimental condition. Fourth, the survey experiment and its related hypotheses were not pre-registered. As the survey was carried out during the first phase of the COVID-19 pandemic, the research group who carried out the survey was strongly focused in rapidly designing instruments aimed at assessing people's reactions to the pandemic (Vezzoni et al., Reference Vezzoni, Ladini, Molteni, Dotti Sani, Biolcati, Chiesi, Guglielmi, Maraffi, Pedrazzani and Segatti2020), thus neglecting such an important practice. Future experiments aimed at focusing on the Francis effect on public opinion are invited to undertake the pre-registration procedure for the sake of transparency of research findings and a stronger connection between theoretical expectations and experimental findings.
Finally, we would like to highlight that our study has been carried out in a country, Italy, where the presence of the Pope in the public debate is particularly relevant. Therefore, studies aimed at applying similar research designs in other contexts are more than welcome to understand to what extent our conclusions can be generalized.
Notwithstanding some limitations, our results offer new evidence supporting the theoretical perspective of heuristic processing in opinion formation (Mondak, Reference Mondak1993) by showing that exposure to the opinion of a recognized leader impacts people's opinions. Indeed, we have shown that people can use external cues when expressing their opinion on a certain issue, by referring to the positions of those sources—here, Pope Francis—they approve.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1755048324000166
Data
Survey data come from the research project ResPOnsE COVID-19 (available at https://doi.org/10.13130/RD_UNIMI/IJDSVS).
The project was carried out and funded by Sps Trend Laboratory “Hans Schadee” at the Department of Social and Political Sciences, University of Milan (Italy). The research infrastructure was designed as part of the “Departments of Excellence 2018–2022” project promoted by the Italian Ministry of University and Research and is supported by funding from the Cariplo Foundation.
Syntax and dataset for the replication analysis can be found at https://doi.org/10.13130/RD_UNIMI/BM8WDU
Acknowledgments
The authors wish to thank the two anonymous reviewers and Giulia Dotti Sani for their careful reading of the paper and the insightful suggestions. They also wish to thank Ferruccio Biolcati and Francesco Molteni, who designed with Riccardo Ladini the module on religiosity of the ResPOnsE COVID-19 survey, and the Sps Trend laboratory of the University of Milan for allowing the inclusion of the experiment in the survey.
Financial support
None.
Competing interests
None.
Riccardo Ladini is an assistant professor of Political Sociology at the Department of Social and Political Sciences, University of Milan, where he is member of the Sps Trend laboratory. He is currently interested in studying the influence of religiosity on social and political attitudes and the public opinion on climate change.
Lucia Faggiana has obtained a Master degree in International Relations (curriculum International Cooperation and Human Rights) at the University of Milan. She collaborated with the Sps Trend laboratory through an internship, aimed at studying the influence of Pope Francis on Italian public opinion.
Appendix
Table A1 Regression models with support for the anti-poverty measure (1–4 scale) as dependent variable
Table A2. Regression models with support for the anti-poverty measure (1–4 scale) as dependent variable, controlling for the other moderating variables
Open access
