Abstract
In this study we conduct a least-likely case study in order to assess the analytical power of the ideational approach to populism. We do so by testing the direct and conditional effects of populist attitudes on vote choices in Argentina. We examine whether populist attitudes are associated with the Peronist vote, as more essentialist interpretations would lead us to expect, or, on the contrary, linked to vote for right parties, an expectation that is more consistent with thin-ideological approaches. Our data consists of an original online survey carried out in September 2020, a specific juncture at which the Peronist government had to deal with widespread popular discontent caused by intense economic crisis and the COVID-19 pandemic. Our findings reveal that populist attitudes are positively associated with voting for right parties and that the effects of such attitudes are conditioned by ideological preferences. These results underscore the explanatory power of ideational approaches to the electoral activation of populist attitudes.
Introduction
Populist discourse and ideas have played a crucial role in the formation and evolution of Argentine mass politics since at least the birth of Radicalism. By directly appealing to the people, rejecting pacts with entrenched Conservative elites, and advocating radical political change, Hipólito Yrigoyen started a political tradition aiming at popular hegemony that would characterize Argentine politics throughout the 20th century (Aboy Carlés, 2013). After the 1940s, the Peronist movement constituted the most effective and enduring embodiment of populist appeals in the Argentine party system. This association has no doubt been strengthened by the political economy use of the term populism to label specific combinations of public policies (Bonvecchi and Novaro, 2021; Gerchunoff et al., 2020; Grigera, 2017) that are furthermore prone to lead to macroeconomic imbalances. The fact that such economic policy combinations have often been implemented by Peronist governments reinforces the historically grounded associations between populism and Peronism. It is then not surprising that the mere mention of populism leads analysts of Argentine politics to focus their attention on Peronism (Germani, 1973; Di Tella, 1965).
The taken for granted associations between populism and Peronism become problematic, however, if we focus on the presence, uses, and effects of populist political ideas and attitudes in Argentina, therefore leaving aside the political economic connotations of this term. This is particularly the case if, following the ideational approach (Hawkins et al., 2018), we understand populism as a contingently articulated thin-centered ideology (Mudde, 2004) or discursive frame (Aslanidis, 2016; Bonikowski and Gidron, 2016a) emphasizing the Manichaean opposition between a virtuous people and a conspiring and self-serving elite. This discursive frame is not the stable feature of any particular political actor, but can be articulated by groups located at different, even opposing, locations of the political spectrum (Bonikowski and Gidron, 2016b).
It is building upon this rationale that we assume that the articulation of populist appeals will be possible both for Peronist and anti-Peronist forces, depending on the constellations of circumstances characterizing each historical conjuncture. In particular, and contrary to common belief, we expect anti-Peronist forces to articulate populist discursive elements and to electorally activate populist attitudes when they are in the opposition, provided that a situation favorable to this type of political strategy prevails (Hawkins et al., 2018). This may have been the sort of scenario confronted by the Argentine government in 2020. In March 2020, at the beginning of the COVID-19 pandemic, Alberto Fernández’s government enjoyed high approval rates (Goyburu, 2020). However, seven months later, Argentine citizens began to express their discontent with the long compulsory quarantine and its economic consequences. In the final months of 2020 there were demonstrations against the government of Alberto Fernández to reject the prolongation of mobility restrictions. These grievances were mixed with the discontent triggered by the bad economic situation and other specific policies implemented in this period like the plans for judicial reform and the taxes on foreign currency purchases.
The application of the ideational theory to the Argentine political situation leads to the main expectations that inform our study, which are that, under the presidency of Alberto Fernández, populist attitudes may have had a positive impact on vote intentions for opposition candidates and parties and that such effects will be positively conditioned by the ideological (conservative) preferences of Argentine voters. To our knowledge, there are no previous analyses on the voting effects of populist attitudes in Argentina. However, the importance of falsifying these expectations goes beyond the historical interest of the Argentine case. Interpretations underlining the structural, permanent, and path-dependent associations between populism and a specific political movement (Peronism) would lead us to expect a negative or (at the most) null association between populist attitudes and voting for anti-Peronist parties. By contrast, finding a strong and positive association between these two social phenomena would be hard to explain without resorting to a theory emphasizing the thin-centered ideological and/or rhetoric and discursive character of populist ideas. We understand that some of the political features of Argentina make this country approach the characteristics of a ‘least-likely’ case for the ideational theory (Gerring, 2007) and an adequate context for testing ‘the Sinatra inference’ (Levy, 2002) (‘if it can make it here it can make it anywhere’). Least-likely cases have been defined as instances in which presumed causes are weakly present and counteracting factors are strong (Schimmelfennig, 2014: 105). We assume that the very strong historical association between populist discursive appeals and Peronism is a powerful counteracting factor weakening the association of populist attitudes and anti-Peronist vote. Furthermore, the very recent experience of a center-right government and the absence of a coordinated and coherent anti-Peronist opposition in the aftermath of President Macri’s electoral defeat must have limited the resonance and credibility (as discussed in Bonikowski and Gidron, 2016b) of center-right populist appeals. Given the presence of strong counteracting factors and the plausibly limited resonance of center-right populist appeals, we would expect Argentina to show a comparatively low probability of confirming the theory we are putting to test, thus approaching, in an intuitive and non-formalized way, the characteristics of least-likely case study (Rohlfing, 2012: 84).
In the first section of this paper we examine the historical roots of populism in Argentine politics. In the second section we present the ideational theoretical framework and the main expectations that guide this research. We devote the following section to presenting our data and methods and to discussing our empirical findings. The results of our unconditional models show that populist attitudes increase the probability of voting for center-right and right parties over the Peronist party. And the results from our conditional models show that this relationship is conditioned by the position of voters on the ideological dimension. In our conclusion we explore the theoretical implications of these findings, which, in our view, are far more consistent with the ideational approach than with conventional frameworks emphasizing the stable associations between populism and Peronism.
The historical roots of Argentinian populism
Populism has been considered a classical feature of the Argentinian mass-politics. The presidencies of Hipólito Yrigoyen (1916–1922 and 1928–1930), the leader of Unión Cívica Radical (UCR), and Juan Domingo Perón (1946–1955) have been often labeled as populist and interpreted as such within the frame of the modernization theory (Di Tella, 1965; Germani, 1973; Portantiero, 1977), according to which rapid industrialization and urbanization processes made migrant workers available for the disruptive mobilization efforts led by new political entrepreneurs.
The success of Yrigoyenist radicalism in 1916 represented the conquest of political rights for urban middle classes who challenged the oligarchic regime by demanding expanded democratic participation. The democratic transformations led by Yrigoyen also involved a process of homogenization and nationalization of the political community. With a discourse emphasizing ‘national unity,’ effectively embodied in the national executive power, Yrigoyen presented himself as the recipient of a plebiscitary mandate from the people (Aboy Carlés, 2013).
Three decades later, the Peronist movement embodied in an even stronger manner some of the classical elements of populist movements, such as the anti-elitist appeal to the people, the proposal of systemic change, and the popular devotion towards its charismatic leader, Juan Domingo Perón, who, as Labor Secretary of the 1940s’ military government, led an ambitious program of social, economic, and legal changes aimed at strengthening labor unions and improving the situation of the working classes (Collier and Collier, 1991; Ostiguy, 1997). In addition to launching legal and economic changes, Peronism included new cultural and symbolic appeals to the lower sectors of society (Ostiguy, 1997), articulating in a new way Argentine national-popular cultural elements (Portantiero and De Ipola, 1981) and giving rise to powerful popular identities that were infused with class and anti-elite content (Ostiguy, 1997, 2009). These discursive characteristics went hand in hand with organizational structures characterized by strong mass linkages and low routinization (Levitsky, 2003: 23). As shown by Levitsky (2003), these structures provided the Peronist party with a very high level of strategic flexibility and allowed it to embrace different policy agendas.
Despite the sharp changes experienced by Peronism over the last few decades, scholars have continued to find populist features in the presidencies of Carlos Menem, Nestor Kirchner, and Cristina Fernández (Muñoz and Retamozo, 2008; Rodríguez, 2014). In the case of President Menem, the image of the redeeming leader was a central device structuring a populist representative link with a population that was terrified by hyperinflationary chaos (Rodríguez, 2014). As for Menem’s economic reforms, they were not based on a symbiotic articulation of populist appeals and neoliberal programmatic orientations. Menem’s agenda was rather based on the conviction that the catastrophic economic situation demanded extraordinary measures and a radical departure from classical Peronist policies (Palermo and Novaro, 1996: 147; Palermo and Torre, 1992). In fact, as shown by Levitsky (2003: 148–155), most national and local Peronist leaders did not share neoliberal ideas, and the party members and voters who accepted the neoliberal program did so based on pragmatic considerations (Levitsky, 2003: 155), not on the articulation of populist appeals and a neoliberal agenda. As for the presidencies of Néstor Kirchner and Cristina Fernández, it has been emphasized that the performative and constitutive power of Néstor and Cristina Kirchner’s discourse aimed at social inclusion (Muñoz and Retamozo, 2008). The persistence and programmatic versatility of Peronist populist appeals have stimulated very rich theoretical debates among post-structuralist and post-Laclauian scholars on the contents, articulation, and limits of populist discourses and identities (Aboy Carlés et al., 2013; Retamozo, 2017).
In sharp contrast to Peronism, the Argentine liberal right never succeeded at building a transversal and stable political coalition including sizable segments of the popular sectors. Historically, the classical liberal right was not able to find popular leaders nor to build stable and popular partisan networks. However, from 2008–2009 onwards, Mauricio Macri (President from 2015 to 2019) and his Party PRO (Propuesta Republicana) were able to transcend the electoral limits of the classical Argentinian right. Leaning on emotional appeals to the people which could be traced back to his experience as president of the popular Boca Juniors football club, the multimillionaire Macri was able to build an effective electoral campaign based on an image of joy, empathy, consensus, and personal closeness. Furthermore, the political and discursive strategy of PRO underwent a deep transformation after the 2008 agrarian crisis, when the Argentine society split into two opposing blocks (Casullo, 2019). From that moment on, his discourse became openly oppositional and antagonistic. Since 2009, Macri’s political rhetoric offered an epic/moralizing alternative: the project to eliminate Kirchnerist populism from the Argentine political system (Casullo, 2019: 171). However, despite the presence of some people-centered and Manichaean elements, President Macri’s discourse never included anti-elitist elements and thus failed to have a fully populist character. In comparative terms, in the years 2018–2019, the PREPPS Survey (Wiesehomeier et al., 2021) ranked Peronist forces as more people-centered, more Manichaean, and more anti-elitist than President Macri and his party. Differences were particularly accentuated in the dimensions of anti-elitism (with values of 2.98 for Macri and 3.27 for his party, in contrast to the scores of 9.85 for the PJ (Partido Justicialista) and 16.69 for the also Peronist coalition FPV (Frente para la Victorial) – the higher the score in this 20-point indicator, the higher the anti-elitism) and Manichaeism (scores of 9.01 for Macri and 8.15 for his party, in contrast to 9.85 for the PJ and 5.18 for the FPV – lower scores in this 20-value indicator reveal more Manichaeism).
The articulation of populist elements by the Argentine right experienced a sharp rise during the year 2020, under Alberto Fernández’s presidency. It is quite likely that this transformation was favored by the combination of an intense economic crisis and the COVID-19 pandemic. From May 2020 onwards, in a bottom-up mobilization process, mass protests against mobility restrictions and other governmental policies extended all over the country. Demonstrators named their protests banderazos, thus making the national flag (bandera) the rallying cry for their mobilizations and merging populist and nationalist claims, although not in the ethnically exclusionary way analyzed by Jenne et al. (2021) for North American and European cases. Some, but not all, center-right opposition leaders endorsed the anti-quarantine demonstrations (Goyburu, 2020), sometimes performing clearly populist speech acts. Among the most prominent was Patricia Bullrich, president of former President Macri’s party, PRO, who tweeted her support for mass mobilizations ‘in the fight against privilege and the K oligarchy.’ 1 Bullrich called ‘to march in peace. . .for the anguished older people, for the feeling of injustice of those who are on the front line against COVID, and for the indignation provoked by the appropriation of the vaccine by the K oligarchy.’ 2 Other legislators of the center-right coalition Cambiemos endorsed the mobilizations and criticized Alberto Fernández’s presidency using the term infectadura (a combination of the words infect and dictatorship). The term had been coined by some Argentine intellectuals, who in an open letter had referred to the ‘illegal use of sanitary terror as a tool to subject the population.’ 3 The words used by libertarian leaders who had also participated in the banderazo demonstrations were even harsher. In a political advertisement of the Libertarian Party, the political outsiders José Luis Espert and Javier Milei referred to politicians as ‘sociopaths,’ ‘thieves,’ and ‘monsters’. 4
Former President Macri’s declarations were more measured. In 2020 he accused the national government of infringing the national constitution (Macri, 2020), criticized corporations, uncompetitive businessmen, and politicians who upheld an inefficient status quo, 5 and expressed his joy to see public demonstrations against government abuses. 6 But overall his speeches avoided anti-elitist themes, defended pluralism, and eluded tragic Manichaean overtones. Even more moderate was the Chief of Government of the city of Buenos Aires, Horacio Rodríguez Larreta, also a prominent leader in Macri’s party, who refused to participate in popular mobilizations. The different positions adopted by leaders of the Argentine right and center-right were consistent with the findings of Bonikowski and Gidron (2016b) regarding the importance of the outsider status for the adoption of a populist discourse. They also reflected the powerful but uncoordinated and somewhat inconsistent character of the center-right populist appeals in the 2020 political scenario.
Theory: populist attitudes at the individual level
Although different theoretical approaches to populism have been proposed by the literature, there is a growing consensus on the strong hermeneutic and heuristic power of the ideational approach (Hawkins et al., 2018). According to this theoretical framework, populism must be understood as a set of ideas that posit a morally Manichaean struggle between the unequivocal will of the people and a conspiring and self-serving elite (Hawkins et al., 2018). ‘The people’ is defined here as a homogeneous and virtuous entity, whereas the elite is understood as the group that threatens the well-being of the sovereign people (Mudde, 2004). According to this approach, populism must be understood as a multi-dimensional concept consisting of essential components or subdimensions (Akkerman et al., 2014; Castanho Silva et al., 2020; Schulz et al., 2018; Van Hauwaert et al., 2020; Wuttke et al., 2020). These main components or subdimensions identified by the literature are people-centrism, anti-elitism, and Manichaeism. The first one refers to the people as a virtuous entity that shares key characteristics and has a clearly defined popular will. The second is related to the hostility towards elites, conceived as a minority that controls key decisions and conspires against the interests of the people (Mudde, 2004). Finally, the third component referred to by the literature is Manichaeism, which portrays political divisions in a highly charged moral vocabulary, pitting the good people against the evil elites.
From a different perspective, Aslanidis (2016) has proposed an understanding of populism as an anti-elite discursive frame appealing to the sovereign people. Aslanidis (2016) has emphasized the situational, flexible, and strategic character of populist discourses, and rejected the use of the terms ideology and thin-centered ideology to characterize it . Despite these important theoretical differences, as Bonikowski and Gidron (2016a) point out, ‘ideological’ and ‘discursive’ approaches are not mutually exclusive. They both emphasize the role of the same ideational elements, and they are compatible with an empirical and fine-grained analysis of populism at the supply- and demand-side levels. Also, in both approaches, populist ideas can be associated with different political and ideological alternatives, depending on contingent constellations of factors.
At the level of individuals, populist attitudes function as a latent attitude or disposition that reflects whether citizens share anti-elite, people-centered, and Manichaean views of politics. In order to have political effects, these latent attitudes must be made salient and activated by linguistic and emotional cues articulated by political activists and leaders in an appropriate context (Hawkins et al., 2018). For example, the political activation of populist attitudes can be favored by both long-term historical legacies and conjunctural combinations of factors (economic crises, crises of representation, corruption, etc.) (Bornschier, 2018; Hawkins et al., 2018). Additionally, populist attitudes are unlikely to matter if populist ideas do not connect with grievances and issue positions that are important to voters. It is because of these unsatisfied popular demands that the populist discursive frame can be persuasive and politically effective (Andreadis et al., 2018; Hawkins et al., 2018; Marcos-Marne, 2020). However, neither the magnitude nor the substantive political effects of populist attitudes can be taken for granted. For instance, statistical analyses on the Bolivian case (Andreadis et al., 2018: 259, 260) have revealed that voters with populist attitudes were more likely to support Evo Morales in 2008. In 2016, however, populist attitudes were not statistically associated, directly or indirectly, with support for Morales’s party.
The possibility that populist ideas are articulated by very different political actors depending on historical conjunctures sheds light on the contingent character of populist appeals in Argentina. Peronism has been traditionally understood as the quintessential populist movement as a result of its socioeconomic policies, its ideological characteristics, and the nature of its links to Argentine citizens (Germani, 1973; Ostiguy, 1997, 2009; Di Tella, 1965). However, if the ideational approach is right, nothing prevents non-Peronist actors from the possibility of articulating populist discourses and activating latent populist attitudes scattered in specific segments of the political spectrum. In fact, as we point out in our historical section, this may have been the case in the year 2020. It is on the basis of these transformations that we want to put to the test the expectation that populist attitudes will have a positive impact on the vote intention of right and center-right candidates and parties.
If we assume, along the lines of ideational theory, that populist ideas tend to be associated with different substantive and ideological frames, partly as a result of the deliberate and strategic action of political leaders, then the voting effects of populist ideas should be conditional on, or enhanced by, other political preferences and orientations (Andreadis et al., 2018; Marcos-Marne, 2020). In this respect, different analyses have revealed the importance of both socio-economic and socio-cultural divisions in explaining the divides between Peronists and anti-Peronists in general and between Kirchnerists and anti-Kirchnerists in particular (Alessandro, 2009; Ostiguy, 1997, 2009; Sendra, 2021). As shown by Luna (2014) for the cases of Chile and Uruguay, under conditions of segmented representation, programmatic linkages are in fact compatible with the existence of clientelistic linkages targeting specific social segments. It is also the case that orientations in substantive dimensions tend to be associated with positions in a more encompassing, left–right, ideological super dimension (Sendra, 2021) and that, according to previous research, left–right orientations may specifically condition the policy expectations of the voters of conservative parties (Calvo and Murillo, 2013). Given the substantive importance of this ideological superdimension, and since this survey was conducted at a time when the main opposition force to the Peronist government was the conservative Juntos por el Cambio, led by the former President Mauricio Macri, in our analyses we put to the test the expectation that the effects of populist attitudes on vote choices will depend on the ideological orientations of Argentine voters. In particular, our statistical analyses intend to assess the possibility that the more rightist the placement of voters in the left–right dimension, the stronger the effects of populist attitudes on the support for Juntos por el Cambio.
Data and analysis
Our data comes from an original survey that was implemented online by the company Netquest during September 2020. To avoid bias in the distribution of age, gender, education, and territorial distribution of respondents, the 1003 respondents were selected using quota sampling (see details in Online Appendix – Table OA.1).
Our dependent variable is vote intention, based on the answer to the following question: ‘If a presidential election were held tomorrow, which party would you vote for?’ We grouped the different existing parties in the two coalitions that competed in the last presidential elections of October 2019 (see Table 1). PJ (Partido Justicialista) and FR (Frente Renovador) were recoded as ‘Frente de Todos’; PRO (Propuestas Republicana), UCR (Unión Cívica Radical), and CC (Coalición Cívica) were aggregated in the category ‘Juntos por el Cambio.’ Partido Libertario 7 and Partido Valores para mi País 8 were aggregated in the category ‘Minor right parties.’ Lastly, those who answered they would vote null or would not vote were grouped in the category ‘Null.’
Vote intention (if elections were held tomorrow), Argentina, September 2020.
Source: Our own elaboration.
Partido Libertario (24) + Partido Valores para mi País (15).
Would vote null (156) + Would not vote (67).
In order to measure our key independent variable, populist attitudes, we rely on Akkerman et al.’s (2014) battery of six items, the most common standard in the measurement of populist attitudes. To this battery we have added two additional items proposed by Van Hauwaert et al. (2020) (see Table 2). The initial Akkerman et al. (2014) battery captures three evaluative reactions to the set of constitutive elements of populism: people-centrism (questions 1, 2, and 3); anti-elitism (questions 4 and 5); and Manichaeism (question 6). Recent analyses have shown the validity and informative capacity of this battery of items (Castanho Silva et al., 2020; Van Hauwaert et al., 2020; Wuttke et al., 2020). The average value for these six items in Argentina equals 4, slightly above the averages reported in Andreadis et al. (2018) for Chile of 3.9, Spain 3.9, Greece 3.7, and Bolivia 3.7 (see Supplemental Figure 1). The two new items strengthen the original anti-elitism items from the Akkerman et al. (2014) scale by asking about the negative and ‘antagonistic’ relationship between the people and the elites (Van Hauwaert et al., 2020). Respondents were asked to rate their agreement with each item on a Likert scale ranging from 1 (very much agree) to 5 (very much disagree). For the purpose of these analyses we inverted these scales, higher values now indicating maximum agreement and, therefore, more populist attitudes.
Battery items tapping political attitudes.
In order to conduct our empirical analysis, we have followed the practice of using a single higher-order construct. For the aggregation of items into one scale, we have followed the method developed by Wuttke et al. (2020) and treated the three subdimensions – people-centrism, anti-elitism, and Manichaeism – as non-substitutable. Using the Goertz correction, the eight original items were first standardized to a mean of zero and a standard deviation of one. Afterwards, the minimum value of each respondent on the three subdimensions was taken to determine the final populist attitudes score with a range from −4.56 to 0.61 (mean = −1.11; standard deviation = 1.05). This procedure ensures that all three theoretical subdimensions are treated as necessary non-compensatory conditions, thus avoiding the risk that high values in one of these dimensions compensate for low values in other dimensions.
Besides our key independent variable (populist attitudes), we included a number of control variables. Drawing on the existing literature we controlled for the general assessment of the economic situation (1 – Very good to 5 – Very bad), self-placement in the left–right scale (ranging from 1 to 10), and the standard set of socio-demographic variables, namely age (continuous), gender (male as reference), education (basic and university, taking high school as reference), and social class (ranging from 1 – upper class to 5 – lower class). We further included two indicators to control for individuals’ preferences regarding economic-distributive and socio-cultural issues 9 (for descriptive statistics see Table OA.2). We assume that public evaluations related to COVID-19 might have affected the development of populist attitudes, and it is even possible that COVID-19-related perceptions have conditioned the effects of populist attitudes on vote choices. Unfortunately, however, our survey lacks indicators to control for these associations. We also cannot control, as per Carty and Torcal (2020), for the role of attitudes towards political leaders in predicting vote choices. To conduct our analysis, we model respondents’ vote intention using multinomial logistic regression, taking the Frente de Todos, the ruling party, as the baseline category.
Results
Results in Table 3 10 show that, after controlling for other potential explanatory variables, populist attitudes have a positive and statistically significant effect on the probability of voting for Juntos por el Cambio over the Peronist Frente de Todos. Populist attitudes are also a good predictor of vote intention for minor right parties (Partido Libertario and Partido Valores para mi País) as well as, to a lesser degree, for the radical-left party FIT (Frente de Izquierda y de los Trabajadores), over the Frente de Todos. These results confirm our expectation that populist attitudes have been electorally activated in Argentine party competition, even if we cannot pin down the specific moment and the causal paths (supply- or demand-side driven) that led to this outcome.
Unconditional model of vote intention, Argentina, September 2020.
Standard errors in parentheses.
p < 0.05. **p < 0.01. ***p < 0.001.
As we expected, left–right ideology exerts a statistically significant effect (at the .01 level) on vote intention for Juntos por el Cambio (positive) and FIT (negative). Additionally, the respondents’ socio-economic preferences are highly associated with vote intention in the expected direction. Those more oriented to the pro-market pole of this dimension are more likely to vote for Juntos por el Cambio, Partido Libertario and Partido Valores para mi País. Socio-cultural preferences are a statistically significant predictor only in the case of the minor right parties, with conservative individuals being more prone to support them. This makes sense if we take into consideration that the party ‘Valores para mi País’ may fit the single-issue party label, since it was created to oppose abortion and defend traditional values. Finally, the economic evaluations variable also exerts a statistically significant influence (at the .01 level) on support for Juntos por el Cambio and for the minor right parties. Those who think that the general economic situation is bad are more prone to support these political options.
Figure 1 shows the size of the statistical effects of populist attitudes on voting (Frente de Todos being the base category). The voting probability for Juntos por el Cambio reaches its maximum level, more than 30%, at the highest populist attitudes level (all other variables being set at the average values). By contrast, the predicted support for Frente de Todos reaches its lowest level (about 10%) at the highest level of populist attitudes.
Predicted probabilities of vote intention, by populist attitudes (95% confidence interval).
A limitation of our current indicators of populist attitudes lies in the fact that they over-specify anti-elitism by focusing on political elites and are therefore likely to miss the hostility towards other kinds of social elites (Jungkunz et al., 2021). Jungkunz et al.’s analyses show that these limits could have important effects when populist parties are in power. Despite this caveat, we understand that finding positive effects for our current indicator of populist attitudes on voting for center-right, anti-Peronist forces has theoretical and substantive importance, and should not be strictly interpreted as a failure of our indicators of populism. Such a finding is consistent with the theory that populist rhetoric can be used by all kinds of forces. In particular, this finding endorses the expectation that citizens of all thick-ideological persuasions can develop or be cued into a populist understanding of national political contexts, even if the substantive character of such populist frames may vary depending on the political orientations of parties and citizens. Still, in order to provide a robustness check for our analyses, we have rerun them after building a new two-dimensional non-compensatory indicator that excludes the items focusing on anti-elitist attitudes (see Table 2). Despite its limitations, this indicator can reduce the probability that populists with Peronist leanings will display low scores due to their lack of anti-political elites attitudes. This new indicator is very strongly associated with our three-dimensional indicator (Pearson correlation of +0.94). Results from our new analysis (see Table OA.5) reveal that this indicator works in the same direction as the previous one, and that it increases the probability of voting for center-right and right parties over the ruling coalition Frente de Todos.
Finally, as indicated above, we have also tested whether ideological preferences condition the effects of populist attitudes on vote choices. Following Andreadis et al. (2018), we have tested the conditioning role of ideology by creating an interacting term based on populist attitudes and left–right positions. We report the model results in Table OA.6. Following Kam and Franzese (2009), instead of focusing on the significance levels of the interaction term, which can be substantively meaningless for this analysis, we use graphs to assess and interpret conditional effects. Figure 2 shows that populism has positive and statistically significant effects on the voting probabilities for Juntos por el Cambio and minor right parties over the ruling party just for voters whose left–right position is located at six or higher in the left–right dimension (see also Table OA.7 for the predicted marginal effects on each point of the left–right scale). On the other hand, there are no conditional effects of populist attitudes for the voters of FIT and Frente de Todos. The effect of populist attitudes in this model is therefore conditional on specific ideological positions on the left–right scale. We interpret these results as a corroboration of our expectation regarding the conditional effects of populist ideas on vote choices, and, in particular, of the fact that the effects of such attitudes on voting for right parties depend on the rightist ideological locations of voters.
Predicted average marginal effects of holding populist attitudes, conditional on left–right self-placement (95% confidence interval).
Conclusion
The importance and continuity of populism in Argentine politics has been asserted from multiple perspectives. These perspectives have emphasized the connections of populist appeals with the rise of the two main Argentine political movements in the 20th century: first Yrigoyenist Radicalism and then Peronism. In particular, since the birth of Peronism, Argentine populism has become inextricably linked to this political actor. However, in contrast to a static interpretation of Argentine politics, the application of the ideational approach has led us to propose a more flexible and historically contingent interpretation of the uses of populism in that country, one in which populist attitudes can also be activated in an anti-Peronist direction when the political situation makes that outcome possible. In our view, that was the case with the Argentine political conjuncture the first year of Alberto Fernández’s presidency, in a situation characterized by intense economic crisis, the COVID-19 pandemic, and anti-government mobilizations. The anti-Peronist activation of populist attitudes was also rendered possible by the legacies, and also quite possibly the learning effects, of the intense and Manichaean political polarization experienced under Cristina Fernández’s presidency. Furthermore, the fact that Alberto Fernández chose a more moderate and inclusive political strategy from the very moment he started building up his electoral coalition (even incorporating former anti-Cristina Fernández leaders) had to attenuate the populist components of the President’s discourse, and thus hinder the pro-Peronist activation of populist attitudes.
Given these characteristics of the Argentine context, and based on the ideational approach to populism, we formulated the expectation that an electoral activation of populist ideas and attitudes by right and center-right parties was possible. The likelihood of this association was constrained, however, by the presence of strong counteracting factors limiting the resonance of populist center-right appeals. Key among them were the long-term strong associations between populism and Peronism and also the position and characteristics of the current Argentine center-right forces, marked by their recent governmental responsibilities and their lack of political coordination. These counteracting factors reduced the probability of confirming our expectations and thus made Argentina approach the characteristics of least-likely case.
The results from our statistical analyses were consistent with these expectations. They showed that populist attitudes increase the probability of voting for center-right parties over the ruling Peronist coalition, even when we control for other crucial independent variables (such as economic judgments and ideological and programmatic preferences). We had similar results when, in order to avoid the anti-political elites bias of our populism indicator, we used a limited, two-dimensional predictor that combined just people-centrism and Manichaeism.
Furthermore, also along the lines defined by the ideational theory, we found that when we used interactive terms the effects of populist ideas on right and center-right vote choices became statistically significant just for voters located at the right side of the ideological spectrum, a result that confirmed that the political effects of populist attitudes depended on substantive orientations and preferences.
Overall, our findings endorse the ideational approach by showing that the electoral activation of populist attitudes is historically contingent and open to forces located at very different positions in the ideological spectrum, including parties for which such activation would have been initially deemed as highly unlikely. Our work has not analyzed, however, the causal processes leading to the emergence of populism-based electoral linkages in the Argentine case, nor has it tested the validity of our findings in other comparable cases. Achieving these goals would greatly benefit from different research strategies. The causal processes leading to the activation of populist attitudes could be addressed by using both qualitative analyses of discursive frames and quantitative analyses on panel data. Testing the validity of our inferences in other cases could be addressed by developing both structured comparisons and comparative statistical analyses. All these strategies should prove important for the objective of refining and expanding our knowledge of populism.
Supplemental Material
sj-do-2-ips-10.1177_01925121211044924 – Supplemental material for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina
Supplemental material, sj-do-2-ips-10.1177_01925121211044924 for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina by Mariana Sendra and Iván Llamazares in International Political Science Review
Supplemental Material
sj-docx-4-ips-10.1177_01925121211044924 – Supplemental material for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina
Supplemental material, sj-docx-4-ips-10.1177_01925121211044924 for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina by Mariana Sendra and Iván Llamazares in International Political Science Review
Supplemental Material
sj-dta-1-ips-10.1177_01925121211044924 – Supplemental material for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina
Supplemental material, sj-dta-1-ips-10.1177_01925121211044924 for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina by Mariana Sendra and Iván Llamazares in International Political Science Review
Supplemental Material
sj-pdf-5-ips-10.1177_01925121211044924 – Supplemental material for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina
Supplemental material, sj-pdf-5-ips-10.1177_01925121211044924 for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina by Mariana Sendra and Iván Llamazares in International Political Science Review
Supplemental Material
sj-R-3-ips-10.1177_01925121211044924 – Supplemental material for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina
Supplemental material, sj-R-3-ips-10.1177_01925121211044924 for The reversal of populism. Assessing the explanatory power of the ideational approach in contemporary Argentina by Mariana Sendra and Iván Llamazares in International Political Science Review
Footnotes
Acknowledgements
The authors thank José Ignacio Antón, Ryan Carlin, Hugo Marcos Marné, Juan Antonio Rodríguez Zepeda, and Mariano Torcal for help and ideas on previous versions of this article. They also thank anonymous reviewers of this journal for their helpful comments and suggestions.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was made possible by the financial support provided by the Spanish Ministry of Innovation and Universities (Research Grant PGC2018-098207-B-I00).
Supplemental material
Supplemental material for this article is available online.
Notes
Author biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
