What is Liberalism? A Mixed-Method Study of Ideology and Representation in Latin American Party Systems ^*

O que é Liberalismo? Um Estudo de Método Misto sobre Ideologia e Representação nos Sistemas de Partidos Latino-Americanos.

Qu’est-ce que le Libéralisme ? Une Étude Mixte des Idéologies et de la Représentation dans les Systèmes de Partis d’Amérique Latine

¿Qué es Liberalismo? Un Estudio de Método Mixto sobre Ideología y Representación en los Sistemas de Partidos Latinoamericanos

Abstract

What is the concept of liberalism? Despite being a conceptually contested term, political parties still label themselves liberal and seek legitimacy by joining Liberal International (LI). In this paper, I adopt a mixed-methods strategy to assess what this ideology means in Latin America. First, I rely on economic and political theory to propose four potential components of liberalism: private property, liberal democracy, non-conformism, and social justice. Then, I search for these components in the declaration of principles of all the region’s LI members. Next, I assess liberals’ relative support for these components by comparing the attitudes of their elites and voters to those of conservatives and socialists in Paraguay, Honduras, and Nicaragua. This paper finds that liberal democracy is the only core component of liberalism in Latin America. Even though non-conformism and social justice are widely mentioned in political documents, their support among elites and voters is context-dependent. These results emphasize the contestability of liberalism while shedding light on what unites liberals in Latin America.

essentially contested terms; liberalism; party ideology; manifesto analysis; survey analysis; elite-masses congruence

Resumo

Qual é o conceito por trás do liberalismo? Apesar de ser um termo conceitualmente contestado, os partidos políticos ainda se rotulam como liberais e buscam legitimidade ao se juntarem à Internacional Liberal (IL). Neste artigo, adoto uma estratégia de método misto para avaliar o que essa ideologia significa na América Latina. Primeiro, baseio-me na teoria econômica e política para propor quatro componentes potenciais do liberalismo: propriedade privada, democracia liberal, não conformismo e justiça social. Em seguida, procuro por esses componentes na declaração de princípios de todos os membros da IL na região. Depois, avalio o apoio relativo dos liberais a esses componentes comparando as atitudes de suas elites e eleitores com as de conservadores e socialistas no Paraguai, Honduras e Nicarágua. Este artigo constata que a democracia liberal é o único componente central do liberalismo na América Latina. Embora o não conformismo e a justiça social sejam amplamente mencionados em documentos políticos, seu apoio entre elites e eleitores é dependente do contexto. Esses resultados enfatizam a contestabilidade do liberalismo, ao mesmo tempo que lançam luz sobre o que une os liberais na América Latina.

termos essencialmente contestados; liberalismo; ideologia partidária; análise de manifesto; análise de pesquisa; congruência elites-massas

Résumé

Quel est le concept derrière le libéralisme ? Bien que ce terme soit contesté sur le plan conceptuel, les partis politiques continuent de se qualifier de libéraux et cherchent légitimité en rejoignant l’Internationale Libérale (IL). Dans cet article, j’adopte une stratégie mixte pour évaluer ce que cette idéologie signifie en Amérique Latine. Tout d’abord, je m’appuie sur la théorie économique et politique pour proposer quatre composantes potentielles du libéralisme : la propriété privée, la démocratie libérale, le non-conformisme et la justice sociale. Ensuite, je recherche ces composantes dans la déclaration de principes de tous les membres latino-américains de l’IL. Ensuite, j’évalue le soutien relatif des libéraux à ces composantes en comparant les attitudes de leurs élites et de leurs électeurs à celles des conservateurs et des socialistes au Paraguay, au Honduras et au Nicaragua. Cet article constate que la démocratie libérale est la seule composante essentielle du libéralisme en Amérique Latine. Bien que le non-conformisme et la justice sociale soient largement mentionnés dans les documents politiques, leur soutien parmi les élites et les électeurs dépend du contexte. Ces résultats soulignent la contestabilité du libéralisme tout en mettant en lumière ce qui unit les libéraux en Amérique Latine.

termes essentiellement contestés; libéralisme; idéologie des partis; analyse des manifestes; analyse des enquêtes; congruence élites-masses

Resumen

¿Cuál es el concepto detrás del liberalismo? A pesar de ser un término conceptualmente en disputa, los partidos políticos aún se autodenominan liberales y buscan legitimidad al unirse a la Internacional Liberal (IL). En este documento, adopto una estrategia de métodos mixtos para evaluar lo que representa esta ideología en América Latina. En primer lugar, me baso en la teoría económica y política para proponer cuatro posibles componentes del liberalismo: propiedad privada, democracia liberal, inconformismo y justicia social. En seguida, investigo estos componentes en la declaración de principios de todos los miembros de LI en la región. Después, evalúo el apoyo relativo de los liberales a estos componentes comparando las actitudes de sus élites y los votantes con las de los conservadores y socialistas en Paraguay, Honduras y Nicaragua. Este artículo concluye que la democracia liberal es el único componente central del liberalismo en América Latina. Aunque el inconformismo y la justicia social se mencionan ampliamente en documentos políticos, su apoyo entre las élites y los votantes depende del contexto. Estos resultados hacen énfasis en la disputabilidad del liberalismo, al tiempo que arrojan luz sobre lo que une a los liberales en América Latina.

términos esencialmente controvertidos; liberalismo; ideología partidista; análisis de manifiestos; análisis de encuestas; congruencia élite-masas

Introduction

Although all of the participants shared the same basic values, they were by no means agreed on how to counter the attack on those values, or on the policies required to implement them. As a result, our sessions were marked by vigorous controversy (...). I particularly recall a discussion of this issue, in the middle of which Ludwig von Mises stood up, announced to the assembly ‘You’re all a bunch of socialists,’ and stomped out of the room, an assembly that contained not a single person who, by even the loosest standards, could be called a socialist.

Milton and Rose Friedman (1998:160-1) on the first meetings of the Mont Pelerin Society.

The Mont Pelerin Society’s anecdotal reference suggests how hard it is to understand the concept of liberalism, even when discussed solely from the perspective of free markets and by scholars who share very similar academic backgrounds. The problem of defining liberalism only grows when considering more extended periods of time. There are signs of essential contestability already in the nineteenth century, though mainly in the twentieth ( Abbey, 2005 ). After all, the concepts of liberty and freedom, which the term derives from, are already vague ( Gray, 1978 ). On top of these problems of political theory, some issues arise from comparative politics. Once a concept crosses borders, it may acquire new meanings that further blur its already contested understandings ( Sartori, 1970 ). This is the case with liberalism. As the concept traveled to the United States (US) or Latin America, for example, how politicians, voters, and social sciences students understood it varied considerably.

However, many political actors still use the term and identify with this ideology. More than that, they form domestic and international associations that claim to fight for similar ideas. One example is Liberal International (LI)—the world federation of liberal parties. Although its history has been rather Eurocentric, it has recently sought to strengthen bonds with other regions. This attempt becomes evident if one looks at LI Congress Meetings in the 2000s: Dakar (2018), Mexico City (2015), Abidjan (2012), Manila (2011), Cairo (2009), Marrakesh (2006), and Dakar once again (2003). These cities followed a long list of European and North American hosts since the Federation’s 1947 foundation in Oxford.

If Liberal International is expanding its reach and membership basis while liberalism remains an important label for political actors worldwide, then studying the basic values underlying the term’s usage makes sense. Toward that end, one must ask: what is the current concept of liberalism? More specifically, what is the concept of liberalism outside of the Euro-dominated political field? I address this question based on several innovative strategies.

First, I concentrate on Latin America. Following the example of many European nations, the region’s original partisan spectrum structured itself along the conservative-liberal divide ( Dix, 1989 ). In several cases, these parties disappeared or migrated to alternative families of ideology—survival rates of political parties in the region are often meager ( Cyr, 2017 ). Yet, liberals have developed into a significant political force in at least Paraguay, Honduras, and Nicaragua, providing interesting case studies.

Second, I look at this contestability (or the concept’s decontestation) from the perspective of partisan actors. Politicians and their parties will not necessarily abide by theorists’ expectations. However, these actors represent liberalism to the average voter. After reviewing how political and economic theory explains liberal values, I investigate which of these values Latin American liberal parties defend.

Initially, I collect and analyze the declaration of principles of all parties affiliated with Liberal International in the region. The analysis of political manifestos is standard in European politics ( Gemenis, 2013 ; Klingemann; Budge, 2013 ) and has been on the rise in Latin America (Jorge et al ., 2018, 2020). I find that liberals converge on their views regarding democratic governance and non-conformism. Support for liberal democracy and respect for individual rights and desires are the only components of this ideological tradition in all manifestos. The pursuit of equality through distributive policies and access to education , two core characteristics of social liberalism, appear in 78% of the declaration of principles. References to classical liberalism, such as reduced taxation or the minimum State, are less common.

I, then, look for evidence of whether politicians and voters also identify with these values. Furthermore, I test if these values are unique to liberals or whether other party families in the region also share them—a necessary step to ensure that these are the components of liberalism rather than a common-sense feature among major Latin American parties. I rely on surveys conducted with masses and elites in Paraguay, Nicaragua, and Honduras. Liberal voters often hold similar beliefs to those of other parties. Liberal legislators, however, are significantly more supportive of liberal democracy than their Paraguayan and Honduran counterparts.

Despite the lack of elite-mass convergence, these approaches identify liberal democracy as the core element of liberalism in the region, which follows John Stuart Mill’s (2001 , 2004 ) understanding of liberalism. Besides democracy, all manifestos defend individual liberties. Nevertheless, liberal elites and masses from Paraguay, Honduras, and Nicaragua do not support some individual liberties (compared to other partisans), such as the right to marry someone of the same sex. Furthermore, Latin American liberals defend private property or social justice (not both), reinforcing a State vs. market divide within the international liberal community.

This paper contributes to different streams of research. After the pink tide, Latin America saw the rise of right-wing and far-right political leaders ( Oliveira, 2019 ; Payne; Santos, 2020 ). This blue tide includes Mauricio Macri (Argentina), Pedro Pablo Kuczynski (Peru), Sebastián Piñera (Chile), Iván Duque (Colombia), Jair Bolsonaro (Brazil), Guillermo Lasso (Ecuador), and Lacalle Pou (Uruguay). As with the left wave, the new tide came in different tones of blue: for instance, it is hard to claim that Lula and Chávez or Macri and Bolsonaro are ideological twins. Power and Rodrigues-Silveira (2018) focus on Brazil and propose four types of right-wing politics whose scope should be broader in Latin America. This paper studies only one of these politicians’ parties (Macri’s Unión por Todos ); still, it contributes to understanding the nuances behind the term liberal often linked to Latin American right-wing policies.

In addition to the study of the Latin American right-wing, this paper further clarifies how the region understands liberalism. As mentioned, I demonstrate that Latin American liberals are not necessarily economic right-wingers—party manifestos cite social justice more often than private property, which differs from how the term is commonly used. Sallum Jr. (2011) classified Collor de Mello’s pro-market reforms as liberal; Rocha (2019) used the hyperbolic term ultraliberal to talk about supporters of the Austrian and Chicago schools of economics. To a certain extent, I offer a reply to Carlos Estevam Martins’s (2003) discussion in Dados two decades later. Like Martins, I acknowledge the term’s complexity and propose how to conceptualize it.

This paper also offers an empirical contribution. I follow Martins’s (2003) steps and look for the components of liberalism in political and economic theory and then search for them in the discourse of political elites and their voters, which resembles Rosas’s (2005 , 2010 ) study of the meaning of right and left-wing politics in Latin America. However, rather than concentrating on this one-dimensional spectrum, I investigate the ideological cohesion of a party family. Despite the region’s weak party systems and the term’s contestation, I find a minimum level of convergence: Latin American members of Liberal International defend liberal democracy and individual rights. This result suggests that party federations may be significant beyond developed democracies.

The paper develops as follows. First, I discuss the conceptual contestability of liberalism among partisan actors. I rely on this debate to propose that despite substantive differences, these parties still opt to work together globally. Next, I rely on political and economic theory to identify four potential components of liberalism. In the following two sections, I conduct a series of empirical tests to identify which of these elements Latin American liberal parties actually share. My strategy begins with an analysis of the declaration of principles of said parties. I focus on three cases to assess whether these values are unique to liberal parties or also feature in other political families. Finally, I conclude the paper by discussing the findings and proposing future agendas.

Ideology, political parties, and representation

The contestability of liberalism as a concept ( Abbey, 2005 ) is apparent in disputes between and within political parties. The way members of Denmark’s Radikale Venstre and Germany’s Freie Demokratische Partei describe liberalism is considerably different, with the former advocating for more social justice and the latter emphasizing private property. This divide also occurs between liberal parties within the same country, like the Dutch classical liberal Volkspartij voor Vrijheid en Democratie and the social liberal Democraten 66 . Humphreys and Steed (1988:396) argue that “the problem arises because the Liberal family itself asserts that it exists; yet even the most superficial glance at the Liberal group in the European Parliament or at the Liberal International finds it to be a most heterogeneous collection.”

This heterogeneity has always been part of Liberal International’s platform, the global federation of liberal parties. Its origin dates back to 1946 when the Belgian Liberale Partij invited the liberal and radical parties from Great Britain, France, Denmark, the Netherlands, Switzerland, Sweden, Italy, and Spain to attend its centennial in Brussels. During the meeting, which was also motivated by the end of World War II, all attendees co-signed a broad declaration vouching for world peace, the liberty of man, economic freedom, education, and the value of character. One year later, liberal parties from nineteen different countries assembled in Oxford, where they created Liberal International and co-signed the Oxford Manifesto. Since then, the Federation has reformed its guiding principles at least five times so as to accommodate incoming members and the changing nature of time.

Many of these parties were born in Europe as dissidents of conservative groups ( Kalyvas, 1996 ). In nineteenth-century Britain, Liberals agreed with the Whigs that the Crown should hold less power. Yet they also wanted to reduce the power of the Church, avoid wars, and expand free trade. These ideas motivated Lord John Russell (the United Kingdom’s prime minister between 1846-1852 and 1865-1866) and colleagues to form the new bloc with prominent politicians like William Gladstone (prime minister between 1868-1874, 1880-1885, 1886, and 1892-1894) and the philosopher John Stuart Mill (a member of parliament between 1865-1868).

These ideas traveled to Latin America and adapted to the region. According to Dix (1989: 24), the cleavages between liberals and conservatives in nineteenth-century Latin America mirrored the liberal bourgeoisie vs. rural conservative divide. Liberal elites “could usually be found advocating federalism, disestablishment of the church, and the defense of commercial interests, often including the advocacy of free trade.” However, “the very designations conservative and liberal at times appeared to be mere labels, adopted by one or another caudillo to enhance his image or legitimacy” ( Dix 1989: 35).

There are different reasons why ideological labels become blurred in party politics. The first concerns conceptual stretching that is believed necessary to adapt ideas to regions and the nature of time ( Sartori, 1970 ). In the United Kingdom, Leonard Hobhouse’s attempt to make the British Liberal Party more progressive occurred as a response to the growth of pro-labor ideologies in the early twentieth century (Hobhouse, 1919). Ludwig von Mises (1985: 3), a fierce supporter of non-interventionism, complained that “nothing is left of liberalism [in the Liberal Party] but the name” due to its endorsement of “the nationalization of railroads, of mines, and of other enterprises, and even (...) protective tariffs.”

This discrepancy does not necessarily mean that Hobhouse or Mises is “less liberal.” The concept’s evolution allowed for divergences even within the same country and period because parties may adopt different strategies to attract voters. More recently, the Volkspartij voor Vrijheid en Democratie ’s (VVD) leadership became more averse toward immigrants in response to the growing Dutch far-right ( Bale, 2003 ), which distanced VVD even more from the other liberal party’s ( Democraten 66 ) ideology that strongly support immigrants. Still, both Dutch parties are active in Liberal International and recognize each other as legitimate members of the same party family.

In Latin America, ideological shifts and disagreements should be even more likely given the region’s weak party systems. Lupu (2016: 172) shows that “since the mid-1990s, a quarter of the established political parties in Latin America have broken down. From one election to the next, they became irrelevant.” Furthermore, in many countries, voters choose members of congress from open lists provided by political parties and directly vote for the Head of State. It means that in some cases, support depends significantly on the candidate, making the party’s image and ideology therefore irrelevant, perhaps, to electoral politics ( Deegan-Krause, 2007 ).

Again, changes over time and politicians’ opportunistic behavior do not mean that parties do not have any ideological beliefs. Hawkins and Mogersntern (2010) and Rosas (2005 , 2010 ) find some degree of ideological coherence among Latin American parties, at least regarding selected policies or the left-right one-dimensional spectrum. Furthermore, the fact that parties still opt to join international federations suggests that they are interested in belonging to a party family: after all, this means going through a bureaucratic, lengthy, and sometimes costly political process, which includes foreign parties judging whether they consider candidates to be legitimate representatives of the ideology in their respective countries. This reinforces the idea that liberals believe in and wish for something in common despite not being ideological twins. Therefore, I propose that:

H ₁: Although liberal parties are not ideological twins, they share a minimum level of agreement.

Four potential components of liberalism

Now, the challenge is to identify what exactly liberals agree on. In this section, I rely on political and economic theory to map out four components of liberalism. Some of these components are contradictory, but this is plausible that liberal parties will adhere to a combination of them or at least one (as a whole). These components will serve to test which features are associated with the current agreed-upon concept of liberalism within Latin American political parties.

Abbey (2005: 463) affirms that “while liberalism might dominate contemporary English-language political theory, its ascendancy has not cast a pall of grim uniformity and conformity over that theorizing.” She uses Gallie’s (1956) methodological tools to study essentially contestable concepts to infer that liberalism, although often pointed out as hegemonic (for instance, Laclau, Mouffe, 1985), has varying meanings that differ significantly from each other. Indeed, this view is not unanimous ( Gray, 1978 ). However, even critics like Gray (2000) tend to acknowledge concurrent concepts of liberalism.

Adam Smith’s (2007) work is typically portrayed as a seminal or a watershed piece in the birth of classical liberalism ( Buchanan, 2000 , 2005 ; Rothbard, 2006 ). Like John Locke, Smith advocated for life, liberty, and property ( Wolfe, 2003 ). This mode of liberalism—also proposed by scholars like David Ricardo—paid special attention to the free market system and individualism.

This political stance is not to be confused with praise for selfishness or egoism—Smith emphasizes empathy as a virtue in The Theory of Moral Sentiments ( Smith, 1982 ). In The Wealth of Nations , Smith (2007) uses self-love , which is compatible with his view of individualism. He claimed that by seeking to maximize one’s utility, individuals could potentially achieve higher gains, increasing the aggregate welfare level in their respective societies.

The Austrian School of Economics follows the same line of thought. The conceptual discussion offered by Mises (1951) begins by approaching the meaning of ownership , reinforcing the liberal significance of private property. He does so to discuss the many forms that organized society, meaning the State, could implement socialist policies. According to his arguments, socialism would not end with the State expropriating the means of production. It could extend to the form of legislation passed to regulate private property, as it would distort the meaning of ownership. Nozick (1974) concurs—he considers that ownership exists only when something is acquired through a voluntary transfer of holdings. On these grounds, he rejects even taxation, which he compares to forced labor.

Ownership and voluntary transfers are also critical to Hayek (1945 , 1966 ), who discusses these elements as components of the market order (spontaneous order or liberal social order). He posits that “the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form, but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess” ( Hayek, 1945: 519). Based on this assumption, it becomes more reasonable, as Adam Smith previously suggested, to rely on the aggregate of individuals’ voluntary actions and subjective evaluations who, through the best use of their incomplete knowledge, contribute to supplying the needs of society.

Following their understanding of liberalism, the first possible component (C ₁ ) of liberalism is private property. I describe it as:

Private Property (C ₁ ): Liberals advocate for individualism in terms of private property. The State should not intervene in the economy, especially in matters of ownership.

Before moving on, I highlight that this component is designed as an ideal type (to follow Weber’s terminology). Some of the following components also adopt this strategy, which is done on purpose since this paper aims to assess the meaning of liberalism in Latin America. Thus, by utilizing components as ideal types, I test whether each is identified en masse and possibly simultaneously in Latin American liberal parties.

John Stuart Mill (2001 , 2004 ) also advocated for a limited State respectful of individualism. Nonetheless, there are mixed signs in his view on economic interventionism ( Mill, 2004 ). His defense of the individual was centered around a different perspective: the struggle against the tyranny of the majority. His seminal work discusses “the nature and limits of the power which can be legitimately exercised by society over the individual” ( Mill, 2001: 6). Devigne (2006: 163) sustains that “liberal thinkers have generally feared that a standard of what is best could easily grant political authorities the unconditional right to impose beliefs and practices on the citizenry to that end.”

Indeed, several authors combined Smith’s defense of private property and Mill’s struggle against the tyranny of the majority. For example, Friedman (1951: 92) argued that neoliberalism—or the twentieth-century version of nineteenth-century liberalism—regarded a system where “the citizens would be protected against the state by the existence of a free private market; and against one another by the preservation of competition.”

This view is fairly similar to the one of ordoliberalism: an alternative form of the liberal ideology promoted by Walter Eucken, Franz Böhm, and Leonhard Miksch, among other scholars from the Albert-Ludwigs-Universität Freiburg in the post-II World War period (Dullien, Guérot, 2012). They believed that competition was of utmost importance, yet it could only exist with the State’s protection, which should intervene to prevent cartels and monopolies. While Hayek, Nozick, and Mises supported the free market as a result of the absence of State power, ordoliberalism believed the in free market as a result of State power ( Bonefeld, 2013 ). ¹ In fact, Hayek (1966: 601) claimed that “liberalism and democracy, although compatible, are not the same. The first is concerned with the extent of governmental power, the second with who holds this power.”

Nevertheless, Mill’s theory merged liberalism and democracy as “a more balanced reconciliation of elite competence and mass participation” ( Krouse, 1982: 511), which lies between radical theories of participatory democracy and conservative theories of elite democracy. In this sense, the State does not exist to hold people submissive but to allow their individualism to flourish. Following this logic, I contend that liberal democracy should be a potential component of liberalism.

Liberal Democracy (C ₂ ): Liberals defend liberal democracy as the best way to organize society.

There is another concept of the struggle against the tyranny of the majority. While liberal democracy is an institutional proposition of organizing society, there is also the micro-component of how individuals should relate to one another. This dimension is not foreign to other classical authors, such as Smith (1982) , when he refers to tolerance and empathy in The Theory of Moral Sentiments . It is much more evident in Mill (2001) , who challenges the notion of conformism. Tulloch (1989) exemplifies it with what she calls Mill’s liberal feminism: the author’s support for sexual equality. Thus, this view of democracy also refers to defending the right not to conform to hegemonic behavioral standards.

This proposition is an alternative view of liberalism that focus on minorities and democracy rather than on the free market. The State must ensure that individuals can pursue their interests despite not conforming to the majority’s way of life. I derive another potential component of liberalism from this assumption:

Non-Conformism (C ₃ ): Liberals advocate for the right of individuals not to conform to the way of life of the majority.

Although John Stuart Mill is also cited as a free market advocate, the author advocates for some degree of State interventionism. One example is public control over wages and direct cash transfers from rich to unemployed citizens (Mill, 2004). He acknowledges that desirably regulated salaries could generate unemployment when set above competitive rates. However, he believes proper deliberation could reach ideal rates of a ‘negative income tax’ (to use Friedman’s later term) that would ultimately improve living standards. Therefore, he suggests that the population should force wealthier members of society to contribute to the welfare of the worse-off citizens through a wage-fund.

It is possible to find various examples of this renewed version of liberal utilitarian rupture with Smith’s laissez-faire liberalism. This argument prevailed, especially in the transition from the nineteenth to the twentieth century. This new liberalism (not to be confused with neo liberalism) emerged as both a theoretical and a political platform. Turner (2008: 36-7) affirms that intellectuals

realized that inequalities of economic power and the acuteness of the social and economic problems that the latter produced could constrain individual liberty as much as political power. Individual freedom, the new liberals argued, could only be realized within a social order. (...) This required a substantial increase in state activity in the name of what came to be called ‘social justice.’

Ross (1991) presents a similar narrative. In her view, liberalism began in Britain, opposing the remains of feudalism and mercantilism and defending “justice, representation, and economic activity on individualistic basis” (1991:10). These refer to the humanistic (i.e., liberal democratic, non-conformist) and commercial (i.e., private property) nature of liberalism. As the liberal model faced accelerated American capitalist industrialization, she suggests that economic inequalities and class confrontation led to the emergence of progressivism, which generated this new liberalism as a form of social justice without socialism (see also Clark, 1914 ).

Higgs (2013) and Harvey (2007) suggest that the 1929 Crisis helped speed the transition of the United States’ liberal tradition. President Franklin Delano Roosevelt’s New Deal included a set of interventionist policies that heavily contrasted with those of his predecessor Herbert Hoover. This new way of thinking about politics in the Democratic Party motivated a wave of interventionism that became dominant at least until the Republican Ronald Reagan became president ( Harvey, 2007 ; Higgs, 2013 ).

This process is similar to what happened in the United Kingdom. The Liberal Party scholar and activist Leonard Hobhouse (1919) led the campaign for policies that could enhance individual economic sovereignty by implementing anti-trust legislation and other forms of economic control. He believed this would benefit citizens and eliminate what he saw as social injustice. He had a substantive role in shifting the now-Liberal Democrats toward adopting some interventionist policies.

Throughout the twentieth century, theoretical bases of new liberalism developed under the labels of political or egalitarian liberalism. Rawls (1971) suggests that the State should create a regulatory setting that grants adequate living standards for every citizen, especially those worst off. Within this setting, they would have equal opportunities and be free from any culturally conservative intervention restricting their individual rights.

Dworkin (2000) develops this egalitarian notion further and focuses his concept of liberalism on an ideal distribution of resources that allows for every citizen to be equally satisfied with his or her achievements given the effort he or she put into obtaining them ( Wolfe, 2003 ). This view of liberalism leads to the fourth potential component of liberalism: social justice. I frame it as:

Social Justice (C ₄ ): Liberals advocate for State interventionism to reduce inequality between individuals to assure individual liberties.

Before moving to the next section of this paper to discuss the intersection between liberal theory and party politics, I summarize the four potential components of liberalism in Table 1 . This includes private property, liberal democracy, non-conformism, and social justice. In the empirical section of this work, I will look for these elements in Latin American liberal parties. While some may appear contradictory (for instance, private property and social justice), I expect that parties will often combine more than one component at the same time (e.g., liberal democracy and non-conformism).

Table 1
: Four potential components of liberalism

C	Component
C 1	Private Property
C 2	Liberal Democracy
C 3	Non-Conformism
C 4	Social Justice

Manifesto analysis

Empirical Strategy

I rely on manifesto analysis to identify the presence of these four components in the liberal parties of Latin America. This approach follows a long tradition of research developed especially following the seminal work of Robertson (1976) , who collected, coded, and analyzed British party manifestos from 1920 to 1974 to assess ideological changes as a tool of electoral strategy (Budge, Meyer, 2013). Three years later, the Manifesto Research Group, now the Manifesto Research on Political Representation, engaged in a large initiative mapping political documents from all over Europe ( Gemenis, 2013 ).

In the mid-2010s, the same research program also began mapping manifestos in Latin America (Volkens, Bara, 2013). However, the project focuses on elections, meaning that multiparty coalitions produced many of these documents rather than single parties. Furthermore, electoral documents tend to be more oriented toward specific domestic policies or contexts, and may not reflect similar ideological principles.

I select single-party declarations of principles instead of multiparty manifestos. These documents are created when political parties are founded and are often revised in the general assembly of these organizations. Rather than focusing on specific elections, they provide the guiding principles that orient the organization. I, then, conduct a content analysis with these documents. This method allows testing hypothesis 1 by identifying each of the four potential components of liberalism in the declarations of principles of Latin American liberal parties.

The challenge now is to identify the best approach to select liberal parties in Latin America. Mair and Mudde (1998) suggest four alternative methods for mapping party families: (1) origins, (2) international affiliation, (3) policy affinity, or (4) party name. A party’s historical framework is critical to understand its roots and its evolution—one example is the brief discussion about the British Liberal Party (today’s Liberal Democrats), which was born as a dissident of the Whigs and moved toward liberal progressivism in the early nineteenth century.

However, in some cases, this approach misses the point. One example is the Romanian Partidul National Liberal (PNL). Among the oldest liberal parties in the world, it dates back to 1875 when it appeared as the bourgeois alternative to Partidul Conservator . The Liberals were dissolved in 1950 and returned to politics after the 1989 Romanian Revolution. After over two decades as the leading liberal representation in the country, the party decided to identify as Christian Democrat in 2014. It no longer has ties to other liberal parties and has been an active member of the European People’s Party and the Centrist Democrat International (CDI) since then.

We should expect to find more cases like Romania’s PNL in Latin America than that of the British Liberal Democrats. In the region, party systems are weak ( Deegan-Krause, 2007 ), many parties die ( Cyr, 2017 ), and others adapt their brands to improve electoral chances ( Lupu, 2016 ). Recently, two members of Liberal International in the region decided to follow PNL’s path and joined the CDI. They are the Brazilian Democratas (formerly Partido da Frente Liberal ) and the Costa Rican Movimiento Libertario .

Policy affinity flies away from this paper’s raison d’être , as its objective is to identify the ideological framework that defines a party as liberal. Had I selected this approach, it would mean identifying all the parties that adopt at least one of the potential components of liberalism. In other words, rather than investigating what liberal parties have in common, this paper would search for parties that believe in what scholars see as the components of liberalism, regardless of how parties themselves identify.

Party name is misleading for several reasons. Consider three examples: in Brazil, far-right Jair Bolsonaro coopted a small party called Partido Social Liberal to launch his presidential campaign. He left the party after being elected. It would be difficult to defend that his ideas were, in fact, social liberal ideas. There are some components of liberalism in Colombia’s Partido Liberal , most notably its support for social justice. The party was also the main opponent of Colombian conservatives when created in 1848. However, Colombian liberals prefer a different label: they joined the Socialist International in 1999. Finally, one false negative: the Mexican Partido Nueva Alianza (PNA). Founded in 2005 as a dissident from the catch-all Partido Revolucionário Institucional , the PNA self-identified as liberal in its statutes and quickly joined Liberal International. This party would not have been selected through the party name approach.

Finally, association with an international family, i.e., Liberal International, allows selecting parties that self-identify as liberals and are accepted as such by a long-standing global liberal community. This solves the problem of parties that do not have “liberal” in their name or which opt to identify with another party family despite their name. Furthermore, it also addresses the issue of party origin: while some parties were born liberals and decided to change platform, others opted to become liberal after inception. In any case, belonging to an international party family means once again that the party leadership wants to identify as liberal and possesses enough ideological coherence for international peers to acknowledge them as such (and as a member of the ideological family).

Liberal International had ten Latin American members in 2016. ² I collected each one’s declaration of principles ( Table 2 ). As explained previously, this document is different from the one often used in manifesto analyses. Many scholars use the manifesto that parties produce for each election, i.e., the electoral platform. However, this is not always available in Latin America as it often comes as a multiparty document and presents short-term orientations. As a manner of capturing the long-term ideological framework of each party, I use the declaration of principles, which is a fixed document usually reformed only during party congresses. In this paper, the terms party manifesto and declaration of principles are used interchangeably.

Table 2
: Declarations of principles selected for analysis

Country	Party	Source
Argentina	Unión por la Libertad	Poder Judicial de la Nación
Chile	Partido Liberal de Chile	Own website
Cuba	Partido Solidaridad Democratica	Liberal International
Cuba	Unión Liberal Cubana	Own website
Guatemala	Movimiento Reformador	Party leadership
Honduras	Partido Liberal de Honduras	Tribunal Superior Electoral
Mexico	Nueva Alianza	Instituto Nacional Electoral
Nicaragua	Partido Liberal Independiente	Facebook Official Page
Paraguay	Partido Liberal Radical Auténtico	Tribunal Superior de Justicia Electoral

The nine documents were re-typed in plain format. Then, I follow Bardin’s recommended text-analysis approach (2011). The lines are numbered and each quasi-sentence is highlighted in alternative colors. These are matched to 49 units of analysis selected through an initial review process. The units represent the manifestos’ ideological features. For instance, a quasi-sentence highlighted in yellow in lines 1858-1859 was coded as ‘Transparent Governance.’ It is part of Unión por Todos ’s declaration of principles and consists of “the promotion of transparency and fight against corruption in all its forms.” ³

After coding, I created a table of frequencies, where I assign 1 for “presence” and 0 for “absence” matching each unit of analysis with each of the nine documents. With this, the final frequency of each unit of analysis was calculated by the sum of such values. The percentage was calculated by dividing this sum by the total number of documents analyzed. This approach allows for identifying the features that are common among all Latin American liberal parties (i.e., high frequency) while separating those which are valid only within specific countries (i.e., low frequency).

Findings

Many manifestos present ideological features related to their own historical or political frameworks. Other than representing the party family’s beliefs, they portray the roots of certain political movements and their relationships to the country’s circumstances. This is the case of Partido Solidaridad Democratica ’s references to the Cuban nineteenth-century national hero, José Martí, and to former Catholic Pope John Paul II during a recent visit to the island. The same is true for Nueva Alianza ’s reference to protecting Mexican citizens who migrated to the United States. To avoid including these elements in the regional conception of liberalism, I set a threshold of a minimum number of parties referring to the same ideological features. This mechanism also helps avoid units of analysis that are not internationally recognized as components of liberalism, even if they may be significant nationally. The selected threshold is seven out of nine manifestos, or 78% of the documents ( Table 3 ).

Table 3
: Frequency of most frequent units of analysis

Unit of Analysis	Component	Frequency
		N	%
Liberal democracy	Liberal Democracy (C 2 )	9	100%
Respect for individual rights and desires	Non-Conformism (C 3 )	9	100%
Transparent governance	Liberal Democracy (C 2 )	8	89%
Peaceful means and/or promotion of peace	Liberal Democracy (C 2 )	8	89%
Respect the rule of law and strong institutions	Liberal Democracy (C 2 )	8	89%
Equality pursued through distributive policies	Social Justice (C 4 )	7	78%
Access to education	Social Justice (C 4 )	7	78%
Modernization/efficiency of public administration	Mixed	7	78%

• Note: Most frequent units of analysis. N reflects the number of documents with which the unit of analysis was identified. Maximum N is 9.

Only two units of analysis were found in every document: liberal democracy and respect for individual rights and desires, derived from Mill’s perspective and support the validity of liberal democracy (C ₂ ) and non-conformism (C ₃ ) as components of Latin American liberalism. The first category usually appears as the support for political pluralism and free elections. The Unión Liberal Cubana , struggling against the national socialist dictatorship, affirms that “it is not the government, in summary, that must watch over the people. It is the contrary. It is not the government that must drive the people. It is the inverse” (Appendix I, lines 1357-1358). Similarly, the Partido Solidaridad Democratica believes in “the tolerance and political pluralism, in representative and participatory democracy, where the citizen actively and efficiently engages through civil society in the public life” (Appendix I, lines 1474-1475). The Partido Liberal de Honduras has a strong position in this respect: it “only accepts as a legitimate source of public power the will of the people expressed in free and honest campaigns; it rejects continuity, anti-democratic and exclusionary practices against civic participation” (Appendix I, lines 1199-1201).

The Movimiento Reformador holds individual rights in high esteem—an element that I categorized as non-conformism (C ₃ ). The party “looks for the strengthening of the mechanisms (...) to observe rights and liberties of the citizens” (Appendix I, lines 12-15), while Nueva Alianza praises the political leader Benito Juaréz and his colleagues, who “built the Republic over the grounds of individual guarantees” (Appendix I, lines 39-40), and the Flores Magón brothers who fought for the civil rights in the 1857 Constitution. The Partido Liberal de Chile affirms that “we, the liberals, need to especially appreciate diversity. Each human life constitutes an original invention, an essay on how to live in the right manner; drawings with their own aesthetics, their dignity, their special sense” (Appendix I, lines 1078-1080).

Transparent governance, peaceful means, and the rule of law appear in eight out of nine manifestos. I categorized these elements as part of liberal democracy (C ₃ ). Concerning the first element, the Partido Liberal Radical Auténtico advocates for “the promotion of transparency and the fight against corruption in all its forms” (Appendix I, lines 1858-1859), while the Unión Liberal Cubana idealizes a “government subject to constant auditing performed by the citizens” (Appendix I, lines 1371-1372). Relating to peace, the Unión por la Libertad advocates for “peace for all the individuals that inhabit the Argentine land” (Appendix I, lines 1902-1903) and “world peace and the compromise with the fight against terrorism” (Appendix I, line 1885). Finally, the Partido Liberal de Honduras defines defending an inclusive democracy as the need to “strength the Rule of Law based on civic participation, and to recognize the rights of civil society actors, to organize themselves freely and to develop their activities under the protection of the law” (Appendix I, lines 1275-1277).

There are two additional elements in seven out of nine manifestos that refer to equality through distributive policies. They are closely connected to the social justice component of liberalism (C ₄ ). The Unión por la Libertad supports “social security that grants to the ones in need all the basic benefits” (Appendix I, lines 1888-1889) while Partido Solidaridad Cubana “considers the conquest of authentic social justice as the main duty of a democratic government” (Appendix I, lines 1761-1762).

The last element mentioned in seven out of nine manifestos is the modernization/efficiency of public administration. This could be categorized in different ways—the first is private property. The assumption would have been that increasing efficiency in the public sector would allow providing the same services with fewer resources, thus reducing taxation ( Hood, 1991 ). However, Aligica et al . (2019) discuss how the notion of efficiency often brought up in New Public Management is incompatible with the classical liberal tradition. Similarly, a modern public administration could improve social justice by providing better public services. Again, if framing this as making the government resemble the business sector, some authors could challenge this proposition ( Abrucio, 2007 ). Therefore, I do not label it.

If we set the threshold at 78% of the documents, most Latin American political parties appear to share only three components of liberalism: liberal democracy, non-conformism, and social justice. It suggests that Mill’s view against the tyranny of the majority unites Latin American liberals. Seventy-eight percent of manifestos defend social justice—often associated with US Democrats and other left-leaning liberal parties in developed democracies—through two different elements, which contradicts the common usage of the term that has been connected to pro-market reforms and preferences (see Rocha, 2019 ; Sallum Jr., 2011).

One interesting aspect of the manifestos is that social justice and private property are not necessarily contradictory. Most often, social liberals praise property rights, while classical liberals support egalitarian societies. In some cases, the same statement will speak to both components. One example comes from the Partido Liberal Radical Auténtico: the Paraguayan party affirms that the “extension and diffusion of private property and goods to all citizens of the Republic (...) reduce social and economic inequalities” (Appendix I, lines 1850-1852). In this sense, it uses social justice to reinforce private property, even though not by the same standards advocated by Mises or Nozick. This kind of sentence is common within the selected manifestos. Here, the point is that while parties may still defend most of the selected components, they will emphasize some and not others. This emphasis speaks to their view on liberalism and helps identify the meaning of this ideology within Latin America.

Survey analysis

In the previous section, I demonstrate that all members of Liberal International in Latin America defend liberal democracy (C ₂ ) and non-conformism (C ₃ ). Furthermore, most of them support social justice ideas (C ₄ ). The other potential component of liberalism, private property (C ₁ ), seems less consensual among these parties. However, identifying those features within the political documents of liberal parties does not mean that these elements define liberalism. They may regard features widely supported in Latin America, regardless of whether the party belongs to a liberal, conservative, or socialist family.

The working hypothesis of this paper is that although liberal parties are not ideological twins, they share some common ground. However, parties may agree on certain basic issues even if not belonging to the same party family. It is possible, for instance, that all parties in a country support democracy despite belonging to different ideological families.

The question, then, is: do Latin American liberal parties especially support these components, or are these components simply part of the regional political discourse? This question works to verify how robust the results of the previous section are. That is, if I find that liberal parties are more supportive of liberal democracy and non-conformism (and potentially the other two components), then it is possible to support this paper’s working hypothesis and suggest the core components of Latin American liberalism.

Empirical Strategy

I answer this question based on the analysis of elite and mass surveys. For the former, I consider the Parliamentary Elites of Latin America (PELA) project. This is a Universidad de Salamanca initiative, which consists of surveys conducted with legislators all over Latin America. Since the early 1990s, PELA has sent teams of researchers to interview many representatives after every election in 17 Latin American countries. The questions in this survey are comparable to those in the Americas Barometer (also known as LAPOP), coordinated by Vanderbilt University. Their team follows a similar logic, fielding nationally representative mass-level surveys every two years in almost all Latin American countries.

There are only three countries where it is possible to identify a large enough sample size of liberal legislators and voters: Paraguay, Honduras, and Nicaragua. They are also ideal cases as they contain the largest Latin American party members of Liberal International. To ensure that the data is comparable across datasets, I select compatible survey years in both cases.

These cases are ideal also because they offer opponents who are either conservative or socialist. Therefore, it is possible to test whether the components of liberalism in question are stronger among liberal parties than their opponents, regardless of the opponent’s ideological family. The out-parties are the conservative Asociación Nacional Republicana (Paraguay) and Partido Nacional de Honduras (Honduras) and the socialist Frente Sandinista de Libertación Nacional (Nicaragua). I summarize the selected cases in Table 4 .

Table 4
Cases selected for the survey analysis

	Paraguay	Honduras	Nicaragua
PELA (legislature years)	2013-2018	2014-2018	2012-2017
LAPOP (survey year)	2016	2016	2012
Liberal party	PLRA	PLH	PLI
Share of legislative seats	32.8%	24.2%	28.9%
Main opponent	ANR	PNH	FSLN
Opponent’s ideology	Conservative	Conservative	Socialist

Furthermore, each of the three liberal parties is considerably different in terms of the history and context in which the surveys were conducted. The Partido Liberal Radical Auténtico is the oldest party in Paraguay, founded as the Centro Democrático in 1887. In the same year, the governing authorities created the Partido Nacional Republicano (today, Asociación Nacional Republicana –ANR). In 1890, Centro Democrático became the Partido Liberal and, one year later, tried to oust the formerly allied ANR’s president, initiating a political divide that still exists today. The liberals governed the country from 1904 to 1936, when a series of coups and authoritarian governments, mostly under ANR’s leadership, characterized national politics until the late 1980s. Since then, the ANR has won almost every election. The only exception is 2008 when socialist Fernando Lugo won the election with a coalition of liberals and Frente Guasú . Lugo left the government to the liberal Federico Franco after his impeachment in 2012. When the surveys were carried out, ANR had the country’s president, and the liberals controlled 32.8% of the legislative house.

Liberalism in Honduras dates back to Dionisio Herrera, its first Head of State in 1824. Nonetheless, the Partido Liberal de Honduras ’s (PLH) history began in 1866 with Celeo Arias, president between 1872 and 1874 who founded the Liga Liberal . The party was given its current name in 1891 while under the leadership of Policarpo Bonilla of the revolutionary government that ousted the conservative General Domingo Vásquez through the mid-1890s Revolución Liberal. Since then, the PLH has elected more than a dozen presidents, including Manuel Zelaya (2006-2009), who was ousted and exiled after his political rapprochement with Venezuelan socialist leader Hugo Chávez. The liberals controlled 24.2% of the Honduran Congress when PELA and LAPOP fielded their interviews.

Dionisio Herrera was also elected Nicaragua’s Head of State (1830-1833) but by then, at least five liberal politicians had governed the country since 1825. Nonetheless, the Partido Liberal Independiente was founded only in 1944 as a dissident of the Partido Liberal Nacionalista (1912-1979). The latter governed Nicaragua for more than four decades, beginning in 1936 with mostly members of the Somoza family. In 1979, the Marxist Frente Sandinista de Liberación Nacional ousted the government with the support of Partido Liberal Independiente . Daniel Ortega led the socialists in office until 1990 when the liberals joined a coalition that defeated the Marxist group. Ortega became president once again in 2007 and has remained in office until now. Nicaraguan liberals controlled 28.9% of the country’s congress at the time of interviews.

I selected four survey questions that proxy the four potential components of liberalism, that are comparable across PELA and LAPOP, and that had answers from liberals and their opponents. For each question, dataset, and country, I run a two-way t-test. This approach is comparable to an OLS regression with no covariates, with the benefit of yielding the group’s average, aside from the difference across groups (equivalent to a regression’s coefficient) and its p-value. Therefore, it is possible to assess whether group averages are high/low and whether they differ across parties.

To test support for private property (C ₁ ), I examine a question that asks whether the State should own important public companies. Respondents answer based on a 1 to 7 scale, in which 1 represents support for private ownership and 7 for State ownership—the question is identical across surveys.

The same is not true for the liberal democracy (C ₂ ) proxy. Elites were asked if democracy is the best form of government and could only answer yes or no. Voters faced a 1 to 7 scale, in which responses varied from no support for democracy (1) and strong support for democracy (7). To ensure both scales are comparable, I recoded voters’ answers to 1 if they answered 7 (i.e., unconditional support), and 0 otherwise.

The third component, non-conformism (C ₃ ), could be tested in many ways since parties may choose different individual rights to advocate for. I conduct a conservative test on the hypothesis by selecting the right to marry someone of the same sex. This choice aligns with authors who discuss Mill’s support for gender equality ( Tulloch, 1989 ) and addresses the potential confusion between social and economic liberalism on moral politics ( Bohigues, 2021 ). Respondents were asked whether they support the State’s recognition of same-sex marriage (on a 1 to 10 scale, where 10 means ‘firmly approve’ and 1 ‘firmly disapprove’).

Finally, I test support for social justice (C ₄ ) based on a question on the State’s role in reducing economic inequality. On a 1 to 7 scale, responses vary from full support for State intervention in the matter (7) to no support at all (1). I re-scale all responses to a 0 to 1 range in this and all other cases, which makes comparing across variables easier and the analysis more straightforward while avoiding any loss in data quality.

Findings

In the manifesto analysis, private property rights did not appear as a common component of liberalism in Latin America, which does not mean that liberals do not support this right. Among the three selected countries for the survey analysis, Honduran and Nicaraguan liberal elites favor private ownership of large firms ( Table 5 ). The Nicaraguan response is unsurprising since its main opponent is the socialist FSLN.

Table 5
: Elites’ attitudes toward the public ownership of firms (C 1 )

	Paraguay	Honduras	Nicaragua
Liberals	0.2807 (19)	0.1765 (17)	0.1778 (15)
Conservatives	0.3086 (27)	0.5167 (30)	-
Socialists	-	-	0.3602 (31)
Difference	-0.02794	-0.3402	-0.1824
P-value	0.7612	0.0009	0.0968

• Note: Group averages; sample size between parentheses.

The case of Honduras, however, is less obvious. Liberals and conservatives are 34.02 percentage points apart—the most significant difference in the survey analysis. Conservative Honduran legislators favor public firms more than Nicaraguan socialists—a trend seen among elites but not voters. This suggests that the gap exists due to the PLH’s strong position on the issue and the Honduran conservatives’ left-leaning approach to property.

Among voters ( Table 6 ), this issue only divides Nicaraguan partisans. Again, this is the most significant difference among the four selected issues—22.44 percentage points. As with manifestos, the data suggest that while private property is not inherent to liberalism within the region, it is still critical for some liberal parties.

Table 6
: Voters’ attitudes toward public ownership of firms (C 1 )

	Paraguay	Honduras	Nicaragua
Liberals	0.4658 (151)	0.4976 (70)	0.3252 (103)
Conservatives	0.4880 (319)	0.5464 (280)	-
Socialists	-	-	0.5496 (723)
Difference	-0.0222	-0.0488	-0.22443
P-value	0.4982	0.3234	0.0001

• Note: Group averages; sample size between parentheses.

If only considering manifestos, liberal democracy is the main component of liberalism. Nonetheless, none of the two-way t-tests yielded statistically significant differences when comparing liberal parties to their main opponents in Paraguay, Honduras, and Nicaragua, which may be related to the nature of the data—its small sample size and binary nature may hide significant differences. Consider elites’ attitudes presented in Table 7 . All liberal legislators, with no exception, said that democracy is the best form of government. While support for democracy is also high among conservatives and socialists, it is not unanimous, ranging from 86.7% (Honduras) to 96.8% (Nicaragua). Thus, the differences may not be statistically significant at conventional levels, but it remains remarkable that all liberal elites provided identical answers within and between countries.

Table 7
: Elites’ attitudes toward liberal democracy (C 2 )

	Paraguay	Honduras	Nicaragua
Liberals	1.000 (19)	1.000 (17)	1.000 (16)
Conservatives	0.9630 (27)	0.8667 (30)	-
Socialists	-	-	0.9677 (31)
Difference	0.0370	0.1334	0.0323
P-value	0.4076	0.1206	0.4786

• Note: Group averages; sample size between parentheses.

These attitudes are more heterogenous among voters ( Table 8 ). First, unconditional support for democracy is low: roughly 15% of Paraguayan liberals and conservatives gave democracy the highest grade (liberals being only half p.p. higher than conservatives). In Nicaragua, where averages are substantively higher, the difference of means is 8.30 p.p.—yet again not statistically significant.

Table 8
: Voters’ attitudes toward liberal democracy (C 2 )

	Paraguay	Honduras	Nicaragua
Liberals	0.1583 (150)	0.1972 (71)	0.5500 (100)
Conservatives	0.1483 (317)	0.2283 (276)	-
Socialists	-	-	0.4670 (728)
Difference	0.0051	-0.0311	0.0830
P-value	0.8864	0.5754	0.1196

• Note: Group averages; sample size between parentheses.

Finally, the case of Honduras seems somewhat contradictory. The country has the highest difference of means among legislators, with liberals being 13.34 p.p. more supportive of democracy than conservatives. Among voters, however, conservatives are slightly more supportive of democracy than liberals. The difference, again, is not statistically significant.

Liberal manifestos commonly cited the third component: non-conformism. All parties stated that they support individual rights. Therefore, an important test addresses whether these parties support rights such as same-sex marriage more than their opponents. However, results show the contrary. The difference of means was statistically significant in only one case, i.e., voters in Nicaragua; the significant results showed socialists as 16.19 p.p. more supportive of this individual right than liberal voters. Most of the other differences of means are negative (Paraguayan elites being the exception), which contradicts these parties’ declaration of principles.

Table 9
: Elites’ attitudes toward same-sex marriage (C 3 )

	Paraguay	Honduras	Nicaragua
Liberals	0.2026 (17)	0.0850 (17)	0.1373 (17)
Conservatives	0.1496 (26)	0.1667 (30)	-
Socialists	-	-	0.2153 (32)
Difference	0.0530	-0.0817	-0.0780
P-value	0.5342	0.3250	0.3518

• Note: Group averages; sample size between parentheses.

Table 10
: Voters’ attitudes toward same-sex marriage (C 3 )

	Paraguay	Honduras	Nicaragua
Liberals	0.2100 (154)	0.1944 (72)	0.0651 (58)
Conservatives	0.2362 (327)	0.2005 (287)	-
Socialists	-	-	0.2270 (371)
Difference	-0.0262	-0.0061	-0.1619
P-value	0.4293	0.8892	0.0011

• Note: Group averages; sample size between parentheses.

The fourth potential component of liberalism is social justice, which should be more salient when liberals oppose conservative parties rather than socialist—which holds true in the selected countries ( Table 11 ). Among Paraguayan elites, liberals are 18.06 p.p. (p < 0.05) more supportive than the conservative party of the State’s role in reducing economic inequality.

Table 11
: Elites’ attitudes toward reducing economic inequality (C 4 )

	Paraguay	Honduras	Nicaragua
Liberals	0.8596 (19)	0.9118 (17)	0.8125 (16)
Conservatives	0.6790 (27)	0.8778 (30)	-
Socialists	-	-	0.8978 (31)
Difference	0.1806	0.0340	-0.0853
P-value	0.0221	0.6410	0.1109

• Note: Group averages; sample size between parentheses.

The difference of means is smaller and not statistically significant in Honduras, yet the average among liberals is higher than among conservatives. It is curious that while Honduran conservatives favor public ownership of important industries significantly more than liberals, they are less supportive of State interventionism to reduce economic inequalities.

In Nicaragua, where liberals oppose the Sandinistas , socialists are 8.53 p.p. more in favor of State intervention in this area. Even if this result is unsurprising, the difference is not statistically significant at conventional levels, possibly because the liberal average is considerably high (0.8125).

Interestingly, liberal voters support this remedy to social injustices less than their opponents in the three countries ( Table 12 ). This divergence, which also occurs in other issues, suggests that liberal voters and their elites are not ideologically connected—at least not by the core elements of the liberal doctrine. The only convergence occurs in Nicaragua, where socialist voters and their representatives advocate for State intervention to reduce economic inequality.

Table 12
: Voters’ attitudes toward reducing economic inequality (C 4 )

	Paraguay	Honduras	Nicaragua
Liberals	0.6362 (153)	0.6981 (69)	0.8236 (103)
Conservatives	0.6723 (323)	0.7304 (290)	-
Socialists	-	-	0.8774 (738)
Difference	-0.03362	-0.0324	-0.0537
P-value	0.2489	0.4417	0.0248

• Note: Group averages; sample size between parentheses.

Discussion

These research strategies present complementary pieces of evidence. First, the manifesto analysis shows that liberal democracy, non-conformism, and social justice are part of liberal parties’ agendas. Nonetheless, different Latin American party families may equally defend these ideas. The survey analysis helps address this issue (see the summary of findings in Table 14 ). Liberal democracy, present in all liberal manifestos, receives unconditional support from all liberal legislators interviewed in Paraguay, Honduras, and Nicaragua. While conservatives and socialists also strongly support this type of regime, it is not a consensus among them that it is the best one. Therefore, it is possible to confidently affirm that liberal democracy is a core component of liberalism in Latin America.

Table 14
: Support for potential components of liberalism in Latin America

C	Component	Manifestos	Paraguay		Honduras		Nicaragua
			Elites	Voters	Elites	Voters	Elites	Voters
1	Private Property	No	No	No	Yes	No	Yes	Yes
2	Liberal Democracy	Yes	S*	No	S*	No	S*	No
3	Non-Conformism	Yes	No	No	No	No	No	No
4	Social Justice	Yes	Yes	No	No	No	No	No

• Note: Yes means support; No means no support; S* means some support.

Non-conformism, also mentioned in all liberal manifestos analyzed, does not have the same level of support among voters or elites. In the survey analysis, voters and elites of conservative and socialist major parties were either more or equally in favor of same-sex marriage (one example of individual rights) than liberals were, which suggests that it is not a sufficient condition for a component to be included in the declaration of principles to predict relative attitudes in Latin America.

At the beginning of this paper, I mentioned that some components could be seen as contradictory, such as social justice and private property. Parties focusing on Hayek’s or Mises’ understanding of private property may attach little importance to social justice policies. More closely related to authors like Dworkin or Rahls, social justice policies could lead to centralized solutions prioritizing public solutions over private property—which is not necessarily true. One illustration is the PLRA’s declaration of principles. The Paraguayan liberal party emphasizes the role of private property as a means of social justice. Furthermore, while Honduran elites support private property more than their conservative counterparts, they also view State solutions to economic equality more favorably (on average, despite the lack of statistical significance).

Most Latin American liberal parties’ manifestos emphasize social justice but not so much private property. The surveys help explain the puzzle: when liberals face conservatives, they become more likely to demonstrate support for State intervention to help differentiate parties. This scenario holds true in Paraguay. When the main opponent is a socialist party, this scenario does not apply. Nicaragua is an example: both elites and voters favor private property more but are less supportive of policies that address social injustices.

The mixed-methods approach also sheds light on a problem commonly found in Latin American party systems: the lack of cohesion between the masses and elites. In almost all cases, the elites’ support for a potential component of liberalism did not mean that voters held the same beliefs. In one case, the elites and voters viewed a given component more favorably than their counterparts of the opposing political party (private property in Nicaragua). This study doesn’t seek to explain why such a discrepancy exists. However, this result confirms the thesis that the region has an ideological disconnect between its elites and masses.

Final remarks

This study aimed to answer what the current concept of liberalism is or more specifically what the concept of liberalism is outside of the Euro-dominated political arena. I extracted four potential components of this ideology from classic and modern liberal readings. Then, I searched for them in the declaration of principles of Latin American liberal parties. Finally, I tested them in comparative terms through survey analyses based on the attitudes of elites and voters from Paraguay, Honduras, and Nicaragua, where liberals hold a significant place within national legislatures.

Liberal democracy is the core element of Latin American liberalism. It appears in all manifestos of liberal parties studied and is unanimous among party elites. Furthermore, parties tend to defend social justice or private property, though not necessarily both. These features are context-dependent and specially affected by the party system’s composition. Non-conformism seems a significant principle in theory, but not when liberals are asked about their preferences and compared to their main rivals.

These findings highlight the contested nature of liberalism and reinforce the path dependence of the ideology in Latin America. The pro-modernization nineteenth-century urban bourgeoisie created liberal parties. However, their core beliefs adapt to political competition. While liberals always support democracy, the relative defense of private property and social justice depends on whether the main rival is a conservative or a socialist party.

In sum, if Ludwig von Mises could see the Latin America’s current liberal parties, he would probably repeat his memorable statement, “You’re all a bunch of socialists.” As Friedman noted for the Mont Pelerin Society cases, Mises would probably be wrong or, at least, have (greatly) exaggerated. Nonetheless, this perception would confirm that even though Latin American liberal parties have at least liberal democracy in common, other core values face disputes across and within countries.

These results contribute to understanding political preferences in Latin America from the perspectives of both elite-mass linkages and international relations (mainly due to the use of Liberal International as a parameter for case selection). Still, its main contribution lies in improving the conceptualization of liberalism by focusing on its democratic aspect. This is an important advancement to the students of political theory and political parties, as well as to practitioners of politics in the region.

References

• Abbey , Ruth . Is Liberalism Now an Essentially Contested Concept? New Political Science , v. 27 , n. 4 , p. 461 – 480 , 2005 .
• Abrucio , Fernando Luiz . Trajetória recente da gestão pública brasileira: um balanço crítico e a renovação da agenda de reformas . Revista de Administração Pública , v. 41 , p. 67 – 86 , 2007 .
• Aligica ; Boetke , Peter J. ; Tarko , Vlad. Public governance and the classical-liberal perspective: Political economy foundations . Oxford : Oxford University Press , 2019 .
• Bale , Tim . Cinderella and Her Ugly Sisters: The Mainstream and Extreme Right in Europe’s Bipolarising Party Systems . West European Politics , v. 26 , n. 3 , p. 67 – 90 , 2003 .
• Bardin , Laurence . Análise de Conteúdo . Lisboa : Edições 70 , 2011 .
• Bohigues , Asbel . Élites, radicalismo y democracia: Un estudio comparado sobre América Latina . Madrid : Centro de Investigaciones Sociológicas , 2021 .
• Bonefeld , Werner . Adam Smith and ordoliberalism: on the political form of market liberty . Review of International Studies , v. 39 , n. 2 , p. 233 – 250 , 2013 .
• Buchanan , James . The Soul of Classical Liberalism . The Independent Review , v. 5 , n. 1 , p. 111 – 119 , 2000 .
• Buchanan , James . Why I, Too, Am Not a Conservative: The Normative Vision of Classical Liberalism . Cheltenham : Edward Elgar , 2005 .
• Budge , Ian ; Meyer , Thomas . The Best Tools to Tackle the Job . In: VOLKENS , A. et al . ( Org .) . Mapping Policy Preferences from Texts: Statistical Solutions for Manifesto Analysts . Oxford : Oxford University Press , 2013 . p. 9 – 32 .
• Clark , John B . Social Justice Without Socialism . Boston : Houghton Mifflin , 1914 .
• Cyr , Jennifer . The Fates of Political Parties: Institutional Crisis, Continuity, and Change in Latin America . Cambridge : Cambridge University Press , 2017 .
• Deegan-Krause , Kevin . New Dimensions of Political Cleavage . In: DALTON , R. ; Klingemann , H.-D. ( Org .) . Oxford Handbook of Political Behavior . Oxford : Oxford University Press , 2007 . p. 538 – 556 .
• Devigne , Robert . Reforming Liberalism: J. S. Mills’ Use of Ancient, Religious, Liberal, and Romantic Moralities . London : Yale University Press , 2006 .
• Dix , Robert . Cleavage Structures and Party Systems in Latin America . Comparative Politics , v. 22 , n. 1 , p. 23 – 37 , 1989 .
• Dullien , Sebastian ; Guérot , Ulrike . The Long Shadow of Ordoliberalism: Germany’s Approach to the Euro Crisis . European Council on Foreign Relations: Policy Brief , v. 49 , p. 1 – 15 , 2012 .
• Dworkin , Ronald . Sovereign Virtue: The Theory and Practice of Equality . Cambridge : Harvard University Press , 2000 .
• Friedman , Milton . Neo-Liberalism and its Prospects . Farmand , v. 17 , pp. 89 – 93 , 1951 .
• Friedman , Milton ; Friedman , Rose . Two Lucky People: Memoirs . Chicago : University of Chicago Press , 1998 .
• Gallie , Walter . Essentially Contested Concepts . Proceedings of the Aristotelian Society , v. 56 , pp. 167 – 198 , 1956 .
• Gemenis , Kostas . What to Do (and Not to Do) with the Comparative Manifestos Project Data . Political Studies , v. 61 , n. S1 , pp. 3 – 23 , 2013 .
• Gray , John . On Liberty, Liberalism and Essential Contestability . British Journal of Political Science , v. 8 , n. 4 , pp. 385 – 402 , 1978 .
• Gray , John . The Two Faces of Liberalism . Cambridge : Polity Press , 2000 .
• Harvey , David . A brief history of neoliberalism . Oxford : Oxford University Press , 2007 .
• Hawkins , Kirk ; Morgenstern , Scott . Ideological Cohesion of Political Parties in Latin America . In: Kitschelt , H. et al . ( Org .) . Latin American Party Systems . Cambridge : Cambridge University Press , 2010 . pp. 145 – 168 .
• Hayek , Friedrich . The Constitution of Liberty . The Collected Works of F.A. Hayek, Volume XVII ed . Chicago : University of Chicago Press , 2011 .
• Hayek , Friedrich . The Principles of a Liberal Social Order . Il Politico , v. 31 , n. 4 , pp. 601 – 618 , 1966 .
• HAYEK , Friedrich . The Use of Knowledge in Society . The American Economic Review , v. 35 , n. 4 , pp. 519 – 530 , 1945 .
• Higgs , Robert . Crisis and Leviathan: Critical Episodes in the Growth of American Government . Oakland : Independent Institute , 2013 .
• Hobhouse , Leonard . Liberalism . 2 nd. ed. London : Williams and Norgate , 1919 .
• Hood , Christopher . A Public Management of All Seasons? Public Administration , v. 69 , n. 1 , pp. 3 – 19 , 1991 .
• Jorge , Vladimyr L. et al. Análise dos programas eleitorais dos candidatos a presidente em 2014: o posicionamento ideológico do PT e do PSDB . Revista de Sociologia e Política , v. 26 , n. 67 , pp. 1 – 20 , 2018 .
• Jorge , Vladimyr L. ; Faria , Alessandra Maia Terra de ; Silva , Mayra Goulart da. Posicionamento dos partidos políticos brasileiros na escala esquerda-direita: dilemas metodológicos e revisão da literatura . Revista Brasileira de Ciência Política , v. 33 , pp. 1 – 44 , 2020 .
• Kalyvas , Stathis . The Rise of Christian Democracy in Europe . Ithaca : Cornell University Press , 1996 .
• Klingemann , Hans-Dieter ; Budge , Ian . Using the Manifesto Estimates to Refine Party Family Placements . In: Volkens , A. et al . ( Org .) . Mapping Policy Preferences from Texts: Statistical Solutions for Manifesto Analysts . Oxford : Oxford University Press , 2013 . pp. 49 – 65 .
• Krouse , Richard . Two concepts of democratic representation: James and John Stuart Mill . Journal of Politics , v. 44 , n. 2 , pp. 509 – 537 , 1982 .
• Laclau , Ernesto ; Mouffe , Chantal . Hegemony and Socialist Strategy: Towards a Radical Democratic Politics . 2 nd. ed. London : Verso , 1985 .
• Lupu , Noam . Party brands in crisis: Partisanship, brand dilution, and the breakdown of political parties in Latin America . Cambridge : Cambridge University Press , 2016 .
• Martins , Carlos Estevam . Liberalismo: O Direito e o Avesso . Dados: Revista de Ciências Sociais , v. 46 , n. 4 , pp. 619 – 660 , 2003 .
• Mill , John Stuart . On Liberty . Kitchener : Batoche Books , 2001 .
• Mill , John Stuart . Principles of Political Economy: With Some of Their Applications to Social Philosophy . Cambridge : Hackett Publishing Company , 2004 .
• Mises , Ludwig . Liberalism in the Classical Tradition . 3 ^rd . ed. New York : Foundation for Economic Education , 1985 .
• Mises , Ludwig . Socialism: An Economic and Sociological Analysis . New Haven : Yale University Press , 1951 .
• Nozick , Robert . Anarchy, State, and Utopia . Oxford : Blackwell , 1974 .
• Oliveira , Felipe A. The rise of the Latin American far-right explained: dependency theory meets uneven and combined development . Globalizations , v. 16 , n. 7 , pp. 1145 – 1164 , 2019 .
• Payne , Leigh A. ; Santos , Andreza A. The right-wing backlash in Brazil and beyond . Politics & Gender , v. 16 , n. 1 , 2020 .
• Power , Timothy J. ; Rodrigues-Silveira , Rodrigo . The Political Right and Party Politics . In: AMES , B. ( Org .) . Routledge Handbook of Brazilian Politics . Abingdon : Routledge , 2018 . pp. 251 – 268 .
• Rawls , John . Theory of Justice . Cambridge : Harvard University Press , 1971 .
• Robertson , Bruce . A theory of party competition . Hoboken : John Wiley & Sons , 1976 .
• Rocha , Camila . “Imposto é Roubo!” A formação de um contrapúblico ultraliberal e os protestos pró-impeachment de Dilma Rousseff . Dados: Revista de Ciências Sociais , v. 62 , n. 3 , pp. 1 – 42 , 2019 .
• Rosas , Guillermo . Issues, Ideologies, and Partisan Divides: Imprints of Programmatic Structure on Latin American Legislatures . In: KITSCHELT , H. et al . ( Org .) . Latin American Party Systems . Cambridge : Cambridge University Press , 2010 . pp. 70 – 95 .
• Rosas , Guillermo . The Ideological Organization of Latin American Legislative Parties: An Empirical Analysis of Elite Policy Preferences . Comparative Political Studies , v. 38 , n. 7 , pp. 824 – 849 , 2005 .
• Ross , Dorothy . The origins of American social science . Cambridge : Cambridge University Press , 1991 .
• Rothbard , Murray . Economic Thought Before Adam Smith: An Austrian Perspective on the History of Economic Thought . Auburn : Ludwig von Mises Institute , 2006 . v. I .
• Sallum Jr. , Brasilio . Governo Collor: o reformismo liberal e a nova orientação da política externa brasileira . Dados: Revista de Ciências Sociais , v. 54 , n. 2 , pp. 259 – 288 , 2011 .
• Sartori , Giovanni . Concept Misformation in Comparative Politics . American Political Science Review , v. 64 , n. 4 , pp. 1033 – 1053 , 1970 .
• Smith , Adam . An Inquiry into the Nature and Causes of the Wealth of Nations . Lausanne : MetaLibri , 2007 .
• Smith , Adam . The Theory of Moral Sentiments . Indianapolis : Liberty Fund , 1982 .
• Tulloch , Gail . Mill and Sexual Equality . Boulder : Lynne Rienner Publishers , 1989 .
• Turner , Rachel . Neo-Liberal Ideology: History, Concepts and Policies . Edinburgh : Edinburgh University Press , 2008 .
• Volkens , Andrea ; Bara , Judith . Presidential Versus Parliamentary Representation: Extending the Manifesto Estimates to Latin America . In: Volkens , A. et al . ( Org .) . Mapping Policy Preferences from Texts: Statistical Solutions for Manifesto Analysts . Oxford : Oxford University Press , 2013 . pp. 277 – 298 .
• Wolfe , Christopher . Liberalism at the Crossroads . Oxford : Rowman & Littlefield Publishers , 2003 .

Notes

• Dataset: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/3C84D6
• 1
  . Friedrich Hayek defended the absence of State power but was not an anarcho-capitalist. He proposed that public policies should pass a generalizability test, in which all individuals should equally benefit from State interventions (Hayek, 2011).
• 2
  . When collecting the texts, I could not find the manifesto of the Partido Liberal de Cuba . This case was excluded, explaining the use of nine parties only.
• 3
  . I translate the original text in Spanish to English every time I quote the manifestos.
• *
  I am thankful for the comments provided by Barry Ames, Nigel Ashford, Asbel Bohigues, Scott Morgenstern, Ferdinand Müller-Rommel, Christian Welzel, and the participants of the annual conference of the Latin American Studies Association (Barcelona, 2018), the Ohio Latin Americanist Conference (Columbus, 2018), the annual conference of the Postgraduates in Latin American Studies (Leeds, 2017), and Science Po’s Graduate Conference in Political Theory (Paris, 2016).

Publication Dates

• Publication in this collection
  01 Sept 2023
• Date of issue
  Jul-Sep 2024

History

• Received
  4 Jan 2021
• Reviewed
  7 June 2022
• Accepted
  15 July 2022