Introduction
It is often argued that enduring democratic rule is a necessary condition for parties to institutionalise, building strong territorial organisations, developing partisan ties in the electorate and acquiring a value of their own, independent from individual party leaders. According to this view, party institutionalisation should increase as democracies age and parties compete for voters’ support, election after election.Footnote 1 Over four decades into the Third Wave of democratisation, however, successful party-building efforts have been more of an exception than the rule in Latin America.Footnote 2
If party consolidation has been overall a relatively infrequent phenomenon in post-transition Latin America, cases of successful conservative party institutionalisation have been even rarer. Historically, right-wing elites have preferred to invest in non-partisan forms of political action, including state corporatism, non-partisan clientelistic networks and, at times, support for military coups.Footnote 3 This general trend of conservative party underdevelopment has persisted and even deepened after the democratic transitions.Footnote 4 Although the overthrow of democratic regimes no longer seems to be a feasible option for right-wing elites lacking strong parties, the upper classes that constitute the core constituency of the Right have often engaged in non-partisan strategies – such as lobbying elected officials and supporting populist candidates – that have been deleterious for the quality of democracy.Footnote 5
Comparative research has demonstrated that the Latin American Right was most successful at party-building in those instances where there existed one or more authoritarian successor parties (henceforth ASPs) that benefited from association with the previous authoritarian regime.Footnote 6 In many cases, conservatives inherited valuable resources from the dictatorship (e.g. a party brand or a territorial organisation), which helped them to survive and even thrive under democracy.Footnote 7
In this article, I build on the literature on ASPs to explain why strong and highly institutionalised conservative parties emerged after the democratic transition in some Latin American countries but not in others. By institutionalised parties, I mean parties displaying the following properties: i) permanent and stable organisations, embodying a stable and well-established set of decision rules and procedures (routinisation); ii) substantial commitment of party members to the organisation, beyond their own personal interests (value infusion); iii) strong party brand and consistent ideological positioning (external institutionalisation).Footnote 8
Although I agree with previous research in that parties’ roots in authoritarianism are an important element in the explanation of conservative party consolidation, I argue that authoritarian inheritance in isolation is insufficient to produce strong and enduring conservative parties. This is especially so because parties’ roots in the authoritarian past involve both benefits, in terms of access to resources, and reputational costs associated with the systematic violation of political and civil rights by the authoritarian regime. Regime insiders may prefer to obfuscate their authoritarian past and, thus, avoid such costs by defecting to new parties or by eschewing the construction of parties and party brands altogether, developing instead non-partisan, individualistic ties to voters.
The central claim of the article is that strong conservative parties were most likely to consolidate in the post-transition period where, in addition to inheriting valuable political assets from the dictatorship, they experienced ideologically driven, violent conflict involving the supporters and opponents of the dictatorship during their early years. Party-building incentives should be greatest when authoritarian incumbents perceive these conflicts as a threat to their values and ideals and to their own political survival, thus fostering strong bonds among party members.Footnote 9 Violent political conflict, such as civil war, persecution or assassination of party leaders, breeds retrospective loyalty and facilitates the development of durable and clearly distinctive identities based on members’ shared memories and history.Footnote 10 On the other hand, where political conflict during the dictatorship or the transition period did not play a relevant role in shaping members’ identities throughout the party's early years, right-wing political actors had much weaker incentives to invest in party-building and tended to opt instead for a short-term electoral survival strategy, which resulted in much lower levels of institutionalisation.
In terms of its scope conditions, the theory developed in this article applies to a specific subset of right-wing parties, created throughout the authoritarian regime and the transition period. Arguably, the latter face challenges and opportunities that are different from those faced by parties created many years after the transition to democracy (or in previous democratic eras). Thus, my theory does not apply to cases of conservative parties created after a new, post-transition party system emerged. In these instances, the opportunities and constraints faced by party founders tend to be substantially different, because once parties occupy the policy space and mobilise previously unattached voters (or rebuild partisan identities from previous democratic eras), new parties tend to have a harder time in obtaining electoral success.Footnote 11 Transition periods offer greater opportunities for political entrepreneurs, and especially so in those cases where authoritarian rulers succeeded in weakening or erasing previously existing partisan attachments.
Based on the above criteria, the argument developed in this article applies to right-wing parties created during the authoritarian regime or during the transition to democracy in countries such as Argentina, El Salvador, Bolivia, Chile and Brazil. The theory can account for variation in institutionalisation across ASPs − e.g. Alianza Republicana Nacionalista (Nationalist Republican Alliance, ARENA) in El Salvador, Acción Democrática Nacionalista (Nationalist Democratic Action, ADN) in Bolivia − and new right-wing parties lacking roots in the authoritarian past and in previous democratic periods − e.g. Partido Liberal (Liberal Party, PL) in Brazil, Unión del Centro Democrático (Union of the Democratic Centre, UCEDE) in Argentina.
In terms of empirical strategy, I rely on a most-similar systems comparative design. I compare two countries – Chile and Brazil – where conservative elites benefited substantially from authoritarian inheritance. These two cases are similar in that the military and its civilian supporters relied on the discretionary implementation of social funds by regime allies at the subnational level to solidify clientelistic networks that would later serve as the basis for the creation of competitive right-wing parties under democratic rule.Footnote 12 Also, in both Chile and Brazil former regime insiders were able to negotiate the transition in terms that were favourable to them and yet impose on the new democratic regime a series of institutional biases that secured the over-representation of conservative interests.Footnote 13
Despite these similarities, the degree of institutionalisation of the partisan Right substantially differed across cases. Whereas Chile's major ASP, the Unión Demócrata Independiente (Democratic Independent Union, UDI), was extremely successful in cultivating stable sources of electoral support, building an enduring organisation and developing a party brand, in Brazil the former supporters of the military regime failed to build strong parties. Instead, they dispersed across a myriad of small, poorly institutionalised party organisations.Footnote 14
The article contributes to a recent literature that emphasises the importance of the legacies of authoritarianism for party and party-system institutionalisation.Footnote 15 It also contributes to the literature on ASPs in two aspects. First, the article moves beyond the issue of electoral survival, focusing instead on party institutionalisation. Moreover, I rely on a multidimensional approach to the conceptualisation and measurement of party development that can adequately deal with the oft-observed unevenness in party-building processes. For instance, parties may create strong and stable organisations, and yet rely on weak brands.Footnote 16 Second, whereas the ASP literature lacks a clear set of hypotheses to explain why authoritarian inheritance leads to party-building and party institutionalisation in some cases but not in others, the theory developed here indicates that the presence/absence of foundational conflicts shapes to a great extent the impact of authoritarian inheritance on ASP institutionalisation.Footnote 17 Finally, because the parties of the Latin American Right have received scarce attention from scholars, and even less so in comparative perspective, the article makes an empirical contribution by comparing the trajectories of the major conservative parties of Brazil and Chile.Footnote 18
Defining Party Institutionalisation
Most scholars tend to agree that Samuel P. Huntington's definition has provided the basic foundation for all subsequent studies of party institutionalisation.Footnote 19 He defines institutionalisation as the process through which organisations acquire value and stability. Value infusion implies that party members will have a stake in the perpetuation of the party organisation, even when its original goals change or are met. Poorly institutionalised parties, on the other hand, are valued by their members only in instrumental terms. That is, members will consider the party valuable as long as it is useful in achieving specific goals.Footnote 20 In Huntington's terms, an organisation institutionalises when it develops a value of its own, independent of the tasks it may perform at any given time.Footnote 21
The structural dimension of institutionalisation refers to the scope, density and regularity of the interactions that constitute the party as a structure.Footnote 22 An institutionalised party should have a stable organisational structure that is present throughout the territory.Footnote 23 However, institutionalisation is not identical with the party's development in pure organisational terms. Rather, it requires the development of formal decision rules and participatory mechanisms to integrate members, factions and sub-units within the party structure, thus leading to routinisation of behaviour.Footnote 24 Routinisation implies that party leaders and members are constrained by a stable set of rules that cannot be easily changed or circumvented.Footnote 25
Vicky Randall and Lars Svåsand argue in favour of further differentiating institutionalisation into an internal and an external dimension.Footnote 26 The latter refers to the party's relations with voters and with the broader competitive environment. Building on Randall and Svåsand's conceptual scheme, Thomas Kestler et al. define external institutionalisation as the process through which parties differentiate themselves from other competitors, by occupying a clear position in the ideological space, and develop stable links to voters, based on party identification and ‘reification’.Footnote 27 The development of a distinctive party brand involves a process of ‘reification’, through which the party's existence becomes gradually established in public imagination.Footnote 28 Party brand development and programmatic consistency are closely related, because when a party adopts distinctive and consistent policy positions over time, voters are more likely to develop a clear image of their prototypical partisans and the brand becomes stronger. A strong brand, in turn, fosters party identification, by allowing voters to more easily evaluate to what extent they resemble the party's prototypical partisan.Footnote 29
To sum up, my definition of party institutionalisation very closely follows the conceptual scheme proposed by Randall and Svåsand and revised by Kestler et al. It is comprised of both an internal and external dimension. The internal dimension requires routinisation and value infusion. The external dimension requires that parties differentiate from their competitors through consistent ideological positioning and develop stable links to voters based on party identification and reification.
Political Conflict, Uneven Party Institutionalisation and the Political Right
In post-transition Latin America, party institutionalisation has often proceeded unevenly across parties, as the political Right lagged behind left-wing and centrist parties. In programmatic terms, right-wing parties across the region differ from both the Left and the Centre in that they are much more likely to defend neoliberal reforms and a smaller role for the state in the economic realm; at the same time they score high on moral conservatism and thus oppose progressive policies on homosexuality, divorce and abortion.Footnote 30
In some cases, however, conservative elites benefited from the legacy of military rule to build strong party organisations. In these instances, ASPs created by the incumbents of authoritarian regimes survived and even thrived after the transition to democracy.Footnote 31 One important element in the success of ASPs under democratic rule is authoritarian inheritance. That is, conservative elites may enjoy a substantial competitive advantage in the electoral arena by inheriting the political assets – party brand, territorial organisation, sources of funding and clientelistic networks – controlled by former authoritarian incumbents.Footnote 32 The comparative evidence gathered by Steven Levitsky, James Loxton and Brandon Van Dyck in their edited collection actually indicates that all cases of successful conservative party-building in post-transition Latin America resulted from authoritarian inheritance.Footnote 33
Authoritarian inheritance facilitates party consolidation because it provides party leaders with resources that enable the task of party-building. For instance, a conservative party inheriting a large territorial organisation comprised of several local party offices in all regions of the country would be required to invest much less time and resources to build a nationalised and stable organisational structure under democracy than any other party needing to build a new organisation from the ground up. Authoritarian inheritance can also facilitate the long-term task of building a strong party brand. When authoritarian regimes provide goods that people value, such as economic growth or political stability, they may be able to obtain broad popular support, despite political repression.Footnote 34 In other words, by associating themselves with the political and economic achievements of the dictatorship, ASPs can more rapidly build a distinctive party brand than any newly created conservative party lacking roots in the authoritarian past.Footnote 35
Although dictatorships vary widely in what concerns their performance and their ability to provide valuable political assets, ASPs will typically enjoy substantial competitive advantages at the critical moment of the democratic transition when new parties are being formed. These early advantages tend to cumulate over time, thus allowing former regime insiders to more effectively survive electoral competition from ideologically similar parties disputing the support of the same constituencies. However, although parties’ roots in authoritarianism are an important element in the explanation of successful conservative party-building, authoritarian inheritance in isolation is unlikely to produce strong parties. Former regime insiders may or may not mobilise the resources inherited from the authoritarian regime, such as a party brand and a territorial organisation, to create institutionalised parties under democratic rule.
To the extent that authoritarian regimes exclude the opposition from the state, either by not holding elections or by holding uncompetitive, incumbent-biased elections, the parties created by dictators are likely to monopolise access to patronage resources. For this reason, authoritarian ruling parties will often recruit their members disproportionately among office-seeking, pragmatic activists who strive for a political career or government patronage.Footnote 36 Due to their privileged access to state resources, ASPs will often emerge as opportunistic coalitions whose members are strongly motivated by the rewards of political office.Footnote 37 Moreover, parties’ roots in the authoritarian past involve both benefits, especially in terms of access to resources, and the ‘authoritarian baggage’, understood as the reputational costs associated with the political legacies of the dictatorship. In general, the systematic violation of political and civil rights by the authoritarian regime will likely foster strong negative feelings among voters, including among those who approve of the regime's economic achievements, therefore imposing relevant reputational and electoral costs on ASPs.Footnote 38 Former supporters of the dictatorship may prefer to avoid such costs by defecting from the regime party or by developing non-partisan ties to voters (e.g. by delivering particularistic, material benefits to local constituencies), especially when they expect to obtain short-term electoral gains.
In order to explain why right-wing elites in post-transition Latin America built strong parties in some cases but not in others, one needs to look at factors other than the availability of resources inherited from the dictatorship. I argue that ASPs are most likely to institutionalise when party members develop a deeply ingrained, common identity as a result of ideologically driven, violent conflict involving the supporters and opponents of the dictatorship at the time of the party's foundation or during its early years of existence. This hypothesis builds on the literature that has highlighted the key role of foundational political struggles in the development of strong political parties.Footnote 39 To the extent that ASPs are born and bred during dictatorships, political conflicts between the authoritarian elite and opposition forces tend to be crucial for the development of these parties. Party-building incentives will vary not only depending on the actual duration and intensity of conflict, but also, and even more importantly, depending on whether authoritarian incumbents believe that these conflicts represent a real threat to the values and ideals of the regime and to their own political survival. Writing on dominant party endurance in Africa and Asia, Dan Slater and Nicholas Rush Smith make a similar point: elites’ disposition to engage in counter-revolutionary party-building efforts is conditional on the perception that those engaging in contentious mobilisation pose a credible threat to their elite status.Footnote 40
Violent political conflict is especially effective in producing retrospective loyalty and in creating durable and clearly distinctive partisan identities because, very often, it involves trauma, understood as a dramatic political experience shared by party members, such as persecution or assassination of party leaders. Trauma fosters members’ sense of common identity and it facilitates cooperation among individuals, while at the same time reducing the potential for defection.Footnote 41
Overall, where party members have experienced violent, polarising conflict, they are more likely to frame choices about cooperation or defection in terms of loyalty rather than material calculations.Footnote 42 Moreover, polarising conflict hardens partisan boundaries and fosters ideologically based voting.Footnote 43 In these settings, defectors have a greater likelihood of being punished by voters and, therefore, regime insiders have greater incentives to stick with ASPs and invest in their consolidation.Footnote 44
In contrast, where foundational, early political struggles were absent or they did not involve ideologically driven conflict and/or violent contestation perceived as threatening to authoritarian incumbents, one would expect former regime insiders to face much weaker incentives to invest in party-building. In these settings, the parties created by authoritarian regimes tended to recruit mostly careerist politicians motivated by the rewards of political office rather than ideologically motivated activists. Therefore, authoritarian elites were likely to pursue a short-term electoral survival strategy, avoiding both the costs of direct association with the military regime and the costs of long-term party-building.
Conservative Party Institutionalisation and ASPs in Chile and Brazil
I apply the theoretical framework outlined in the previous section to further an understanding of the distinct trajectories of right-wing parties in Chile and Brazil. The party systems that emerged in these two countries after the democratic transition have been among the most stable in Latin America, different from the many instances of party-system erosion and breakdown throughout the region.Footnote 45 From 2014 onwards, however, levels of electoral volatility have substantially increased in Brazil. In Chile levels of volatility have been on the rise since 2017.Footnote 46 Chile and Brazil also differ in that some of the major Chilean centre and left-wing parties are extremely old organisations that emerged before the Second World War, whereas virtually all the largest Brazilian parties did not exist previous to the authoritarian regime (1964–85).
In spite of these differences, the major right-wing parties in both countries are new parties, in the sense that they organised throughout the final days of the dictatorship and the transition period and they developed, to a substantial extent, independently from the partisan loyalties that existed in previous democratic eras. Moreover, in both Brazil and Chile the political Right benefited extensively from association with the authoritarian regime.
Brazil's bureaucratic-authoritarian regime was peculiar in that the military decided to maintain a façade of political pluralism by creating a two-party system formed by ARENA, which represented the authoritarian government, and the opposition Movimento Democrático Brasileiro (Brazilian Democratic Movement, MDB). Institutional rules were heavily biased in favour of the pro-military ARENA, which left the MDB with the role of a party of largely symbolic protest.Footnote 47
During the democratic transition, the party of the military government, now called Partido Democrático Social (Social Democratic Party, PDS) split in two due to factional disputes on the nomination of the regime's presidential candidate in indirect elections to be held in 1985. A dissident PDS faction comprised of governors from the Northeast region created the Partido da Frente Liberal (Liberal Front Party, PFL). The PDS and the PFL not only survived democratic competition, but they also succeeded in electing large delegations to the two chambers of the Brazilian Congress at least until the early 2000s.Footnote 48
Different from the Brazilian case, Chile's authoritarian regime banned parties and elections. In the early 1980s, however, General Pinochet started working with his closest allies in order to build a political party that could defend the regime's moral and economic values. In 1983, Jaime Guzmán, a former political aide to Pinochet, created the UDI. The party originated from the gremialista (trade unionist) movement led by a group of conservative intellectuals who emerged from a student movement at the Catholic University of Chile.Footnote 49
Since the early years of democratic rule, the UDI has disputed the conservative vote with the Renovación Nacional (National Renovation, RN), which was created in 1987. Although the RN did benefit from authoritarian inheritance in that several of its elected representatives had been mayors under military rule, it had a much weaker connection to the military regime than the UDI, and its leaders had a more ambiguous position with regard to the legacies of the dictatorship.Footnote 50 Moreover, part of the early generation of RN leaders comprised former members of the National Party that competed in elections in the 1960s and early 1970s; thus, the party had roots in the previous democratic period as well.Footnote 51
Measuring the Institutionalisation of the Partisan Right: Cross-Country and Within-Country Comparisons
Below I present party institutionalisation scores for each of the major right-wing parties in Brazil and Chile. I also calculated scores for the major left and centre parties in both countries, to allow for inter-bloc comparisons. By ‘major parties’, I mean all parties that obtained, on average, at least 3 per cent of the national vote in the four lower-chamber elections held immediately after the transition to democracy.Footnote 52 The ideological classification of parties was created by using the 2006, 2011, 2015 and 2018 waves of the Political Representation, Executives and Political Parties Survey (PREPPS).Footnote 53 I obtained data for the period previous to 2006 by relying on the Brazilian Legislative Surveys (BLS) and on Andy Baker and Kenneth F. Greene's extended ideological classification of Latin American parties.Footnote 54 I classified parties by defining equally sized, seven-point intervals for both the Left and the Right categories within the 20-point scale created by Nina Wiesehomeier and Kenneth Benoit.Footnote 55 The Centre was defined as a residual category. Parties were classified according to their mean ideology scores over time.
I estimated scores ranging from zero to one for each of the dimensions of institutionalisation and for each of the major left, right and centre parties. I measured the internal dimensions of institutionalisation – value infusion and routinisation – relying on the comparative dataset created by Nicole Bolleyer and Saskia P. Ruth, based on the Democratic Accountability and Linkages Project (DALP).Footnote 56 I extended their original operationalisation by calculating separate measures to account for brand recognition and parties’ programmatic consistency. Aggregate external institutionalisation was calculated as a simple mean of the scores obtained in each of these dimensions. I operationalised programmatic consistency by relying on DALP survey questions on parties’ positions on distinct issue dimensions. Brand recognition was calculated by relying on survey data on party identification. The operationalisation of each dimension is explained in detail in the Appendix.
A summary institutionalisation score was calculated for each party as the arithmetic mean of three dimensions: value infusion, routinisation and external institutionalisation. Results for right-wing parties in Brazil and Chile are presented in Table 1. Recall that Brazil's ASPs have changed their original names. The PFL went through rebranding in 2007 to become the Democratas (Democrats, DEM). The PDS experienced a series of fusions with other right-wing parties before it adopted its current name in 2003: Partido Progressista (Progressive Party, PP). Throughout the article, I utilise the original names when I analyse the circumstances of their foundation and initial development. Otherwise, I use the original names followed by the most recent denominations to refer to these two parties (e.g. PFL/DEM). The same procedure is applied to the PL, which changed its name to Partido da República (Party of the Republic, PR) in 2006.
Table 1. Brazil and Chile: Party Institutionalisation Scores for Major Right-Wing Parties
Sources: Author's elaboration based on DALP 2008/2009; Bolleyer and Ruth, ‘Elite Investments in Party Institutionalization in New Democracies’; Latin American Public Opinion Project (LAPOP) 2006; Centro de Estudios Públicos (CEP) 2005.
Note: a indicates an ASP.
As seen in Table 1, Brazil's largest right-wing parties are, on average, substantially less institutionalised than the parties of the Chilean Right. Within the group of ASPs, the UDI exhibits the highest aggregate institutionalisation score (0.68), with a difference of 0.17 and 0.34 points compared to the PFL/DEM and the PDS/PP respectively. Brazil's ASPs obtain some of the lowest scores in the value infusion dimension, a pattern similar to that observed for the PL/PR. These results indicate that the partisan Right in Brazil has mostly failed to create organisations that have a value of their own, independent from any specific tasks they may perform.
There are also substantial differences between Chilean and Brazilian right-wing parties in what concerns the development of distinctive party brands. The extremely low scores obtained on the brand recognition dimension indicate that the political Right in Brazil has failed to cultivate mass partisan attachments. In Chile, although voters’ identification with all parties has steeply declined since the return to democracy, both the RN and the UDI have been able to count on the support of a small yet significant core of partisan voters.Footnote 57 Note, however, that the high score calculated for the RN on brand recognition likely reflects the fact that the party inherited the rural strongholds of the old National Party and does not necessarily indicate higher elite investment in brand development in the recent democratic period.Footnote 58
Table 1 also indicates that the institutionalisation of the Brazilian Right has been rather uneven. In particular, Brazil's major ASP, the PFL/DEM, scores substantially better in the routinisation dimension than in any other dimension. The score of 0.81 is the second-highest among all parties selected for analysis. The UDI is also more institutionalised in the routinisation dimension, but the differences between the latter and the other dimensions are not as substantial.
One potential difficulty involved in the comparison of levels of party institutionalisation across countries is that part of the variation shown in Table 1 may relate to fixed or mostly invariant country characteristics, such as the electoral system and other institutions that may either increase or decrease politicians’ incentives to build strong parties. For instance, it could be argued that the differences in the institutionalisation of the partisan Right in Brazil and Chile reflect broader differences in the institutionalisation of the party system as a whole. In particular, one influential interpretation of Brazil's party system argues that the country's institutional mix, which combines robust federalism and open-list proportional representation with extremely high district magnitudes, fragments the party system and fosters individualistic campaign strategies.Footnote 59 In Chile, in contrast, the centrifugal tendencies associated with open-list proportional representation that are detrimental to party development would likely be reduced by the binominal system that existed until 2015 and by the lack of politically autonomous state governments.
If this interpretation were correct, we should not find substantial within-country differences in levels of party institutionalisation. That is, if the political Right is poorly institutionalised in Brazil because the electoral system and federal arrangements are inimical to strong parties, then left and centre parties should also exhibit similarly low levels of institutionalisation. Moreover, within-country differences should be much less substantial than cross-country differences in institutionalisation.
Figure 1 shows the mean institutionalisation scores in each dimension for the major right, left and centre parties in Brazil. The left parties included in the analysis are the Partido dos Trabalhadores (Workers’ Party, PT), the Partido Socialista Brasileiro (Brazilian Socialist Party, PSB) and the Partido Democrático Trabalhista (Democratic Labour Party, PDT). The centre bloc includes the Partido da Social Democracia Brasileira (Brazilian Social Democracy Party, PSDB) and the Partido do Movimento Democrático Brasileiro (Brazilian Democratic Movement Party, PMDB).
Figure 1. Brazil: Mean Party Institutionalisation Scores by Ideological Bloc
Sources: Author's elaboration based on data from DALP (2008/2009); Bolleyer and Ruth, ‘Elite Investments in Party Institutionalization in New Democracies’ (2018); LAPOP (2006); CEP (2005).
Figure 1 reveals that Brazil's major right-wing parties are, on average, substantially less institutionalised than both centre and left parties. These differences are especially large regarding the value infusion dimension. When one looks at individual scores for Brazilian parties (presented in Table A.1 of the Appendix), the PT appears as the most institutionalised party, which is consistent with most scholarly evaluations. The parties of the Right, on the other hand, figure among the least institutionalised.
Figure 2 presents average institutionalisation scores for the major Chilean parties, grouped by ideological bloc. The centre parties included in the analysis are the Partido Demócrata Cristiano (Christian Democratic Party, PDC), the Partido por la Democracia (Party for Democracy, PPD) and the Partido Radical Socialdemócrata (Radical Social Democratic Party, PRSD). Because the DALP survey did not include the Chilean Communist Party, data for the left bloc refers only to the Partido Socialista (Socialist Party, PS).
Figure 2. Chile: Mean Party Institutionalisation Scores by Ideological Bloc
Sources: Author's elaboration based on data from DALP (2008/2009); Bolleyer and Ruth, ‘Elite Investments in Party Institutionalization in New Democracies’ (2018); LAPOP (2006); CEP (2005).
Figure 2 shows that average levels of institutionalisation for Chilean right-wing parties are close to country mean scores. In the case of Brazil, in contrast, the political Right scores below the party-system average on all dimensions, as seen in Figure 1. The PS, which is the only party in the Left category, is the most institutionalised party in Chile. This is not too surprising considering the PS is the second-oldest Chilean party.Footnote 60 On average, Brazil's centre and left parties obtain aggregate institutionalisation scores similar to country mean scores estimated for Chile. This indicates that cross-country differences in institutionalisation are not especially substantial when one controls for ideological bloc. In particular, because Brazil's right-wing parties are significantly less institutionalised than both left and centre parties, we can rest assured that differences in the institutionalisation of the political Right in Brazil and Chile are not a mere result of system-level differences associated with electoral and federal institutions.
Political Conflict, Authoritarian Inheritance and Conservative Party-Building in Chile and Brazil
I argue that differences in institutionalisation among the major right-wing parties in Brazil and Chile are very strongly related to the interplay of authoritarian inheritance and levels of perceived and actual political conflict throughout parties’ foundation and early years. Although it is hard to operationalise the saliency of conflict along the authoritarian−democratic divide for parties’ early development, I account for systematic differences across cases by looking at : i) the occurrence of violent, ideologically motivated political conflict throughout the party's foundation and early years; ii) authoritarian elites’ motivation to create the party; iii) the positions adopted by party members in its early years with regard to the authoritarian−democratic divide. When ASPs are founded in a context of polarising, violent conflict and, in addition, the foundation of the party is directly or indirectly related to conflict between the opponents and supporters of the authoritarian regime, one would expect the authoritarian−democratic divide to be highly relevant for defining the party's identity and values. Moreover, when party members adopt clear and consistent positions in favour of defending the legacies and the values of the authoritarian regime, thus differentiating clearly from the opposition, this is further evidence of the saliency of the authoritarian−democratic cleavage for the party's early development.
In the Brazilian case, the electoral successes of the opposition and the regime's decaying legitimacy triggered a slow and carefully controlled political liberalisation, starting in the Geisel presidency (1974–9).Footnote 61 Brazil's liberalisation culminated in a transition to civilian rule through an indirect presidential election in 1985, within an electoral college dominated by conservatives. Despite the massive demonstrations that took to the streets of Brazil's major cities between 1983 and 1984, a constitutional amendment that would have instituted direct presidential elections by November 1984 did not pass in the lower chamber.Footnote 62
Brazil's ASPs emerged mostly because of factional disputes within the pro-military party (the PDS) regarding the nomination of the regime's presidential candidate. A dissident PDS group, formed mostly by governors from Northeastern states, created the Partido da Frente Liberal (Liberal Front Party, PFL) in reaction to the nomination of Paulo Maluf as the PDS candidate.Footnote 63 Relying on archival data and interviews, Ricardo Luiz Mendes Ribeiro found that a central motivation for the creation of the PFL was the fact that Maluf, in order to be nominated as the PDS presidential candidate, bypassed state governors and state party leaders, and bargained directly with the delegates that would vote in the national convention. Fearing that Maluf's victory in the electoral college would threaten their control over regional party branches, various PDS governors decided to leave the party and work together to defeat the regime's presidential candidate.Footnote 64
The PFL founders sought to distance themselves from the authoritarian regime by sealing the so-called ‘Democratic Alliance’ in support of the opposition candidate in the 1985 indirect presidential elections, Tancredo Neves.Footnote 65 As it became clear that the PMDB−PFL alliance had the votes to win, politicians affiliated with the party of the military regime migrated massively to the PFL and even to the PMDB. In the first years of the democratic regime, the authoritarian elite dispersed across several right-wing parties, although a substantial share of the ARENA cohort chose the PFL.Footnote 66
Overall, the organisation of Brazil's ASPs occurred in a context of low levels of conflict between authoritarian incumbents and the opposition to the regime. Demonstrations in favour of direct elections in 1984 were mostly pacific and the few episodes of political violence throughout the transition resulted mostly from military hardliners’ resistance to the opening of the regime.Footnote 67 Whereas in Chile left-wing parties had created strong roots in society, mobilising voters along class lines before the 1973 coup, the Brazilian Left was rather weak in organisational terms, and especially so outside the major urban centres.Footnote 68 Moreover, the major left parties in Brazil were rather moderate in comparison with Salvador Allende's Unidad Popular (Popular Unity) coalition, and Marxist parties obtained only a tiny portion of the national vote by the early 1960s. By the mid-1970s, left guerrillas had been decimated or demobilised by the Brazilian military regime. The legal opposition organised within the MDB (later renamed PMDB) was largely comprised of moderate politicians.Footnote 69
The authoritarian elite never perceived the opponents of the dictatorship as representing an actual challenge to their fundamental interests throughout the transition process.Footnote 70 Indeed, by the end of the regime in the early 1980s, military strategists were convinced that the return to multipartyism would fragment the opposition among several small parties, whereas pro-regime forces would hold together, thus favouring continuity between authoritarian and civilian rule.Footnote 71
The unimportance of ideological conflicts between authoritarian incumbents and the opposition in shaping the identity of Brazil's major ASP – the PFL – is evident in the fact that the party's initial manifestos emphasised its commitment to democratic values and principles, and to the full restoration of democracy, in a clear attempt at denying the party's authoritarian roots.Footnote 72 But the party's programme was inconsistent with its actual behaviour in the Constituent Assembly (1987–8), when the PFL and other conservative parties supported controversial military policies such as unpublished decrees and a blanket refusal to acknowledge human-rights abuses after 1964.Footnote 73 This inconsistency between the party's programme and its legislative behaviour was also evident in the case of the PDS/PP. Although the PDS/PP opposed the PMDB−PFL government (1985–90) after being defeated in indirect presidential elections, it pursued a systematic strategy of ideological obfuscation by adopting programmes and manifestos that belied its conservative and authoritarian roots.Footnote 74
In the case of Chile, the Chilean Communist and Socialist parties had been very successful in building a strong grassroots organisation in poor urban neighbourhoods that proved to be extraordinarily resilient throughout the dictatorship.Footnote 75 Although party competition in post-transition Chile has been mostly centripetal, the party system organised around a regime cleavage opposing supporters and opponents of Pinochet, whereas in Brazil such divisions were largely absent.Footnote 76
By the early 1980s, the regime faced a major economic crisis: in 1982 and 1983, GDP decreased 13.4 per cent and 3.5 per cent respectively. This severe economic downturn led to the eruption of mass protests against the Pinochet government that were brutally repressed.Footnote 77 The active participation of the Marxist Left in the organisation of the protests challenged the military's claim to have reshaped loyalties in Chile. This period also witnessed an upsurge in left insurrectionary activities, as the Communist Party decided to join forces with the members of the urban guerrilla group Movimiento de Izquierda Revolucionaria (Movement of the Revolutionary Left, MIR).Footnote 78
It was in this context of economic crisis, reorganisation of the opposition and violent political conflict that the gremialista movement decided to create the UDI. The founders of the UDI understood formation of their party as an essentially defensive act. UDI party leaders interviewed by Loxton said they were convinced, at the time, that the project of the military regime was under threat, due to Pinochet's decision to introduce more statist economic policies and initiate a political opening in response to the economic crisis of 1982–3.Footnote 79 UDI founders strongly defended the continuity of the regime's neoliberal economic policies, and they were in favour of a transition to a ‘protected democracy’ that would secure military tutelage over civilian authorities. Thus, the UDI's creation was understood by party founders as a reaction against a series of political and economic events that threatened the basic values of the gremialistas and their own political survival. They feared that political liberalisation would result in a return to the pre-coup party system, an outcome that was unacceptable in their view.Footnote 80
Societal polarisation and ideological conflict shaped the UDI's identity to a very substantial extent. The party's founders viewed themselves as ‘political warriors’ fighting against Marxism. The killing of Senator Guzmán by leftist extremists in 1991 reinforced this self-perception.Footnote 81 Because Guzmán had a crucial role in establishing the party's worldview, his assassination marked a whole new generation of UDI activists. This shared, traumatic experience contributed to reinforce members’ loyalty to the party, fostering a sense of common identity.Footnote 82
Whereas the UDI's early development was strongly shaped by political violence and by ideological conflict opposing supporters and opponents of the authoritarian regime, these factors had little if any relevance in the case of the RN. The party was first organised as a political front that included several right-wing groups in preparation for the 1988 plebiscite that would decide on the continuity of Pinochet's rule for an additional term.Footnote 83 The right-wing coalition included the UDI, the Frente del Trabajo (Workers’ Front) and the Movimiento de Unión Nacional (National Unity Movement, MUN). The RN was officially created in 1987 after the UDI abandoned the right-wing front. Although several of the early-generation members of the RN had occupied government positions during the dictatorship, the party, as already mentioned, had roots in the previous democratic period, as some of its leaders were remnants of the National Party.
Whereas UDI founders refrained from accepting any major changes to the 1980 Constitution throughout the transition, the RN was much more willing to negotiate with opposition parties.Footnote 84 The UDI and RN clearly differed in terms of their initial platforms: whereas the former party fully defended Pinochetismo and its legacies, the latter had a much more critical position toward the regime's political and human-rights records, thus adopting a more ambiguous position regarding the authoritarian−democratic divide.Footnote 85 Not only was the ‘cultural war’ against Marxism relatively unimportant in defining the RN's identity but also the party leaders and the rank and file never experienced trauma.Footnote 86
Despite their connections to the Pinochet regime, RN leaders did not perceive the reorganisation of the political Left during the 1980s as a threat to their own political survival in the same way as UDI founders did. The crucial role of counter-revolutionary struggles in the formation of the UDI's early identity, which was reinforced by the killing of Guzmán in 1991, allowed the party to develop a highly cohesive organisation, whereas the RN lacked such a source of cohesion.
Conservative Party Institutionalisation and the Long-Lasting Impact of Party Origins and Early Trajectories
In both Brazil and Chile, the political Right has obtained substantial electoral success after the transition to democracy, obtaining national vote shares ranging from roughly 30 per cent to 50 per cent. Still, in the Brazilian case the electoral performance of the right-wing bloc is largely unrelated to the consolidation of right-wing parties in the electorate. In other words, conservative politicians as individuals have achieved substantial success in the electoral arena despite the persistent weakness of the partisan Right. Figure 3 shows the evolution of the national vote share of the two largest right-wing parties in Brazil and Chile since the time of the transition. These are the PFL/DEM and PDS/PP for Brazil and the UDI and RN for Chile.Footnote 87
Figure 3. Largest Conservative Parties’ Share of the National Vote in Lower-Chamber Elections, Brazil and Chile, 1986–2018, by Percentage
Sources: Author's elaboration based on data from Tribunal Superior Eleitoral (www.tse.jus.br); Servicio Electoral de Chile (www.servel.cl).
Figure 3 clearly reveals the electoral decay of Brazil's major right-wing parties in the last 20 years and, in particular, the huge electoral losses suffered by the PFL/DEM from 2002 onwards. In the case of Chile, longitudinal and cross-party variation in electoral results notwithstanding, the UDI and RN have consolidated as the most important parties of the Right in the recent period.
Among Brazil's two ASPs, the PFL/DEM was the major beneficiary of authoritarian inheritance, in that the party absorbed most of the territorial organisation and party cadres of the pro-military ARENA. According to Timothy Power, ten years after the transition ex-Arenistas represented an impressive 60 per cent of the PFL delegation in the Chamber of Deputies, whereas only 31 per cent of the PDS delegation was comprised of former members of the party of the military regime.Footnote 88
Although the political assets provided by authoritarian inheritance greatly facilitated the task of party-building, the fact the party emerged mostly from factional, unideological conflicts reduced the incentives to strengthen the party brand. Rather, the PFL leadership prioritised the short-term goals of maximising votes and winning office. As the party sought to obfuscate its authoritarian past by joining the Democratic Alliance and supporting the Sarney government (1985–90), ex-Arenistas enjoyed continued access to valuable patronage resources that could be used to grease their subnational political machines.Footnote 89
In stark contrast with the PFL/DEM, the UDI opted for embracing the past and building a distinctive party brand around the legacy of Pinochetismo. In the transition election, the party supported the presidential candidacy of Pinochet's minister of finance, despite the fact that his association with the regime's policies was seen as a liability, including by RN leaders who were defeated in the negotiation of a pre-electoral pact around a single right-wing candidate.Footnote 90 Although the UDI gradually moderated its discourse by downplaying association with the authoritarian regime, the party's identity was largely built around the regime cleavage opposing supporters and opponents of the dictatorship, which allowed it to obtain consistent support from voters that were most sympathetic to Pinochet and to the military regime.Footnote 91
Whereas the UDI succeeded in unifying its leaders and rank and file around a common set of values and principles, and in differentiating itself from other parties, the Brazilian Right was clearly unsuccessful at creating strong party brands and infusing parties with a value of their own. In the case of the PDS/PP, in particular, the party leadership deliberately sought to dilute the party brand inherited from the authoritarian regime by adopting programmes and manifestos that appeared far more progressive than the party ever was. Power found that among the major Brazilian parties, the PDS/PP was characterised by the largest differences between the party's ideological position, according to elites’ perceptual data, and the content analysis of the Manifesto Research Group.Footnote 92 In a clear demonstration of the irrelevance of the authoritarian−democratic divide in Brazilian politics, the PDS/PP supported the PT administrations headed by presidents Lula (2003–10) and Dilma Rousseff (2011–16), both of whom had been active members in the opposition to the military.
The PFL, on the other hand, struggled to survive while in opposition to the PT governments. In an attempt at repositioning the party in the electoral market, the PFL leadership hired a team of political consultants to prepare a rebranding strategy in 2007. As a national survey showed that voters associated the party with the military regime and old-style, clientelistic politics, the national executive committee of the party decided to adopt a new name: Democrats (DEM). But rebranding efforts largely failed and did not prevent the party's electoral decline in subsequent elections.Footnote 93 Without the federal patronage resources that had kept the party factions together until then and suffering from chronically low levels of value infusion, the party was unable to avoid massive defection.Footnote 94
In comparison with both the PDS/PP and the smaller right-wing parties that emerged during the transition – the PL/PR and the PTB – the PFL/DEM has been more successful in the construction of a highly nationalised, centralised and stable organisation.Footnote 95 As seen in Table 1, the PFL/DEM has the highest score in the routinisation dimension among these four parties. This indicates that authoritarian inheritance has likely increased the party's ability to institutionalise in the organisational dimension. Note that both the PL/PR and the PTB received a smaller influx of ex-Arenistas in the aftermath of the transition as compared to both the PFL/DEM and the PDS/PP, and they lacked a strong connection with the authoritarian regime. Moreover, the party split that led to the creation of the PFL allowed the latter party to benefit more decisively from authoritarian inheritance than the PDS/PP.
A comparison between the UDI and RN reveals a similar pattern. The RN benefited to a much lower extent from authoritarian inheritance, as it lacked the organic connections with the authoritarian regime that characterised the UDI. Since its foundation, the RN has been divided between a liberal faction, who advocated a ‘progressive centrism’, dissociated from the authoritarian regime, and hardline conservatives who had been appointed to top government positions during the dictatorship. The party has mostly failed to develop stable and effective decision-making rules, and it has lacked in intermediate organisations between voters and the national party executive.Footnote 96 In contrast, the UDI has taken advantage of the territorial organisation and clientelistic networks inherited from the authoritarian regime to secure a strong presence at the grassroots level.Footnote 97
The ambiguous position of the RN regarding the legacies of the authoritarian regime has been detrimental to the construction of a distinctive, shared ideological identity that might have induced members to develop strong bonds to the organisation. As ideological conflicts and political violence played no role in shaping the party's early development, the RN has remained susceptible to factional struggles and defection. According to Fernando Rosenblatt, the RN was the Chilean party that suffered the largest number of defections between 1989 and 2009.Footnote 98 In sum, although the Chilean political Right did not experience the extreme fragmentation of Brazil's conservative parties, the RN was more similar to Brazil's ASPs than to the UDI, to the extent that members’ loyalty to the party has remained low and the party has regularly suffered with defection.
Conclusion
Within- and cross-country variation in party trajectories is largely consistent with the central claim of the article, namely that strong right-wing parties are more likely to emerge after a transition to democracy in the presence of two conditions: i) the party benefits extensively from authoritarian inheritance; ii) the party experiences violent and ideologically driven conflict that is perceived by its leaders as threatening to their values and/or to their own political survival during its early years.
Authoritarian inheritance in both Chile and Brazil was associated with higher levels of organisational consolidation or routinisation. But the comparative evidence also indicates that strong roots in the authoritarian past do not necessarily lead to value infusion and external institutionalisation. Chile's UDI is the only right-wing party in our sample of six parties to have succeeded in obtaining medium to high scores in all dimensions of institutionalisation. Although the UDI and the PFL/DEM are similar in what concerns authoritarian inheritance, they critically differ regarding the presence/absence of early political conflicts perceived as threatening by authoritarian incumbents. The case of the UDI is thus congruent with the claim that conservative parties are most likely to institutionalise when they inherit substantial resources from the previous authoritarian regime and, also, their identities are forged throughout violent, ideologically driven conflict. In the absence of these conditions, right-wing parties tend to institutionalise rather unevenly, as revealed by the cases of the PFL/DEM and RN.
The RN did not inherit a large territorial organisation from the authoritarian regime as did the PFL/DEM (or the UDI), and it obtained lower scores than both of these parties in the routinisation dimension. Still, it might be the case that the RN and the UDI inherited a somewhat stronger brand than Brazil's ASPs. Indeed, the Chilean economy experienced high economic growth during the final years of the regime – GDP growth averaged 6.4 per cent between 1984 and 1989 – whereas the Brazilian dictatorship ended amidst a deep economic crisis that combined a 10 per cent decline in per capita income and high inflation between 1983 and 1984.Footnote 99 Note, however, that the long-term performance of the Brazilian authoritarian regime was substantially better than that of the Chilean dictatorship.Footnote 100 Moreover, there is no evidence that citizens’ identification with Brazil's ASPs was substantially lower in comparison with Chilean ASPs in the initial years of democratic rule. According to national surveys conducted by the Instituto Brasileiro de Opinião Pública e Estatística (Brazilian Institute of Public Opinion and Statistics, IBOPE) in Brazil and by the Centro de Estudios Públicos (Centre for Public Studies, CEP) in Chile two years after the democratic transitions (1987 and 1991), 4.7 per cent and 3.6 per cent of Brazilian voters identified with the PDS and the PFL, respectively, whereas 8.1 per cent and 3.5 per cent of Chilean voters identified with the RN and the UDI, respectively.Footnote 101 In both Brazil and Chile, ASPs counted on a much smaller partisan base than the parties that had opposed the military regime, such as the PDC (37.5 per cent) and the PMDB (25 per cent), according to the same opinion polls. In sum, in what concerns the strength of their brands relative to major competitors, one can say that ASPs in Brazil and Chile faced similar challenges in their early years.
The case of Chile provides evidence of the ambiguous impact of parties created under authoritarianism on democratic politics. On the one hand, when these parties institutionalise and become major players in the post-transition party system, they are likely to foster continuity between authoritarian and democratic rule. Consistent with this view, the Chilean transition led to consolidation of an ‘elite-biased’ democracy, characterised by the persistence of authoritarian enclaves and by severe institutional obstacles to large-scale redistribution.Footnote 102 On the other hand, once former authoritarian incumbents succeed in constructing highly institutionalised and stable parties, they tend to play an important role in solidifying and structuring the party system.Footnote 103 Moreover, the consolidation of ASPs may favour democratic stability by leading the former supporters of the dictatorship to have a stake in the preservation of the rules of the democratic game. Indeed, the successful trajectory of the UDI resulted in the incorporation of the authoritarian sectors of the political Right, at both the elite and mass levels, into electoral politics.
Where former regime insiders fail to build institutionalised parties, they may resort to non-partisan forms of political action that are, more often than not, detrimental to the quality and stability of democracy and to the consolidation of the party system, as demonstrated by the case of Brazil. The persistent weakness of the partisan Right and its shallow roots in society have created opportunities for conservative populists supported by personalistic electoral vehicles, with deleterious consequences for party-system consolidation. The two most successful presidential candidates of the Right – Fernando Collor and Jair Bolsonaro – relied on an anti-system and moralising rhetoric against traditional politicians and their corrupt practices to win the presidential race in 1989 and 2018, respectively. The election of authoritarian populist and former army captain Bolsonaro in 2018 is especially telling of the persistent and unresolved legacies of Brazil's dictatorship, as his campaign was explicit in the celebration of Brazil's authoritarian past and he mobilised the support of politically intolerant voters, with a low commitment to democratic norms and institutions.Footnote 104
Overall, ASP institutionalisation matters for the quality of democracy and for party-system stability, and this is especially the case in the Latin American context, where authoritarian regimes, with few exceptions, have been established by the most powerful elites and social forces that constitute the core of the political Right.
Acknowledgements
This research benefited from funding provided by the Federal District's Research Foundation (FAPDF) in support of a visiting scholar appointment at the Centre for Iberian and Latin American Studies at the University of California, San Diego (August 2019–February 2020). I am grateful to the three anonymous reviewers and to JLAS editors for providing very extensive and detailed comments throughout the peer-review process. I am also thankful to Lucio Rennó, Adrián Albala, Fernando Bittencourt and Gabriel Squeff for their criticisms and suggestions on a previous version of this article presented at the University of Brasília in March 2020.
Appendix
Measuring Party Institutionalisation in Brazil and Chile
The measures of value infusion and routinisation were obtained from Bolleyer and Ruth (2018). They used two questions from the 2008/2009 DALP survey to measure the strength of party organisations. The questions (A1 and A3) focus on the existence of permanent local offices and local party intermediaries, accounting for both formal and informal organisation. Higher expert ratings indicate that parties have permanent local offices with paid staff in most districts and that they also rely on local intermediaries to maintain contact with voters, organise electoral support and voter turnout, and distribute party resources in most constituencies. The routinisation index was constructed by normalising each indicator between zero and one, with higher values indicating higher levels of routinisation. Mean expert ratings were obtained for each party and each indicator. The final score is a simple mean of the measures of formal and informal organisation.
Bolleyer and Ruth (2018) measured value infusion by utilising a question from the DALP expert survey (question E4) on the importance of the party's origins or achievements of historical leaders, or references to party symbols and rituals for the party's campaign strategies. This indicator is also a mean of expert ratings normalised between zero and one.
Parties’ programmatic consistency was operationalised as the mean of two indicators. The first is a simple mean proportion of valid expert responses to a series of DALP survey questions on parties’ positions on five distinct issue dimensions. Theoretically, parties with consistent and easily recognisable positions should obtain a higher proportion of valid responses. The second indicator uses the mean standard deviation of experts’ ratings to each issue dimension for each party. These raw scores were normalised between zero and one by using the full distribution of responses to the DALP expert survey, which includes 88 countries and hundreds of parties worldwide. Finally, I subtracted the normalised scores from one, attributing higher (lower) values to parties with lower (higher) variation in experts’ ratings. A summary programmatic consistency score was obtained for each party by averaging these two indicators.
The strength of party brands in the electorate was calculated in a similar manner to that proposed by Basedau et al. (2000): I obtained the percentage of respondents who declared they sympathised with the party and then divided it by the party's vote in the lower-chamber race. To secure temporal consistency across indicators, I considered the lower-chamber elections held immediately before the DALP 2008/2009 expert survey. The intuition behind this operationalisation is that a higher partisans-to-voters ratio likely indicates that most of the party's voters do distinguish the party brand, instead of voting on the basis of alternative, non-partisan voter−candidate linkages.
I relied on party identification data obtained from the 2006/2007 LAPOP survey for Brazil and on the 2005 CEP opinion poll for Chile. These are the sources covering the greatest number of parties. Electoral results for the 2006 (Brazil) and 2005 (Chile) lower-chamber elections were obtained from Brazil's TSE (www.tse.jus.br) and from the Political Database of the Americas (pdba.georgetwon.edu). External institutionalisation was calculated as the mean of party scores on brand recognition and programmatic consistency.
Table A.1 presents a list of all parties included in the analysis and the respective scores for each dimension. Parties were grouped by country and then rank, ordered from the most institutionalised to the least institutionalised.
Table A.1. Brazil and Chile: Party Institutionalisation Scores for Major Parties
