What explains equity-enhancing reforms under centre-right governments? Evidence from Brazil

Abstract

Inequality levels declined across Latin America between the 1990s and 2010s, but research remains incipient on how equity-enhancing initiatives sometimes emerged in centre-right governments. This article checks which mechanisms enabled education reform and cash-transfer schemes during the PSDB-led administration in Brazil (1995–2002), considering electoral competition, left-wing legislative strength, social mobilisation, and coalition dynamics. By combining multiple data sources, it is possible to observe the president promoting his party’s electoral ‘brand’ and cultivating a large multiparty alliance as the equalising policies passed through Congress. Meanwhile, leftist legislators comprised a relatively weak group, and bottom-up pressures were almost non-existent. As validity checks, the study first engages with two alternative explanations: personal leadership and bureaucratic action. Second, it proposes applying competitive elections and cross-party cooperation as possible explanatory factors for the persistence of the two analysed redistributive programmes in the governments of the left-wing Workers’ Party (2003–2016) and the far-right Jair Bolsonaro (2019–2022).

Introduction

Between the late 1990s and early 2010s, many countries in Latin America experienced a significant reduction in inequality levels while income distributions became increasingly unequal almost everywhere else in the world (Birdsall et al., 2012). Following the development of more inclusionary welfare states (Levy and Schady, 2013), Gini coefficients declined across the region in cases of left- and right-led governments (López-Calva and Lustig, 2010). Although there is extensive literature on the political determinants of declining inequality in Latin American countries where left-wing parties were in power or on time-specific comparisons between the performances of left- versus right-leaning incumbents with respect to social policy outcomes (Cornia, 2010; Huber and Stephens, 2012; Mauro, 2022), fewer works have so far focused on right-of-centre administrations and how equity-enhancing reforms sometimes advanced in such contexts.

This article builds on existing case studies of Argentina (2015–2019), Chile (2010–2014), and Mexico (2000–2012), as conducted by Fairfield and Garay (2017) and Niedzwiecki and Pribble (2017), to check which enabling mechanisms may also be observed in the processes leading to the proposition and approval of two consequential redistributive policies in the crucial case of Brazil during the presidency of Fernando Henrique Cardoso (PSDB), who led a centre-right governing coalition between 1995 and 2002.

What would eventually become, in the 2000s, a substantial decrease in Brazil’s Gini coefficient received an essential push in the PSDB-led administration with the launch or consolidation of key equalising measures, such as macroeconomic stabilisation, the implementation of constitutional mandates to enact a public healthcare system and non-contributory pensions, as well as the institution of nationwide targeted cash-transfer schemes and new rules increasing federal spending on state (free) schools (Hagopian, 2019; Melo, 2008; Power, 2016).

While the PSDB occupied the centre of the political spectrum in the 1980s and early 1990s, the party steadily moved rightward during its time in office (Power and Zucco, 2009). It is broadly understood in the literature of social policymaking in Latin America and Brazilian politics that the governing coalition the party led sat on the centre-right once its main allies were representatives of long-dominant conservative elites (Garay, 2016: 120–164; Garmany and Pereira, 2018: 72). Skidmore (2010: 215) writes that Cardoso’s (PSDB) lasting partnership with the PFL party, first established in 1994 for that year’s presidential election, ‘would identify him with one of the most traditional (and one of the most reactionary) political images in Brazil, opening him up to fierce attacks from the left’.

The article analyses the policy cases of the 1996 education reform that boosted public funding to primary education (known as FUNDEF) and the federalisation of social welfare payments in 1997–2001. Both initiatives have exerted substantial downwards pressure on poverty and inequality levels (Ferreira et al., 2010; Lustig et al., 2013). The study considers the roles that electoral competition for socially vulnerable voters, left-wing legislative strength, social mobilisation capacity, and coalition dynamics might have played as contributing factors for the emergence of the two redistributive programmes.

By combining Congress archives, news pieces, and secondary sources, qualitative observations reveal electoral and coalitional incentives shaping the behaviour of political actors. In contrast, the relevance of the leftist bloc in the legislature and bottom-up pressures appear less pronounced. On the one hand, evidence shows President Cardoso (PSDB) signalling to the electorate an exchange of benefits for votes and showcasing his electoral ‘party brand’ of fiscal responsibility while the education reform and social welfare payments made their way through the legislature. At the same time, in both policy cases, the incumbent, whose political party was minoritarian in Congress, worked to keep coalition partners on board and dictate the discourses that allied legislators should advance when backing the proposals. On the other hand, the left remained relatively small, divided, and ineffective during policy processes in Congress, as proposals were enacted mostly without its direct input. Social mobilisations to influence reforms appeared not to have been even speculated.

As validity checks, the article first engages with two alternative explanations: personal leadership and bureaucratic action. Second, I set out directives for a future case study about the apparent continuing relevance of electoral competition and coalition dynamics as FUNDEF and social welfare payments have persisted over time during the subsequent left-wing Workers’ Party (PT) government (2003–2016) and the far-right presidency of Jair Bolsonaro ¹ (2019–2022). It is possible that presidents after Cardoso (PSDB) also attempted to claim electoral credit for maintaining and boosting these programmes, and the cross-party cooperative arrangements that first supported them stayed in place for the following decades.

The article makes two contributions to the politics of inequality literature. Empirically, it reinforces with a crucial case the theory predicting that under certain conditions, both the left and right are likely to propose greater redistribution – which in turn weakens expectations that the emergence of equity-enhancing reforms would depend on left-wing protagonism. Theoretically, the study incorporates coalition dynamics and cross-party cooperation as a potential enabling mechanism for effective social policymaking in multiparty systems.

The next section discusses theoretical developments about the four mechanisms under consideration and how they are expected to affect income distribution. An explanation of the article’s methodological approach, data collection, and analytical procedures follows. In the two last sections, I present the case study and end the article with a discussion and concluding remarks.

Redistribution-enabling mechanisms in centre-right governments

Fairfield and Garay (2017) argue that left-wing parties are not the only political actors interested in enacting redistributive policies. When the electorate is formed by a large share of low-income groups, competition for socially vulnerable voters creates distributional incentives for all political parties pursuing electoral victory. From this perspective, to attend to the preferences of the poorer segments of society, in return for votes, parties make social policy promises regardless of their position in the ideological spectrum. Consequently, even those leaning to the right could actively propose expanding redistribution.

Considering that incumbents spousing diverse ideologies may equally strive to stay in power or elect an ally and to the extent that policy decisions affect the electoral prospects of the current group heading the government, promises of more benefits can help candidates attract or retain voters’ support (Diaz-Cayeros, 2008). Even if a leftist opposition is relatively weak and incapable of exerting pressure on the executive, as the example of Mexico (2000–2012) indicates, electoral incentives could still prompt a right-leaning administration to improve welfare programmes (Garay, 2016). Rightist partisans might agree with pro-poor measures if they are perceived as conducive to winning. As such, the expectation that competitive elections would generate positive socioeconomic outcomes connects with the literature according to which democratic regimes foster social spending (Avelino et al., 2005).

While agreeing about democracies’ inclusive potential, Niedzwiecki and Pribble (2017) contend that advancing redistributive policies in centre-right governments also depends on the left-wing opposition being strong in terms of seats in the legislature and capacity to mobilise societal actors. The more legislative seats left-wing parties hold, the heavier their influence over the design and approval of inequality-reducing policies. Thus, a centre-right government aiming at passing legislation but facing off with numerous leftist legislators could be pressed to compromise with the left around more progressive programmes or to refrain from rolling back existing benefits. An example of this case is Argentina (2015–2019), where the left-wing opposition controlled the Senate.

Regarding social mobilisation capacity, left-wing parties may seek to shape policy by calling for popular movements to take to the streets (Niedzwiecki and Pribble, 2017). While Fairfield and Garay (2017) find that societal actors do not necessarily need to activate party-society linkages to pressure governments, what matters the most is whether mobilised groups can succeed in imposing their demands into the incumbents’ agenda – and collaboration from political parties should be reinforcing in that respect. Protests in Chile in 2011 represent an instance of bottom-up mobilisations pushing a centre-right administration to increase redistribution (Fairfield and Garay, 2017; Niedzwiecki and Pribble, 2017).

The importance of the strength of the left in opposition is based on the theory that as left-wing parties become more resourceful, their growing political power should translate into better social protection for the working-class groups that the left typically represents (Korpi, 1983). Commonly known as power resources theory, it identifies the political left as a protagonist of welfare state development. Left-wing parties should embody the preferences of underprivileged citizens for enlarged social policies, whereas centrist parties might prioritise economic modernisation and improved governance, and the right would back social schemes in which the rich enjoy higher-quality services (Huber and Stephens, 2012: 32–34).

The crucial contention in power resources theory is that electoral competition is not enough to drive non-left incumbents to commit to higher redistribution. In such a view, the stronger the left-wing opposition is in its ability to mould policy outcomes by leveraging on a significant legislative presence or social mobilisation, the likelier it is that equity-enhancing policies will emerge. Therefore, the approval of redistributive proposals during right-leaning governments should be associated with leftist groups intervening in the policy process.

However, in addition to considerations about whether non-left administrations would promise more redistribution to earn the votes of the less affluent, assuage a strong leftist bloc in the legislature or attend the demands of street protesters, because of Latin America’s prevailing multiparty presidential systems, forming and managing cross-party coalitions may also be vital for the success of the incumbent’s policy agenda. Legislatures in the region can and do amend and reject executive-sponsored programmes (Alemán and Tsebelis, 2016).

As for the cases of Argentina (2015–2019), Chile (2010–2014), and Mexico (2000–2012), the largest party in government never controlled a majority in the lower house (Cruz et al., 2021), and, apart from the Mexican case between 2007 and 2012, executives were always formed by more than one party. When the chief executive’s party does not enjoy a majoritarian position in the assembly, (minority) presidents often seek cross-party support to acquire some control over legislative works. In that way, they can better protect their mandate from impeachment and increase their chances of passing legislation (Chaisty et al., 2018). Therefore, policy approval could be predicated on the executive being able to cultivate a sufficiently large political alliance.

In the case of Brazil, where the party of the president usually holds less than one-fifth of assembly seats and multiparty governments are the rule, besides tracing the role that electoral competition, left-wing legislative strength, and social mobilisation could have played in enabling two consequential redistributive policies, coalition dynamics are also taken into account. Winning a presidential election does not guarantee passage to the (minority) president’s policy agenda, as proposals must get through a fractured legislature. Under this condition, maintaining a majoritarian coalition could allow the executive to disregard pressures coming from either the (left-wing) opposition in Congress or social movements on the streets. In other words, if non-left parties have electoral incentives to pledge greater redistribution and, once in office, they are able to form alliances around their (social) policy programme, the emergence of equity-enhancing initiatives might not require the protagonism of the left.

Qualitative observations as (dis)confirming evidence

This article aims to contribute to the literature on the politics of inequality in Latin America by identifying which enabling mechanisms would hold superior explanatory power for a crucial case of a centre-right government that promoted equity-enhancing reforms. Through a detailed analysis of the behaviour of political actors during the processes leading to the approval of two consequential policies, it is possible to set theoretical expectations against the historical record. In this sense, the article is a causes-of-effects type of study, adopts thick descriptions to uncover associative relationships between explanations and results, and relies on qualitative methods to provide confirming or disconfirming evidence of mechanisms in operation (Gerring, 2017; Mahoney and Goertz, 2006).

Building on Gerring’s (2007: 231) definition, according to which ‘[a] case is crucial if the facts of that case are central to the confirmation or disconfirmation of a theory’, Brazil is considered a crucial case because it represents a critical test for political theories ambitioning to explain changes in inequality levels across Latin America. On the one side, the country is home to about a third of the regional population and is often described as one of the world’s largest democracies (Garmany and Pereira, 2018); hence, existing explanations about socioeconomic progress in Latin America could be challenged if supporting empirical evidence is unfounded for impactful initiatives emerging in a country where income distribution equalisation benefitted millions of people. On the other side, as the emergence of education reform and federal social welfare payments during the PSDB-led centre-right administration has not yet been systematically studied from the perspective of the politics of inequality literature, finding the most relevant mechanisms associated with these positive outcomes can provide for case accumulation and analytical generalisation.

In relation to data collection, primary sources cover a long and deep terrain. Evidence was collected from the daily registers of Brazil’s upper and lower legislative houses, and news articles came from two of the country’s most-read national newspapers (Folha de São Paulo or Folha and O Estado de São Paulo or Estado). Data from legislative files regard the daily registers of plenary sessions in both houses of Congress referring to a constitutional amendment and an ordinary law about education reform and a constitutional amendment and an ordinary law about social welfare payments. These bills are described in the analysis section below, and their official reference numbers are provided in the Supplementary Material. These four pieces of approved legislation authorised the enactment of FUNDEF and the federalisation of cash-transfer schemes. Legislative files reveal the voting counts and the arguments that legislators adopted to support or reject these reforms. This data was collected from the day the related legislation arrived in a plenary session to the day it had its final vote, thus covering the full extent of Congressional scrutiny.

For data collection in newspaper archives, I searched for news pieces by only using words that gave the policies their names, and these words are listed in the Supplementary Material. It is reasonable to expect that news containing the names of the two selected policy initiatives are data points of interest if their content alludes to electoral competition, left-wing legislative strength, social mobilisation, or coalition dynamics, as newspapers regularly expose the political actions and reactions surrounding meaningful social proposals. Newspaper archives for the education reform were searched from the middle of the year (01 July) when the first related legislation was sent to Congress in 1995 until the last day of the following year when the second piece of legislation gained final consent (31 December). For social welfare payments, newspapers were searched for the entire year (01 January to 31 December) when the related legislation was enacted, covering four whole years (1997, 2000, 2001 and 2002) and considering multiple cash-transfer initiatives and their supporting laws. ² Hence, collected newspaper pieces for both policies and their components cover at least 12 months of data from proposition to approval.

Concerning analytical procedures, content analysis and manual coding generated the data points that undergird thick descriptions. After converting the primary data into electronic documents, they were uploaded to the computer-assisted qualitative data analysis software NVivo. To carry out the content analysis, I created codes in NVivo representing the four mechanisms being investigated. In the next step, while reading Congress and newspaper pieces, I manually selected document extracts and collated them under the appropriate coded mechanism. Once all documents were examined, I organised these extracts into the policy cases, demonstrating the more or less relevant roles that different mechanisms played in enabling each reform.

The Supplementary Material presents a table of sources and searches describing how the data was collected, including timeframes and keywords. It also displays examples of coded extracts and all legislative files and newspaper pieces consulted. Besides, to augment accuracy, primary evidence is often combined with secondary literature in the development of the policy cases. Therefore, this article adopts transparency for replicability and a combination of data sources to enhance reliability.

Finally, in addition to the policy cases, the analysis section below contains two extra components designed as validity checks. After identifying which mechanisms enabled the success of education reform and the federalisation of social welfare payments, I discuss two possible alternative explanations: personal leadership and bureaucratic action. ³ Subsequently, I set out a potential case study about the continuation of these equity-enhancing initiatives after the end of the PSDB government in 2002 to explore the plausibility that the identified mechanisms have retained explanatory power over time. The left-wing Workers’ Party government under Luiz Inácio Lula da Silva (PT, 2003–2010) ⁴ substantially expanded FUNDEF and social welfare payments, and the two policies were again relatively propped up during the far-right presidency of Bolsonaro (2019–2022).

Equity-enhancing reforms during the centre-right PSDB government (1995–2002)

Ameliorating social indicators in Brazil

During Cardoso’s (PSDB) tenure, Brazil’s Gini coefficient decreased from 59.6 to 58.1, and, at that moment, an unprecedented reduction in inequality began, with the indicator falling uninterruptedly between 1997 and 2011 (World Bank Data, 2024a). Moreover, from 1995 to 2002, the share of the population living in extreme poverty declined by approximately 20% (World Bank Data, 2024b), the average total years of schooling for the adult population increased by around 30% (Our World in Data, 2022), and the proportion of children attending primary school raised from 86% to 97% (IPEA Data, 2022). Lustig et al. (2013: 134–136) associate gains related to education reform and social welfare payments with a sizable income equalisation in 2000–2010.

In 1996, the PSDB-led administration approved in Congress the creation of a fund that significantly increased federal financing for the public primary education system. The fund boosted spending for pupils between seven and 14 years old in state and municipal schools around the country, and it became known as FUNDEF (Fundo de Mantenção e Desenvolvimento do Ensino Fundamental e de Valorização do Magistério in Portuguese). Melo (2008: 164) estimates total expenditure on public education grew by about one-third between 1995 and 2000. FUNDEF set a national minimum spending per pupil, equalised expenditure between primary schools within the same states, and increased teacher salaries (Draibe, 2003: 79). The fund paid up expenses in localities that did not have sufficient resources to meet the national target and incentivised mayors to improve enrolment rates. Plus, 60% of the fund’s resources had to be directed to the training and remuneration of teaching staff. Mello and Hoppe (2005) show that enrolment rates in primary education for Brazil’s North and Northeast regions, the poorest parts of the country, went from 80%-85% in 1994 to 95% in 2002, and teachers’ remuneration increased nationally by 38% between 1997 and 2001.

Similarly, regarding social welfare payments, cash-transfer schemes that already existed in some cities started to be federalised after Congress voted in a 1997 ordinary law to institute a national basic income programme and a constitutional amendment to create an anti-poverty fund, which passed in 2000, greatly invigorated the initiative. These reforms gave origin to a federal school grant, Bolsa Escola, in 2001, replacing the initial scheme instituted four years earlier. Two other programmes, Bolsa Alimentação (nutrition grant) and Auxílio Gás (kitchen gas grant), were established in the final 15 months of the PSDB-led administration. The federal budget for social assistance increased by five percentage points between 2001 and 2002 after the approval of the anti-poverty fund and the launch of Bolsa Escola (Hall, 2006: 693). The national spread, growing budget, and enhanced targeting of such schemes are among the causes of a decrease in Brazil’s Gini coefficients between 1993 and 2004 (Ferreira et al., 2008: 222) and for 2001–2007 (Barros et al., 2010: 154). The school grant reached five million families in 2002, four million more than what the basic income programme accounted for a year earlier, and the related social spending increased from R$ 162 million to R$ 2 billion year-on-year (Lício, 2004: 42).

The 1996 education reform (FUNDEF)

Cardoso (PSDB) campaigned successfully in two consecutive presidential elections, 1994 and 1998, after leading the development of the macroeconomic stabilisation plan that ended hyperinflation in the early 1990s. The plan was fully implemented in the year of the first victory, and its popular endorsement showcased inflation as a considerable issue for low-income voters (Kingstone and Ponce, 2010). In office, the newly inaugurated president promised his administration would strive for socioeconomic inclusion by keeping prices under control and making spending more effective so that redistributive measures could target the most vulnerable without triggering the type of debt-driven recessions Brazilians had become used to in previous years (Souza, 1997).

Accordingly, from its first legislative steps and onwards, FUNDEF was portrayed by members of the governing coalition as a reform to make public education spending more efficient and equitable. ⁵ Members of the executive and allied legislators argued that FUNDEF’s improved governance would reduce inequality by making more funds available for pupils and teachers. ⁶ In a speech after FUNDEF’s constitutional amendment passed Congress, President Cardoso (PSDB) emphasised that the new rules set out with the fund were integral to the government’s electoral commitment to fairer spending and branded FUNDEF as a fiscal revolution. ⁷

Maintaining a large cross-party coalition facilitated the executive dictating pro-reform behaviour to allied legislators while collecting their votes. For almost the entire duration of Cardoso’s (PSDB) two terms, political parties with cabinet positions accounted for more than two-thirds of the seats in both houses of Congress (Amorim Neto, 2019: 303), above the three-fifths threshold required for constitutional amendments. Whereas the PSDB alone controlled less than one-fifth of the legislature, the governing coalition was formally large, and the president was powerful when the alliance was held together.

The constitutional amendment creating FUNDEF was sent to Congress in 1995, and its principal legislative sponsors were coalition members that recurrently reproduced the education minister’s arguments to promote the project. In 1996, FUNDEF was approved in the lower house by 350 to 94 and 358 to 82, ⁸ and in two rounds in the Senate, just one senator cast ‘no’ votes. ⁹ In the lower house, the three main coalition parties (PSDB, PMDB, and PFL), which by themselves controlled 282 (55%) of the seats there, had only 16 ‘no’ votes or abstentions, adding the two voting rounds together. The following ordinary law that regulated the fund was approved by acclamation in both houses at the end of the year. ¹⁰

During plenary sessions about the constitutional amendment, when pressed by left-wing parties, legislators on the government’s side acknowledged they would have liked to dedicate even more resources to FUNDEF if that was fiscally feasible – consistent with the executive’s prudent discourse. ¹¹ Meanwhile, the opposition was still ineffective in approving changes to the proposal, and some leftists in Congress actually praised the fund. ¹² Similarly, in the process of passing the ordinary bill regulating FUNDEF, Paulo Renato Souza, minister of education, PSDB founder, and responsible for authorising concessions and managing the coalition around education reform, had again his work applauded by some opposite partisans. ¹³ For this bill, the minister was defeated by his own coalition partners in his intention to roll out the fund nationally in 1997, but it was launched with compulsory status one year later. ¹⁴ Allies were satisfied.

As the government managed to amass a sizable legislative base, the left-wing bloc in Congress ended up with a low share of seats during the scrutiny of FUNDEF, and two members of the leftist PDT party, in fact, spoke from the podium on voting days to commend the reform. ¹⁵ Attempts in the Chamber of Deputies to contest the government’s proposal ended up exposing divisions within the left. Workers’ Party and PCdoB party members presented vigorous criticisms against FUNDEF. Their deputies described the amendment as an illegal confiscation of funds from states and municipalities, argued that the federal government was trying to reduce its spending obligations, and complained that FUNDEF did not go far enough. ¹⁶ However, a leftist PDT deputy confronted these critiques, defending the fund as a positive step forward for public education. ¹⁷ Remarkably, the Workers’ Party and PCdoB not only dropped their most controversial claims after the first voting round in the lower house but also gave FUNDEF some support afterwards – especially in the Senate, where the former directed a ‘yes’ vote to its senators. ¹⁸ Even critics recognised that FUNDEF’s improved spending governance should result in fewer funds being lost on the way from Brasília to local schools. ¹⁹

Regarding the involvement of social and labour movements with FUNDEF, there is little evidence of how they could have exerted political pressure on the government in order to shape the policy. As an example, the constitutional amendment that created FUNDEF set up municipal, state, and federal councils formed by representatives of governments, civil society, unions, school associations, and parents and students to monitor spending and formulate educational guidelines. ²⁰ While their insertion and configuration in the constitutional amendment sometimes resulted from concessions made by the bill’s final sponsor to left-wing legislators, the attribution of local councils to monitor FUNDEF’s expenditures had government support, and the regulation was designed by the ministry of education. ²¹ Furthermore, as an additional illustration of the reduced role of social mobilisation, a union leader from São Paulo state complained about some parts of FUNDEF but approved others even though the left had little policy input in the reform. ²² Large-scale social mobilisations were not contemplated.

On the one hand, FUNDEF increased federal transfers to the country’s poorest areas while the governing coalition framed the reform around improved fiscal governance. Such a position was embedded in the PSDB’s electoral discourse, and it would help to re-elect Cardoso (PSDB) in 1998. It is also possible to observe how the executive took advantage of its large legislative base and got its coalition partners to fall in line during the votes in Congress. On the other hand, the opposition’s bloc in Congress was disunited on the education issue, and protests were absent in the data. Hence, left-wing legislative strength and social mobilisation capacity cannot be described as relevant enabling mechanisms in this policy case to the same extent as electoral competition and coalition dynamics.

The federalisation of social welfare payments

During the scrutiny of both the 1997 bill launching a national basic income programme and the 2000 constitutional amendment creating the anti-poverty fund, the government pushed a message that coalition members consistently repeated on the floor of Congress. The message for social welfare payments, similarly to the FUNDEF case, preached gradualism and fiscal prudence. The coalition defended that the basic income scheme and the anti-poverty fund, even under such framing, remained significant steps towards tackling poverty and inequality. ²³ President Cardoso (PSDB) publicly stated that the effectiveness of social spending would be more important than the total amount being disbursed by the government. ²⁴ Both coalition members, the proposer of the basic income bill in the lower house and the sponsor of the anti-poverty fund in the Senate, justified their works on the pro-poor measures in lines congruent to the president’s discourse. ²⁵

The executive created Bolsa Escola in 2001 and other grants in 2002 by provisional executive decree – an agenda-setting instrument that enacts legislation at its issuance while pending Congress verdict. Cardoso (PSDB) was determined in the run-up to the 2002 national election to convince voters his government was paying Brazil’s ‘social debt’. ²⁶ Increasingly, the president and ministers sought to underscore in public speeches that fiscal responsibility ²⁷ would go hand-in-hand with the coalition’s growing number of welfare initiatives. Notably, the launch of a nutrition grant was announced by a minister who would become the PSDB’s presidential candidate in 2002. ²⁸

In addition, the executive kept the legislative process under control in both the 1997 bill and the 2000 constitutional amendment as the coalition nominated the main sponsors, rejected changes proposed by the opposition, and displayed no internal dissent. Consequently, the government enjoyed comfortable victories while coalition members consistently repeated in Congress the arguments that the president and ministers prepared for the enabling legislation of social welfare payments. The 1997 bill was approved by acclamation in both legislative houses, ²⁹ and the 2000 constitutional amendment passed the Senate with 87% of the votes in two rounds ³⁰ and an even bigger victory in the Chamber of Deputies. ³¹

Critically, Cardoso (PSDB) worked to keep allies happy. A pact uniting the three main parties in the coalition, PSDB, PFL, and PMDB, was based on the first leading the executive while the other two would split the presidencies of Congress’ lower and upper houses. ³² This agreement came under pressure during the succession of the then Senate head, Antônio Carlos Magalhães from PFL, who was also the proposer of the anti-poverty fund and did not want to accept just any PMDB senator to replace him. To remedy the situation, the executive promised to celebrate Magalhães (PFL) for the anti-poverty fund ³³ and to shield him from potential reputational damages that might have arisen in a parliamentary inquiry. ³⁴ Later, his supporters would benefit from cabinet appointments and budget allowances. ³⁵ About PMDB, Cardoso (PSDB) dealt with threats that the party could paralyse the government’s bills in Congress if he interfered in the Senate run against their candidate and eventual winner, Senator Jader Barbalho (PMDB), even though Barbalho (PMDB) was not the president’s favourite choice. ³⁶

As for left-wing legislative strength, Bethell and Nicolau (2008) show that the leftist bloc remained relatively weak in Congress throughout the 1990s. Election results in 1994 and 1998 for the Chamber of Deputies gave left-wing parties about 21% of the seats on both occasions. In the Senate, the left accounted for just 14% of the seats at the start of the 1995 legislature and 17% four years later. In the context of the 1997 ordinary bill and the 2000 constitutional amendment, left-wing parties either backed the policies or allowed their legislators to vote as they wished, despite the opposition having failed to approve important proposed changes – some prominent Workers’ Party senators even voted against the anti-poverty fund in protest. For instance, during the legislative process that eventually enacted the anti-poverty fund, deputies and senators from the Workers’ Party wanted existing social councils at the municipal, state, and federal levels, which counted on civil society members, to gain legal powers to develop guidelines and monitor the fund’s finances. ³⁷ In the Senate, where the constitutional amendment started, the bill’s sponsor accepted only the creation of a new consultive council with civil society participation but without legal powers. ³⁸ Once the amendment arrived in the Chamber of Deputies, the left-wing opposition made a fresh effort to approve these changes but could not gather enough votes.

In addition, observations of social mobilisation with respect to the proposals undergirding cash-transfer programmes are scarce in both Congress archives and newspapers. As an illustration, mentions of traditional societal actors, like labour unions, are almost non-existent in the data. During discussions on the national basic income programme in 1997, the president of the country’s largest union confederation, a close ally of the Workers’ Party, wrote an op-ed at Folha but only cited the initiative inappreciably. ³⁹ There are no apparent signs of social mobilisation, let alone of a large-scale nature.

For the case of social welfare payments, key evidence highlights electoral competition and coalition dynamics in the policy process. Cardoso (PSDB) made efforts to publicise the launch of the national basic income programme and state grants and sought to claim credit for them around the time of the 1998 and 2002 elections. Also, the government directed coalition partners to frame the policies on the basis of redistribution with fiscal responsibility – its preferred electoral discourse. Meanwhile, the left-wing bloc in Congress remained ineffective in making substantial changes to executive-sponsored proposals, and social mobilisation is again relatively absent in the data.

Alternative explanations: Personal leadership and bureaucratic action

As a validity check against the findings that electoral competition for socially vulnerable voters and coalition dynamics may have enabled redistributive reforms during the PSDB-led centre-right administration, I engage with two alternative explanations present in the existing literature. The first explanation points to Cardoso’s (PSDB) leadership as a renowned sociologist likely holding progressive preferences (Fausto and Fausto, 2014: 331–365). The second calls attention to Brazil’s state bureaucracy and its evolving capabilities since re-democratisation in 1985, including during the PSDB administration, which could have facilitated the development of effective social policies (Taylor, 2020: 194–228).

Before entering politics in the late 1970s, when Brazil’s military dictatorship (1964–1985) began a slow opening process, Cardoso (PSDB) had dedicated much of his academic career to the study of democracy and development. Already a senior politician in the 1980s, he founded PSDB with other intellectuals and technocrats inspired by European social-democratic parties trying to combine a robust welfare state system with an efficient market economy. It is possible, then, that Cardoso’s (PSDB) personal vision helped spark redistributive policies. To this point, the nomination of internationally recognised specialists to work in social assistance, health, and education areas, as is the case of Cardoso (PSDB) placing his close ally Souza (PSDB) in the education ministry, may underscore a transmission of individual beliefs to policymaking. Yet, other authors also highlight Cardoso’s (PSDB) political negotiation skills while stating they were grounded in an institutional matrix in which potential allies were more likely to back programmes with good electoral prospects, and the executive could activate tools to cultivate cross-party coalitions (Alston et al., 2016).

Regarding bureaucratic action, existing literature focuses on the role of federal officials in providing impetus and protection to socioeconomic initiatives in Brazil. From this perspective, the creation of autonomous, coherent, and talented bureaucratic teams, coupled with state structures that became gradually more resourceful and able to reach low-income citizens directly, should be conducive to the emergence of social programmes. In this respect, Draibe (2004: 403–404) argues that ‘top-down reformers’ within the education ministry conducted FUNDEF following the leadership of Minister Souza (PSDB), himself a former bureaucrat. In relation to cash-transfer schemes, Bolch (2022) attributes their success in Brazil to growing state capacity, with policy development and implementation requiring the government to hire specialists and construct monitoring and delivery networks connecting states and municipal authorities to federal ministries. Still, Melo (2008: 173–179) suggests that as the enactments of federal education and anti-poverty funds depended on the approval of constitutional amendments, the government’s eagerness that these reforms would attend popular demands combined with the executive’s large legislative base served as a key to unlock FUNDEF and social welfare payments.

Building on Mahoney and Vanderpoel (2015), this discussion suggests that personal leadership and bureaucratic action could be understood as additional elements of a package of conditions in which, nonetheless, competitive elections and cross-party cooperation would remain the most relevant. From this perspective, the four components might not be necessary or sufficient individually but could be parts of a sufficient package allowing the emergence of equity-enhancing reforms under the centre-right PSDB government. As for the higher relevance of electoral competition and coalition dynamics, the policy cases above show that Cardoso (PSDB) ensured his party’s favourite election discourse was conveyed to voters and that allies understood this strategy had electoral potential. The evidence here aligns with other findings showing the PSDB-led administration believed the electorate wanted inclusion with fiscal responsibility (Alston et al., 2016). Moreover, Power’s (2010) argument that Cardoso (PSDB) wrote the operating manual of Brazil’s coalitional presidentialism implies not only that he was willing to compromise but also reveals the president sought to activate the instruments available to him, like the distribution of cabinet appointments, to cultivate the alliances that made many of the PSDB government’s reforms possible before a fragmented Congress.

FUNDEF and social welfare payments in the following decades

In this sub-section, I set out the contours of a case study that could further check the validity of electoral competition and coalition dynamics as relevant enabling mechanisms for equity-enhancing initiatives, considering that subsequent administrations of opposing ideological leanings continued FUNDEF and social welfare payments after the PSDB left office. As the discussion below suggests, the two mechanisms seem to have preserved their explanatory power throughout the following decades.

During his first stint in the presidency, Lula da Silva (PT, 2003–2010) united existing and new cash-transfer schemes in a single programme, launching Programa Bolsa Família in 2004 as Brazil’s flagship anti-poverty policy, and in 2006, before FUNDEF expired, his government proposed to replace it with a similar federal fund which would now encompass financing to pre- and high schools. Both initiatives significantly strengthened the federal government’s social and public education provisions (Huber and Stephens, 2012: 190–192). Importantly, however, the new programmes built on their predecessors’ technical foundations, which left-wing parties had previously had limited input over or shown partial support for, while legislative proceedings exhibit consistent support for the executive-sponsored proposals coming from former right-leaning coalition partners of Cardoso (PSDB), like the PMDB party (Hall, 2006; Melo, 2008). The ordinary law creating Programa Bolsa Família was approved by acclamation just as the PMDB was about to enter Lula da Silva’s (PT) cabinet, and the 2006 constitutional amendment enacting FUNDEF’s successor passed in Congress with minimal opposition at a time when PMDB members headed three ministries and were recently given the government’s leadership post in the Senate (Alves, 2024).

The pull of elections before an electorate formed by a high share of underprivileged citizens and minoritarian presidentialism may explain the Workers’ Party’s transformation during the 1990s into a professional vote-seeking organisation with a newly found willingness to reach out to potential partners across the political spectrum (Hunter, 2010). Pressure for adaptation, in turn, could shed light on why education reform and social welfare payments appear to have taken a continuation path from the PSDB years to the left’s first administration. The Workers’ Party achieved its inaugural win for the executive in 2002 after moving from the left to the centre on macroeconomic policy as the party’s candidate, Lula da Silva (PT), realised that being portrayed as a reliable manager of the public purse would bring electoral returns – as it eventually did, thus reverberating Cardoso’s (PSDB) theme of fiscal prudence (Montero, 2014: 39–40). Also, on the way to its 2002 presidential victory, the Workers’ Party decided to agree on an unprecedented pact with a centre-right party, which then appointed the vice-president post in Lula da Silva’s (PT) ticket, signalling wide-ranging coalitions had finally been embraced by the left’s leading representative (Flores-Macias, 2012: 124–133).

Regarding the far-right Bolsonaro administration (2019–2022), while it represented ‘four years of chaotic governance’ and ‘massive policy failures’ (Hunter and Power, 2023: 129–130), a narrower look at the federal education fund and social welfare payments might still reveal some continuity, which is possibly associated with the enduring effects of electoral competition and coalition dynamics. In 2020, as FUNDEF’s successor approached expiration, a new fund was created and made permanent for the first time. It did not carry significant regulation changes, and compulsory federal spending contributions to the new fund were set to increase (Barros and Machado, 2022). Likewise, although Bolsonaro had opposed cash-transfer programmes for most of his political career, Programa Bolsa Família continued during his government, albeit with a different name, and its monthly stipend was even scaled up during the COVID-19 pandemic and again near the 2022 presidential race. ⁴⁰

The victory of Bolsonaro’s radical candidacy in 2018 can be attributed to a wide anti-systemic sentiment among the electorate as, at that point, Brazilians had been dealing for some years with the impacts of a prolonged recession in 2014–2016, vast corrupt scandals, and record levels of street violence (Hunter and Power, 2019). Bolsonaro was effective in exploiting voters’ disenchantment with traditional parties, such as the Workers’ Party, the PSDB, and PMDB. Yet, the pull of elections and minoritarian presidentialism did not dissipate when the new incumbent entered office. His anti-system discourse made the cultivation of a majoritarian cross-party coalition unattainable, and the executive branch experienced the lowest success rate in the legislature since 1995 (Melo and Pereira, 2024). Facing competitive elections, Bolsonaro dealt with increased pressure to change tactics, explore popular welfare initiatives, like social welfare payments, and not oppose other policies that counted on broad legislative backing, such as the federal education fund. Legislators from left and right, including those from the PSDB, PMDB, and the former PFL parties, ⁴¹ the main coalition partners during the Cardoso (PSDB) years, voted overwhelmingly to augment cash-transfer programmes and install a new education fund. ⁴²

Discussion and concluding remarks

Distinct enabling mechanisms were recognised as theoretical possibilities explaining how Brazil’s PSDB-led centre-right administration advanced two significant equity-enhancing reforms between 1995 and 2002. Electoral competition, left-wing legislative strength, and social mobilisation capacity were identified by existing literature as contributing factors to the proposition, continuation, or even expansion of social policies in the cases of right-leaning governments in Argentina (2015–2019), Chile (2010–2014), and Mexico (2000–2012), as demonstrated by Fairfield and Garay (2017) and Niedzwiecki and Pribble (2017). Besides, considering that policy outcomes in multiparty presidential systems can be affected by coalition dynamics (Chaisty et al., 2018), this mechanism was also checked in the Brazilian case.

The study followed a causes-of-effects rationale, and thick descriptions based on content analysis and the manual coding of Congress files and newspaper archives undergirded the policy cases that unearthed (dis)confirming evidence about which hypothesised mechanisms would have enabled two consequential inequality-reducing initiatives. While limitations may remain, several steps were taken to enhance the reliability of analytical procedures and the validity of results. The Supplementary Material provides for transparency and replicability; primary and secondary sources were combined to improve accuracy; and, in the policy cases, there is a comprehensive engagement with alternative explanations and an examination of the identified mechanisms in different timelines.

In this sense, based on qualitative observations from the policy processes leading to the growth in funds for public primary schools and the launch of federal social welfare payments, it is plausible to conclude that competitive elections for socially vulnerable voters and the possibility of cultivating a sufficiently large governing coalition enabled greater redistribution during the centre-right PSDB-led administration. In the case of FUNDEF, the PSDB’s electoral brand of incrementalism and fiscal prudence permeated the policy, and despite some resistance from opposition legislators, the constitutional amendment was approved in both legislative houses without dissent among coalition partners. Similarly, the creation and implementation of federal cash-transfer schemes were accelerated around election time, with the government making explicit efforts to publicise them while coalition partners in Congress demonstrated cohesion in their discourses and votes.

However, regarding its presence in the legislature, the left held a small share of seats in both houses of Congress in 1995–2002, and, in the specific case of FUNDEF, left-wing parties demonstrated disagreements over the merits of the education reform. Parties of the left were often ineffective in approving amendments to the executive-sponsored bills. Similarly, the data shows that organised bottom-up pressures seeking to shape the centre-right administration’s plans for education and social welfare payments did not come to fruition. Low levels of societal mobilisation had already been noted in studies by Draibe (2004) about FUNDEF and Garay (2016) about PSDB’s cash-transfer schemes.

Also, two alternative explanations were considered: personal leadership and bureaucratic action. The first relates to Cardoso’s (PSDB) individual preferences and to what extent his background as a progressive academic could have sparked his government’s redistributive policies. The second highlights the bureaucracy and whether Brazil’s growing state capabilities in the 1980s and 1990s would be conducive to the emergence of inequality-reducing initiatives. These explanations could be interpreted as components of a package of conditions whose leading factors would still be electoral and coalitional incentives. Crucially, the analysed policies passed Congress while the president kept them under his favourite electoral framing of a fiscally sound social inclusion (Alston et al., 2016) and used coalitional tools to maintain alliances before a highly fragmented legislature (Power, 2010, 2016).

Furthermore, I proposed the contours of a case study about the continuation of the education fund and cash-transfer programmes after the PSDB left office in 2002, once the policies persisted during two ideologically divergent administrations. The centre-left Lula da Silva (PT, 2003–2010) government expanded FUNDEF and social welfare payments, but the policies also survived the far-right Bolsonaro (2019–2022) presidency. Pressures from competitive elections and minoritarian presidentialism may explain why the former settled for expansion through continuation, even though Workers’ Party members had previously shown some opposition to Cardoso’s (PSDB) social policy agenda. In a similar vein, Bolsonaro (2019–2022) might have refrained from frontally opposing FUNDEF and social welfare payments once he realised the credit-claiming electoral benefits in play for him and the broad cross-party backing these policies had in Congress.

In addition, it is important to note that the lack of empirical support for the hypothesis that left-wing legislative strength and social mobilisation capacity would have represented relevant enabling mechanisms for the emergence of equity-enhancing reforms during the centre-right PSDB government should not mean that the political left did not play a role in the development of social welfare payments. Prominent Workers’ Party members were early advocates of such initiatives and were deeply involved in their related intellectual debates and sub-national experimentation (Pereira, 2015). The present article only reveals that electoral and coalitional incentives offer a more robust explanation for the federalisation of cash-transfer programmes.

Although the country’s Gini coefficient fell more sharply during the Workers’ Party government (2003–2016) when compared to the Cardoso (PSDB) years (1995–2002), the policy cases above give cautious support for the thesis in Hagopian (2019) that inequality levels would have still declined in the 2000s had the PSDB stayed in power for longer. Elections continued to feature a large share of low-income voters, and cross-party alliances could be harnessed to improve redistribution, particularly at a point in time, at the beginning of the 21st century, when international economic conditions became significantly more favourable for commodity exporters like Brazil.

Finally, on the one hand, electoral competition could be distorted by political parties’ clientelistic behaviour (Gherghina and Marian, 2024), and corrupt practices may negatively affect inequality-reducing efforts (Wong, 2023). In this context, Brazilian presidents have been found using controversial favour-exchanging tools to approve their proposals in Congress (Taylor, 2020: 121–156). On the other hand, several of the equalising policies launched in Brazil during the PSDB and Workers’ Party governments, including those analysed here, deviate from the traditional logic of clientelism as they established transparent rules for redistribution, centralised federal oversight, and the anonymity of beneficiaries (Arretche et al., 2019). Future research can shed light on what factors might contribute to the institutionalisation of social programmes in the face of potentially long-standing opposing forces.

As the article’s main empirical contribution to the politics of inequality literature, the equity-enhancing role played by electoral competition in the Brazilian case (1995–2002) approximates it to the examples of centre-right governments in Argentina (2015–2019), Chile (2010–2014), and Mexico (2000–2012). Thus, this article adds strength to predictions that elections could drive political parties from different sides of the ideological spectrum to promise more benefits if the number of less-affluent voters is high. In turn, findings here weaken theoretical expectations that the emergence of inequality-reducing programmes would depend on the protagonism of the left, as, unlike the Argentinian and Chilean cases, key pro-poor measures advanced by the PSDB administration were not affected by left-wing legislative strength or social mobilisation.

As for theory building, the analyses of education reform and social welfare payments in the Brazilian case demonstrate that the existence of tools providing a minoritarian executive with a chance to cultivate sufficiently large alliances can be an essential factor in enabling long-term socioeconomic achievements in multiparty systems. This claim incorporates into the politics of inequality literature the proposition that policies should acquire resoluteness and persist over time if, despite several existing veto points, cooperation remains within reach (Armijo et al., 2006). Therefore, more research should be carried out about how coalition dynamics may have also mattered in fostering a fairer income distribution in other Latin American countries.

Supplemental Material

sj-pdf-1-pol-10.1177_02633957241310577 – Supplemental material for What explains equity-enhancing reforms under centre-right governments? Evidence from Brazil

Supplemental material, sj-pdf-1-pol-10.1177_02633957241310577 for What explains equity-enhancing reforms under centre-right governments? Evidence from Brazil by Daniel H Alves in Politics