Social Policy-Making under Populism

Social policy is one of the most visible policy areas and directly affects the pocketbooks of much of a country’s population, particularly the poor. It constitutes a clear way for politicians to signal their priorities, expand their base and reward supporters. Social policy thus offers a particularly useful lens through which to analyse how governments wield power and make policy.

Populism is the textbook example of a quintessentially contested concept. Colloquially used to discredit politicians as opportunistic and policies as unsustainable, recent political science research has sought to build consensus around empirically verifiable definitions. Following Julio Carrión, populism is defined as ‘a political strategy for seeking and exercising power that has three constitutive features: (1) it exhibits a personalistic style of leadership – meaning it seeks to establish direct and unmediated links with supporters, (2) it is anti-pluralistic and confrontational toward opponents, and (3) it has a general distrust of institutional checks and balances’.Footnote ¹¹

Because populism is a political strategy rather than a cohesive political ideology with a concrete governing agenda, there are no inherently ‘populist’ policies, social or otherwise. Populism is a practice, a ‘means’ of achieving policy, not an ‘end’ or a particular policy outcome.Footnote ¹² As populism can be combined with other ideologies, it should be compatible with a wide range of specific policies. In practice, populists tend to be flexible in their beliefs and policy preferences, quickly adjusting them to better achieve their objectives.Footnote ¹³ Nevertheless, populism’s distinctive features should be expected to influence in predictable and clearly identifiable ways how policies are designed and implemented.

Thus, there should exist a populist style of making social policy that is independent of ideology. This article follows Alan Knight in defining populism as a ‘political style’ or way of doing politics that is not unique ‘to a specific ideology, period, or class alliance’.Footnote ¹⁴ However, in contrast to Knight’s seminal work, it argues that this populist style extends beyond the realm of electoral politics into the policy-making process.

Based on prior research, the remainder of this section derives a series of predictions regarding the defining features of this populist style of social policy-making. Populism’s predicted impact on social policy is then contrasted with the better-understood role of traditional Left/Right ideology.

Rejection and Rebranding

Taking ownership of social policy is of particular importance to populist leaders. As their connection to voters is based on direct personalist appeals, rather than disciplined programmatic parties or long-standing patronage networks, populists run the risk of having their support evaporate early in their tenures.Footnote ¹⁵ It is therefore imperative for a populist to simultaneously deepen voters’ connection to ‘the leader’ and further discredit the defeated establishment. Attacking existing programmes and claiming credit for replacing (or at least rebranding) them advances both objectives.

Thus, populists should be particularly prone to overhauling social policy and dismantling existing programmes, even popular and/or successful ones. Given their need to consolidate support, tendency to paint the situation they inherit as a ‘crisis’Footnote ¹⁶ and frequent claims that their predecessors are corrupt and/or benefited ‘undeserving’ groups,Footnote ¹⁷ populists should be expected to reform social policy early in their terms.Footnote ¹⁸ As Kurt Weyland notes, ‘when they [populists] have the requisite funds, there is little that can stop them’.Footnote ¹⁹ Further, as ‘relative outsiders … not as beholden to established interest groups and better-off social constituencies’,Footnote ²⁰ populists find it easier to adopt sweeping reforms.Footnote ²¹

These trends should be particularly relevant to transfer programmes because the latter are highly visible, closely associated with presidentsFootnote ²² and the electoral benefits they provide to incumbents seem to decline over time.Footnote ²³ Thus, it can be expected that:

Design and Implementation

Social policy-making under populism ‘tends to have a significantly faster tempo and a shorter duration with frequent episodes of accelerations and an unpredictable timing’.Footnote ²⁴ This, however, ‘comes at the expense of good policy design, careful implementation, and fiscal and political consolidation’.Footnote ²⁵ Decisions tend to be centralised around the chief executive, circumvent established institutions and ignore the views of the opposition and civil society.Footnote ²⁶ As Benjamin Moffitt notes, ‘consensus, deliberation and negotiation’ are anathema to populist policy-making.Footnote ²⁷

This approach centred on ‘powering through’ rather than ‘carefully sequenced incremental changes’ increases the likelihood of performance problems, backlash and reversibility under future governments.Footnote ²⁸ As outsiders sceptical of expertise, populists distrust career civil servants and technocrats, dismissing them as representatives of ‘the old, corrupt system’ that ‘must be side-lined or replaced’ with ‘true believers’.Footnote ²⁹ Those loyalists, however, tend to be inexperienced. Reflexive rejection of existing policies and disdain for expertise result in policy-making that strays ‘significantly and systematically’ from ‘technical rationality, evidence-informed, and pluralistic standards’.Footnote ³⁰ Indeed, redistributive policy under Latin America’s classical populists was defined by its ‘improvisation’.Footnote ³¹ New programmes and reforms to existing programmes are often enacted before crucial details are hammered out. Dorottya Szikra and Kerem Gabriel Öktem describe social policy decision-making under Viktor Orbán and Recep Tayyip Erdoğan as ‘chaotic’, with policies requiring multiple amendments to correct their initial problems.Footnote ³²

Latin American populists have an equally worrisome history of failing to ensure the long-term sustainability of their policies. Classical populists, such as Argentina’s Juan Perón (1946–55; 1973–4) pursued redistribution at the cost of unsustainably high inflation and deficits. During the 1990s, Peruvian neoliberal populist Alberto Fujimori (1990–2000) relied on windfall profits from privatisation and loans from multilateral banks to fund anti-poverty programmes. In the 2000s, left-wing populists Hugo Chávez (2002–13) of Venezuela and Rafael Correa (2007–17) of Ecuador relied on oil rents. Eventually debts had to be repaid, funds from privatisation were exhausted and oil windfalls evaporated, making painful retrenchment necessary. From this, the second set of predictions can be derived:

Clientelism and Personalisation

Whereas inefficiency and corruption are unintended consequences of populist policy-making, clientelism is a feature, not a bug.Footnote ³³ As Jan-Werner Müller puts it, populists ‘can engage in such practices openly and with public moral justifications, since for them only some people are really the people and hence deserving of the support by what is rightfully their state’.Footnote ³⁴ Clientelism is central to populist strategies to consolidate power. Populists often rise to power seemingly out of nowhere amid crises that discredit the political establishment. Given the urgent need to expand their base, populists will seek to ‘personalise’ social policy by presenting benefits as products of their personal largesse. Other examples of clientelistic practices include politicised (rather than rules-based) selection of beneficiaries, purposeful exclusion of groups opposed to the government and even conditioning of benefits on supporting the government. There is well-documented evidence of Latin American and Eastern European populists targeting beneficiaries on political, rather than programmatic, grounds.Footnote ³⁵

Populists also tend to ramp up social policy prior to elections. Chávez launched the clientelistic misiones (missions) to cement his popularity among the poor in the run up to a 2004 recall referendum.Footnote ³⁶ Similarly, Orbán and Erdoğan tended to enact major social policy expansions before facing re-election.Footnote ³⁷ Thus, it can be expected that:

Populism versus Ideology

As a political strategy, populism is compatible with different political ideologies and governing agendas. Traditional ideology can thus be conceptualised as a dimension of policy priorities and design that runs parallel to populism. Right-wing populist leaders should enact right-wing policies but in a populist manner. Left-wing populists will enact left-wing policies in a similar manner to their right-wing counterparts.

Left-wing governments, both populist and otherwise, have played a decisive role in the adoption and expansion of social policy worldwide. In line with power resource theory, the relative strength of the Left and organised labour constitutes the dominant explanation for the adoption of and cross-national variation in social policy in Latin America.Footnote ³⁸ Focusing on cash transfers, Fabián Borges finds that left-wing governments adopt more generous and universalistic programmes with more permissive conditionality enforcement. Left-wing presidents also tend to reform the programmes they inherit in line with the goals mentioned above.Footnote ³⁹ Most research on the influence of ideology over social policy, however, has not distinguished between the effects of populist and non-populist governments.Footnote ⁴⁰ From this, a final set of predictions can be derived:

The next section explains the choice of cases that will be analysed in the empirical section.

Case Selection and Methodology

The predictions presented in the previous section will be assessed through in-depth case studies tracing the evolution of cash transfer policy in Mexico under left-wing populist AMLO and Brazil under right-wing populist Bolsonaro, from the moment both presidents assumed office until the end of their terms. The case studies engage in both within-country comparison (prior to versus under populism) and comparison between the two countries (Mexico versus Brazil).

Assessing the consequences of populism on policy design and implementation is complicated by the fact that populism never exists in a pure form and is always attached to a more cohesive ideology.Footnote ⁴¹ The consequences of populism, however, can be teased out through careful comparisons that allow for variation in ‘host’ ideologies. Most research on populism and social policy has focused on Western European radical right-wing parties. Given their primarily working-class bases and hostility toward immigration, these parties typically espouse welfare chauvinism. This stance, however, is not inherently populist.Footnote ⁴² Furthermore, analyses of the effects of populism based on such cases are complicated by the fact that these parties have mostly been limited to serving as partners in broader coalitions. Latin American and Eastern European countries where populists have won full control of government offer ‘unfiltered’ examples of populist policy-making. However, because the single-party populist governments of Poland (2014–23) and Hungary (2010–) shared a right-wing and nativist agenda, it is difficult to disentangle populism from ideology in those cases. Research comparing left-wing Latin American populist governments suffers from the same problem. Cross-regional research raises the challenge of comparing countries with very different histories and political systems.

Cash transfer policy under AMLO and Bolsonaro offers a particularly useful lens through which to tease out the consequences of populism. Within a month of each other, at the end of 2018, populist presidents from opposite ends of the political spectrum took office in two large Latin American federal democracies with similar levels of income and human development.

Both leaders meet the definition of populism. They exhibited personalistic political styles that sought to establish direct and unmediated links to their supporters. Bolsonaro relied heavily on social media to communicate with supporters. AMLO spoke directly to the voters and sought to control the media narrative through his daily morning press conferences (mañaneras). Both exhibited anti-pluralistic rhetoric, framing rival political forces as corrupt enemies who actively conspire against ‘the people’. Both also resisted checks and balances, clashing with their countries’ Supreme Courts and questioning the validity of elections they lost. This behaviour elicited concern from several scholars and pundits, who warned that populism would severely damage Mexican and Brazilian democracy.Footnote ⁴³

With regard to ideology, there is a general consensus on Bolsonaro being coded as right-wing. His disdain for the political left and progressive causes as well as his social conservatism leave few doubts about his ideological orientation. AMLO’s leftism is more open to debate. Citing his combination of social conservatism and statism, Rodrigo Castro Cornejo classifies AMLO’s politics as ‘leftism without progressivism’.Footnote ⁴⁴ Cognisant of these idiosyncrasies, AMLO is coded as a leftist because he promised statist economic policies, especially in the natural resource sector, promised to expand social policy in a universalist direction, constantly criticised neoliberalism and showed affinity toward other left-wing presidents, even undemocratic ones.Footnote ⁴⁵

In Prospera and Bolsa Família, AMLO and Bolsonaro inherited comparable programmes covering similar shares of their respective countries’ populations. In 2018, these programmes covered 24.8 per cent of Mexicans (31.2 million) and 26.8 per cent of Brazilians (57.1 million).Footnote ⁴⁶ Both were long established, having persisted largely unchanged for nearly two decades through multiple non-populist administrations from across the ideological spectrum. The next section compares the design of these programmes prior to the election of populist leaders.

The Baseline: Pre-Populist Mexican and Brazilian programmes

Mexico’s Prospera and Brazil’s Bolsa Família were the product of technocratic policy design processes and subject to carefully sequenced incremental refinement during their roughly two decades of operation under multiple non-populist administrations.

The blueprint for Prospera was a 1991 World Bank research paper by economist Santiago Levy.Footnote ⁴⁷ A pilot version of the programme was launched in randomly selected rural municipalities in 1995 under Ernesto Zedillo (1994–2000), a centre-right member of the hegemonic Partido Revolucionario Institucional (Institutional Revolutionary Party, PRI). Findings from experimental evaluations were incorporated into a national version of the programme in 1997. Positive evaluations led Vicente Fox (2000–6) of the right-wing Partido de Acción Nacional (National Action Party, PAN) to continue the programme and expand it to urban areas in 2002.Footnote ⁴⁸ The programme later survived the PRI’s return to power under Enrique Peña Nieto (2012–18) with only minor modifications.

Bolsa Família’s origins date back to policy proposals made by Brazilian academics in the early 1990s.Footnote ⁴⁹ The first local-level transfer programmes were adopted in 1995 before spreading across the country.Footnote ⁵⁰ The success of these programmes led centrist Fernando Henrique Cardoso (1995–2002) of the Partido da Social Democracia Brasileira (Brazilian Social Democracy Party, PSDB) to adopt a series of national-level programmes in 2001. Originally dismissive of these programmes, Luiz Inácio Lula da Silva (Lula) (2003–10; 2023–) of the left-wing Partido dos Trabalhadores (Workers’ Party, PT) merged and significantly expanded those programmes under the name Bolsa Família.Footnote ⁵¹ Following the impeachment of Lula’s hand-picked successor Dilma Rousseff (2010–16), the centre-right administration of Michel Temer (2016–20) continued the programme.

Both programmes adhered to strict rules limiting political discretion and clientelism.Footnote ⁵² To minimise the involvement of potentially corrupt subnational politicians, management was centralised at the federal level with stipends deposited electronically to beneficiaries’ accounts. Clear eligibility rules all but eliminated the need for local intermediaries.Footnote ⁵³ Despite this, there is robust evidence from both countries that these programmes increased electoral support for presidential candidates from the incumbent parties.Footnote ⁵⁴

Despite their similarities, these programmes were not identical. There is an ‘inherent tension between the twin goals of reducing poverty in the short term and increasing human capital over the long run’.Footnote ⁵⁵ Design can be tailored to prioritise one goal over the other. Borges argues that whereas Prospera emphasised human capital, Bolsa Família emphasised poverty reduction and that government ideology explains those design choices.Footnote ⁵⁶ Brazil’s programme was more expensive as a share of GDP than Mexico’s and had broader targeting. Whereas Mexico’s Right-controlled programme relied on a stringent means test to determine eligibility, Brazil’s Left-controlled programme relied on self-reported incomes. Where Mexico adopted a strict (critics would say punitive) approach to conditionality enforcement where non-compliance led to immediate loss of benefits,Footnote ⁵⁷ noncompliance in Brazil was treated as a sign of family vulnerability and met with warnings and intervention from social workers.Footnote ⁵⁸ The next section analyses Mexico’s cash transfer programme experience under AMLO.

Mexican Cash Transfers under AMLO

AMLO had long alleged that Prospera did not reach the poor, was used to buy votes and had been used to illegally deprive him of the presidency in 2006. In early 2019, two months into his presidency, AMLO replaced Prospera with a series of unconditional transfer programmes. The changes were enacted in a rushed manner without evidence-based justification.Footnote ⁵⁹ Cash transfer policy has since become significantly more discretionary and moved away from Prospera’s technocratic rules-based approach.

Rejection and Rebranding

AMLO long claimed (without evidence) that Prospera was used to buy votes during the 2006 presidential election, which he lost by a razor-thin margin and never conceded. While occupying Mexico City’s Zócalo square in protest, AMLO asked ‘how is it anything other than fraud if they used all the social programmes, including the beneficiary rolls of Oportunidades [later renamed Prospera] and the scholarships, to benefit the candidate of the Right [Felipe Calderón (2006–12)] …?’Footnote ⁶⁰

Beyond his personal grievances, AMLO’s critiques of Prospera were left-coded – the programme did not reach those who needed it. Once in office, he justified dismantling Prospera by claiming (also without evidence) that its rolls were inflated by half and that three quarters of stipends were stolen. In February 2019, he claimed that his administration had ‘found ghost beneficiaries in the programme … They falsified signatures and stole the money, or they [beneficiaries] were supposed to receive MX$1,000 and were given MX$250’.Footnote ⁶¹ AMLO later claimed that the previous administration ‘was supposedly providing support to six, seven million families but the censuses showed that only half that number actually received benefits’. He said he would send a report to the Prosecutor’s Office.Footnote ⁶² No report was sent.

AMLO replaced Prospera with the unconditional Becas para el Bienestar Benito Juárez (Benito Juárez Wellbeing Scholarships). The new programmes increased the share of students eligible for transfers. Prospera only covered children in extreme poverty but paid individual transfers to each eligible child. Becas targets poor households with children in basic education (up to grade nine) and all children in upper secondary education (grades 10–12). The former receive a single bimonthly stipend, regardless of number of children. The latter is paid per child.

Actual programme coverage is less impressive. Whereas Prospera targeted students across the country based on household conditions, Becas targets students based on their community’s socioeconomic conditions. Transfers are guaranteed to eligible students from schools in ‘priority areas’ – communities that are indigenous, have less than 50 people and/or have high marginalisation.Footnote ⁶³ Benefits for qualifying students from non-priority schools are subject to budget availability. Thus, needy students from better-off areas may be excluded.Footnote ⁶⁴ The scholarships for upper secondary students, which the government touts as ‘universal’, do cover all students in a given school, but only if that school is in a priority community. This has been a major source of confusion among parents and administrators.Footnote ⁶⁵

In practice, the new programmes cover significantly fewer students. Whereas in 2018 Prospera covered 17.6 per cent of households, Becas reached only 12.1 per cent of households in 2022. And, whereas 52 per cent of households in the bottom decile received Prospera in 2018, less than a third received Becas in 2022. In fact, Prospera in 2018 reached a larger share of the poorest tenth of households than all of AMLO’s programmes combined in 2022.Footnote ⁶⁶ Further, in 2022 Becas transfers were on average 45 per cent smaller than the average 2017 Prospera stipend.Footnote ⁶⁷ Families with multiple children in basic education were particularly hard hit. Upper secondary school stipends decreased significantly, especially for girls, who used to receive more money than boys.

Reform Process

Luis David Ramírez Benítez argues that ‘the substitution of the Prospera programme was not a public policy decision based on evidence; there was no study suggesting that such actions would benefit society’.Footnote ⁶⁸ Máximo Ernesto Jaramillo-Molina, a long-time Prospera critic, argues that, faced with a choice between continuing a programme whose failings were understood and correcting those failings, AMLO opted for a third, and worse, option – cancelling the programme altogether and replacing it with one that failed to address its predecessor’s failings.Footnote ⁶⁹ More broadly, Mauricio Dussauge-Laguna goes so far as to assert that, ‘all reform initiatives implemented by the López Obrador administration have shared one key feature: a lack of solid background studies, diagnoses, data, or evidence showing their adequacy or potential usefulness’.Footnote ⁷⁰ The programmes, he adds, lack ‘basic elements of good policy design or monitoring and evaluation criteria’.Footnote ⁷¹

The desire to move quickly ensured a chaotic transition.Footnote ⁷² Many Prospera households, particularly in urban areas, were excluded from the new programmes without justification.Footnote ⁷³ Civil society groups have repeatedly criticised the programmes’ lack of transparency, warning of potential for corruption.Footnote ⁷⁴ The programmes lack fixed payment schedules,Footnote ⁷⁵ while AMLO used press conferences to personally announce payment dates. In the run up to the 2024 election, AMLO frequently handed out benefits during presidential trips.Footnote ⁷⁶

These and other problems have been detailed in evaluations conducted by the autonomous Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Social Development Policy Evaluation Council, CONEVAL).Footnote ⁷⁷ AMLO has repeatedly clashed with the institution and a reform abolishing it was approved in the interregnum between Sheinbaum’s election and the start of her presidency.Footnote ⁷⁸ In 2021, AMLO vehemently rejected the agency’s finding that poverty increased during his first two years in office, claiming he had ‘other data’.Footnote ⁷⁹

Discretionality and Politicisation

Transfer policy became more discretionary. Katya Rodríguez Gómez states it bluntly: ‘the manner in which these new programmes have been operating recently is more clientelistic than technocratic’.Footnote ⁸⁰ The programmes, informally known as Becas AMLO (AMLO Scholarships), are closely associated with the figure of the president, something that was explicitly avoided under Prospera.Footnote ⁸¹ Notably, by 2023, 55 per cent of survey respondents attributed transfers to AMLO himself, up from 32 per cent in 2020.Footnote ⁸²

The administration’s difficulty in reaching the poorest households can be traced back to the ‘Wellbeing Census’, an opaque survey of the population’s needs for social policy purposes, which was used to enrol beneficiaries. The ‘census’ was conducted by the so-called ‘Servants of the Nation’, volunteer political operatives from AMLO’s Movimiento de Regeneración Nacional (National Regeneration Movement, Morena) party. It was never explained why this was necessary – comprehensive registries existed – or why the unqualified Servants, rather than the well-regarded National Institute of Statistics and Geography, were tasked with the process. The Servants interviewed 24.1 million people,Footnote ⁸³ far short of the 73.3 million in existing registries.Footnote ⁸⁴ Regardless, Becas launched months before the census concluded.

The census was criticised for being a public relations exercise meant to link the new programmes directly to AMLO and make Mexicans see him as ‘the great benefactor’.Footnote ⁸⁵ In violation of Mexican law, the Servants wore clothing with AMLO’s likeness and stated that they had come ‘on behalf of the president of the republic, Licenciado Andrés Manuel López Obrador … to provide support [to your family]’.Footnote ⁸⁶ Although a court ruled that the Servants broke the law, AMLO and his ministers were not legally liable.

This began a pattern of politicising social policy. In 2021, the Servants, now government employees, asked COVID-19 vaccine recipients for their voting credentials and suggested that they write AMLO thank you letters.Footnote ⁸⁷ And, in the run up to the April 2022 mid-term recall referendum, teachers and retirees reported receiving calls and even visits from Servants warning them that AMLO’s defeat would result in their benefits being revoked.Footnote ⁸⁸

The arbitrariness and politicisation of social policy under AMLO was also on display during the COVID-19 pandemic and the 2024 election. Mexico stood out among Latin American countries for its refusal to increase the coverage and generosity of cash transfers, let alone extend emergency benefits to those excluded from existing policies.Footnote ⁸⁹ This is explained by AMLO’s long-standing aversion to government borrowing and his personalistic control over Morena, which at the time had a legislative supermajority.Footnote ⁹⁰ Morena legislators ignored proposals from the opposition and demands from the business community, civil society and the Catholic Church.Footnote ⁹¹ The result was a sharp increase in poverty compared to other Latin American countries.Footnote ⁹²

AMLO abandoned austerity during his final year in a successful bid to ensure the election of his chosen successor, Claudia Sheinbaum. The 2024 budget called for a 27 per cent increase in social spending financed by the largest fiscal deficit since 1988.Footnote ⁹³ The additional funds were used to more than double the number of schools covered by the Becas. This would allow 2.4 million students to join the programme in addition to the 5.6 million in basic education and the 5.8 million upper secondary students receiving benefits at the end of 2023.Footnote ⁹⁴ This likely contributed to Sheinbaum’s resounding victory. A February 2024 poll found that 67 per cent of people in benefit-receiving households and 72 per cent of direct beneficiaries supported AMLO, compared to just 38 per cent of non-beneficiaries.Footnote ⁹⁵

A Leftist Social Policy?

AMLO promised to put the poor first (primero los pobres). As would be expected from a left-wing government, transfer policy saw a narrowly targeted programme with strict conditionality enforcement replaced by more broadly targeted, unconditional programmes. However, the push toward universality came at the cost of reduced progressivity. AMLO’s Becas pay less generous stipends and are more likely to exclude the poorest households, including many previously covered by the ‘neoliberal’ Prospera.Footnote ⁹⁶

In line with the Left’s more expansive social policy ambitions, the share of the population eligible for benefits increased. Whereas Prospera was limited to children in extreme poverty, Becas extended cash transfers to all upper secondary school students and, starting in 2020, to moderately poor households with students in basic education. Further, in line with a rights-based view of social policy, in March 2020 the government secured approval of a constitutional amendment enshrining the right of students to cash transfers (with priority given to the poor) and prohibiting future governments from reducing the real value of these stipends. The Becas were also complemented by comparatively generous cash transfers for young adults, including university scholarships for low-income and indigenous students and paid apprenticeships for the unemployed.

The new programmes represented a potential boon for those previously excluded, namely upper secondary students not in extreme poverty, moderately poor parents with children under 15 and young adults. However, the expansion of transfers to upper secondary students and the move to a single stipend per household with children in basic education marked a clear de-prioritisation of young children, who are disproportionately poor, in favour of older students, who tend to be better off. Notably, only 34 per cent of upper secondary students come from the bottom half of the income distribution and only 17 per cent are from rural areas.Footnote ⁹⁷ Electoral motivations might explain the prioritisation of upper secondary students, who are close to voting age, and the creation of benefits for voting-eligible young adults.

AMLO’s transfer policy was the victim of two contradictions: one between universality and progressivity and another between expanding eligibility and pursuing fiscal austerity. Without significantly increasing spending levels, expanding eligibility would naturally lead to less generous benefits and rationing of benefits. Although by 2022 both total social spending and the share of the population covered by social programmes had reached record levels, relative to Prospera the average cash transfer was half as generous, coverage among the poorest 30 per cent was lower and coverage among the poorest 10 per cent had declined by almost a quarter.Footnote ⁹⁸ This persistent inability to reach the neediest is also likely a consequence of the shoddy methodology of the Wellbeing Census.Footnote ⁹⁹ Discussion now turns to the Brazilian case.

Brazilian Cash Transfers under Bolsonaro

Cash transfer policy under Bolsonaro can be divided into two phases: before and after COVID-19. A long-time critic of Bolsa Família, during his first year Bolsonaro considered replacing the programme and enacted cuts that were politically targeted against the north-east, the country’s poorest region and a bastion of support for the opposition.Footnote ¹⁰⁰ Surging approval ratings following the country’s generous social policy response to the pandemic convinced Bolsonaro that expanding cash transfers offered his clearest path to re-election.Footnote ¹⁰¹ Circumventing a constitutionally mandated spending cap and a ban on election-year reforms to social policy, he significantly expanded the coverage and generosity of cash transfers in a failed bid to secure re-election.

Rejection and Rebranding

In contrast to AMLO, Bolsonaro’s long-standing critiques of Bolsa Família were right-coded: cash transfers made beneficiaries idle, beneficiaries remained in the programme indefinitely and the PT used it to buy votes, particularly in the north-east. As a congressional backbencher in 2010, Bolsonaro claimed that ‘if nowadays I give Reais$10 to someone and I am accused of those R$10 being used to buy a vote, I will be impeached. Now, the Federal Government … gives 12 million families around R$500 per month, under the name Bolsa Família, and comes out ahead 30 million votes’.Footnote ¹⁰² The following year, Bolsonaro called for ending the programme because it damaged beneficiaries’ ‘work ethic’.Footnote ¹⁰³ While running for office in 2018, Bolsonaro claimed, without evidence, that 30 per cent of beneficiaries underreported their incomes and should be expelled.Footnote ¹⁰⁴

However, after it became clear that he had a real chance of becoming president, Bolsonaro promised to continue Bolsa Família, claiming that reports of his opposition to the programme were ‘fake news’.Footnote ¹⁰⁵ This change was confirmed on October 7, when, after winning the first round of voting, Bolsonaro uploaded a video directed at north-eastern voters promising a thirteenth monthly payment in 2019. After winning the second round comfortably, Bolsonaro announced the bonus would be paid in lieu of a pending inflation adjustment with savings obtained from a crackdown on Bolsa Família fraud. When this failed to free up sufficient resources, funds were secured at the last minute via legally dubious delays in the approval of new pension applications.Footnote ¹⁰⁶ The bonus turned out to be a one-off electoral ploy and was not paid in subsequent years.

During this time, Bolsonaro’s economic team, led by Minister of the Economy Paulo Guedes (2019–22), a technocratic free-market economist, proposed streamlining Brazilian anti-poverty policy without increasing the deficit by simultaneously increasing Bolsa Família’s stipends and coverage and eliminating several smaller programmes. During the administration’s first two years, the economic team carried out studies and simulations aimed at determining what to cut.Footnote ¹⁰⁷

The COVID-19 pandemic transformed Bolsonaro into a strong backer of cash transfers. In contrast to Bolsonaro’s dismal federal public health response, Brazil’s social policy response was exemplary.Footnote ¹⁰⁸ Under the name Auxílio Emergencial (Emergency Aid), Bolsa Família benefits were increased and extended to all informal and unemployed workers. Conditionality enforcement also temporarily stopped. At its peak in mid-2020, 66 million people were enrolled and half of all households had at least one beneficiary. The cash transfers, which equalled 78 per cent of the average monthly income of self-employed workers, cut poverty by 60 per cent to its lowest recorded level.Footnote ¹⁰⁹

Bolsonaro deserves, at most, partial credit for this. His administration initially proposed transfers of R$200 per person per month for three months. Congress raised that amount to R$600 and R$1,200 for single parents. With the end of the pandemic nowhere in sight, the administration later proposed two additional months at R$200. Congress again tripled it. Benefits were subsequently extended for four additional instalments of R$300 per month until December. After a hiatus, benefits for non-Bolsa Família households resumed at a lower level between April and October 2021.

Auxílio Emergencial significantly boosted Bolsonaro’s popularity among the poor at a time when his mishandling of the pandemic and corruption scandals had cost him middle-class support.Footnote ¹¹⁰ His approval rating peaked in December 2020 with support among the poor having seen a 70 per cent year-on-year increase.Footnote ¹¹¹ This further convinced Bolsonaro that cash transfers held the key to his re-election. Bolsonaro publicly disavowed Guedes’s plans to fund the expansion of cash transfers for the poor by cutting other social programmes in late 2020.Footnote ¹¹² Expanding transfers, however, would require the circumvention of a constitutionally mandated spending cap before an election-year ban on social policy reforms came into effect.Footnote ¹¹³

Reform Process

Under the new name Auxílio Brasil (Aid Brazil), Bolsonaro substantially increased the coverage and generosity of cash transfers. By election day, the new programme was two and a half times more generous in real terms and covered nearly 50 per cent more households than Bolsa Família when Bolsonaro took office.Footnote ¹¹⁴ Auxílio and its subsequent updates, however, were highly improvised and paid little regard to technical criteria.Footnote ¹¹⁵ They were also rolled out suddenly with minimal consultation, sparking public confusion.Footnote ¹¹⁶

The pandemic changed Bolsonaro’s political strategy. Facing multiple impeachment motions for his mismanagement of the pandemic, and mired in corruption scandals, in 2020 Bolsonaro made an alliance with the Centrão (big centre), a loose block of legislative parties whose defining characteristic is its willingness to back whoever holds the presidency in exchange for pork and patronage opportunities.Footnote ¹¹⁷ In exchange for protection from impeachment, the executive branch delegated much of its control over spending priorities to the Centrão-controlled Congress through rapporteur’s amendments to the federal budget, a process colloquially known as the ‘secret budget’ (orçamento secreto). This opaque but legal mechanism grants legislators funding for local projects with minimal oversight. Although a long-standing feature of Brazilian politics, this type of behaviour became supercharged under Bolsonaro.Footnote ¹¹⁸

With pandemic relief set to expire and elections approaching, in August 2021 Bolsonaro announced Auxílio Brasil. The new programme was to award each household at least R$300 a month – 50 per cent more than the average Bolsa Família stipend at the time – and would be extended to everyone on the waiting list for the programme.Footnote ¹¹⁹ In October, to the surprise of congressional allies, Bolsonaro announced a R$400 minimum. Programme design decisions were made without access to up-to-date budget data.Footnote ¹²⁰ Social policy experts derided the process as ‘improvised’, ‘irresponsible’ and ‘embarrassing’.Footnote ¹²¹

Auxílio Brasil ‘launched’ in November, six weeks before funding was secured. As a result, Bolsa Família continued unchanged under the new name.Footnote ¹²² This caused chaos at welfare agencies, with confused beneficiaries camping out overnight to get information. As one official noted, ‘people are uninformed, and even we, employees, don’t know either. There is no information from the federal government’.Footnote ¹²³ Funds were secured two weeks before the year-end deadline, allowing per-household stipends to start in January. The amount secured was significantly larger than what was needed for Auxílio with the rest going toward ‘secret budget’ initiatives.Footnote ¹²⁴ During the first seven months of 2022, an additional 3.6 million households were enrolled, representing a 25 per cent increase.Footnote ¹²⁵

The introduction of a flat minimum for per-household cash transfers was widely seen as a step backward from the carefully designed Bolsa Família. Marcelo Neri, an authority on Brazilian poverty, went as far as to argue that ‘the country unlearned the basic lesson, which is to give more to those who have less’.Footnote ¹²⁶ A single, easy-to-calculate household payment made sense during the pandemic when the imperative was to quickly get money to the most people possible. But during normal times it made stipends less progressive – larger households, which tend to be poorer, received less money per capita. This also incentivised two-parent households to register as separate households, something single-parent households, which tend to be poorer, could not do.

Discretionality and Politicisation

As in Mexico, Brazilian cash transfers became increasingly personalised and politicised. Bolsonaro bragged of being ‘the president who made the world’s largest social programme’.Footnote ¹²⁷ This marked a complete U-turn relative to his original position on cash transfer policy and actions during the first months of his administration.

Upon coming to office, Bolsonaro’s administration launched a crackdown on supposed Bolsa Família fraud that disproportionally targeted the PT-backing north-east of the country. Between January and October 2019, an average of 5,400 households were enrolled in the programme each month. However, between January and May, each month only about 26 households from the 200 poorest municipalities were enrolled. Between June and October, only one household was enrolled in 37 of those municipalities, and in another 64 none were enrolled.Footnote ¹²⁸ Discrimination continued after admissions picked up in January 2020. Of the 100,000 households enrolled that month, only 3,035 (3 per cent) were from the nine north-eastern states representing a quarter of Brazil’s population. The more prosperous south and south-east, which Bolsonaro conveniently won, received 75 per cent of the allocations.Footnote ¹²⁹ North-easterners later made up 61 per cent of the 158,000 households expelled in March 2020.Footnote ¹³⁰ North-eastern governors successfully appealed the cuts before the Supreme Court, but by then the pandemic had begun, making the cuts a moot point.Footnote ¹³¹

Transfer policy took another discretionary turn during Bolsonaro’s re-election campaign. Having already bent budget rules to expand cash transfers, in 2022 Bolsonaro blatantly ignored rules preventing presidents from ramping up social spending prior to elections. In July 2022, down in the polls and facing high fuel prices following Russia’s invasion of Ukraine, Bolsonaro returned to Congress. Once more, in exchange for access to the ‘secret budget’, Centrão legislators reformed the constitution. By declaring a state of emergency, the reform, which was approved in less than a week, allowed Bolsonaro to increase stipends to an unprecedented R$600 until the end of the year and further increase enrolment in the programme.Footnote ¹³²

In all, the administration injected close to US$4 billion into the economy in after-budget expenditures during the three months before the election.Footnote ¹³³ Between the programme launch and election day, Auxílio Brasil enrolled 6.6 million households, a 45.7 per cent increase on its predecessor Bolsa Familia.Footnote ¹³⁴ Bolsonaro outperformed the polls, finishing a strong second in the first round on 2 October 2022 against centre-left former president Lula. The following day, in a last-ditch effort, he moved up that month’s payments.Footnote ¹³⁵ Bolsonaro narrowly lost in the second round on 30 October 2022. Lula restored the name Bolsa Família but kept the flat per-household stipend.

A Right-Wing Social Policy?

As noted, Bolsonaro’s cash transfer policy can be divided into pre- and post-pandemic. Prior to the pandemic, the administration followed a right-wing strategy of retrenchment with the caveat that the cuts made disproportionally targeted the north-east. Post-pandemic, Bolsonaro carried out political manoeuvres to circumvent legal impediments to increasing the generosity and coverage of cash transfers before he faced the voters. Bolsonaro’s populist instincts ultimately trumped his right-wing ideology.

Pre-pandemic, the administration aggressively monitored Bolsa Família’s rolls with the intent of weeding out households with higher incomes and outdated or incomplete information. Between December 2018 and March 2020, 1.1 million families were removed from the programme’s rolls, a 7.6 per cent decline.Footnote ¹³⁶ New admissions were also restricted, causing the programme’s waiting list to balloon.Footnote ¹³⁷ Retrenchment was set to continue in 2020. The administration’s budget called for an 8 per cent cut in Bolsa Família spending, projected that the number of beneficiaries would remain flat and did not fund a bonus payment.Footnote ¹³⁸

The pandemic changed everything. After Bolsonaro’s popularity spiked amid the government’s generous social policy response, retrenchment plans went out the window. In line with Weyland’s predictions about social policy under populism, ‘the opportunistic goal of self-perpetuation won out over programmatic commitments’.Footnote ¹³⁹ It should, however, be noted that, unlike AMLO’s Becas, Auxílio remained conditional. Enforcement, which was suspended during the pandemic, resumed in early 2022.

Discussion and Findings

The near-simultaneous election of populist presidents from opposite ends of the political spectrum in Latin America’s two biggest countries at the end of 2018 offered a unique opportunity to study how populists make public policy. Crucially, the fact that Mexican left-wing populist López Obrador and Brazilian right-wing populist Bolsonaro both inherited and sought to overhaul their respective countries’ large and long-standing cash transfer programmes made it possible to isolate the impact of populism on policy and policy-making and to distinguish it from the impact of the host ideologies that accompanied each leader’s distinctive brand of populism.

This article’s main finding is that, despite AMLO and Bolsonaro’s diametrically opposed ideologies, rather than diverging, Mexican and Brazilian cash transfer policy converged on multiple fronts. Populism led to more improvised, politicised and clientelistic transfer policy in both countries. Ideology, though far from insignificant, played a secondary role in shaping the evolution of Mexican and Brazilian cash transfer policy relative to populism. In sum, the case studies reveal that populism trumps ideology.

The case studies demonstrate that there exists a distinct and coherent populist style of policy-making. Both populists were highly critical of existing programmes, accusing them (without evidence) of failing to reach ‘the right people’ and being used to buy votes. In office, both sought to put their own spin on transfer policy. AMLO fully scrapped Prospera two months into his term. Bolsonaro launched Auxílio Brasil during his third year, but only after concluding that generous and expansive transfers held the key to his re-election.

As predicted, policy-making became more erratic under populist rule. Auxílio changed repeatedly, often at the last minute and through legislative workarounds to circumvent laws ensuring fiscal prudence and preventing political manipulation of social policy during election years. Politics outweighed good governance as Bolsonaro sought to provide politically popular flat per-household benefits to the greatest possible number of households, to the detriment of ensuring the neediest received the most, as had been the case under Bolsa Família.

The transition to AMLO’s social programmes was more chaotic still. AMLO went further than Bolsonaro, completely replacing Prospera with untested programmes. Becas launched before their design was finalised and even before the politicised Wellbeing Census concluded. The rollout was opaque and saw some of the poorest Prospera beneficiaries being excluded from the new programmes without justification.Footnote ¹⁴⁰ To the confusion of families and educators, upper secondary school scholarships are still trumpeted as ‘universal’ even though only certain locales are covered.Footnote ¹⁴¹ Cash transfers lack set payment schedules and have been subject to delays.Footnote ¹⁴² Today, although more Mexicans than ever benefit from cash transfers, these benefits reach a smaller share of those most in need.Footnote ¹⁴³

Policy shifted in a clientelistic direction, though there is no evidence in either country of benefits being explicitly conditioned on support for the government. Early on, Bolsonaro weaponised cash transfers to punish the north-east, an opposition stronghold.Footnote ¹⁴⁴ He was forced to abandon this practice amid political backlash. And, after pandemic relief policies boosted his popularity, Bolsonaro’s strategy pivoted to finding legal workarounds to expand transfers prior to the 2022 election.

Though Bolsonaro’s motives were cynical, coverage and generosity increased to levels never before seen under prior administrations, including left-wing ones operating under more favourable economic conditions. Such an expansion required a bold and likely opportunistic politician. This offers an example of the positive potential of populism.Footnote ¹⁴⁵ Because of Bolsonaro’s brazenness, left-wing Lula inherited cash transfer coverage and generosity levels he may not have been able to achieve otherwise. Lower-income Brazilians came out on top, though only if current spending levels remain sustainable.

Like Bolsonaro, AMLO dramatically ramped up spending and coverage in the run up to presidential elections.Footnote ¹⁴⁶ But, in contrast to the Brazilian case, there is no evidence that AMLO’s programmes sought to discriminate against opposition strongholds. Still, the so-called Servants of the Nation broke the law in trying to associate the new programmes with the president.Footnote ¹⁴⁷ AMLO cut the value of benefits for non-voting school-age children while creating comparatively generous benefits for those of or close to voting age. Further, his Servants of the Nation personally threatened beneficiaries that their benefits could end if AMLO lost the recall vote.Footnote ¹⁴⁸

Despite this populist convergence, there is evidence of ideological divergence. Although both leaders had long derided their countries’ programmes, their critiques differed along ideological lines. Left-wing AMLO claimed that much of Prospera’s budget went to ‘ghost’ beneficiaries and ‘intermediaries’ and not to the poor.Footnote ¹⁴⁹ In contrast, Bolsonaro claimed that many beneficiaries should be expelled for being too well off and that cash transfers made them lazy.Footnote ¹⁵⁰ Once in power, AMLO created unconditional programmes targeting a broader segment of the population. This, however, came at the expense of less generous benefits for Prospera beneficiaries, who tended to be the poorest Mexicans. Prior to the pandemic, Bolsonaro retrenched cash transfers by cracking down on better-off beneficiaries and, more controversially, on the north-east. However, Bolsonaro completely jettisoned this agenda in favour of massively expanding transfers after his approval ratings spiked with the launch of generous pandemic relief policies. Still, in contrast to AMLO’s programmes, Auxílio Brasil remained conditional on school attendance.

Beyond ideology, the seeming ease with which populist presidents were able to overhaul the programmes they inherited constitutes an empirical puzzle. Both programmes were internationally lauded and had persisted largely unchanged for two decades through multiple non-populist administrations from across the ideological spectrum. Loss aversion and status quo bias should have made beneficiaries leery of changes, even ones promising more generous benefits.Footnote ¹⁵¹ There was reason to believe that these were consolidated programmes whose beneficiaries would mobilise to defend their entitlements.

On the political side, populist leaders were particularly determined to rebrand cash transfers. These are highly visible programmes, attributed to presidents, that have been widely shown to boost support for incumbents.Footnote ¹⁵² They are also cheap and easy to set up and expand compared to more contentious improvements to education and healthcare quality.Footnote ¹⁵³ Further, cash transfers are a policy area where presidents have historically had significant unilateral authority. Regardless, at the time of the reforms, both leaders had working legislative majorities.

On the societal side, James E. Mahon Jr. posits that the potential for mobilisation in defence of these programmes may be limited by the heterogeneity of the informal sector (to which most beneficiaries belong) and the fact these programmes originated in government ministries rather than being the result of bottom-up mobilisation.Footnote ¹⁵⁴ Another possible obstacle is that having children in school, unlike being elderly, is a transitory phase. Parents cycle in and out of eligibility depending on their children’s age and enrolment status. Once their children exit the educational system, parents lose their reason to mobilise to defend these programmes.

The Mexican case is more puzzling than the Brazilian one. Mexico’s policy changes were deeper and more sudden than Brazil’s. Also, all Bolsa Família beneficiaries ended up better off under Auxílio, which was not the case for Prospera beneficiaries under Becas. Based on interviews, Viviana Ramírez and Ricardo Velázquez Leyer (2023) argue that strict conditionality enforcement made Prospera beneficiaries apathetic toward its demise and supportive of unconditional transfers, even if less generous.Footnote ¹⁵⁵ These are intriguing findings, but they are derived from the opinions of just 20 women in a semi-rural community in Puebla.

Counter to the many attacks AMLO and Bolsonaro hurled at them, Prospera and Bolsa Família were highly effective policies that were largely shielded from political manipulation. It is concerning that, because of populist rule, the gold standards in the region’s fight against poverty might degenerate into two more in the long line of clientelistic anti-poverty initiatives that used to come and go with each new president. Conditional cash transfers revolutionised social policy. These innovative, home-grown Mexican and Brazilian programmes diffused across Latin America and the world.Footnote ¹⁵⁶ The possibility that the recent pernicious trends in Mexico and Brazil detailed in this article might also diffuse across the region is a major cause for concern.