After Democratic Erosion: Institutional Resilience and Democratic Distortion in Brazil

Image: Focus Pix / Shutterstock

This paper discusses Brazilian democracy which survived Bolsonaro, but did not return to its previous equilibrium. Breakdown was avoided through a combination of contingency and institutional resistance: a permissive fiscal moment that ultimately proved insufficient, an international signal that a government born of a coup would not be recognized, and, decisively, a judicial-electoral response concentrated in the Supreme Federal Court (STF), the Superior Electoral Court (TSE), and above all in Justice Alexandre de Moraes. The outcome was neither restoration nor breakdown, but a redistribution of institutional power under distorted conditions: an executive more constrained, by institutions that are themselves less accountable. The paper asks what a failed episode of democratic erosion leaves behind. Three questions organize the analysis. How did the institutional response that contained Bolsonaro redistribute power among Brazil’s political institutions? What kind of democratic defect did that redistribution produce? And why should we doubt that the result is a settled arrangement at all?

Contents

• Summary
• 1. The problem: survival without restoration
• 2. Democratic erosion and the conditions of containment
• 3. Congressional government and the accountability problem
• 4. Judicial guardianship, ministrocracia, and the accountability trap
• 5. Politicized state capacity and electoral opportunism
• Conclusion: contained but unresolved

Summary

Brazilian democracy survived Bolsonaro, but it did not return to its previous equilibrium. Breakdown was avoided through a combination of contingency and institutional resistance: a permissive fiscal moment that ultimately proved insufficient, an international signal that a government born of a coup would not be recognized, and, decisively, a judicial-electoral response concentrated in the Supreme Federal Court (STF), the Superior Electoral Court (TSE), and above all in Justice Alexandre de Moraes. That response was essential, and it was also extraordinary, stretching the ordinary boundaries of institutional action. Congress, for its part, used presidential vulnerability to bring under its own control powers that Brazil’s post-1988 system had shared with the presidency, above all over the federal budget. The outcome was neither restoration nor breakdown, but a redistribution of institutional power under distorted conditions: an executive more constrained, by institutions that are themselves less accountable.

The democratic problem is no longer only the threat of backsliding from the presidency. It is the institutional distortion produced by the very mechanisms that contained that threat.

This essay asks what a failed episode of democratic erosion leaves behind. Three questions organize the analysis. How did the institutional response that contained Bolsonaro redistribute power among Brazil’s political institutions? What kind of democratic defect did that redistribution produce? And why should we doubt that the result is a settled arrangement at all?

1. The problem: survival without restoration

The conventional interpretation of Brazil after Bolsonaro emphasizes democratic resilience. This interpretation is not wrong, but it is incomplete. Bolsonaro challenged democratic institutions, attacked the electoral system, politicized state agencies, relied on military symbolism, weakened environmental and scientific institutions, and questioned the legitimacy of electoral defeat. Yet he failed to overturn the 2022 election, remained unable to shut down Congress or the courts, and ultimately failed to remain in power.

From this perspective, Brazilian institutions appear to have worked. The problem is that democratic survival is not the same as democratic recovery. Avoiding breakdown does not necessarily restore democratic equilibrium. Democratic erosion changes the conditions under which political actors operate. It reshapes expectations, incentives, institutional boundaries, and repertoires of action.

The distinction matters conceptually as well as empirically. In practice, resilience is usually judged by containment: a democracy is counted as resilient when its institutions prevent rupture or autocratisation. This essay proposes a more demanding standard, on two grounds. First, survival is one dimension of resilience, but it should not be conflated with the restoration of democratic quality or of institutional balance: the mechanisms of containment can themselves durably alter the regime they defend. Second, and more fundamentally, resilience is a retrospective concept. To judge a democracy resilient is to treat an episode as closed—a shock arrived, institutions responded, and the system either recovered or did not. Brazil’s episode is not closed. The conflict among the institutions that contained Bolsonaro is unresolved and escalating, and the political project they contained remains electorally competitive. Under those conditions, to call the Brazilian case resilient is not to describe an outcome but to assume one.

In Brazil, Bolsonaro’s presidency weakened norms that had previously constrained executive behaviour. But the institutional consequences of this erosion were uneven. The institutions that controlled the executive became stronger in the process of resisting him. The STF and TSE became more central because Bolsonaro’s strategy directly targeted elections and the legitimacy of electoral defeat. Congress became more powerful because Bolsonaro was politically vulnerable and needed legislative support to protect himself from impeachment, although this process of congressional empowerment had been underway since the Dilma Rousseff years.

What emerged was not a return to the previous democratic order. It was a different configuration of power. Practices once treated as exceptional, transgressive, or institutionally unacceptable became more available within ordinary politics. Judicial intervention became more central to political conflict. Congressional control over the legislative agenda and budget execution became more entrenched. Technical agencies became more vulnerable to politicization. Electoral-year fiscal opportunism became easier to justify.

Brazil’s post-Bolsonaro democracy is therefore best understood as a transformed democracy. It should not be understood as a new equilibrium: the transformation is unfinished, and no settled configuration can yet be assumed. Formal institutional functioning was restored, but the practices, incentives, and institutional scars of authoritarian erosion remained embedded in the system. The central argument of this essay is that Brazil’s democracy proved resilient but not restorative: it survived Bolsonaro by strengthening institutions of executive control, especially Congress and the judiciary, in ways that now produce new democratic distortions.

2. Democratic erosion and the conditions of containment

Before turning to Bolsonaro, it is worth making explicit the conception of resilience this essay works with, and where it departs from current usage. Merkel and Lührmann (2021) define democratic resilience broadly, as the ability of a democratic system—its institutions, actors, and citizens—to prevent or react to challenges, and they distinguish three possible reactions: withstanding a challenge without change, adapting through internal change, and recovering, in each case without losing the democratic character of the regime and its core institutions. Read as a closed episode, Brazil would belong unambiguously to the second category: it neither withstood Bolsonaro unchanged nor recovered its prior configuration, but adapted, and the adaptation was substantial. The claim of this essay is that the classification is premature rather than wrong.

The difficulty is not that the classification misdescribes what happened; it is that each of the three reactions Merkel and Lührmann describe—withstanding, adapting, recovering—is a terminal state: each presupposes that the challenge has passed and that the system’s response to it can be assessed. The Brazilian case does not support that presupposition. Averting breakdown removed an incumbent; it did not establish that the country has left the trajectory it was on. The institutions that contained Bolsonaro are now in open conflict with one another, the accountability deficits described below are deepening rather than settling, and the political project the containment was meant to defeat is competitive in the coming election on a platform of amnesty. The United States, where the containment of 2020 was followed by a return to office rather than by a restoration, is the cautionary case (Kaufman 2026). Resilience is a verdict that requires an endpoint. This essay describes a phase, and makes no claim to know how it ends.

The Brazilian case suggests that democratic resilience should be understood not only in terms of whether democracy survives an authoritarian challenge, but also in terms of the institutional configuration that survival produces. Mechanisms of democratic defence may prevent regime breakdown while simultaneously generating enduring distortions in accountability, the separation of powers, and institutional balance.
Resistance to an aspiring autocrat is a set of choices made under duress, and this essay asks what those choices leave behind. Where the residue falls on horizontal accountability—the capacity of state institutions to check one another, in O’Donnell’s (1998) sense—the democracy that survives is not the democracy that was defended.

This places the Brazilian case at the intersection of two literatures: democratic resilience and regime quality. Rather than asking simply whether Brazil remained democratic or crossed into hybrid or illiberal territory (Merkel 2004), the case draws attention to transformations within democracy itself—institutional adaptations that preserve electoral competition while weakening other dimensions of democratic accountability.
Democracies need not cross a categorical threshold to undergo lasting qualitative change. Deterioration can occur inside the regime, and it can be produced, in part, by the very institutions that defended it. Applying that standard to Brazil requires beginning before Bolsonaro, with the conditions that made his project possible in the first place.

Bolsonaro did not emerge in a vacuum. His victory in 2018 was made possible by an exceptional conjuncture (Campello 2022): an economic collapse that began with the reversal of commodity prices in 2011–12 and deepened after 2014, a corruption investigation that implicated much of the political class, and the impeachment of a sitting president. The result was a profound rejection of established politics and politicians. By October 2017, eighty-two percent of Brazilians said they would support a candidate from outside the political establishment. Bolsonaro was unusually well positioned to capture that sentiment: after twenty-seven years as a congressional backbencher with little national prominence, he could plausibly present himself as an outsider to the very system in which he had long served. Polarization did not create these conditions, but it amplified them—and it has not receded. Comparative work on backsliding treats polarization as one of its central enabling mechanisms (Haggard and Kaufman 2021); in Brazil, as Power and Zucco (2026) show, it now structures political elites around competing understandings of democracy itself. Two implications matter for what follows. The institutional vulnerabilities Bolsonaro exploited were not of his making, and they did not disappear with his defeat. Nor did the polarization that made his project viable; it remains the environment in which the transformed institutional configuration described below operates. The damage traced in this essay, then, is not a self-contained episode of democratic erosion, but an additional layer of institutional distortion imposed on vulnerabilities that predated Bolsonaro and are likely to outlast him.

Bolsonaro’s presidency represented a sustained challenge to Brazilian democracy. His administration did not dismantle democracy through a single, sudden rupture. Instead, it weakened institutional constraints, delegitimized opponents, attacked oversight bodies, politicized technical agencies, and undermined trust in elections.

The attack on the electoral system was the most visible expression of this process. For years, Bolsonaro questioned electronic voting, suggested that electoral fraud was systemic, and cast doubt on the legitimacy of the institutions responsible for administering elections. These attacks were not merely rhetorical. They aimed to create the political conditions under which defeat could be contested as illegitimate.

This assault on electoral legitimacy was accompanied by the politicization of the state. Bolsonaro’s government undermined the autonomy of technical agencies and bureaucratic offices in areas such as environmental enforcement, public health, education, culture, science, and human rights. Institutions that depended on professional expertise were treated as ideological obstacles. Environmental agencies such as IBAMA, Brazil’s federal environmental enforcement agency, and scientific institutions such as INPE, the national space research institute responsible for satellite-based monitoring of deforestation, became emblematic of this process, as technical information and enforcement capacity were subordinated to political conflict.

This politicization also reached the Prosecutor General’s Office (PGR). Bolsonaro ignored the traditional triple list of the Federal Public Prosecutor’s Office [1] and appointed Augusto Aras as Prosecutor General, breaking with a practice followed since 2003. Although constitutionally permissible, the appointment weakened the perceived autonomy of the office at a moment when the PGR would have been central to criminal accountability involving the president and his allies. This matters because the later expansion of judicial action by the STF cannot be understood without considering the perceived paralysis or cooptation of the ordinary prosecutorial channel.

Bolsonaro also militarized and personalized executive power. He repeatedly invoked the armed forces as political arbiters and surrounded his government with military personnel. The appointment of thousands of military officers to civilian positions blurred the boundary between civilian administration and military authority. Even where formal military intervention did not occur, the association between executive authority and military support placed pressure on civilian institutions and normalized military participation in government. As a result, a preference for or rejection of praetorianism—defined by Power and Zucco (2026) as the acceptance of military intervention or tutelage in civilian politics—emerged as a powerful new cleavage in Brazilian politics.

Bolsonaro’s use of incumbency powers further reshaped the electoral playing field. This strategy became possible under a favourable and exceptional fiscal environment. Bolsonaro entered the final stretch of 2022 amid the commodity-price boom and fuel-price shock that followed Russia’s invasion of Ukraine. Congress then approved a state of emergency, formally justified as a response to rising fuel prices, which relaxed ordinary fiscal constraints and created space for extraordinary pre-election spending. Under these conditions, the administration accelerated the expansion of Auxílio Brasil, included new beneficiaries, created gas vouchers, granted subsidies to truck and taxi drivers, authorized payroll loans to welfare beneficiaries, and adopted other emergency measures.

These policies were accompanied by administrative actions that raised further concerns about electoral fairness, including efforts to obstruct free transportation for voters and Federal Highway Police operations that disproportionately affected roads in the Northeast on election day.

The conditions under which this strategy failed were contingent as well as institutional. The commodity windfall and resulting fiscal space proved insufficient to erase the political damage accumulated by Bolsonaro’s administration, much of it produced by the combination of governance failures and adverse external conditions that marked his presidency (Campello 2022). Nevertheless, the rapid acceleration of his polling numbers in the final months of the campaign and the extreme closeness of the final vote suggest that the timing of the election mattered. Had the election taken place only a few weeks later, the fiscal windfall and the distributive measures it enabled might have been enough to secure his reelection.

The international environment also mattered. The Biden administration made clear to Brazilian military leaders that Washington would neither support nor recognize a government born of a coup. Had Donald Trump been in office at the time, Bolsonaro would likely have faced a more permissive international environment.

Domestically, the decisive form of containment came from the judicial-electoral arena, especially the STF and the TSE. This was not accidental. Bolsonaro’s principal strategy of erosion targeted the electoral process itself: electronic voting, electoral administration, the legitimacy of the TSE, and the credibility of an eventual defeat. The institutional response therefore concentrated in the bodies responsible for defending electoral integrity.

Within this response, Alexandre de Moraes became the central figure. His dual position in the STF and the TSE allowed him to combine constitutional authority, electoral oversight, investigative tools, and punitive capacity in ways that raised the costs of anti-democratic escalation. The STF and TSE did not merely issue routine legal decisions. They moved aggressively to defend the electoral process, contain disinformation, punish attacks on electoral authorities, and signal that attempts to delegitimize or overturn the result would face institutional consequences.

This resistance was essential, but also extraordinary. To contain Bolsonaro’s attack on elections, the judicial-electoral system stretched ordinary boundaries of institutional action. It acted with unusual speed, visibility, and concentration of authority. The containment of Bolsonaro was therefore both institutional and highly personalized.

Congress played a different and more ambivalent role. It did not operate as a principled democratic counterweight to Bolsonaro. At various points, congressional actors enabled the government, bargained with it, or benefited from its weakness. Yet Congress did not become a unified vehicle for democratic rupture.
Bolsonaro’s need to maintain a legislative majority to shield himself from impeachment gave congressional leaders extraordinary leverage over the executive. Rather than simply constraining the president from outside, Congress used Bolsonaro’s vulnerability to expand its own control over legislation, budget execution, and distributive politics. Its role was therefore less one of democratic resistance than of institutional self-strengthening under conditions of executive dependence.

This distinction matters. The STF and TSE acted as guardians of electoral democracy, but in doing so they expanded the political centrality and discretionary power of the judiciary. Congress strengthened itself not because it consistently defended democratic principles, but because it used presidential vulnerability to bring into Congress powers that had historically been shared with the presidency. In both cases, containment altered the distribution of power.[2]

3. Congressional government and the accountability problem

One major transformation of Brazilian democracy is the rise of a far more powerful Congress. This transformation did not begin with Bolsonaro. It dates back at least to the Dilma Rousseff years, when the foundations of executive dominance over the budget began to erode. As Bonfim, Luz, and Vasquez (2023) note, “Until 2013, the Brazilian executive branch had control over the execution of all discretionary public spending.” Legislators could amend the budget proposal prepared by the government, but they did not control whether those amendments would be executed. This changed with the institutionalization of mandatory budget amendments, first for individual amendments and later for caucus amendments. These changes made the execution of parliamentary amendments less dependent on presidential discretion and reduced executive dominance over the budgetary cycle.

Bolsonaro did not create this congressional empowerment, but his presidency accelerated and radicalized it. The mechanism was not principled legislative resistance to executive authoritarianism. It was Bolsonaro’s political need to maintain a congressional majority capable of shielding him from impeachment. In exchange for support and survival, congressional leaders expanded their control over the legislative agenda, policy production, budget execution, and distributive politics.

This produced a structural shift in the balance between executive and legislative power. Brazil’s political system has always required coalition management and legislative bargaining, and the presidency has historically shared legislative power with Congress. Yet the presidency also retained substantial agenda-setting authority and control over budget execution. The new pattern went beyond ordinary coalition presidentialism. Congress acquired not only influence over legislation, but direct leverage over the resources and implementation instruments through which policy is made.

The key mechanism was the congressional capture of parts of the federal budget, especially through opaque instruments associated with the so-called ‘secret budget’ and rapporteur amendments. This was not simply a transfer of budgetary authority from one branch to another. It was a transfer toward opaque and highly parochial forms of allocation, through which resources could be channelled to localities according to legislators’ preferences, with limited transparency, weak national planning, incomplete public information about authorship and beneficiaries, and little connection to coherent federal programs or national policy priorities.

This shift produced several democratic distortions. Public spending became more fragmented, localistic clientelism was strengthened, and accountability weakened because voters struggle to identify responsibility when legislators control spending while the executive remains formally responsible for governing. The legislature became a primary governing actor without being clearly accountable as the government. Congress brought into its own hands powers over legislation and the budget that, in the post-1988 system, had historically been shared with or coordinated by the presidency, but it did not assume equivalent responsibility for national policy outcomes.

Correction is difficult because the actors who would need to approve reforms are the same actors who benefit from the current arrangement. Congressional leaders benefit from control over the legislative agenda, budget execution, and negotiations with the executive. Rank-and-file legislators benefit from geographically targeted resources. Local political machines benefit from discretionary transfers. Presidents may resent congressional power, but they increasingly depend on Congress to govern. The resulting outcome is therefore self-reinforcing: many actors criticize the system, but few have incentives to dismantle it.

4. Judicial guardianship, ministrocracia, and the accountability trap

The judiciary, especially the STF and electoral courts, played the central role in containing Bolsonaro’s attack on democracy. This role built on a longer process through which the STF had become a central actor in Brazilian politics, a phenomenon Oscar Vilhena Vieira (2008) famously described as supremocracia. During Bolsonaro’s presidency, however, this institutional centrality acquired a more defensive and politically charged form. The Court acted against attacks on the electoral system, constrained executive abuses, and responded to what Vieira (2022) describes as the authoritarian use of infralegal instruments by the executive. Judicial power was therefore essential to democratic survival.

The problem is that this expansion of judicial power did not end with the crisis that produced it. The STF became a central political actor, and the authority accumulated during the Bolsonaro years remained available after Bolsonaro left office. Courts can protect democracy, but when they become central political actors, they also blur the boundary between constitutional guardianship and political intervention.

Brazil’s judiciary faces a particular problem because the expansion of judicial authority has not been matched by equally clear rules of conduct, recusal, transparency, and horizontal accountability. Supreme Court justices exercise substantial individual power through unilateral decisions, agenda control, public speech, informal interactions with political actors, and control over the timing of collective deliberation. Some of these practices may have been defensible under exceptional circumstances. But once normalized, they generate unpredictability, politicization, and perceptions of arbitrariness.

This is the problem described by Arguelhes and Ribeiro (2018) as ministrocracia [3]: a judicial order in which individual justices, rather than the Court as a collective body, exercise broad discretionary power over politically consequential decisions. Monocratic decisions, requests for review that delay collective deliberation, individualized agenda control, and public political signalling all contribute to the personalization of judicial power.

This expansion of judicial authority cannot be understood apart from the weakening of the ordinary prosecutorial channel. Under normal conditions, the Prosecutor General’s Office would have played a central role in assessing criminal allegations involving the president and other federal authorities. Under Bolsonaro, however, this channel was politically weakened. The president ignored the traditional triple list of the Federal Public Prosecutor’s Office and appointed Augusto Aras, an ally, as Prosecutor General. Although the president was not legally bound to follow the list, the decision broke with a post-2003 practice that had helped preserve the institutional autonomy of the office. Aras’s tenure was widely criticized for its reluctance to pursue allegations against Bolsonaro and his allies. In this context, the STF’s expansion was not only a manifestation of judicial ambition; it was also a response to the perceived cooptation or paralysis of the institution that would ordinarily have filtered, initiated, or advanced criminal accountability against the executive.

The institutional innovation was therefore not simply that the STF decided politically consequential cases. It was that the Court, and especially Alexandre de Moraes, came to occupy multiple positions within the same accountability chain. The Inquérito das Fake News was opened ex officio by the president of the STF and assigned to Moraes to investigate fake news, threats, and attacks against the Court and its members.[4] The Court later upheld the inquiry, grounding it in Article 43 of its internal rules. This was a departure from the ordinary model in which criminal investigations are initiated and advanced through police and prosecutorial channels before reaching judges as adjudicators. In the new arrangement, the Court helped initiate, supervise, and later adjudicate processes linked to the defence of democratic institutions.

This pattern expanded during the Bolsonaro years. Moraes opened or oversaw investigations involving attacks on the electoral system, digital militias, disinformation networks, and anti-democratic mobilization. Some of these initiatives were triggered by requests or materials from the TSE, the Federal Police, Congress, or the PGR itself; others developed through evidence-sharing and procedural connections across inquiries. What changed was the degree of concentration. Judicial authority, electoral oversight, investigative supervision, sanctioning power, and later adjudicatory authority became increasingly centred in the same institutional arena, and often in the same individual justice. This concentration raised the costs of anti-democratic escalation and helped contain Bolsonaro. But it also blurred the line between investigation and adjudication, creating a form of judicial guardianship that was more active, more personalized, and harder to subject to ordinary horizontal controls.

The Bolsonaro years thus intensified ministrocracia by joining two dynamics that had previously been more distinct: the personalization of judicial power and the judicialization of democratic defence. During the crisis, this concentration could be defended as necessary because the ordinary accountability chain appeared compromised and the executive was actively attacking electoral legitimacy. After the crisis, however, the same concentration became part of the new configuration. A judiciary empowered to initiate, supervise, sanction, and adjudicate political conflict is more capable of blocking authoritarian escalation, but also more difficult to monitor without provoking claims that oversight itself threatens democracy.

The result is an accountability trap. The STF became indispensable to the defence of democracy, which makes criticism of the Court politically dangerous. Bolsonaro’s allies and other actors on the right have used the language of accountability to threaten and intimidate the Court, including through calls for impeachment of justices. Their goal is not necessarily to construct a more transparent judiciary, but to retaliate against the institution that helped block Bolsonaro’s authoritarian escalation. This contaminates the accountability debate: reforms that may be democratically necessary can be absorbed into an anti-STF agenda.

At the same time, the Court’s defenders often avoid the accountability question because they fear weakening the judiciary in the face of the far right. This leaves the STF in a position where its expanded authority is protected by its democratic role, even when that authority raises legitimate concerns. Judicial accountability becomes both necessary and politically dangerous.

Recent corruption and financial scandals involving politically connected economic actors illustrate this problem. The recent Banco Master controversy [5] is especially revealing not because the legal responsibility of individual justices has been established, but because it exposed the institutional vulnerability of a Court that combines extraordinary political authority with weak behavioral constraints.

The Banco Master case exposed reported proximity between financial elites and members or former members of the STF. Dias Toffoli was initially responsible for the case but later stepped aside after scrutiny over alleged links to Daniel Vorcaro, the bank’s controlling shareholder. Reports based on Federal Police material referred to Toffoli’s name in data extracted from Vorcaro’s phone and to business connections involving actors close to Banco Master and a resort associated with members of Toffoli’s family. Toffoli denied having received payments or maintaining financial ties with Vorcaro, and the Court defended the propriety of his conduct. Yet the episode revealed a basic vulnerability: when a justice with broad monocratic powers over a major financial investigation becomes the object of questions about proximity to one of the investigated actors, the Court’s authority depends too heavily on personal assurances and too little on pre-established rules of recusal, transparency, and conduct.

The controversy surrounding Alexandre de Moraes points to a related problem. Moraes became one of the most important judicial figures in the defence of Brazilian democracy during and after Bolsonaro’s presidency. Precisely because of that role, allegations involving professional ties between Banco Master and the law firm of his wife, Viviane Barci de Moraes, carry broader institutional significance. Reports have described multi-million-real legal contracts between the bank and her firm, while Moraes and his wife have denied irregularities and emphasized that he does not participate in her professional activities. The concern is not that such contracts prove illegality. The concern is that, in a context where the STF has become a central political arbiter, even indirect familial or professional entanglements with major economic actors can erode confidence in judicial impartiality. André Mendonça’s subsequent role as rapporteur further illustrates the concentration of authority in individual justices. After Toffoli’s departure, Mendonça became the central judicial figure overseeing the case, including decisions related to preventive detention and the handling of plea-bargain material that reportedly touches actors across multiple branches of government. This is not in itself improper. But it reinforces the broader problem of ministrocracia: politically consequential investigations, emergency measures, secrecy decisions, plea-bargain homologation, and detention orders can depend heavily on the authority of a single justice before meaningful collective deliberation occurs.

The concern about Moraes’s own entanglements sharpened on September 1, 2026, when a Federal Police report drawn from the analysis of Vorcaro’s phone, documenting his contacts with the justice, became public and calls for Moraes’s impeachment became widespread.[6] The institutional mechanics of the disclosure matter more here than the allegations themselves. The inquiry that produced the material was opened by Mendonça, appointed to the Court by Bolsonaro, acting on his own initiative and without the participation or prior knowledge of the PGR—that is, on the precedent that Moraes himself helped establish and that the Court upheld in the Inquérito das Fake News. The instrument built to defend the Court against external attack is now available for use against its own members, and by the same monocratic route. This is the argument of this essay in miniature: the residue of containment is not inert. It enters the ordinary repertoire, and it does not remain in the hands of those who created it.

The conflict is now no longer simply between courts and an authoritarian executive. It is a broader clash among empowered institutions. An emboldened Congress has attempted to initiate impeachment proceedings against Supreme Court justices. It has also, for the first time in Brazil’s recent democratic history, rejected the name proposed by the president for a seat on the STF. These moves do not amount to a coherent agenda of judicial accountability. They are part of a more ambiguous struggle over power after Bolsonaro: Congress has gained leverage over the executive, while sectors of the right use the language of oversight to weaken or intimidate a Court they associate with the defeat of Bolsonaro’s authoritarian project.

The STF, on the other side, has also become more difficult to scrutinize. Because the Court played a central role in protecting democracy against Bolsonaro, accusations against its members are often framed as attacks on democracy itself. This defensive posture is not simply rhetorical. Some justices have reportedly displayed discontent with President Lula for not intervening politically in processes in which they face accusations of corruption or conflict of interest. This reveals one of the central distortions of the post-Bolsonaro configuration: the judiciary’s role as democratic guardian can generate an expectation of political protection, even when the issue at stake is not democratic defence but accountability.

The Banco Master controversy intensifies this problem. Bolsonaristas and other actors on the right, some of whom have themselves been mentioned in connection with the scandal, have used the case to frame the STF as a source of distrust in Brazilian democracy. This does not make their critique democratically neutral. In many cases, attacks on the Court remain connected to resentment over the judiciary’s role in containing Bolsonaro. But it also does not eliminate the underlying accountability problem. When empowered justices face allegations of proximity to financial elites, conflicts of interest, or improper influence, and respond by treating scrutiny as an attack on judicial independence or democracy itself, the Court’s role as democratic guardian becomes harder to separate from judicial self-protection.

This is the institutional core of the post-Bolsonaro judicial problem. During Bolsonaro’s presidency, the STF was pushed into an extraordinary role because the executive became a threat to constitutional democracy and the ordinary prosecutorial channel had been politically weakened. But once the Court assumed that role, its internal safeguards did not evolve at the same pace as its political authority. The result is a Court powerful enough to protect democracy from executive authoritarianism, but insufficiently constrained by rules that protect democracy from judicial personalization, conflict-of-interest risks, agenda manipulation, and opaque individual decision-making.

The democratic distortion is therefore not simply judicial activism. It is the collision between judicial empowerment and weak horizontal accountability. A Court that exercises extraordinary power over democratic politics, electoral competition, major corruption investigations, financial crises, and intergovernmental disputes must also maintain credibility as an impartial institution. When empowered justices resist oversight precisely as allegations against them intensify, the judiciary’s role as democratic guardian becomes harder to separate from judicial self-protection.

5. Politicized state capacity and electoral opportunism

A further legacy of democratic erosion concerns the politicization of technical offices and the militarization of civilian administration. Bolsonaro’s presidency weakened the autonomy of institutions responsible for environmental enforcement, public health, education, science, culture, and regulatory policy. Technical expertise was often treated as ideological opposition; bureaucratic agencies were pressured, defunded, or staffed with loyalists.

The appointment of military personnel to civilian posts intensified this process. Military presence in government was not merely symbolic. It created an alternative governing network, reduced reliance on ordinary political negotiation, and blurred the boundary between civilian administration and military authority. Even without a formal military rupture, the normalization of military participation in civilian government lowered democratic safeguards against the politicization of the state.

The dismantling of technical agencies had lasting effects. Institutions such as IBAMA and INPE do not recover automatically when an anti-democratic president leaves office. Expertise is lost, careers are demoralized, and enforcement capacity deteriorates. Data production becomes politically contested, and the threshold for political interference declines. Once governments learn that technical autonomy can be violated without immediate collapse, future governments may also be tempted to use appointments, budgets, and administrative pressure strategically.

The problem did not disappear with Lula’s return. Lula is not Bolsonaro, and the democratic significance of their administrations is not equivalent. But Lula governs in the institutional world Bolsonaro helped create. One lesson of the Bolsonaro period is that technical and control institutions can be politicized more aggressively than previously assumed. Once this boundary is crossed, future democratic governments may also absorb parts of the repertoire, including strategic appointments to institutions of control, partisan use of state capacity, and distributive spending designed to preserve political support.

This is visible in Lula’s own appointments. Lula nominated Cristiano Zanin, his former personal lawyer, to the STF in 2023, and later nominated Flávio Dino, a former minister of justice and political ally, to another seat on the Court.[7] Both nominations were constitutionally valid and approved by the Senate, but they also show how presidents may place trusted figures in institutions that can shape investigations, accountability disputes, and impeachment-related conflicts A similar pattern appeared in the appointment of Paulo Gonet as Prosecutor General. Lula did not follow the traditional triple list of the Federal Public Prosecutor’s Office, even though the practice of choosing from that list had begun under Lula himself in 2003 and had been treated as an important informal guarantee of prosecutorial autonomy until Bolsonaro broke with it.[8] Here again, the issue is not formal illegality. It is the normalization of appointments that privilege personal trust, political alignment, or strategic protection over informal practices designed to preserve institutional insulation.

Another legacy of the Bolsonaro period is the normalization of using incumbency powers to reshape the electoral playing field. Bolsonaro did not succeed in overturning the 2022 result, but he demonstrated how close an incumbent could come to reshaping electoral competition through public spending, administrative pressure, and institutional permissiveness. The point is not that social spending is inherently illegitimate. The problem lies in the strategic deployment of emergency fiscal measures and state capacity to improve the incumbent’s electoral prospects under conditions of democratic stress.

This logic did not disappear with Lula’s return. The difference in democratic stakes between the two governments is real: the current government has not combined pre-election spending with a frontal attack on the electoral system or preparations to contest defeat. Yet Lula’s own pacote de bondades shows how electoral-year distributive politics has become part of the post-Bolsonaro repertoire. Even if less unconstrained than Bolsonaro’s 2022 package, it has been enacted in an election year and has an evident electoral dimension. Measures such as debt renegotiation, subsidized cooking gas through Gás do Povo, reinforcement of Minha Casa, Minha Vida, Crédito do Trabalhador and other credit lines, and credit for truck drivers, taxi drivers, or app-based drivers all operate as distributive instruments with potential electoral effects.[9] The democratic issue is not each policy’s substantive merit. It is the normalization of governing through electorally timed fiscal and credit measures designed to improve the incumbent’s position.

Conclusion: contained but unresolved

Brazilian democracy survived Bolsonaro, but survival did not restore the previous democratic order. The process of resisting executive authoritarianism strengthened the institutions capable of controlling the executive. Congress became more powerful over the legislative agenda, the federal budget, and policy implementation. The judiciary became more central to electoral politics, constitutional conflict, and the policing of democratic life.

These developments helped prevent democratic breakdown. They also left a worse and unsettled distribution of institutional power. The executive is more constrained, but not necessarily by more accountable institutions. Congress exercises governing power without clear responsibility for governing. The judiciary exercises extraordinary authority without equivalent horizontal oversight. Both institutions can invoke democratic defence, institutional autonomy, or separation of powers to resist scrutiny.

The central problem after Bolsonaro is therefore not only the risk of renewed executive backsliding, although that risk remains real, especially under conditions of electoral competitiveness by Bolsonaro-aligned actors. It is the institutional distortion created by the struggle against that backsliding. Brazil’s democracy now operates through a tense balance among empowered institutions that distrust one another and use accountability selectively. Congress can frame confrontation with the Court as oversight while pursuing retaliation, including through impeachment threats against justices and the rejection of presidential nominees to the STF. The Court can frame scrutiny as an attack on judicial independence or democracy itself while resisting accountability in cases involving allegations against its own members. The executive, meanwhile, depends on both: on Congress to govern, and on the Court as a barrier against renewed authoritarian escalation.

The persistence of strategic appointments and electoral-year distributive packages under Lula reinforces this point. Even without Bolsonaro’s authoritarian attack on electoral institutions, the nomination of trusted personal or political allies to institutions of control, together with the use of fiscal and credit instruments with clear electoral timing, suggests that one legacy of the crisis was to lower the threshold for incumbents to treat institutional appointments and state resources as instruments of political protection and electoral repositioning.

Institutional resistance to Bolsonaro was essential. Without the judiciary, electoral authorities, the press, the international environment, and the resistance of a minority opposition within Congress, Brazilian democracy might not have survived. But the institutions that contained executive authoritarianism did not return to their previous roles. They emerged stronger, more politicized, and harder to constrain.

Whether the configuration described here constitutes a new equilibrium is doubtful, and the doubt is worth stating plainly. An equilibrium implies that the forces in play have settled into a self-sustaining arrangement. What Brazil has instead is an open and escalating conflict among empowered institutions, with no arbiter above them and no established procedure for resolving it. Congress has attempted to impeach Supreme Court justices and the Senate has rejected a presidential nominee to the Court for the first time in 132 years; the Court has treated scrutiny of its own members as an attack on democracy. Neither side can subordinate the other, and neither has an incentive to accept a rule that would bind itself. That is not stability. It is a standoff whose outcome is unknown.

The electoral horizon makes this sharper. The project that the containment was meant to defeat has not been retired from Brazilian politics: Flávio Bolsonaro, campaigning on a general and unrestricted amnesty for those convicted of the January 8 attack—his father among them—is statistically tied with Lula in second-round polling, and he has tied that promise to the Senate races, asking voters to elect the majority that would pass it.[10] That prospect is not idiosyncratic to Brazil. In each of the three most recent presidential runoffs in South America the right defeated the left: José Antonio Kast in Chile in December 2025, Keiko Fujimori in Peru and Abelardo de la Espriella in Colombia in June 2026. The runoff is where these contests have been decided, and it has been decided the same way three times running. A Brazilian second round within the margin of error would therefore unfold against a recent regional backdrop in which the right has won three consecutive South American presidential runoffs.[11] If that is the immediate future, the institutional configuration analyzed here is not the settled residue of a resolved crisis. It is the terrain on which the crisis will be fought again, and the terms of the next round will be set by institutions that are more powerful, more mutually hostile, and less trusted than those that met the first.

The appropriate conclusion is therefore not that Brazil has arrived at a worse equilibrium, but that it has not arrived anywhere. Democratic erosion was contained without being resolved; the containment redistributed power without settling it; and the distribution now stands to be contested by an actor whose defeat was supposed to be its premise.

Brazil’s post-Bolsonaro democracy therefore proved resilient in the limited sense of survival, but not restorative. Its main vulnerability no longer lies only in an overreaching executive. It also lies in the distorted distribution of power produced when the institutions that checked the executive become centres of power insufficiently subject to democratic accountability.

REFERENCES

Arguelhes, Diego Werneck, and Leandro Molhano Ribeiro. 2018. “O Supremo Individual: mecanismos de atuação direta dos ministros sobre o processo político.” Novos Estudos CEBRAP 37(1): 13–32.

Bateman, David A. 2026. “Democracy-Reinforcing Hardball: Can Breaking Democratic Norms Preserve Democratic Values?” Comparative Political Studies 59(6): 1191–1225.

Bonfim, Raul, Joyce Hellen Luz, and Vitor Vasquez. 2023. “Mandatory Individual Amendments: A Change in the Pattern of Executive Dominance in the Brazilian Budgetary and Financial Cycle.” Brazilian Political Science Review 17(2).

Campello, Daniela. 2022. “When Incompetence Meets Bad Luck: Bolsonaro’s Third Year in the Brazilian Presidency.” Revista de Ciência Política 42(2): 203–223.

Haggard, Stephan, and Robert R. Kaufman. 2021. Backsliding: Democratic Regress in the Contemporary World. Cambridge: Cambridge University Press.

Kaufman, Robert R. 2026. “Democratic Resilience in the United States in the Second Trump Presidency.” Toda Peace Institute. https://toda.org/publications/policy-briefs-and-reports/democratic-resilience-in-the-united-states-in-the-second-trump-presidency/

Merkel, Wolfgang. 2004. “Embedded and Defective Democracies.” Democratization 11(5): 33–58.

Merkel, Wolfgang, and Anna Lührmann. 2021. “Resilience of Democracies: Responses to Illiberal and Authoritarian Challenges.” Democratization 28(5): 869–884.

O’Donnell, Guillermo. 1994. “Delegative Democracy.” Journal of Democracy 5(1): 55–69.

O’Donnell, Guillermo. 1998. “Horizontal Accountability in New Democracies.” Journal of Democracy 9(3): 112–126.

Power, Timothy J., and Cesar Zucco. 2026. “Varieties of Polarization: Democratic Values and Public Policies Among Brazilian Political Elites.” Working paper, May 14.

Vieira, Oscar Vilhena. 2008. “Supremocracia.” Revista Direito GV 4(2): 441–464.

Vieira, Oscar Vilhena. 2022. “Supremocracia e infralegalismo autoritário: o comportamento do Supremo Tribunal Federal durante o governo Bolsonaro.” Novos Estudos CEBRAP 41(3): 591–605.

NOTES

[1] The lista tríplice is a list of three candidates chosen by federal prosecutors in an internal ballot organized by their professional association. From 2003 until Bolsonaro’s 2019 appointment, every president selected the Prosecutor General from that list. The list is a convention rather than a legal constraint: the Constitution requires only that the appointee be a career member of the office and be confirmed by the Senate, so respecting the list had come to serve as an informal guarantee of prosecutorial autonomy.

[2] Whether Bolsonaro could have been contained by means less costly to democratic institutions is a question this essay raises but does not resolve. The comparison with the United States is suggestive rather than conclusive. There, the effort to hold a former president accountable proceeded through ordinary prosecutorial channels and largely failed (on the United States case, see Kaufman 2026); in Brazil, an assertive and highly personalized judicial response succeeded in blocking the escalation but left the institutional residue described here. Whether that residue was the price of success or an avoidable excess requires a counterfactual this essay cannot supply. On the broader question of when democratic actors should abandon institutional restraint, see Bateman (2026), who argues that the injunction against constitutional hardball may itself constrain the feasibility of the reforms that the anti-backsliding literature proposes.

[3] Ministrocracia—roughly, rule by the individual justices—is the term Arguelhes and Ribeiro (2018) use for a judicial order in which individual members of the Supreme Federal Court, rather than the Court as a collective body, exercise decisive power over politically consequential matters.

[4] The inquiry was opened by STF president Dias Toffoli in March 2019 and assigned to Alexandre de Moraes. See STF, “Presidente do STF abre inquérito para apurar ameaças e fake news que têm a Corte como alvo,” March 14, 2019, https://portal.stf.jus.br/noticias/verNoticiaDetalhe.asp?idConteudo=405790; and Agência Brasil, “Por 10 votos a 1, STF valida inquérito sobre fake news,” June 18, 2020, https://agenciabrasil.ebc.com.br/justica/noticia/2020-06/por-10-votos-1-stf-valida-inquerito-sobre-fake-news.

[5] The Banco Master controversy refers to a major financial-fraud investigation involving alleged irregular credit operations, political connections, and negotiations over the bank’s exposure to public and private financial institutions. Its relevance here lies less in the specific criminal allegations, which remain contested, than in the way the case exposed reported conflicts of interest and proximity between financial elites, political actors, and members or former members of the STF. See Reuters, “Brazil Supreme Court’s Toffoli steps aside from Banco Master case,” February 12, 2026; Associated Press, “Brazil’s Supreme Court orders the arrest of former head of Banco Master,” 2026; and El País, “La sombra del escándalo del Banco Master pisa los talones de Alexandre de Moraes,” April 9, 2026.

[6] On the report and the Court’s first session after its disclosure, see O Globo, “Ministros do STF ignoram crise envolvendo Moraes em primeira sessão após revelação de relatório com mensagens de Vorcaro,” September 2, 2026. Press accounts note that it has not been established whether Moraes replied to the messages.

[7] On Zanin’s nomination and inauguration, see STF, “Cristiano Zanin toma posse como ministro do STF,” August 3, 2023, https://portal.stf.jus.br/noticias/verNoticiaDetalhe.asp?idConteudo=511706&ori=1. On Dino and Gonet’s Senate approval, see Associated Press, “Brazil’s Senate approves Lula ally as new Supreme Court justice,” December 14, 2023.

[8] On the ANPR’s criticism of Gonet’s appointment outside the triple list, see “Associação de procuradores lamenta escolha de PGR fora da lista tríplice,” Gazeta do Povo, November 28, 2023.

[9] For journalistic accounts of Lula’s election-year package, see Folha de S.Paulo, “Bondades do governo Lula já somam R$144 bilhões em ano eleitoral,” May 8, 2026; and CNN Brasil, “Pacote de bondades de Lula injeta R$227 bi na economia em ano eleitoral,” May 17, 2026.

[10] Flavio Bolsonaro ended the first round of the election with 47 percent against 45 percent of the votes for president Lula. On the amnesty pledge, see Diário Carioca, “Flávio pede votos para ‘anistia irrestrita’ ao 8/1,” August 22, 2026, in which he asks voters to elect PL senators in order to approve a “anistia geral, ampla e irrestrita” for those convicted of the January 8, 2023 attack; and CNN Brasil, “Flávio defende anistia: ‘zerar o jogo’ e subir a rampa com Bolsonaro,” on his stated intention to secure a broad congressional amnesty so that his father may attend his inauguration in January 2027.

[11] Kast defeated Jeannette Jara 58 to 42 percent on December 14, 2025; Abelardo de la Espriella defeated Iván Cepeda 49.7 to 48.7 percent in June 2026; Keiko Fujimori defeated Roberto Sánchez in the Peruvian runoff of June 7, 2026. The margins differ sharply—Chile was a landslide, Colombia was decided by a single point—but the direction did not.

---

The Author