Democratic Erosion and the Stress Test of Liberal Democracy

Democratic erosion — also called backsliding or autocratization — is the gradual, typically legalistic weakening of the institutions that sustain existing democracies, initiated by elected incumbents rather than by coups. The defining feature of contemporary erosion is constitutional retrogression: incremental changes that preserve elections and legal forms while hollowing out effective competition, rights of contestation, and the rule of law. The data are unambiguous: V-Dem recorded 92 autocracies against 87 democracies in 2025, with 74% of the world’s population living under autocratic rule; Freedom House documented the 20th consecutive year of global freedom decline in the same period.

For this vault, democratic erosion is the analytical category that frames Bolsonarism, Trumpism, and the broader right-wing populist wave as instances of a general mechanism rather than national anomalies. The key micro-mechanism — affective polarization that reduces voters’ willingness to punish democratic norm violations — connects directly to the vault’s theses on thymos, identity, and the fragility of the political center in Brazil. The Levitsky-Ziblatt framework of informal norms as guardrails is especially relevant for understanding why the Nova República proved structurally vulnerable: institutions that rely on forbearance rather than formal rules are exposed precisely when polarization turns opponents into existential threats.

The field has produced a three-layer causal chain: structural stressors (inequality, economic shocks, status threat) that expand the anti-establishment electorate; cultural-perceptual shifts (polarization, moralization, identity-protective cognition) that lower the cost of norm violation; and institutional exploitation (autocratic legalism, stealth authoritarianism, abusive constitutionalism) that converts electoral victories into durable hybrid regimes. Central authors: Levitsky and Ziblatt (soft guardrails); Bermeo (forms of backsliding); Huq and Ginsburg (retrogression vs. authoritarian reversion); Svolik (partisan-democratic trade-offs); mounk (illiberal democracy vs. undemocratic liberalism); Mudde (populism as thin ideology).

What democratic erosion is and why it is mostly gradual

“Democratic erosion” (often discussed as backsliding or autocratization) is best understood as a state-led weakening of the institutions that sustain an existing democracy, typically initiated by elected incumbents rather than by military takeovers. This framing matters because it shifts attention from dramatic regime ruptures (classic coups) to incremental institutional deterioration—a pattern that recent comparative work treats as increasingly typical.

The empirical baseline is grim and supports the hypothesis that contemporary erosion is commonly gradual and legalistic. The V-Dem Institute reports that by the end of 2025 the world had 92 autocracies and 87 democracies, that 74% of the world’s population lived in autocracies, and that only ~7% lived in liberal democracies. In the same executive summary, V-Dem also highlights surging pressure on freedom of expression (its “most attacked aspect of democracy”) and major declines in multiple indicators compared with two decades earlier.

A parallel picture emerges from Freedom House: its Freedom in the World 2026 report finds global freedom declined for the 20th consecutive year (in 2025), with 54 countries deteriorating and only 35 improving. Freedom House explicitly notes that, outside armed conflict and coups, many status declines are driven by illiberal leaders using repeatable tactics that erode democracy.

This “slow death” view does not deny coups; it reweights them. Nancy Bermeo argues that while coups remain possible, classic open-ended coups are now outnumbered by other forms (including “promissory coups”), and that “more vexing” forms of backsliding have become more common—often legitimated through democratic institutions themselves. Complementing this, Stephan Haggard and Robert Kaufman define backsliding as incremental erosion of democratic institutions, rules, and norms driven by elected governments, emphasizing polarization and incrementalism as core mechanisms.

Finally, the “democratic recession” perspective—associated with Larry Diamond—situates the recent decades as a break from late–20th-century expansion: Diamond notes that around 2006 global expansion of democracy and freedom halted, with subsequent deterioration in average freedom and declining counts of electoral and liberal democracies (depending on measurement).

Institutional pathways and legal mechanisms that convert democracies into hybrid regimes

The institutional model explains how erosion is executed: it identifies a set of repeatable moves that weaken horizontal accountability, tilt electoral competition, and reduce civil and political liberties—often while maintaining electoral rituals.

A useful minimalist institutional core appears in Aziz Huq and Tom Ginsburg. They distinguish two modal paths: authoritarian reversion (rapid collapse) and constitutional retrogression (subtle, incremental erosion) across three “predicates” of democracy: (1) competitive elections, (2) political speech/association rights, and (3) administrative/adjudicative rule of law. Their point is directly relevant to the hypothesis: the contemporary spike is in retrogression—the slow-grind pathway that can preserve formal legality while gutting liberal-democratic substance.

Several legal-theory concepts converge on the same pattern:

• Ozan Varol describes “stealth authoritarianism” as a post–Cold War adaptation in which leaders perpetuate power through legal mechanisms that exist in democracies, cloaking repression in legality and legitimacy.
• Kim Lane Scheppele analyzes “autocratic legalism”: leaders elected through democratic means deploy law to dismantle constitutional systems, entrench themselves, and eliminate meaningful accountability.
• David Landau develops “abusive constitutionalism”: the use of constitutional amendment/replacement procedures to erode democracy while preserving elections and avoiding overt dictatorship—making regimes “less democratic” but not always fully authoritarian.

These frameworks align with the recurring institutional tactics catalogued in contemporary monitoring (and in the political-science mechanism literature): executive aggrandizement, strategic manipulation of elections, capturing/packing accountability institutions (courts, prosecutors, regulators, audit bodies), harassment of civil society, and media control—often sequenced to minimize backlash.

“Judicialization” is a double-edged part of this story. As courts become central political referees, incumbents have incentives to (a) disable courts’ checking capacity through appointments, jurisdiction-stripping, budget levers, intimidation, or “court-packing,” or (b) weaponize courts against opponents (and opponents respond in kind), escalating institutional conflict. Comparative scholarship stresses that court-packing extends beyond simple court expansion to strategies like “emptying” or “swapping,” underscoring how formal legality can mask entrenchment.

The overall institutional lesson is blunt: erosion is rarely one move. It is a portfolio of changes—each individually debatable, many legally defensible—that cumulatively crosses a threshold into a hybrid regime (electoral authoritarianism, competitive authoritarianism, “illiberal democracy,” or other mixed forms).

Polarization, populism, and elite gatekeeping as accelerants

Institutional tools explain how democracies are hollowed out; polarization and populism explain why publics tolerate it and why opponents struggle to coordinate resistance.

The central polarization finding is not merely “people disagree more.” It is that partisan identity and affective hostility can override commitment to democratic principles. Milan W. Svolik argues that when democratic principles are pitted against partisan interests, many voters will trade away the former—supported by comparative evidence spanning multiple cases. A closely related experimental literature finds that partisanship and policy preferences can weaken willingness to punish democratic norm violations.

The micro-foundations of polarization help connect this to “moralization.” Shanto Iyengar and coauthors synthesize evidence that affective polarization—dislike and distrust of the other party—has surged and is rooted in partisanship as a social identity, with consequences beyond politics. Lilliana Mason adds the mechanism of “social sorting,” where race, religion, ideology, and other identities increasingly align with party, reducing cross-pressures and increasing intolerance toward the outgroup.

Political psychology deepens the explanation of why polarization can become norm-permissive. Dan M. Kahan provides evidence and theory for “identity-protective cognition,” where information processing tracks group identity more than accuracy—especially in contested political disputes. Jonathan Haidt and collaborators’ Moral Foundations Theory suggests that political conflict is often experienced as moral conflict, with liberals and conservatives relying on different bundles of moral concerns—creating conditions in which opponents are seen not as rivals but as threats.

On the elite side, the key institutional theme is gatekeeping. The “soft guardrails” view associated with Steven Levitsky and Daniel Ziblatt emphasizes unwritten norms—mutual toleration and institutional forbearance—needed for constitutional democracy to function. Their argument is that when elites abandon gatekeeping and treat rivals as illegitimate, politics becomes a spiral of constitutional hardball—each side using maximal legal tactics, eroding constraints and trust.

Populism becomes especially dangerous when it fuses with polarization. Using the widely adopted ideational definition, Cas Mudde defines populism as a thin-centered ideology that pits “the pure people” against “the corrupt elite” and demands politics express the general will. This moralized anti-elite frame pairs naturally with anti-pluralist moves: if only one side represents “the people,” then checks, opposition, media, and courts become obstacles to be overcome rather than legitimate constraints.

Yascha Mounk sharpens the conceptual tension by arguing that liberal democracy can decay in two directions—toward “democracy without rights” (illiberal democracy) or “rights without democracy” (undemocratic liberalism). In practice, populist leaders often justify weakening liberal constraints (courts, media, minority protections) as restoring democratic responsiveness—especially when publics believe institutions serve only insiders.

Structural drivers that create demand for erosion

Structural explanations do not replace institutional mechanisms; they explain the conditions under which norm-breaking leaders become electorally viable and institutionally effective.

Economic inequality has moved from background condition to measurable risk factor. A 2025 cross-national statistical study in PNAS explicitly links income inequality to the erosion of democracy, placing inequality among the strongest predictors of democratic erosion across cases. This aligns with broader political-economy arguments: highly unequal societies are more prone to (a) perceived or real oligarchic capture, (b) legitimacy crises, and (c) coalitions that accept executive concentration as the price of “getting things done.”

Globalization and distributional shocks are another key channel. Dani Rodrik argues that advanced stages of economic globalization generate predictable political backlash, and that the form of populism varies with the type of shock and the cleavage it activates. Complementing this, Ronald Inglehart and Pippa Norris propose and test a “cultural backlash” framework in which perceived cultural threat interacts with economic insecurity, shaping populist support.

State capacity and bureaucratic integrity influence whether erosion becomes durable. A practical mechanism is: populist incumbents weaken neutral administration (civil service, regulators, procurement, audit), converting the state into a partisan resource. In turn, damaged capacity makes policy failures more likely, which can intensify polarization and justify further exceptionalism—a feedback loop between governance failure and concentration of power.

Finally, structure shapes opportunities for cross-border diffusion. V-Dem’s reporting highlights simultaneous autocratization across many countries and notes patterns like media censorship and civil society repression as common tactics, consistent with the idea that rulers learn from one another.

Cultural legitimacy, democratic disaffection, and the media-technology layer

Cultural explanations focus on the “demand side”: attitudes toward democracy, trust, and legitimacy, plus the information ecosystem that shapes those attitudes.

The most influential “deconsolidation” claim is associated with Roberto Stefan Foa and Yascha Mounk, who, using World Values Survey data, argue that citizens in mature democracies have become less satisfied with democracy and more open to nondemocratic alternatives—a “democratic disconnect.”

Two crucial caveats are needed for rigor. First, there is an active scholarly debate over whether support for democracy is truly collapsing versus changing in interpretation or being overstated in trend narratives. Second, dissatisfaction with “how democracy works” is not the same as preferring authoritarianism; dissatisfaction may reflect performative failure, corruption, or polarization without implying regime rejection.

Still, dissatisfaction is widespread and politically consequential. A 2024 multi-country survey synthesis by Pew Research Center reports a median 59% dissatisfied with how democracy is functioning, 74% saying elected officials don’t care what people like them think, and sizable shares seeing no party representing them. This pattern is consistent with a legitimacy environment in which norm-breaking “anti-system” strategies can be reframed as necessary correction.

Media and technology operate less as a single cause than as an amplifier of polarization and delegitimation. The research landscape is mixed, but several findings are sturdy enough for synthesis:

• Social media lowers barriers to political content production and distribution, including falsehoods, and interacts with declining trust in mainstream media.
• Evidence does not support a simplistic “echo chambers always depolarize exposure” story; experimental work suggests exposure to opposing views can increase polarization for some groups, depending on conditions.
• Monitoring data increasingly treats harassment of journalists, media control, and restrictions on expression as a front-line indicator of autocratization.

In short: cultural disaffection and informational fragmentation do not automatically produce authoritarianism, but they lower the costs of norm-breaking politics and raise the difficulty of coordinated democratic defense.

Comparing institutional, structural, and cultural models and assessing current risks

The “institutions vs. economy vs. culture” question is best answered by separating proximate mechanisms from enabling conditions and legitimation constraints.

Institutional models explain the how with the highest precision. Once a leader has executive power and (often) legislative leverage, the menu of constitutional retrogression is well described: degrade electoral competitiveness, shrink rights of contestation, and weaken rule-of-law constraints, typically via legalistic changes. Institutional models also best explain why erosion can proceed without a single “constitutional moment”: incremental steps confuse publics and divide opposition.

Structural models explain why erosion becomes electorally and socially plausible. Rising inequality, economic shocks, and perceived capture can create the demand for anti-establishment politics and reduce the legitimacy of procedural constraints. Structural factors also shape state capacity—whether institutions can credibly deliver basic governance and whether “anti-corruption” or “anti-system” narratives resonate.

Cultural models explain the permissive environment: they account for whether citizens interpret democratic conflict as normal competition or existential struggle. Here, polarization and moralization are the key bridges. Identity-based conflict, identity-protective cognition, and divergent moral frameworks make it easier for elites to justify norm violations as defense against an evil opponent. The cultural model also clarifies why democratic decline can emerge even when “formal” institutions look intact: legitimacy can evaporate faster than constitutions change.

A synthesis that fits the evidence is a three-layer causal chain:

1. Structural stressors (inequality, shocks, demographic/status threat, weakened party systems, governance failure) expand the electorate for anti-establishment challengers.
2. Cultural-perceptual shifts (polarization + moralization + identity-protective cognition + distrust) reduce punishment for norm violations and increase tolerance for exceptionalism.
3. Institutional exploitation (constitutional retrogression, autocratic legalism, stealth authoritarianism) converts electoral victories into durable entrenchment and hybrid regimes.

This chain tests liberal democracy’s internal tensions. Liberal democracy is not just “elections”; it is a compromise between popular sovereignty and constraints that protect rights and prevent domination. John Locke articulates an early version of this logic: political power is legitimate only via consent and is bounded by law, with legislative power constrained to prevent arbitrary rule and self-dealing. John Rawls reframes stability in pluralist societies around constitutional essentials that citizens can endorse despite deep moral disagreement—an “overlapping consensus” meant to keep conflict within legitimate bounds.

Pluralism is the hinge where the liberal-democratic bargain fails under populist moralization. Isaiah Berlin argues that human values are multiple and often incommensurable; attempts to force monistic reconciliation can justify coercion, and democracy is not automatically aligned with liberty. Judith Shklar (via the “liberalism of fear”) prioritizes preventing cruelty and abuse enabled by asymmetries of power, emphasizing rule of law as a restraint on domination—an explicitly negative liberalism meant to remain realistic about political danger.

These liberal and pluralist insights map directly onto contemporary erosion: when populists claim a single “people,” pluralism becomes treason; when opponents are illegitimate, toleration collapses; when restraint is seen as weakness, forbearance disappears.

The risk assessment is therefore not ambiguous. On monitoring indicators, the world is in a sustained downturn: V-Dem describes democracy for the average global citizen returning to late-1970s levels and emphasizes broad autocratization dynamics; Freedom House records two decades of net decline. The Economist Intelligence Unit explicitly frames a core contemporary problem as widespread disaffection with representative democracy and classifies regimes using categories that institutionalize “hybrid” regimes as a stable outcome, not merely a transition.

The final question—temporary deviation or structural symptom—has an evidence-based answer: both, but more symptom than deviation. The downturn has lasted long enough (roughly since the mid-2000s in multiple datasets and narratives) to rule out a short-lived anomaly. And the mechanisms are not external accidents; they exploit intrinsic tensions of liberal democracy: openness to anti-system actors, majoritarian pressure against constraints, and the fragility of informal norms under identity conflict.

At the same time, “symptom” does not mean inevitability. The same monitoring sources and scholarship also document democratic resilience, reversals, and the possibility of self-correction—especially where opposition coordination, institutional independence, and informational freedom remain strong enough to reimpose costs on would-be authoritarians.

Ver também

• affectivepolarization — affective polarization is the psychological mechanism that explains why voters tolerate democratic norm violations; it bridges the cultural-perceptual layer to the causal chain described here
• direita_radical — the radical right is the most common political vehicle for contemporary erosion; the institutional tools catalogued here (executive aggrandizement, court-packing, media capture) are the instruments that radical-right incumbents deploy
• przeworski_crises_of_democracy_resumo — Przeworski analyzes the structural conditions that produce the crises this entry describes as institutional erosion, with a comparative formal framework
• mounk_people_vs_democracy_resumo — mounk’s book is among the central reference texts for the populist-liberal tension at the core of backsliding: “democracy without rights” vs. “rights without democracy”
• thymos — unmet recognition demand is the psychological fuel of the populist culture that enables erosion; thymos connects the identity dimension to the institutional diagnosis