Democracy Requires Boring Institutions. Populists Know It.

Democracies collapse with theatrical flair—torchlit rallies, emergency decrees, opposition leaders in handcuffs. They recover, when they recover at all, through a process so dull that it rarely makes the news: the redesign of civil-service recruitment, the lengthening of judicial terms, the tedious calibration of electoral thresholds. This asymmetry is not an accident. Autocrats understand that drama erodes institutions faster than legislation. Democrats who wish to rebuild those institutions must accept that the work ahead is less inspiring than the rhetoric that won them power.

Between 2006 and 2024, Freedom House recorded democratic backsliding in 62 countries. Yet the record also contains recoveries. Spain rebuilt democracy after Francisco Franco; Chile after Augusto Pinochet; Poland after martial law; South Korea after military dictatorship. These transitions did not follow a single template, but they share a pattern. Each required the construction—or reconstruction—of institutions capable of constraining power, adjudicating disputes, and surviving changes in leadership. Each took longer than the leaders who initiated them expected. And each depended on elites choosing restraint over retribution.

The Architecture of Restraint

Institutions do not enforce themselves. Spain's transition, often cited as a model, succeeded because elites on both sides made binding commitments. In 1977, King Juan Carlos appointed Adolfo Suárez, a former Francoist official, as prime minister. Suárez legalised the Communist Party eight days before the first democratic elections—a move that enraged the military but signalled to the left that the game was not rigged. The 1978 constitution created a Constitutional Court empowered to strike down laws passed by any government, including the one that drafted it.

This was not magnanimity. It was institutional insurance. Parties that expect to alternate in power have an incentive to build guardrails they will later be grateful for. In 1981, when Lieutenant Colonel Antonio Tejero stormed the Cortes with 200 Civil Guards, the institutions held. Juan Carlos, who had appointed military commanders loyal to the constitution rather than to him personally, ordered them to stand down. They did. The coup failed within 18 hours.

Chile's recovery required a different calculus. Pinochet left power in 1990 but remained commander-in-chief of the army until 1998 and senator-for-life thereafter. The 1980 constitution he designed included designated senators—nine unelected legislators, mostly military officers—who could block reforms. For a decade, Chile's democracy operated under rules written by the dictator it had rejected.

Yet Chile's institutions survived because Patricio Aylwin's government chose not to purge the bureaucracy or pack the courts. Aylwin appointed a Truth and Reconciliation Commission but accepted that amnesty laws would prevent most prosecutions. The left denounced this as capitulation. It was, in fact, a long-term investment. By the time Chile amended the constitution in 2005, removing designated senators and reducing Pinochet's institutional vetoes, the military had accepted civilian authority. The courts, still staffed largely by Pinochet appointees, began authorising human-rights prosecutions. Restraint had bought time for norms to harden.

The Bureaucracy Question

Recovering democracies face a dilemma. The civil service was often complicit in authoritarian rule—enforcing censorship, processing political arrests, administering patronage. Yet firing tens of thousands of bureaucrats creates chaos. Poland's Solidarity government, which took power in 1989, chose selective purges: security services were disbanded, but most ministries were left intact. The gamble was that bureaucrats who had served the communist regime would serve democracy if given no alternative.

It worked, but only because Poland restructured recruitment. A 1998 civil-service law introduced competitive exams, fixed-term contracts, and insulation from political dismissal. By 2004, when Poland joined the European Union, its bureaucracy was among the least corrupt in the former Eastern Bloc. Hungary adopted similar reforms in 1992. Both countries have since regressed—Poland under Law and Justice, Hungary under Fidesz—but the mechanism of backsliding reveals the importance of what was built. Both governments had to pass new laws to re-politicise the civil service. The institutions did not prevent backsliding, but they made it harder and more visible.

South Korea's approach was more aggressive. After the 1987 democratisation, prosecutors launched investigations into the Chun Doo-hwan and Roh Tae-woo regimes. Both former presidents were convicted in 1996—Chun of treason and mutiny, Roh of bribery. The convictions sent a signal: no one was immune. But both men were pardoned in 1997, before serving half their sentences. The message was equally clear: accountability has limits.

This balancing act—prosecutions followed by pardons—became a template. It satisfied demands for justice without eliminating the political opposition or provoking a military backlash. The risk is that it normalises impunity. South Korea has prosecuted every living former president since 1988, yet corruption persists. Institutions constrain power, but they do not eliminate it.

The Electoral Formula

Electoral systems shape the behaviour of elites. First-past-the-post systems, which reward winner-takes-all competition, can entrench polarisation. Proportional representation, which distributes seats according to vote share, encourages coalition-building. Recovering democracies have tended toward the latter—not because it is more democratic in some abstract sense, but because it makes power-sharing a structural necessity.

Spain adopted proportional representation with a minimum threshold of three per cent. Chile used a binomial system—each district elected two representatives, forcing the right and left into coalitions. Poland adopted a five per cent threshold, preventing fragmentation but allowing smaller parties a foothold. None of these systems prevented populism or backsliding, but each created incentives for moderation. In winner-takes-all systems, losing means exclusion. In proportional systems, even opposition parties retain influence.

Yet electoral systems alone do not determine outcomes. Hungary's proportional system did not prevent Viktor Orbán from securing supermajorities and rewriting the constitution. Poland's coalition governments gave way to single-party dominance. Electoral rules matter, but they are not self-enforcing. They require politicians who accept the premise that today's minority could be tomorrow's majority.

What Is to Be Done

Democratic recoveries succeed when leaders prioritise institution-building over symbolic victories. This is unglamorous work. It involves designing appointment processes for regulatory agencies, lengthening the terms of constitutional-court judges, and insulating prosecutorial decisions from executive interference. It means resisting the urge to purge opponents from the bureaucracy or pack the courts with loyalists, even when the opposition did exactly that.

Three reforms stand out. First, independent fiscal councils. Countries from Chile to Sweden have created non-partisan bodies that audit budgets and publish assessments before elections. These councils do not control spending, but they raise the reputational cost of fiscal dishonesty. Second, merit-based civil-service recruitment with protection against political dismissal. Bureaucracies staffed by exam rather than patronage are harder to capture. Third, staggered terms for judges and regulators, ensuring that no single government can reshape the judiciary overnight.

These measures share a logic: they slow politics down. Autocrats thrive on velocity—emergency decrees, midnight votes, constitutional rewrites. Democracies recover when institutions impose delay, transparency, and consultation. Slowing politics is unpopular. Voters reward leaders who promise swift action, not procedural restraint. But the alternative is a system in which every election becomes an existential contest, because the winner controls not just policy but the rules of future competition.

The Patience Problem

Democratic recoveries fail when elites lose patience. Poland's first post-communist government fell in 1990, less than 18 months after taking office, because Solidarity splintered into factions. Spain's transition required eight years of coalition governments before the Socialists won an outright majority. Chile's centre-left coalition governed for 20 years before losing power. In each case, stability depended on elites accepting incremental progress rather than demanding revolution.

This is difficult when voters are impatient. Transitions often occur during economic crises. Spain's unemployment rate was 21 per cent in 1985. Poland's inflation hit 585 per cent in 1990. Voters expect democracy to deliver prosperity, not just process. When it does not, populists offer shortcuts. Hungary's backsliding began in 2010, during the aftermath of the financial crisis, when Fidesz promised to bypass Brussels and restore sovereignty. The institutions that had been built in the 1990s were dismantled in less than a decade.

The lesson is not that institutions are fragile—though they are. It is that they require constant defence. Democracies do not recover and then coast. They recover and then face new challenges: financial crises, migration, polarisation, technological disruption. Each challenge tests whether elites still believe in the rules when the rules produce outcomes they dislike. When the answer is no, backsliding begins.

The Long Game

Democratic backsliding is accelerating. Between 2016 and 2024, more democracies declined than improved for the first time since the Cold War ended. But the recoveries of the past 50 years offer a blueprint. They show that democracies rebuild not through inspirational leaders or mass movements, but through tedious institutional design: independent courts, protected bureaucracies, staggered terms, fiscal councils, proportional representation. They show that restraint pays dividends—but only if practised long enough for norms to harden. And they show that success is measured not in years, but in decades.

The leaders who initiate these recoveries rarely live to see them complete. Adolfo Suárez resigned in 1981, exhausted by coalition politics. Patricio Aylwin left office in 1994, knowing that Pinochet's designated senators still controlled the upper house. Lech Wałęsa lost Poland's 1995 presidential election to a former communist. Democratic recovery is, to put it mildly, a thankless job. But it is the only one that works.