Skip to main content Accessibility help


  • Access


      • Send article to Kindle

        To send this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

        Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Democratic Decline in the United States: What Can We Learn from Middle-Income Backsliding?
        Available formats

        Send article to Dropbox

        To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

        Democratic Decline in the United States: What Can We Learn from Middle-Income Backsliding?
        Available formats

        Send article to Google Drive

        To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

        Democratic Decline in the United States: What Can We Learn from Middle-Income Backsliding?
        Available formats
Export citation


We explore what can be learned from authoritarian backsliding in middle income countries about the threats to American democracy posed by the election of Donald Trump. We develop some causal hunches and an empirical baseline by considering the rise of elected autocrats in Venezuela, Turkey, and Hungary. Although American political institutions may forestall a reversion to electoral autocracy, we see some striking parallels in terms of democratic dysfunction, polarization, the nature of autocratic appeals, and the processes through which autocratic incumbents sought to exploit elected office. These processes could generate a diminished democratic system in which electoral competition survives, but within a political space that is narrowed by weakened horizontal checks on executive power and rule of law.


They are co-authors of The Political Economy of Democratic Transitions (1995), Development, Democracy, and Welfare States (2008), and Dictators and Democrats: Masses, Elites, and Regime Change (2016). They thank Inbok Rhee for research assistance and Haohan Chen, Francisco Garfias, Peter Gourevitch, Blanca Heredia, R. Daniel Kelemen, and David Lake for comments on earlier drafts.

The election of Donald Trump has challenged the widespread assumption that rich, liberal democracies are invulnerable to subversion by autocrats who come to power through electoral means.1 Both in his election campaign and since taking office, Trump has exhibited many autocratic traits. He has stoked underlying ethnic and class divisions, demonized his opposition, attacked the media, weakened protection of civil and political liberties and challenged the independence of the courts and the federal law enforcement and intelligence apparatus.

Is liberal democracy in danger? By “liberal democracy” we mean a political system not only with electoral competition and turnover—a minimalist definition—but also “horizontal” checks on executive authority and robust protection of political and civil rights. Could incremental assault on these constitutional checks on executive power cumulate into a “competitive authoritarian” regime,2 defined as one in which the political playing field has been tipped decisively against meaningful challenges to incumbents?

Such outright reversions are still virtually non-existent in developed countries, and appropriate comparators are thus difficult to find. Nonetheless, we can draw some conclusions by comparing the Trump presidency to democratic backsliding in three middle-income countries: Venezuela, Turkey, and Hungary. All three countries had reached levels of per capita income at which the possibilities of reversal had once seemed highly unlikely.3 Moreover, democratic institutions in Venezuela and Hungary seemed relatively well entrenched, and Turkey’s democracy appeared on a road to consolidation. The United States, of course, is far richer and its political system much more institutionalized. But comparisons with these cases helps identify the causal processes through which democracy deteriorates even under otherwise favorable political and economic circumstances and thus provides insight into developments in the United States in the first twenty months of the Trump presidency.

We conclude that a transition to competitive authoritarianism in the United States is unlikely, although not impossible. In contrast to the middle-income countries discussed later, the American political system has institutional features that pose significant impediments to outright authoritarian rule, including high barriers to constitutional revision. However, we also show evidence in the United States of similar causal processes to those that led to competitive authoritarianism in the three backsliding cases. These developments could signal a gradual erosion of defining features of liberal democratic rule, including institutional checks on the executive, the protection of political rights and civil liberties, and norms of compromise with oppositions.

Theoretical Perspectives

In comparing the sequence of events in the backsliding cases to the United States, we are not seeking to test a single, overarching theory of regime change. Nonetheless, we are guided by two strands of theoretical literature on reversions from democratic rule. One strand focuses on how social polarization and regime dysfunction strain public support for democratic institutions. Political grievances driven by economic stagnation or high inequality have figured prominently in such discussions.4 However, ethnic, racial, and religious cleavages can be equally if not more potent sources of polarization and instability.5

A second, highly influential perspective builds on the seminal work of Juan Linz on democratic failures in interwar Europe.6 This work emphasizes elite polarization and the failure of political institutions to prevent the electoral success of extremists.7

Drawing on these approaches, we identify three interrelated causal processes associated with the reversion from democratic rule in middle-income democracies; they can be conceived as operating in overlapping phases. First, polarizing class or identity cleavages undermine support for centrist political forces and open the door for majoritarian or autocratic electoral appeals. Following McCoy, Rahman, and Somer, we conceive of polarization as a cumulative process through which cross-cutting cleavages are submerged into a single, re-enforcing dimension that pits “Us” versus “Them” on a range of issues.8 Destabilizing class or identity conflicts can originate from above or below, stoked by political entrepreneurs or emerging from underlying economic or cultural grievances in mass publics. The main point for our purposes is that polarization weakens norms of tolerance and self-restraint among competing political elites and increases the likelihood that illiberal majoritarian appeals will generate electoral support.

A second crucial stage in the reversion process centers on how electoral victories of autocrats are converted into dominant legislative majorities that acquiesce to the concentration of executive power. In inter-war Germany and Italy, as well as in a number of contemporary cases, coalitions between outsiders and established political forces (so-called “devil’s bargains”) allowed autocrats to accrete power and ultimately shutter democratic processes.9 In Venezuela, Turkey, and Hungary, however, autocrats exploited electoral victories that swept aside established parties and allowed their political parties to capture both the executive and significant legislative majorities. Supermajorities—whether won under existing rules or “engineered”—proved crucial. They allowed not only delegation to the executive but more fundamental constitutional changes that effectively subordinated the legislature to the executive altogether.

During the third phase, executive powers are used in a step-by-step fashion to weaken institutions of horizontal accountability, oppositions, and political and civil liberties. In contrast to military coups or other abrupt authoritarian seizures of power, the incremental nature of this process makes it difficult to identify any single abuse that tips the balance decisively toward autocracy. But we show how early steps in this process facilitate and normalize later stages.10 Autocrats typically begin by using executive and legislative authority to undermine the independence of the judiciary, law enforcement agencies, and the press. Control over the economic resources of the state and corruption play an important role in building bases of elite support and deterring opposition. This combination of legal, economic, and coercive resources tilts the competitive playing field, but is then turned more directly to the corruption of the electoral process, intimidation of political challengers, and repression of civil and political liberties. In all three of the middle-income countries we examine, these abuses eventually crossed thresholds we associate with democratic rule.

Method and the Cases

To identify the universe of plausible comparators, we drew on the Liberal Democracy Index of the V-Dem data set,11 which has a range from 0 to 1 (least to most democratic). The index includes measures that comport with our conception of liberal democracy: indicators on the integrity of the electoral process—a prerequisite for minimal or “electoral” conceptions of democracy—but also measures of rule of law, judicial and legislative constraints on the executive, and respect for personal liberties.12

Unlike other datasets, V-Dem does not itself stipulate a democratic threshold, although it has been used by others to do so.13 Since no advanced industrial state has undergone a transition to authoritarian rule in the post-Cold War period, we identified all middle-income cases that had achieved a score of at least .5 for 8 years or more from 1992–2016.14 These selection rules eliminated low-income cases, nearly all of which did not in any case meet the .5 threshold. The 8-year threshold captured cases with at least one turnover in government, and that had thus accumulated at least some democratic history.

From that group, we then identified all countries that had seen a statistically significant decline from their peak score during the 1992–2016 period, yielding a list of 11 cases: Bolivia, Brazil, Ecuador, Hungary, Macedonia, Nicaragua, Poland, Serbia, South Korea, Turkey, and Venezuela. Among these countries, Brazil, South Korea, and Poland achieved liberal democracy scores above .75, but for several reasons we did not select them for closer consideration. Neither Brazil nor South Korea fell below the .50 threshold and their electoral systems remained intact. Poland’s decline has followed a path that resembled Hungary’s, but with a more recent history that makes it harder to assess. Most importantly, these three cases exhibited significant margins of error in the V-Dem coding.

We highlight Venezuela, Hungary, and Turkey not only because they are widely recognized as prominent examples of reversion,15 but also because they are difficult cases. As noted, all three had reached levels of per capita income that make them anomalies for modernization theories. Both Venezuela and Hungary exceeded the famous $6,055 threshold (1975 dollars) identified by Przeworski et al.16 While Turkey had not quite reached that level, it was solidly in the upper middle-income category.17

Backsliding in these countries was also surprising because of prior democratic progress. Following the overthrow of a military dictatorship in 1958, Venezuela experienced four decades of continuous constitutional government. Dependence on petroleum, corruption, and the overwhelming influence exerted by party, union, and business elites with access to oil resources raised doubts about the extent of democratization.18 Nevertheless, a long record of stable electoral competition between two deeply-rooted centrist parties provided a marked contrast to the personalist, military, or one-party regimes that had dominated most of the rest of Latin America until the early 1990s. Hungary’s political reforms and its accession to the EU also appeared to mark it as a democratic success story. Democracy in Turkey was the most problematic of the three backsliding cases. Electoral politics during the 1980s and 1990s was distorted by a practice of military vetoes and discrimination against political Islam and the Kurdish minority. However, the election of Erdoğan’s moderate Islamist AKP in 2002 appeared to have launched the country on a more democratic trajectory, before the sharp political U-turn of the late 2000s.

We acknowledge that these countries vary widely on other potentially relevant parameters, from Hungary’s expanding links to the West (which had been expected to strengthen democracy), to Venezuela’s dependence on oil, and the religious and ethnic divisions in Turkey (which work against it). But our objective is not to provide a full account of all the causal factors that might have influenced the outcome. Rather, following Mill’s method of agreement (also confusingly known as a “most different” design), we highlight the effect of the three causal processes outlined above that appear common to the cases despite these other differences. Although idiosyncratic factors may have contributed to the decline of democracy in each individual case, the similarities in the causal processes we identify and the resulting political outcomes suggest that these processes are potentially relevant for understanding reversion more generally; they thus provide a framework for considering developments in the United States.

Democratic Dysfunction: Social Polarization and Political Strain

All three middle-income cases experienced reinforcing cycles of democratic dysfunction, social polarization, and declining support for moderate, democratic political forces and institutions. These stresses on democratic rule were compounded by polarizing political appeals that cast competitors as enemies and even existential threats to the nation and the people.

In Venezuela, social polarization can be linked directly to the failure of the two dominant parties to respond effectively to the debt crisis and oil price shocks of the 1980s. From 1980 to 1990, Venezuela experienced only two years of positive per capita growth, and as can be seen from figure 1, the 1990s were no better; there were three years of negative growth in the five-year period preceding Chávez’s election in 1998.

Figure 1 Growth of GDP Per Capita (1992-2016)

Source: World Bank Group (

Note: vertical lines indicate the elections of Chávez, Erdoğan, and Orbán respectively.

Voters initially reacted by voting out incumbents. Over time, however, public support eroded for the entire political elite and existing democratic institutions. Class polarization accelerated increasingly in the late 1980s,19 particularly after Carlos Andrés Pérez turned abruptly away from his expansive campaign promises and attempted to impose a tough austerity program. The initiative triggered massive protests in Caracas and a bloody response by the police.

Several years later, a dramatic coup attempt catapulted Hugo Chávez into prominence as a strident critic of a corrupt and ineffective democratic regime. But the attempt was not simply a sign of military praetorianism; it also reflected the breakdown of elite willingness to defend the constitution. Although none of the top party leaders supported the coup attempt, few openly condemned it; and Chávez was ultimately released from prison by Pérez’s successor in 1995. Meanwhile, street protests continued, and Pérez was driven from office before the end of his term by accusations of corruption and a vote for impeachment.

Electoral support for the two centrist parties—AD and Copei—deteriorated rapidly during the early 1990s. Between the congressional elections of 1988 and 1993, their combined vote share fell from 81 to 53% in the Chamber of Deputies and from 92 to 60 % in the Senate.20 The space vacated by the political center was filled by smaller left-wing protest parties and anti-establishment candidates.21 In the presidential election of 1993, Rafael Caldera, running as an independent, won with only 31% of the vote. His government veered erratically between populist policies and austerity, and his term ended in 1999 with the economy and the political system in profound disarray. These developments opened the way for Chávez to successfully contest for the presidency on the basis of attacks on “neoliberal” elites and a promise to “refound” Venezuelan democracy. He won with 56% of the popular vote in a fragmented field, while the established party candidates could muster only 11%.

In contrast to Venezuela, Hungary appeared to be an economic as well as a political success story through the mid-2000s. But Hungarian society was deeply divided between the pro-European liberals and Social Democrats, who had engineered the transformations of the post-communist economy, and more religious, nationalist, and conservative voters in the small towns and rural areas. The disaffection of these conservative sectors provided an electoral opportunity for Viktor Orbán. Orbán had entered politics during the transition as a liberal democrat, but by the mid-1990s, he and his Fidesz party aggressively courted this conservative base with increasingly strident nationalist appeals.

Fidesz defeated the Socialist and liberal parties in 1998, but the latter regained office in 2002 and were reelected in 2006. Their political support fell drastically, however, following a sharp economic contraction at the time of the global financial crisis (see figure 1) and a devastating corruption scandal. In 2006, a leaked tape of a private meeting of the Socialist party captured Prime Minister Ferenc Gyurcsány admitting that he had lied repeatedly to the public about the strength of the economy. Orbán, as well as the far-right Jobbik party, mobilized mass demonstrations against the government in the fall of 2006 that quickly turned violent and further polarized the public. When the economy was hit by global financial shocks, Orbán continued his relentless attacks on the government and “the public’s faith in democracy faded along with the economy.”22

In the parliamentary election of 2010, Fidesz won an overwhelming electoral victory with 56% of the popular vote as support for the centrist parties collapsed. The Socialist vote share fell from 43% in 2006 to 19%, which translated into only 15% of the legislative seats, and the LMP, a small green party, captured another 4% of the seats. The center-right MDF and center-left Free Democrats failed even to cross the 5% threshold to enter Parliament, leaving Jobbik as the only other major competitor to gain representation, with 12% of the seats. Once in office, Orbán’s effort to further polarize the electorate continued, with attacks on the EU, on outsiders such as George Soros, and a full-throated exploitation of the European migrant crisis to stoke racial and ethnic anxiety.

In Turkey, the election of Recep Tayyip Erdoğan’s AKP in 2002 also occurred against a fraught backdrop of a financial collapse (see figure 1), strains in the coalition linking state elites and large industrialists, and lingering authoritarian legacies. Unlike Chávez and Orbán, Erdoğan campaigned and governed initially from the center-right, with support from socially-conservative small and medium businesses from the Anatolian interior. Yet social welfare policies allowed him to attract votes from low-income beneficiaries of targeted welfare programs, and subtle Islamist appeals further widened the AKP coalition.23 Somer has argued that the highest priority of this moderate strategy was “state conquest” rather than democratization: the consolidation of control over authoritarian remnants in the civil and military bureaucracy.24 Nevertheless, during Erdoğan’s first term (2002–2007), he did move to strengthen Turkish democracy. His government deepened ties to the EU, opened the economy more widely to international markets, and expanded protection of civil and political liberties. Importantly, Erdoğan also rolled back the military’s historic veto power, prosecuting hundreds of “secularist” officers alleged to have plotted against the AKP government.

These advances, however, occurred within democratic institutions that were considerably weaker than those in Venezuela and Hungary and with latent cultural and religious divisions that Erdoğan could mobilize. In the 2002 election, the party system underwent an unprecedented consolidation. In addition to the crisis, high electoral thresholds favored the AKP; only the center-left CHP (Republican People’s Party) managed to win any parliamentary seats at all. Following Erdoğan’s landslide reelection as prime minister in 2007, AKP appeals increasingly took on majoritarian and Islamist tones,25 with the party as the standard-bearer of the nation and the “virtuous people” against an array of enemies: non-religious Kurds, Alevis, liberals, leftists, and seculars.26 Erdoğan’s dominance of the legislature ultimately set the stage for a weakening not only of authoritarian legacies but of checks on executive power. An incremental attack on secular rivals, civil society opposition, and the media followed.27

Pairing Venezuela with Turkey and Hungary is instructive because it underlines the significance of polarization per se, rather than any particular ideological appeal. Whereas Hugo Chávez appealed to the political left and exploited class cleavages, Erdoğan and Orbán identified with the political right and appealed to religious and rural interests while targeting urban elites, ethnic minorities, and foreigners. But none of these autocrats can be placed easily along a standard left/right continuum. On economic policy, Orbán and Erdoğan, as well as Trump, scrambled these distinctions considerably. Like Chávez, Orbán advanced nationalist economic policies and rejected globalization, and both he and Erdoğan instituted welfare policies favorable to bases of support among the marginalized. Rather, the commonalities can be found in the anti-system and polarizing character of their electoral appeals, which both reflected and accentuated underlying social divisions. All relied on majoritarian promises to overturn corrupt elites in the name of “the people,” and all increasingly demonized their opponents as criminals and even traitors.

Changing the Constitutional Balance of Power: The Role of Electoral and Legislative Majorities

In all three cases, converting votes into large parliamentary majorities proved a crucial step in the expansion of executive power and weakening of horizontal checks. Counterintuitively, majority and particularly supermajority control of the parliament eliminated the legislature as a major source of oversight because it allowed for the delegation of greater formal powers to the executive. Control of the legislature depended in part on disproportionality in the electoral system, features that magnified the legislative effect of the popular vote. And in the case of Venezuela, it required constitutional reforms that restructured the Congress more fundamentally. Once in office, however, these advantages were locked in through further changes to the electoral system and through outright fraud. If control of the legislature was not a sufficient condition for backsliding, all three cases suggest that it was necessary to the establishment of an electoral autocracy.

In Venezuela’s presidential system, Chávez’s strong showing in the 1998 elections was followed by his call for the election of a new constituent assembly that then deactivated the sitting congress and claimed authority to act in its place. Although opposition politicians protested this move, their discredited parties lacked the support to block it, and the Supreme Court reluctantly assented under pressure. Chávez loyalists gained 60% of the seats in the new unicameral legislature elected in August 2000, and in 2005, his coalition captured 100% control when opposition parties boycotted the vote. In 2006 he solidified his power by reorganizing formerly separate Chavista factions into a new ruling party, the United Socialist Party of Venezuela (PSUV). With the concentration of presidential power came vast new decree authority, which the president used to expand control over the educational system, agriculture, and other key sectors of the economy.

During the early 2000s, groups associated with the discredited old order reacted strongly to Chávez’s bid for power: demonstrations, a business lock-out, a short-lived coup by senior military officers in 2002, a three-month stoppage by managers and technical workers in the critical petroleum sector in 2003, and a constitutional referendum in 2004 on his continuation in office.

Chávez might not have survived these attacks if not for a sharp upswing in petroleum prices in 2003–2004 (see figure 1) that enabled him to radically expand social programs and stave off defeat in the 2004 recall referendum. Yet the president’s claim to democratic legitimacy—ratified not only by his own election but by support in the legislature—enabled him to face down these challenges even before the upswing in petroleum prices. Backing for the 2002 coup faded quickly after its civilian leader, Pedro Carmona, announced that he would scrap the 1999 constitution, and Chávez was returned to power. The president faced down the oil strike by firing most of the top and middle management of the national oil company (PDVSA), and—with congressional assent—assuming direct control over petroleum revenues. As we will see, the expansion of presidential power paved the way to attacks on other critical instruments of horizontal accountability, including the courts, law enforcement, and the National Electoral Council.28

In Hungary, Fidesz acquired legislative dominance without the dramatic constitutional struggles visible in Venezuela. However, as noted, the 5% threshold rule drastically reduced the representation of the centrist parties and allowed Fidesz to convert its 56% electoral vote into a 68% legislative supermajority. Orbán, moreover, had long held a dominant leadership role within the party. Through long-standing personal ties, control of nominations, patronage and outright corruption, he could count on the unwavering discipline of Fidesz parliamentarians. The legislature was thus turned into a rubber stamp. In the words of Janos Kornai, it became a “bill factory,” and not only with respect to regular legislation.29 Lendvai shows how crucial changes in parliamentary procedure allowed “urgent” emergency legislation, including constitutional changes, to be passed and become law without deliberation in less than 48 hours.30 Political power resided not in the legislature but in Orbán and his inner circle.

As in Hungary, Turkey’s electoral laws—in this case, a 10% electoral threshold—transformed the AKP’s electoral victories into large legislative majorities. In 2002, the AKP captured 66% of the legislative seats with only 34% of the vote. The main opposition party, the CHP, gained only one-third of the seats, while all of the parties that had comprised the previous government were shut out of parliament entirely.31 Unlike Fidesz, the AKP fell short of the 60% supermajority mark in the 2011 elections. But the party’s electoral pluralities grew throughout the 2000s, and except for a few months in 2015 when a pro-Kurdish party surged over the 10% threshold, it continued to win significant parliamentary majorities.

Following his reelection in 2007, Erdoğan sponsored a crucial constitutional amendment providing for the direct election of the president, paving the way for an AKP ally to take control of the office. Even within the parliamentary system, presidential incumbents wielded important powers over appointments to the civil service and the courts, and AKP control thus proved crucial to expanding executive power, including over the military. In 2014, Erdoğan further concentrated his personal power by assuming the presidency himself, while continuing to maintain de facto control over nominations and other levers of the AKP political machine. Aggrandizement of executive power in turn provided the legal foundations for an onslaught on political liberties and civil rights. We turn next to a fuller account of these crucial final steps in the backsliding process.

The Assaults on the Horizontal Checks, Political Rights, and Civil Liberties

Once executives had been delegated enhanced powers by pliant legislatures, the dismantling of horizontal checks, the weakening of political and civil liberties, and more frontal attacks on the opposition proceeded in a parallel but typically incremental fashion. The independent judiciary and media were typically among the early targets, because weakening them proved crucial to subsequent moves. Elected autocrats also quickly sought to control the agencies of law enforcement and to turn them against opponents and dissenters. The absence of checks further tilted the political playing field by deepening the potential for corruption, which in turn allowed the executive to pay off cronies and clients while persecuting private sector opponents. Governments ultimately reshaped voting laws, undermined independent monitoring of elections, and attacked opposition parties and civil society groups in order to minimize the risks of electoral defeat.

In Venezuela, the assault on the judiciary and the press came after Chávez’s crucial win in the 2004 referendum and his smashing victory in the presidential elections of 2006. The courts, it should be recalled, faced pressure and intimidation dating back to the initial controversies over the constituent assembly. Following the recall victory, however, the independence of the Supreme Court was destroyed by the appointment of twelve additional Chávez loyalists.

Pressure on the media also accelerated. The 2004 Law on Social Responsibility in Radio, Television, and Electronic Media (the Resorte Law) included vague prohibitions against content that “foments citizens’ anxiety or alters public order,” or “disrespects authorities.” In 2007, the government increased the pressure by withdrawing the broadcast license for RCTV, Venezuela’s largest independent news channel. And in 2010, it amended the media law to expand control over the internet and granted sweeping powers to the Venezuelan National Telecommunications Commission (CONATEL)—controlled by Chávez appointments—to cripple networks associated with the opposition.32 The government’s control over the press, the courts, the economy, and the electoral machinery created a highly uneven playing field that allowed Chávez to intimidate or disarm key opposition leaders, including Manuel Rosas, his main challenger for the presidency in the 2006 election.

After Chávez’s death in 2013, however, the gloves came off entirely as oil revenues declined and electoral support eroded. Chávez’s successor, Nicolás Maduro, has relied increasingly on the police, the military, and other more openly authoritarian instruments to stave off wide-spread popular unrest over sharply deteriorating economic conditions. When the opposition parties unseated the Chavistas in the legislature, winning a two-thirds majority in the 2015 elections, the democratic façade quickly fell away. The Supreme Court undercut the supermajority by invalidating the election of several opposition legislators; and when that proved insufficient, Maduro disempowered the Assembly entirely by engineering the election of yet another constituent assembly. Unlike Chávez in 1999, however, Maduro lacked the popular support to succeed in this maneuver without violent reprisals against demonstrators and the arrest of hundreds of political opponents. The long-term effectiveness of these reprisals remains uncertain, but it is clear that Maduro’s government has devolved into an even more openly authoritarian regime than its predecessor.

Hungarian backsliding was “softer” than in the other cases; outright coercion, threats, and police repression played a less significant role. Nevertheless, Orbán used his supermajority to implement measures similar to the ones we have seen in Venezuela. As in the other cases, the independent judiciary was an early target. In 2011, a new constitution expanded the size of the Constitutional Court, packing the new seats with Orbán loyalists. Additional reforms in 2013 placed further restrictions on the Court’s constitutional authority, allowing political appointees in the National Judicial Office to overrule previous constitutional judgments.

Orbán also placed considerable pressure on independent media outlets and on civil society organizations. Shortly following the 2010 elections, the government withdrew advertising to commercial media outlets and required them to register with a regulatory agency empowered to fine or even revoke licenses for infractions.33 Publishers and journalists did not initially face threats of imprisonment or physical assaults as was the case in Turkey. As in Venezuela, however, they have been threatened with penalties for publishing content that is not “balanced, accurate, thorough, objective, and responsible.”34

With parliamentary authorization, the government also removed horizonal checks by seizing control of formerly independent auditing and law-enforcement agencies, enabling wide-spread cronyism within the private sector. Corruption was widespread in Hungary well before the rise of Fidesz, but the weakening of checks and balances after 2010 paved the way to what Balint Magyar has termed a “post-Communist mafia state.”35 Orbán used legal authorities to expand his control of public resources—including through the central bank—and to deploy blackmail and threats of prosecution to force the sale of private firms, the returns from which were subsequently used to buy support from cronies and lower-level government and party officials.

Control of the legislature and courts also permitted a series of electoral “reforms” aimed at locking in Fidesz’s oversized parliamentary majority and placing its traditional rivals at long-term or even permanent disadvantage. New electoral rules encouraged the formation of splinter parties to divide the anti-government vote, established gerrymandered districts, and provided for the transfer of “surplus” votes won in single-member districts to party members competing in multi-member races. The extension of voting rights to ethnic expatriates in neighboring countries generated additional support for Fidesz and provided the margins required to achieve supermajorities.36

Similar legal tools were deployed to harass civil society groups and NGOs, directed especially at those with financial ties to George Soros, the liberal Hungarian-American billionaire. At the same time, the government has generally turned a blind eye to hate crimes committed against the Roma minority, sometimes with the collaboration of Jobbik or even Fidesz politicians themselves.37 Migrants suffered official detention and even more violent abuse, and in 2018, the government began to press for financial penalties against civil society organizations coming to their aid.38

An important puzzle is why the constraints associated with EU membership—a potential external check on the regime—did not impose more of a brake on Hungarian backsliding. An important part of the answer is that Fidesz votes enabled the conservative European People’s Party to maintain a majority in the EU parliament, providing Orbán with powerful supra-national allies.39 Despite some criticism from Brussels, the EU has stopped well short of suspending the massive fiscal subsidies it provided to the country, leaving the government free to pursue its illiberal course.

In Turkey, Erdoğan’s initiative to move from a parliamentary to a presidential system marked the beginning of the turn to outright autocracy. The independent judiciary and law enforcement agencies were early targets in this process. After his victory in the 2007 election—although won with only a plurality—Erdoğan repeatedly made majoritarian arguments to justify his actions, including open criticism of the Constitutional Court.40 Constitutional reforms ratified by a referendum in 2007 included measures to weaken the Constitutional Court and the Supreme Board of Judges and Prosecutors, crucial horizontal checks.

As in the other cases, the press was also an early target, with pressure brought to bear both through restrictive legislation and executive discretion.41 In 2006, anti-terrorism laws were revised to include jail sentences for journalists found to violate their provisions, with chilling effects on reporters. By the end of the 2000s, Turkey was one of the world’s leaders in the imprisonment of journalists. The Doğan company, the largest media group in the country, was subject to repeated attacks, large fines for tax evasion, and forced divestiture of key holdings. Following the coup attempt of 2016, its offices were attacked by pro-AKP mobs, and prosecutors began investigations into allegations of “terrorism propaganda.”

The space for civil society protest also narrowed sharply. In 2013, Erdoğan cracked down on wide-spread anti-corruption protests, leaving five demonstrators dead, over 8,000 injured, and nearly 5,000 detained. In a common pattern, opposition protest was securitized through prosecution under anti-terrorism laws and investigation of NGO finances.42 He also turned state power against rivals within the political and economic elite, launching bribery investigations against potential defectors within the cabinet and the private sector.

In 2016, the defeat of a military uprising provided an opportunity for Erdoğan to consolidate his power even further. The coup—allegedly promoted by Fethulla Gülen, an Islamist rival—was badly organized; and it faced strong opposition from opposition parties and the mass public as well as from other military factions.43 In the wake of the coup, however, Erdoğan engaged in sweeping arrests that caught up tens of thousands of people in the military, the civil service, and academe, apparently drawing on an “enemies list” compiled prior to the coup. Further exploiting this internal threat, the AKP and Nationalist Movement Party staged a constitutional referendum on a reform that substantially expanded formal presidential authority. Although the vote was bitterly contested and marred by accusations of electoral fraud, the government’s control of both the electoral system and the press allowed it to prevail. In 2018 the referendum was followed by a new presidential election that ratified Erdoğan’s claim to the office with a victory over a spirited political opposition. As was the case with Chávez and Orbán, Erdoğan had parlayed his electoral support and control of the legislature into more overt autocracy.

Backsliding in America: The Trump Presidency Twenty Months In

What can we learn from the experiences of middle-income backsliders about the risks to democracy in the United States? Several important differences emerge from this comparative exercise, but one deserves emphasis at the outset. It is highly unlikely that President Trump could obtain Congressional support for the kind of fundamental constitutional changes visible in Venezuela, Hungary, and Turkey. In the middle-income backsliders, formal constitutions lacked strong popular support and were far more malleable. Through popular referenda or even ordinary legislation, rulers could expand executive powers, alter the effective power of the judiciary and legislature, and write entirely new constitutions. In the United States, the “stickiness” of the constitution with respect to key checks such as term limits, the bicameral legislature, and the federal structure of the political system constitute significant barriers to full electoral autocracy.

Yet we also can see parallels with middle-income backsliders that signal an erosion of the liberal features of American democracy. We elaborate them using the framework deployed earlier.

Democratic Dysfunction, Social Polarization and Anti-System Appeals

Similarities are most striking with respect to the increasing polarization of American society around economic, racial, and ethnic grievances. Economic grievances have roots in globalization, technological change, and ultimately in stagnant wage growth, rising inequality, and its effects on political accountability.44 More immediately, they can be traced to the devastating impact of the Great Recession. Wide swaths of the population were left out of the modest recovery of the Obama years. Studies have noted how the Trump vote—resting on distinctive populist appeals—was associated with economic factors such as exposure to trade,45 the problems facing smaller metropolitan areas,46 and social markers such as poor health, lower social mobility, and weak social capital.47

Yet polarization in the United States is also intimately tied to identity politics.48 Parties began to realign over the civil rights movement in the 1960s and subsequently on charged issues such as affirmative action, welfare, and immigration that had deep racial, ethnic, and cultural overtones.49 Racial politics, of course, are deeply embedded in the specificities of American history, but bear at least a family resemblance to the cleavages mobilized by Orbán and Erdoğan in Hungary and Turkey. A cascade of “top-down” factors exacerbated grass-roots resentments over these issues, and fostered popular perceptions that “government doesn’t work.” But the divisive behavior and appeals of elite actors were targeted at a fertile “mass market,” characterized by long-standing racial and cultural divisions and a pronounced trend toward “tribal voting”.50

With polarization came a sharp decline of public trust in political institutions. The sources of this were, again, multiple, including top-down as well as bottom-up factors: policy stalemates in Washington, the perverse incentives of the primary system, and the flow of big money into right-wing organizations and communications networks that emphasized government failures.51 In every Pew poll between 2007 and 2015, trust in political institutions fluctuated between about 20 and 25%, the lowest level in more than 50 years.52 Loss of trust was not limited to particular institutions but extended to declining support for democracy itself.53

For candidate Trump, these conditions were ideal grounds for a campaign that resembled those of his counterparts in Venezuela, Hungary, and Turkey. He evoked the same majoritarian contempt expressed by all three middle-income counterparts for the institutional checks and the play of interests characteristic of a pluralist democracy. Like Orbán and Erdoğan, Trump exploited both economic distress and cultural cleavages, as well as a deepening gender divide. Despite his own problematic behavior toward women, he also posed as a champion of the “traditional values” of his evangelical support base.

The demonization of racial and ethnic minorities, however, was front and center. Drawing on ideas long championed by the so-called “alt-right,” Trump challenged Barack Obama’s citizenship, stoking the “birther” movement, exploited resentments against immigrants, and relentlessly championed the idea of building a border wall. These campaign themes continued once in office, starting at the very outset of his presidency with his effort to impose a complete travel ban on a group of predominantly Muslim countries and continuing with the equivocation around the white supremacist rally in Charlottesville. The administration’s emphasis on the threats posed by immigrants appeared to purposefully blur the line between undocumented and legal immigrants, further stoking resentments against minorities.

Trump’s campaign also had echoes of Chávez’s brand of left populism in the promise to restore national greatness, bring back industrial jobs, and “drain the swamp” of the Washington elite. He appealed particularly to voters in depressed rural areas and small towns and white males without college education and showed no interest in signature Republican issues such as fiscal probity or the reform of entitlements. Opposition to trade agreements—especially NAFTA and the Trans-Pacific Partnership (TPP)—was also a centerpiece of Trump’s campaign. Beginning in 2018, it became an increasingly important focus of his policy agenda, with the threat to withdraw from NAFTA and the unilateral imposition of tariffs on exports from both allies (Canada and the EU) and adversaries (most notably the trade war with China).

In one important respect, Trump’s rise to power differed from that of other populist leaders: his electoral support was weaker and the opposition more robust. Chávez and Orbán were swept into power with significant electoral mandates. Erdoğan won a plurality victory, but in a highly fragmented party system that left his rivals unable to coalesce into an effective electoral or legislative opposition. Trump won a narrow Electoral College victory due to a handful of votes in Michigan, Wisconsin, and Pennsylvania, but he lost the popular vote. And despite disproportionality in the Senate and gerrymandering in the House, he faced a unified Democratic opposition in Congress.

In a “tribal” political context, however, the intensity of the support from Trump’s base partly offset its lack of breadth. As we will discuss in the following section, the continuing loyalty of strong Trump supporters, which is highly concentrated in Republican electoral districts, provided the president with considerable leverage over his congressional majority, including over more traditional leaders such as Paul Ryan and the dwindling number of Republican moderates.

Changing the Constitutional Balance of Power

What is the danger that Trump will be able build on Republican control of the Senate and House to eliminate horizontal checks on his discretion and expand executive prerogatives? Unlike in the middle-income countries, and even with Republican majorities in both houses, the president could not count on the Congress to act entirely as a rubber stamp. Inept management on the part of the White House and deep intra-party divisions blocked important legislation, most significantly on the repeal of Obamacare, spending policy, and on immigration. Moreover, again unlike the middle-income countries, a strong opposition party remained a significant threat to retake Congress.

Despite these constraints, however, Trump could exercise strong influence over Congressional Republicans, both because of his strong popularity among the party’s primary voters and because his interests and those of the Congressional Republicans were closely aligned around at least some key policy and personnel issues. Did this alignment open the door to executive actions that might constitute actual derogations from democratic rule?

One important area of Congressional acquiescence concerns conflicts of interests within the executive branch and the possibility for outright corruption, a hallmark of backsliding in the middle-income cases. The Trump family itself brought extensive conflicts of interest into office and circumvented both the letter and spirit of ethics rules that had bound previous presidents.54 Yet similar lapses extended to over half of the president’s cabinet and other high-level appointments.55 Congressional tolerance for manifest conflicts of interest and discretionary regulatory relief has enhanced opportunities for Trump to build and maintain support from powerful economic interests with direct stakes in tax law, federal contracts, and environmental regulation and trade policy. At the same time, he also showed an unprecedented willingness to go after firms that disagreed with his priorities, singling out companies as diverse as J.P. Morgan, Amazon, and Harley-Davidson.

The most consequential challenge to Congressional oversight of the president concerns the investigations into the possibility of Trump’s collusion with Russia during the 2016 election campaign. Partisan allies, particularly in the House of Representatives, significantly impeded congressional investigations; but even under Republican control, Congress acted with greater independence than we might have expected in the middle-income countries. Backlash over the firing of James Comey and Attorney General Sessions’ recusal led to the appointment of Robert Mueller as special prosecutor and leading Republicans warned strongly against attempts to fire him.

It should be emphasized that this check showed strong signs of erosion over the course of 2018, and Republican support for the investigation wavered substantially as it came closer to the Trump campaign and the president himself. The president’s congressional allies blocked bi-partisan initiatives to require judicial approval of any attempt to fire Mueller, and increasingly trafficked in innuendo and rumor in an attempt to cast doubt on the impartiality of the investigation. Until the Mueller investigation is completed, we cannot know how congressional Republicans will respond if the report reveals unambiguous information of presidential malfeasance.

Whatever the outcome of the Mueller investigation itself, however, Trump’s repeated assaults on the independence of law enforcement agencies is likely to leave lasting damage. Ten days after his inauguration, he fired Acting Attorney General Sally Yates, who had warned that Gen. Mike Flynn might be vulnerable to Russian blackmail. U.S. Attorney Preet Bharara, who had been investigating real-estate fraud and money-laundering, was also dismissed, as were other Obama-era prosecutorial appointees. Trump approached all three top intelligence heads—CIA Director Mike Pompeo, Director of National Intelligence Dan Coats, and NSA Director Mike Rogers—in an effort to wind down the FBI investigation; and he railed publicly against Attorney General Jeff Sessions’ decision to recuse himself from the Russia inquiry. He and his allies in Congress have attacked not only the credibility of the Mueller investigation but that of the intelligence community and the FBI and Justice Department more generally. And in a stunning intervention in August 2018, he suggested that DOJ indictment of two sitting Republican House members for insider trading and campaign finance abuse should have been foregone in the interest of maintaining their seats and the House majority.56

The interests of Trump and Congressional Republicans, finally, converge strongly around executive and particularly judicial appointments that can provide long-term advantages over the opposition. Agencies formerly staffed on a bi-partisan or non-partisan basis are now being politicized. The list of such positions includes, most importantly, professional areas of the Justice Department, the Office of Management and Budget, the Census Bureau, and administrative law judges, all directly relevant for checking executive abuse. Numerous other executive agencies established to collect and evaluate economic and demographic data are also at risk and the winnowing of scientific expertise with respect to the environment and health has been well documented.57

The most serious threat posed by Congressional collaboration on appointments, however, is the opportunity it offers to reshape the judiciary. In the middle-income backsliders, such initiatives were an essential component of the authoritarian playbook. Remaking the judiciary was an early step on the path to undermining the rule of law, the integrity of the electoral system and civil and political liberties. In the next section, we examine how this threat might play out in the United States.

The Assault on the Rule of Law and Civil Liberties

Unlike in the middle-income countries, the American judiciary has so far pushed back strongly against some of the most controversial presidential initiatives. Most notably, district courts initially ruled against President Trump’s signature efforts to block immigration from Muslim countries, to ban transgender personnel in the military, to limit funding to so-called “sanctuary cities,” and to separate immigrant families.

We should not, however, underestimate the opportunities for Trump and his Congressional allies to undermine judicial independence over the long run. Unilaterally, Trump exploited his pardon powers to signal a willingness to extend immunity to those willing to ignore the law and the courts. Collaboration with the Congress, however, centered on judicial appointments. With the elimination of the filibuster rule (a movement originated by Democrats under Obama), the Senate Republicans approved the appointment of Neil Gorsuch to the Supreme Court, restoring the strong conservative tilt that temporarily ended with the death of Antonin Scalia. The implications were increasingly clear in a series of decisions announced in July 2018, most notably, in a 5–4 decision to overturn lower court rulings on the travel ban and a decision to leave gerrymandered congressional districts in Maryland and Wisconsin in place. The nomination of Brett Kavanaugh to replace retiring Justice Kennedy further reflected a strong intent to fill the Court with justices who march in lock step with Trump and the Republican right on key issues such as executive power, voting rights, and protections for women, racial minorities, and LGBT populations.

The opportunity to politicize the judiciary extends to the lower courts as well. At the start of Trump’s term, vacancies and expected retirements placed him in a position to fill an estimated 38% of district and appellate court positions, significantly more than any predecessor going back as far as Ronald Reagan. With further retirements, Trump will conceivably be able to appoint half of the federal bench during his first term.58

It is important to note that the capacity of these new appointments to determine judicial decisions will be constrained by both liberal and mainstream conservative judges appointed under Obama and his predecessors; unlike his middle-income counterparts, Trump cannot rapidly restructure the system or purge his opponents. Even so, there is a serious threat that a constitutionally-created branch of the government—one that is already deeply divided along partisan lines—will become even more politicized and delegitimated.

The most direct threat to American democracy would be judicial acquiescence to restrictions on voting rights. Conservative initiatives to undermine the electoral process were, to be sure, underway well before Trump appeared on the scene. During the Obama presidency, the Supreme Court opened the floodgates to “dark money” in political campaigns with the Citizens United decision,59 and it voided a key provision of the Voting Rights Act that required pre-approval by the Justice Department of changes to voting laws in states with a history of discrimination. But the Court’s acceptance of extreme gerrymandering, noted earlier, provides a strong indication that it will not act as an effective check on efforts by state and local Republicans to suppress the votes of minorities, low income voters, and urban populations through redistricting, registration, and voter ID requirements. As in the middle-income countries, such measures can place opposition parties at a serious, long-term disadvantage and would constitute one of the clearer signs of authoritarian regress.

We turn, finally, to the assaults on the media and civil liberties. Trump’s rhetorical attacks on the press—“fake news,” “enemies of the people”—is unprecedented in the United States, as is his practice of calling out specific networks and even individual journalists for intimidating criticism. As in the other cases of backsliding, moreover, recourse to the “big lie” has contributed to the growing confusion about, or indifference to, verifiable facts. Although the media has fought back, devoting ever greater resources to checking the president’s penchant to bend the truth, the long-term economic difficulties faced by network television and print journalism have increased their vulnerability to such pressures.60

Despite these grave challenges, however, it is highly unlikely that Trump’s assault on the media will suppress its independence. The major news outlets are more diversified and less dependent on the government than those in the middle-income countries; far from succumbing to pressures for self-censorship, they have responded to government attacks with defiant criticism. Moreover, the multiplicity of news sources, both on cable and online, would make it hard for the government to control the flow of information to the same extent as in Venezuela, Turkey, or Hungary.

The larger threat is an acceleration of trends toward the segmentation of social media and cable TV consumers and attendant social and political polarization. Trump’s personal appeal and aggressive use of social media exploits and reinforces a decades-long division of the political community into separate cognitive worlds, with radically different understandings of the challenges facing American society and the range of viable solutions.

Broader civil and political liberties, finally, have also come under increasing pressure. Freedom House has registered a secular decline in rights in the United States since 2010, but the trend accelerated in 2017.61 Moreover, rights groups such as the Southern Poverty Law Center and the American Civil Liberties Union have catalogued important changes in government protections and public discourse since Trump’s inauguration.62 The Departments of Justice and Education have each changed sides in ongoing litigation or reversed commitment to prior consent decrees on a wide range of civil and political rights issues. These include voter ID laws, abusive police practices, the protection of gay and transgender rights (under the claim that such protections might violate religious freedoms), affirmative action policies at high-profile universities, the contraception mandate of the Affordable Care Act, and rights enjoyed by disabled students.

The federal government has undertaken even more visible steps to restrict the due process and non-discriminatory rights that foreigners have traditionally enjoyed under U.S. law. These include the Muslim travel ban (ultimately upheld by the Supreme Court), the protection of so-called Dreamers, and the treatment not only of undocumented immigrants but of asylum seekers and settled refugees as well. The separation of families at border detention facilities became a stark and highly politicized example of these issues in the summer of 2018. State and local governments, district courts, and NGOs have pushed back against many of these actions, but they remain vulnerable to possible Supreme Court decisions favoring the federal government.

Civil liberties, moreover, have been threatened not only by these specific policy shifts, but by Trump’s persistent “dog whistle” appeals to white identity politics and denigration of racial, ethnic, and other vulnerable minorities. These appeals include stereotypes of black communities in the United States, the confrontation with the NFL, the effort to restrict service of transgender personnel in the military, and the charged issue of how local police forces have managed allegations of racial discrimination, excessive use of force, and outright abuse.

Daron Acemoglu has argued that the weakness of institutional checks and balances leaves civil society as the last line of defense against the aggrandizement of executive power.63 And there is reason to believe that the vibrant landscape of advocacy and civil rights groups, unions, and other social organizations can help avert the repression of these actors that occurred in the middle-income backsliders. An indication of these possibilities was provided by the outpouring of over a million protesters across the country in the Woman’s March on the day after the presidential inauguration as well as a welter of more localized mobilizations. Such protests—as well as quieter types of lobbying and litigation—have long been a part of the American political repertoire, and it is unlikely that they can be suppressed as thoroughly or as violently as in the middle-income countries.

On the other hand, in both Venezuela and Turkey, backsliding governments were able to withstand large-scale and sustained mass mobilizations. Moreover, autocrats in all three middle-income countries also engaged supporters in counter-mobilizations that intimidated oppositions and even ended in violence. In the United States, conservative advocacy groups have expanded their power through support for the Tea Party movement, networks of think tanks, and lobbying organizations that rely on professional operatives as much as individual memberships.64 Moreover, in his boisterous political rallies and his winks toward the extremist protesters in Charlottesville, Trump has openly encouraged thuggish behavior; there is ample evidence of an increase in hate crimes under the Trump presidency.65 Rather than civil society acting as an effective check, the mobilization and counter-mobilization of civil society forces could ultimately extend the country’s polarization into the streets, providing opportunities for the invocation of “law and order” tropes or even national security rationales for curtailing civil liberties.

Conclusion: Looking Ahead

Our purpose is fundamentally comparative. We outlined a set of causal antecedents that appear to be associated with reversion from democratic rule and examined them in the context of three prominent middle-income examples. We then reviewed the extent to which similar patterns were visible in the first twenty months of the Trump presidency. What might we glean looking forward?

Although we consider it unlikely, we should not entirely dismiss the possibility that the United States might devolve toward a competitive authoritarian regime that locks the incumbents into power. An American “autocracy scenario” becomes substantially more likely if a strong economy, gerrymandered districts, and “tribalism” carry the Republicans to victory in the 2018 midterm elections and Trump to a second term in 2020. Continuing legislative majorities would allow the administration to sustain its efforts to remake the courts and to restrict voting and civil rights. A second term for Trump would not only increase the president’s control over the other branches of government, but would enhance his capacity to deploy tax, regulatory, and even police powers against political opponents.

But while these threats are real, our comparisons also highlight institutional and political checks in the United States that make an outright reversion to authoritarian rule far less likely than in the middle-income backsliders. Neither Trump nor his allies have tried, like their autocratic counterparts, to alter the formal features of the U.S. constitution. Unlike the backsliders, moreover, a robust opposition party continues to pose a significant challenge, even as a legislative minority; and the modest constraints imposed by the Republican-dominated Congress can be expected to increase exponentially in the event of a Democratic victory in the 2018 midterm elections. As we have seen, finally, the courts, the bureaucracy, media, and civil society have not rolled over. These institutions and other multiple centers of power are still likely to provide significant road blocks to initiatives that move in an openly authoritarian direction.

In the immediate future, the greater likelihood is for a continuation of the democratic dysfunction we have seen in the first twenty months of the Trump administration, characterized by legislative stalemates, fraught confrontations with the Democrats, the press, and civil society, and palace intrigue within the president’s inner circle. This stalemate is likely to be exacerbated by the crisis that will doubtless follow the publication of reports from the Mueller investigation that—as we already know from the public record—are not likely to be fully exculpatory.

But even if Trump’s power is further circumscribed by damning findings from the special prosecutor, a turnover in the Congress, or a defeat in 2020, we strongly doubt that the American system will somehow “snap back” toward a liberal democratic model of principled competition and political compromise. One reason is that Trump’s “abnormal” presidency is itself a product of long-term dysfunction and polarization that led to his election in the first place. These trends will not simply disappear after Trump leaves the scene.

The post-Trump era, moreover, will also bear the marks of long-term damage incurred during Trump’s time in office. Trump’s “takeover” of the Republican party has pushed it increasingly toward positions of intolerance and extremism, and Democrats’ defeat in 2016 has intensified divisions between the moderate and more militant wings of the party. The experience of the middle-income countries shows that division and declining support for established parties rendered them unable to serve as a check on autocratic tendencies and opened the door to anti-system appeals.

As discussed earlier, the most enduring institutional damage is likely to be to the legitimacy and integrity of the judicial system, with serious implications for civil liberties and the integrity of the electoral system. The credibility of the media has also suffered a serious blow that limits its traditional role as the fourth estate.

Perhaps most important, the Trump experience has accelerated the erosion of norms that had long been a bulwark of democratic stability in the United States, a point emphasized by Levitsky and Ziblatt in particular.66 At the elite level, the discourse of the Trump era has erased the boundaries that prevented open appeals to racial and ethnic animosities and the conventions that deterred politicians from portraying their competitors as existential threats. At the mass level, Trumpian populism has deepened the polarization of American society, increasing the tendency of competing “tribes” to set a lower priority on fair democratic procedures than on preventing a victory of the other side.67

The erosion of institutions and political norms may not extinguish “free and fair” electoral competition, as it did in the middle-income backsliders. But it does imply a significant weakening of the rule of law, accountability, and political rights commonly associated with liberal democracy.

Regimes that hold competitive elections but fall short on other dimensions of democratic rule, including effective horizontal checks, robust protection of civil and political liberties, and norms of tolerance have been called “electoral democracies,”68 “illiberal democracies,”69 or simply “feckless.”70 Although the specific definitions of such systems may vary, they share a number of features outlined by Carothers: there are “alternations in power between genuinely different political groupings, but . . . political elites from all the major parties or groupings are widely perceived as corrupt, self-interested, and ineffective [and the] alternation of power seems only to trade the countries problems back and forth from one hapless side to the other.”71

Carothers and the others just cited focus exclusively on relatively new democracies in Latin America, Asia, Africa, and the post-communist world. But as we have argued through this comparative analysis, their concerns are more relevant to trends in the United States than we would care to think. Even after Trump’s political exit, there is a genuine danger that competing parties may lack the capacity to check autocratic tendencies, that special interests will become increasingly entrenched in various institutional centers of power, and that the state will prove increasingly unable to provide legal protection and services to vulnerable sectors of the population.

Such systems are not necessarily vulnerable to outright reversion to authoritarian rule. But in the absence of a more effective and accountable democracy, American society will become increasingly vulnerable to alienation, incivility, and decentralized forms of interpersonal aggression and violence. These more subtle changes are not always well captured by current democracy metrics, but they are important—if underappreciated—components of the concept of liberal democracy.


1 Huq and Ginsburg 2017, Frum 2017, Levitsky and Ziblatt 2018, Sunstein 2018.

2 Levitsky and Way 2010.

3 Przeworski et al. 2000.

4 Ibid., Boix 2003, Acemoglu and Robinson 2006.

5 E.g., Norris and Inglehart 2018, Mudde and Kaltwasser 2017.

6 Linz 1978.

7 Capoccia 2005, Ziblatt 2017, Levitsky and Ziblatt 2018.

8 McCoy, Rahman, and Somer 2018, 18.

9 Levitsky and Ziblatt 2018.

10 Bermeo 2016.

12 To qualify as a democracy, regimes must first fulfill a minimum threshold on the Electoral Democracy Index: a score of at least 0.5 on this index plus thresholds with respect to de facto multiparty elections, that elections are free and fair, and that they allow for substantial competition and freedom of participation. The “liberal” component of democracy is captured by further threshold requirements with respect to measures of respect for personal liberties, rule of law, and judicial as well as legislative constraints on the executive. For discussions of the constructions of these indicators, see Coppedge 2017; Lindberg 2016; Mechkova, Luhrmann, and Lindberg 2017; Luhrmann, Lindberg, and Tannenberg 2017.

13 Coppedge 2017; Lindberg 2016; Luhrmann, Lindberg, and Tannenburg 2017; Mechkova, Luhrmann, and Lindberg 2017.

14 More precisely, we considered all cases that had been middle-income in at least one year, although, as will be seen, all on the list in fact were middle-income for most of the period.

15 For example, Corrales and Penfold 2015 on Venezuela; Lendvai 2016 on Hungary; Önis 2013 and Esen and Gumuscu 2016 on Turkey.

16 Przeworski et al. 2000.

17 In 1998 at the onset of the Chávez period, Venezuela’s per capita GDP stood at $12,574 in 2010 dollars. Adjusting for inflation, that is about the same as the famous $6055 in 1975 dollars identified by Przeworski et al. as a threshold of democratic invulnerability. In the late 2000s, when FIDESZ began to seize control in Hungary, the per capita income was $14, 652, well above the Przeworski threshold. Turkey’s level was somewhat lower in the late 2000s ($10,672), but it was still at a level that should have reduced the odds of backsliding; World Bank Group, available at

18 Coppedge 1997, Karl 1997.

19 Dunning 2008.

20 Ibid., 171.

21 Handlin 2018.

22 Mueller 2014, 16.

23 Gumuscu 2011, Keyman and Gumuscu 2014, Somer 2016.

24 Somer 2017.

25 Keyman and Gumuscu 2014, 45–54.

26 Yabanci 2016, 598.

27 Esen and Gumuscu 2016

28 Corrales and Penfold 2015, 20.

29 Kornai 2015, 20.

30 Lendvai 2016, 101–126.

31 Muftuler-Bac and Keyman 2012, 87.

32 Freedom House 2013.

33 Freedom House 2016a.

34 Kelemen 2017, 12.

35 Magyar 2016.

36 Orenstein, Kreko, and Juhasz 2015.

37 Cernusakova 2017.

38 Kingsley 2018.

39 Keleman 2017.

40 Dinçsahin 2012.

41 Somer 2016, Esen and Gumuscu 2016.

42 Esen and Gumuscu 2016.

43 Esen and Gumuscu 2017.

44 McCarty, Poole, and Rosenthal 2016; Gilens and Page 2014.

45 Autor et al. 2017.

46 Florida 2016.

47 Rothwell and Diego-Rosell 2017.

48 Kuk 2017.

49 For example, Mann and Orenstein 2012; Mickey, Levitsky and Way 2017.

50 Achen and Bartels 2016.

51 Mayer 2017, Neiwert 2017.

52 Pew Research Center 2015.

53 Foa and Mounck 2016.

54 See Sunlight Foundation 2018 for a catalogue of these.

55 Sheck 2018.

56 Baker and Fandos 2018.

57 Union of Concerned Scientists 2018.

58 Katz 2017.

59 Mayer 2017.

60 Anderson Jones and West 2017.

61 Freedom House 2018. With an overall score of 86/100, 51 countries now have higher scores than the United States.

62 Southern Poverty Law Center 2018a; American Civil Liberties Union 2018.

63 Acemoglu 2017.

64 Mayer 2017.

65 Southern Poverty Law Center 2018b.

66 Levitsky and Ziblatt 2018.

67 Aachen and Bartels 2016, Svolik 2017.

68 Mechkova, Luhrmann, and Lindberg 2017.

69 Zakaria 1997.

70 Carothers 2002.

71 Ibid, 10.


Acemoglu, Daron. 2017. “We Are the Last Defense against Trump: America’s Institutions Weren’t Designed to Resist a Modern Strongman. That Leaves Civil Society.” Foreign Policy, January 18. Available at
Acemoglu, Daron and Robinson, James A.. 2006. Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press.
Achen, Christopher H. and Bartels, Larry M.. 2016. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Princeton, NJ: Princeton University Press.10.1515/9781400882731
American Civil Liberties Union. 2018. “Donald Trump: Year One.” Available at
Andersen Jones, Ronnell and West, Sonja R.. 2017 “Don’t Expect the First Amendment to Protect the Media.” New York Times, January 25. Available at
Autor, David, Dorn, David, Hanson, Gordon, and Majlesi, Kaveh. 2017. “Importing Political Polarization? The Electoral Consequences of Rising Trade Exposure.” National Bureau of Economic Research Working Paper No. 22637, December. Cambridge: National Bureau of Economic Research. Available at
Baker, Peter and Fandos, Nicholas. 2018. “In Chastising Sessions over Indictments of Two Republicans, Trump Crosses a Line.” New York Times, September 4. Available at
Bermeo, Nancy. 2016. “On Democratic Backsliding,” Journal of Democracy 27(1): 519.10.1353/jod.2016.0012
Boix, Carles. 2003. Democracy and Redistribution . New York: Cambridge University Press.
Cappocia, Giovanni. 2005. Defending Democracy: Reactions to Extremism in Interwar Europe. Baltimore, MD: Johns Hopkins University Press.
Carothers, Thomas 2002. “The End of the Transition Paradigm.” Journal of Democracy 13(1): 521.10.1353/jod.2002.0003
Cernusakova, Barbora. 2017. “The Roma People’s Hungarian Hell: A Ruling by the European Court of Human Rights Highlights the Pattern of Persecution” Politico, January 25. Available at
Coppedge, Michael. 1997. Strong Parties and Lame Ducks: Presidential Partyarchy and Factionalism in Venezuela. Palo Alto, CA: Stanford University Press.
Coppedge, Michael. 2017. “Eroding Regimes: What, Where and When?” Varieties of Democracy Institute Working Paper Series 2017, 57. University of Gothenburg.
Corrales, Javier 2018. Fixing Democracy: How Power Asymmetries Help Explain Presidential Powers in New Constitutions, Evidence from Latin America. Oxford: Oxford University Press
Corrales, Javier and Penfold, Michael. 2015. Dragon in the Tropics: The Legacy of Hugo Chávez. 2nd ed. Washington DC: Brookings Institute Press.
Dinçsahin, Sakir. 2012. “A Symptomatic Analysis of the Justice and Development Party’s Populism in Turkey, 2007–2010.” Government and Opposition 47(4): 618–40.
Dunning, Thad. 2008. Crude Democracy: Natural Resource Wealth and Political Regimes. New York: Cambridge University Press.10.1017/CBO9780511510052
Esen, Berk and Gumuscu, Sebnem. 2016. “Rising Competitive Authoritarianism in Turkey.” Third World Quarterly 37(9): 1581–606.
Esen, Berk and Gumuscu, Sebnem. 2017. “Turkey: How the Coup Failed.” Journal of Democracy 28(1): 5973.10.1353/jod.2017.0006
Foa, Roberto Stefan and Mounk, Yascha. 2016. “The Democratic Disconnect.” Journal of Democracy 27(3): 517.
Florida, Richard. 2016. “How America’s Metro Areas Voted.” City Lab, November 29. Available at .
Freedom House. 2013. Freedom of the Press: Venezuela. Available at
Freedom House. 2016a. Freedom of the Press in the World 2016: Hungary.
Freedom House. 2016b. Freedom in the World 2016: Table of Country Scores at
Freedom House. 2018. Freedom in the World 2018: United States Profile. Available at
Frum, David. 2017. “How to Build an Autocracy.” The Atlantic, March. Available at
Gilens, Martin and Page, Benjamin I.. 2014. “Testing Theories of American Politics: Elites, Interest Groups, and Average Citizens.” Perspectives on Politics 12(3): 564–81.
Gumuscu, Sebnem. 2011. “The Emerging Predominant Party System in Turkey,” Government and Opposition 48(2): 223–44.10.1017/gov.2012.13
Handlin, Samuel 2018. “The Logic of Polarizing Populism: State Crisis and Polarization in South America.” American Behavioral Scientist 62(1): 7591
Huq, Aziz and Ginsburg, Tom. 2018. “How to Lose a Constitutional Democracy.” UCLA Law Review 65. Available at
Karl, Terry Lynn. 1997. The Paradox of Plenty: Oil Booms and Petro-states. Berkeley: University of California Press.
Katz, Josh. 2017. “Older Judges and Vacant Seats Give Trump Huge Power to Shape the American Courts.” New York Times, February 14. Available at
Kelemen, R. Daniel. 2017. “Europe’s Other Democratic Deficit: National Authoritarianism in Europe’s Democratic Union.” Government and Opposition 52(2): 128.10.1017/gov.2016.41
Keyman, E. Fuat and Gumuscu., Sebnem 2014. Democracy, Identity, and Foreign Policy in Turkey: Hegemony through Transformation. London: Palgrave Macmillan.
Kingsley, Patrick. 2018. “As West Fears the Rise of Autocrats, Hungary Shows What’s Possible.” New York Times, February 10. Available at
Kornai, Janos. 2015. “Hungary’s U-Turn.” Capitalism and Society 10(1): 124.
Kuk, John. 2017. “An Unequal and Polarized Democracy: Why Has Inequality Caused Party Polarization in the American Public.” Unpublished dissertation, University of California–San Diego.
Lendvai, Paul. 2016. Orbán: Europe’s New Strongman. New York: Oxford University Press.
Lepsius, M. Ranier. 1978. “From Fragmented Party Democracy to Government by Emergency Decree and National Socialist Takeover: Germany.” In The Breakdown of Democratic Regimes, Europe , ed. Linz, Juan J. and Stepan, Alfred. Baltimore and London: Johns Hopkins University Press.
Levitsky, Steven and Way, Lucan A.. 2010. Competitive Authoritarianism: Hybrid Regimes after the Cold War. New York. Cambridge University Press.
Levitsky, Steven and Ziblatt, Daniel. 2018. How Democracies Die . New York: Crown.
Lindberg, Staffan I. 2016. “Ordinal Versions of V-Dems Indices: When Interval Measures Are Not Useful for Classification, Description, and Sequencing Analysis Purposes.” Geopolitics, History, and International Relations 8(2): 76111.
Linz, Juan J. 1978. Crisis, Breakdown, and Reequilibration. Baltimore and London: Johns Hopkins University Press.
Luhrmann, Ann, Lindberg, Staffan I., and Tannenburg, Marcus. 2017. “Regimes in the World (RIW): A Robust Regime Type Measure Based on V-Dem.” Varieties of Democracy Institute Working Paper Series 2017, 47. University of Gothenburg.
Magyar, Bálint. 2016. Post-Communist Mafia State: The Case of Hungary. Budapest: Central University Press, in association with Noran Libro Kiado.
Mann, Thomas E. and Orenstein, Norman J. 2012. It’s Even Worse Than It Looks: How the American Constitutional System Collided with the New Politics of Extremism. New York: Basic Books.
Mayer, Jane. 2017. Dark Money: The Hidden History of the Billionaires Behind the Rise of the Radical Right . New York: Anchor.
McCarty, Noland, Poole, Keith T., and Rosenthal, Howard. 2016. Polarized America: The Dance of Ideology and Unequal Riches. 2nd ed. Cambridge: MIT Press.
McCoy, Jennifer, Rahman, Tahmina, and Somer, Murat. 2018. “Polarization and the Global Crisis of Democracy: Common Patterns, Dynamics, and Pernicious Consequences for Democratic Politics.” American Behavioral Scientist 62(1): 1642.10.1177/0002764218759576
Mechkova, Valery, Luhrmann, Anna, and Lindberg, Staffan I.. 2017. “How Much Democratic Backsliding?” Journal of Democracy 28(4): 162–69.
Mickey, Robert, Levitsky, Steven, and Ahmad Way, Lucan. 2017. “Is America Still Safe for Democracy? Why the United States Is in Danger of Backsliding.” Foreign Affairs 96(3): 2029.
Mudde, Cas and Rovira Kaltwasser, Cristobal. 2017. Populism: A Very Short Introduction. Very Short Introductions. New York: Oxford University Press.
Mueller, Jan-Werner. 2014. “Eastern Europe Goes South: Disappearing Democracy in the EU’s Newest Members.” Foreign Affairs 93(2). Available at
Muftuler-Baç, Meltem and Fuat Keyman, E.. 2012. “The Era of Dominant-Party Politics.” Journal of Democracy 3(1): 8599.10.1353/jod.2012.0000
Neiwert, David. 2017. Alt-America: The Rise of the Radical Right in the Age of Trump . London, New York: Verso.
Norris, Pippa and Inglehart, Ronald. 2018. Cultural Backlash: Trump, Brexit, and the Rise of Authoritarian Populism . New York: Cambridge University Press.
Önis, Ziya. 2013. “Sharing Power: Turkey’s Democratization Challenge in the Age of the AKP Hegemony.” Insight Turkey 15(2): 103–22.
Orenstein, Mitchell A., Krekó, Péter, and Juhász, Attila. 2015. “The Hungarian Putin? Viktor Orbán and the Kremlin’s Playbook.” Foreign Affairs, February 8. Available at
Pew Research Center. 2015. “Beyond Distrust: How Americans View Their Government.” Pew Research Center, November 23. Available at
Powell, Jonathan and Thyne, Clayton. 2011. “Global Instances of Coups from 1950–Present.” Journal of Peace Research 48(2): 249–59.
Przeworksi, Adam, Alvarez, Michael E., Cheibub, Jose Antonio, and Limongi, Fernando, 2000. Democracy and Development: Political Institutions and Well-Being in the World, 1950–1990. New York: Cambridge University Press.
Rothwell, Jonathan T. and Diego-Rosell, Pablo. 2017. “Explaining Nationalist Political Views: The Case of Donald Trump.” SSRN. Available at or
Scheck, Tom. 2018. “Ethics Be Damned: More than Half of Trump’s 20-person Cabinet Has Engaged in Questionable or Unethical Conduct.” Marketplace, February 16. Available at
Somer, Murat. 2016. “Understanding Turkey’s Democratic Breakdown: Old vs. New and Indigenous vs. Global Authoritarianism,” Southeast European and Black Sea Studies 16(4): 481503.
Somer, Murat. 2017. “Conquering versus Democratizing the State: Political Islamists and the Fourth Wave of Democratization in Turkey and Tunisia.” Democratization 24(6): 1025–43.
Southern Poverty Law Center. 2018a. “America the Trumped: 10 Ways the Administration Attacked Civil Rights in Year One.” Available at
Southern Poverty Law Center. 2018b. “The Year in Hate and Extremism.” Intelligence Report 164(Spring): 3342.
Sunlight Foundation. 2018. “Tracking Trump’s Conflicts of Interest.” Available at
Sunstein, Cass R. 2018. Can it Happen Here? Authoritarianism in America. New York: Harper Collins.
Svolik, Milan W. 2017. “When Polarization Trumps Civic Virtue: Partisan Conflict and the Subversion of Democracy by Incumbents.” Presented at the Workshop on Backsliding into Authoritarianism, Duke University, April 21.
Union of Concerned Scientists. 2018. “Science under President Trump: Voices of Scientists across 16 Federal Agencies.” Available at
Yabanci, Bilge. 2016. “Populism as the Problem Child of Democracy: The AKP’s Enduring Appeal and the Use of Meso-Level Actors.” Southeast Europe and Black Sea Studies 16(4): 591617.10.1080/14683857.2016.1242204
Zakaria, Fareed. 1997. “The Rise of Illiberal Democracy.” Foreign Affairs , November/December.
Ziblatt, Daniel. 2017. Conservative Parties and the Birth of Democracy. New York: Cambridge University Press.