Why do many autocrats hold at least somewhat competitive elections? And how should democrats engage, if at all, in such elections? Assuming that the answer to the first question is not “out of the goodness of their heart” but that it rather has something to do with how autocrats see elections as serving their interests, the second question may pose a genuine ethical conundrum. We cannot appreciate the complexity of the first question without first exploring the political dynamics of electoral autocracies.
Authoritarianism is not what it used to be. We still tend to associate autocracy with its historically most consequential twentieth-century instances, such as Stalin’s Soviet Union or Hitler’s Germany, in which political opponents and even independent critics were murdered, tortured, imprisoned, or forced into exile, where most forms of political and media pluralism were extinguished, and where violent repression permeated most domains of life. To be sure, some such violently repressive regimes continue to survive and even play a major role on the world stage, such as China or North Korea. However, the majority of contemporary autocracies are not like that. They tolerate a degree of pluralism in both politics and the media, and they maintain their grip on power not by violence (which occurs only episodically rather than systematically) but by partisan capture of key offices, by propaganda, and by vast clientelistic networks that create a system of economic dependence for a sizable share of society. Most characteristically, many contemporary autocracies, including India, Türkiye, or Serbia, hold multiparty, at least somewhat competitive elections in which genuine opposition parties and candidates are allowed to run. What is more, while the playing field is systematically distorted to favor the ruling party, its victory is nevertheless not a foregone conclusion in such autocratic elections. According to one study covering autocratic elections around the world between 1980 and 2002, approximately one in five of them produce the defeat of the ruling party. Another study found that around 30% of authoritarian elections produced significant “liberalizing outcomes,” defined as an improvement of at least three points in Freedom House’s political rights score. While such outcomes should not be equated with democratic regime change, they typically result in such transformations. In other words, such autocratic elections are not mere sideshows, totally meaningless affairs to provide a false facade of democracy for authoritarian rule. They are not like Russia’s elections where only regime-approved token opposition is allowed to compete. They are often fiercely contested and characterized by high turnout.
These features of elections that are held under authoritarian rule raise at least two sets of questions. First, if the elections allow politically meaningful competition in which genuine opposition parties can sometimes win, come into office, and implement significant democratic reforms, then are we in fact justified in calling them autocratic? Second, if such elections pose genuine risk to the autocrat’s grip on power, then why do more and more autocrats decide to hold elections rather than resorting to the measures of violent repression characteristic of old-school autocracies?
In answering the first question, our starting point should be the standard criteria applied to evaluate democratic elections, i.e., whether they are free and whether they are fair. Let’s start with fairness. This criterion is usually understood to require a system of rules that do not systematically favor any of the competing parties and to require that none of the parties enjoy significant resource advantages that are unavailable to others. Autocratic elections are nothing like that. The ruling party enjoys immense material advantages. It is supported by most of the media; the ruling party and the state’s institutions are merged, with the former using the material, personnel, and symbolic resources of the latter without any constraints to its own political advantage; and electoral rules are regularly changed to serve the momentary interests of the ruling party in a shifting political environment. These features constitute not merely an uneven playing field that is often found in flawed liberal democracies as well, but systematic unfairness.
When it comes to the freedom criterion, the analysis is less straightforward, but the conclusion is ultimately no less clear. Elections can be considered free if everyone has the right to vote and run for office and can exercise these rights without undue burdens and viewpoint restrictions. Formally, most electoral autocracies satisfy this condition, because there are no legal restrictions on the right to vote and to run for elected office. However, the absence of formal restrictions is not enough. Individuals must be able to exercise these rights without fear or risk of severe consequences. If some people face coercion, intimidation, or undue pressure to vote in a certain manner, then their vote is unfree. If otherwise viable candidates are threatened or intimidated, if they or their family suffer immediate harms such as job loss, seeing their reputations destroyed by state-orchestrated smear campaigns, or becoming the targets of bogus investigations when they enter the political fray, then they are not really free to run for office. The fact that some people nonetheless decide to accept these sacrifices and run does not show that the elections are free, any more than the fact that Soviet-era dissidents managed to publish samizdat literature shows that the press was free in the Soviet Union. A notion of free elections so thin as to allow systematic, routine intimidation and legal harassment of opposition candidates is not worth taking seriously. But then, all of the above are common in electoral autocracies. Partisan prosecutors harass opposition politicians and independent critics on bogus charges while government cronies engage in serious wrongdoing with impunity. And when the freedom of viable opposition candidates to run for office is significantly undermined, then indirectly, the freedom of everyone to vote for their preferred candidates is also infringed. Therefore, when all these factors are taken together, autocratic elections can be understood as systematically unfair and at least partly unfree.
The fact that unfair and unfree but still competitive elections sometimes still see the defeat of the ruling party leads to the following question: Why do most contemporary autocrats hold elections and tolerate the risk of losing? An influential answer suggested by empirical research on authoritarian politics builds on insights about some of the difficulties that authoritarian rulers face in anticipating, discouraging, and repelling challenges to their power. The so-called “Dictator’s Dilemma” highlights the fact that, in highly repressive regimes, autocrats do not have reliable information about their own popularity. The more repressive the regime, the less likely that unpopularity is observable, that the sincerity of expressions of support is credible, or that facts about unpopularity will be communicated to the leader by fearful subordinates. Knowing this, the autocrat may grow increasingly insecure about his position. Facing this uncertainty, he may choose to treat expressions of support as false and further increase repression, which is politically costly and may backfire. Or he may choose to accept signs of popularity at face value and do nothing, and thus risk missing opportunities to head off challenges to his power. Either way, the unreliability of information about his popularity threatens his survival. The optimal situation for the autocrat is when he has sufficiently reliable information to know when to selectively deploy repression. Now, somewhat competitive elections and the toleration of limited media pluralism may ease the Dictator’s Dilemma by improving the quality of information that reaches the autocrat. Of course, the quality of political information can be assumed to be highest with free and fair elections and full media freedom, but such conditions leave autocrats vulnerable to popular challengers. The best situation for autocratic survival may be when elections are sufficiently competitive to generate meaningful information but also sufficiently manipulated to minimize the risk of successful electoral challenges. When this sweet spot is hit, elections are sufficiently manipulated to make the autocrat’s victory overwhelmingly likely, but at the same time they are sufficiently competitive to make that victory at least minimally credible to the public, the elites, and external actors. In this way, the results discourage serious challenges both from below (the opposition), from within (elite defections), and also from the outside, making external interventions less likely. There is some empirical evidence that citizens are less likely to engage in contentious protest when they believe the system is legitimate, even when they disapprove of its performance. Therefore, holding autocratic elections often serves the autocrat’s interest in maintaining his rule.
This raises the following ethical issue for democrats seeking to end autocracy: Is it morally justified for democrats to participate in autocratic elections as candidates or voters, given that their engagement often serves the autocrat’s purposes? Should they not instead boycott such elections to highlight their illegitimate nature and seek nonelectoral ways of challenging autocracy, such as civil disobedience or resistance? By competing, they contribute to the false facade of democracy. However, by boycotting, they may forgo one of the most effective ways of challenging autocracy. Some studies show that while electoral boycotts may destabilize autocracies, the result is often more repression rather than democratization, and also that when electoral autocracies democratize, they most often do so after a successful electoral challenge. Therefore, autocratic elections confront democrats with a moral and strategic dilemma. The strategic dimension of the dilemma consists in figuring out whether, under the circumstances, electoral challenge is more likely to be effective than its alternatives. The moral dimension of the dilemma is that by participating and playing the autocrat’s game, they often get their hands dirty, morally speaking, and become complicit in the wrong of autocracy. This could be so even if the challenge is successful.
When confronted with this dilemma, few political theorists advocate for boycott. However, the arguments in favor of participation are varied. Some theorists claim that autocratic elections retain some residual value of democratic elections, because participation in them reinforces democratic norms, including the norm of peaceful competition. Therefore, even autocratic elections have some outcome-independent value, i.e., value that is not rooted in their potential to produce democratizing outcomes. Others argue that the only value that autocratic elections have is outcome-dependent, as means of ending autocracy. They do not have any normatively privileged status and are to be compared with nonelectoral options in terms of effectiveness from case to case, in light of facts on the ground. There is nothing inherently wrong with choosing nonelectoral forms of challenge if the evidence suggests that that would be more effective. This contrasts sharply with democratic elections: In a democracy, elections are a normative privileged mechanism of deposing a government, no matter how harmful, and adopting nonelectoral alternatives is usually wrong.
The choice between the answers depends on how one sees the normative continuity (if any) between democratic and autocratic elections. Proponents of the first answer tend to emphasize the competitive and nonviolent nature of elections, whether democratic or autocratic, while proponents of the second approach focus on their systematic unfairness and partial unfreedom. My own view is that while nonviolence is of great moral significance, its value is not distinctively democratic but of a more general character. Peacefulness does not single out electoral challenge from among a range of nonviolent options, including nonviolent resistance. By contrast, fair competition between free and equal persons is close to the normative core of democracy. Therefore, there is no normative continuity between democratic and autocratic elections, and the latter cannot inherit any of the inherent value of the former. Democrats therefore have no democracy-based reasons to choose the electoral route of challenging autocracy if other options promise to be more effective.
Editor’s note: This is an extra edition of the Perspectives on Democracy series for June 2026.
Zoltan Miklosi
Zoltan Miklosi is teaching political philosophy at the Department of Political Science, Central European University in Vienna. His research ranges from foundational problems in egalitarianism and democratic theory to questions of political ethics in flawed democracies and electoral autocracies.
