publisher colophon

After many countries that had embarked upon transitions in the 1980s and 1990s failed to become consolidated democracies, political scientists highlighted the widespread emergence of hybrid regimes, which combine authoritarian and democratic features. Scholars argued such regimes were stable, with some positing that quasi-democratic institutions actually strengthened authoritarianism. But an examination of competitive authoritarianism (CA)—the most prominent of these hybrid types—suggests instability is the norm. Of 35 regimes identified as having been CA between 1990 and 1995, most have either democratized or been replaced by new autocracies. Furthermore, quasi-democratic institutions often contributed to CA’s breakdown. In short, hybrid regimes have not become a new form of stable nondemocratic rule.

In May 2018, Malaysia surprised the world with the peaceful electoral ouster of the United Malays National Organization (UMNO), in power since 1957, and its Barisan Nasional coalition. Equally unexpected was the fall in April of Armenia’s authoritarian government due to mass popular protests. In November 2017, a soft coup in Zimbabwe forced longtime dictator Robert Mugabe from office. Around the same time, Cambodia’s already undemocratic government was making a harsh authoritarian turn after disbanding the main opposition party and charging its leader with treason. These seemingly unrelated events are all part of a little-noticed but significant trend: the decline of competitive authoritarianism as a stable form of government.

In the early 2000s, after many countries that had embarked upon transitions in the 1980s and 1990s failed to become consolidated democracies, political scientists highlighted the emergence of what they labeled “hybrid” regimes.1 These were regimes that combined substantial authoritarian features with democratic ones. Steven Levitsky and Lucan Way broke important analytic ground by identifying perhaps the most common form of hybrid regime: competitive authoritarianism (CA). Competitive authoritarian regimes, they argued, have “formal democratic institutions,” especially elections, but “the playing field is heavily skewed in favor of incumbents. Competition is thus real but unfair.” While some of the CA regimes they examined did later democratize, they noted in 2010 that “19 of our 35 cases remained competitive authoritarian for 15 years or more, a lifespan that is comparable to even the most durable bureaucratic authoritarian regimes in South America. Hence, it appears that many halfway houses do stand.”2

Levitsky and Way were not the only ones to attribute a basic stability [End Page 129] to competitive authoritarian or other similarly described hybrid regimes. Writing about “semiauthoritarianism” around the same time, Marina Ottaway rejected the idea that these “carefully constructed and maintained alternative systems” were transitional, noting that some, such as that in Egypt, had already maintained “stability over a long period.”3 And Marc Morjé Howard and Philip G. Roessler discussed hybrid regimes as having “relatively established institutional forms that are likely to remain for the foreseeable future.”4

A growing body of scholarship that has emerged over the past ten years goes further, arguing that CA and other similar hybrid regime types are stable not in spite of having quasi-democratic institutions, but in part because of these institutions.5 These works contend that by skillful manipulation autocrats can turn political institutions such as elections and legislatures, conventionally seen as inherently democratic, into tools for shoring up their power.6 Some common arguments are that authoritarian rulers may employ quasi-democratic institutions to gain useful information about popular or elite preferences; to signal strength through victories at the polls; and to make credible commitments that they will observe some laws or norms.7

Yet developments over the past ten years demonstrate that most CA regimes have not in fact achieved stability. Of the 35 regimes Levitsky and Way identified as having been CA in the period 1990–95, a large majority have by now either democratized or fallen and been replaced by new CA regimes. A few others have diverged from CA in the opposite direction, with repression heightening to the point where these systems have become fully authoritarian. Only four of the 35 cases have been stable CA regimes throughout the intervening period—that is to say, have not experienced regime change or transition since 1995 (see the Figure below).

Even the four holdout CA regimes that have persisted since the 1990s, all in Africa, do little to demonstrate CA’s supposed strengths. Three are harshly repressive regimes: Cameroon, Gabon, and Tanzania (whose government has grown increasingly draconian in recent years). These states come close to the threshold of full authoritarianism—especially Cameroon, which received the same scores for civil rights and political liberties as Belarus and Iran on Freedom House’s 2018 index. Despite repression, rulers in Gabon and Tanzania have faced increasingly close races against opposition parties in recent elections.8 Gabon’s president Ali Ben Bongo (in office since 2009) has allowed legislative elections to be delayed twice following his own narrow reelection in 2016; critics charge the delays are means to prevent an opposition victory.9 Tanzanian president John Magufuli, since the last general election in 2015, has “banned all political rallies,” except naturally his own, as part of a broader crackdown on dissent.10

These autocrats are not acting as if inclusive legislatures and semi-competitive electoral campaigns are essential sources of legitimation or [End Page 130] somehow useful for social control. Rather, they are treating these institutions as threats to their rule, though perhaps ones with which they cannot dispense entirely—at least not yet. Mozambique is less repressive and still stable, but there is a good case to be made that this is because Frelimo, the dominant party, draws strength and legitimacy from its revolutionary history.11 Similarly, CA’s exceptional stability in Malaysia under UMNO (1957–2018) was likely due to the regime’s politically auspicious origins in “counterrevolutionary” consolidation.12

Figure. C A R T, 1990–2018 Note: This chart is an adaptation of Levitsky and Way, Competitive Authoritarianism, p. 21. I list a country as a democracy if Freedom House calls it an “electoral democracy” using the new, stricter definition for 2018. The only deviation I make is to disagree with Freedom House’s characterization of Tanzania as an electoral democracy (for reasons discussed in the text). *Given the deterioration of Serbian democracy under President Aleksandar Vučić and his Serbian Progressive Party, this country may soon return to the category of unstable CA. **I would argue that Zimbabwe’s soft coup in November 2017, though it did not technically install a new regime, disrupted CA and demonstrates the regime is unstable.
Click for larger view
View full resolution

Competitive Authoritarian Regime Trajectories, 1990–2018

Note: This chart is an adaptation of Levitsky and Way, Competitive Authoritarianism, p. 21. I list a country as a democracy if Freedom House calls it an “electoral democracy” using the new, stricter definition for 2018. The only deviation I make is to disagree with Freedom House’s characterization of Tanzania as an electoral democracy (for reasons discussed in the text).

*Given the deterioration of Serbian democracy under President Aleksandar Vučić and his Serbian Progressive Party, this country may soon return to the category of unstable CA.

**I would argue that Zimbabwe’s soft coup in November 2017, though it did not technically install a new regime, disrupted CA and demonstrates the regime is unstable.

Over the medium-to-long term, CA regimes often see their stability erode precisely because they feature relative political openness and competition, at least compared to fully authoritarian regimes. Over time, the risks to autocrats of losing, even at a game they have rigged, are greater than the uncertain benefits of having better political information [End Page 131] or being able to signal strength or credibility through democratic institutions. There are at least four noteworthy mechanisms by which CA’s relative openness can lead to its breakdown.

Four Factors Leading to Breakdown

First, CA provides opposition forces with legal, if difficult, options for building up public support and challenging the regime; an uneven playing field is better than none at all. As Staffan Lindberg has contended, even elections held in semiauthoritarian contexts can end up aiding democratization.13 The opposition can prepare to challenge the regime at the national level by first winning mayoral or other local elections and demonstrating its ability to govern at the subnational level. This happened, for instance, in Mexico in the later stages of the Institutional Revolutionary Party’s long rule, before the opposition National Action Party finally won the presidency in 2000. Another example would be the opposition Democratic Progressive Party’s control of local governments in Taiwan prior to winning the presidency, also in 2000.14 The public in a CA country becomes accustomed to some local alternation of power and may come to see opposition politicians as providing real policy alternatives. Contrast this with the situation in a fully authoritarian regime such as China, where the vast majority of people have known only the rule of the Chinese Communist Party (CCP). Chinese citizens, even if they are critical of some CCP policies, often cite fear of the political unknown and possible national “chaos” as reasons for their continued support of the regime.

While incumbent parties do enjoy major advantages in CA regimes, the relative openness of these societies means that political opinion can shift too rapidly for autocrats to keep on top of emerging trends or formulate a response. The upset victories of opposition forces in Armenia and Malaysia in the first half of 2018, for instance, surprised even many longtime regime opponents. Even if there are no sudden political shifts, quasi-competitive elections create the possibility that autocrats may lose power if they choose the wrong electoral strategy. Daniel Treisman points to numerous cases in which authoritarian incumbents lost “by mistake” elections that they theoretically should have been able to manipulate.15 Among his examples are several fateful ballotings held by CA regimes: in Croatia in 2000, Kenya in 2002, Nicaragua in 1990, and Senegal in 2000.

Second, a CA regime’s manipulation of elections may succeed, but nevertheless incite public anger and trigger protests that grow into a regime-toppling mass movement. In fact, it may be the case that organizing a vote and then canceling it or blatantly cheating tends to provoke more public anger than going without elections altogether. This kind of violation of expectations has on many occasions been the catalyst for large-scale protests, even if these explosive reactions may also reflect the [End Page 132] presence of other underlying grievances. The Rose Revolution in Georgia, which resulted in the November 2003 ouster of President Eduard Shevardnadze, began with demonstrations sparked by the government’s manipulation of parliamentary elections held earlier that month. In late 2004, reports of electoral fraud and voter intimidation in Ukraine’s presidential runoff between Prime Minister Viktor Yanukovych, the chosen successor candidate of the incumbent president, and challenger Viktor Yushchenko triggered protests that began the Orange Revolution. Official results naming Yanukovych as the winner were eventually annulled, and Yushchenko won a second balloting. It is important to note that despite these and other recent examples, most election-rigging in the world is not met by large-scale public protest. A full explanation of why only certain cases result in mass protests is beyond the scope of this article, but credible international election monitoring may help to galvanize public resistance, as it did in Georgia and Ukraine.16

Third, even if election rigging succeeds and there is no regime-threatening popular protest, quasi-competitive elections may undermine CA by a different route: Electoral gains by the opposition could trigger an autocratic backlash and a turn to full authoritarianism. Unlike the two mechanisms outlined above, this scenario does not end in an opposition victory over the incumbent regime, but the final outcome is still the abandonment of CA. Autocrats who see or predict opposition gains in national elections often take decisive steps to prevent further deterioration of their position. South Korea’s Third Republic (1963–72) was a CA regime with semicompetitive presidential elections held every four years. President Park Chung-hee, who came to power in a coup in 1961, won the elections in 1963, 1967, and 1971. But by the time of the 1971 presidential contest, public opposition to Park’s military regime was on the rise. Even with an assist from illegal electoral manipulation and the regime’s authority behind him, Park defeated opposition leader Kim Dae Jung by fewer than a million votes. Soon thereafter, not coincidentally, Park forced through the dictatorial Yushin Constitution, which ended formal democracy in South Korea and consolidated his personal power. In a more recent example, the previously mentioned authoritarian turn in Cambodia is widely thought to be Prime Minister Hun Sen’s response to the strong showing of the opposition Cambodian National Rescue Party in national elections in 2013 and communal elections in 2017.17

Finally, and most speculatively, quasi-democratic institutions may put CA regimes at a disadvantage in terms of ideological legitimacy. Because CA regimes claim to be democracies, they usually deprive themselves of the alternative bases of ideological legitimacy on which many of the most durable authoritarian regimes rely. These autocracies openly oppose “Western-style democracy,” claiming instead to be based on some superior or more locally appropriate alternative, such as communism, a religious system, or a cultural tradition (Singapore’s former prime minister [End Page 133] Lee Kuan Yew, for instance, famously justified his authoritarian approach as a reflection of “Asian values”). CA regimes, by contrast, run the risk of claiming to be democracies but failing to persuade the public of this, leaving them with no clear ideological basis for their rule.

These regimes may be able to make do without ideological support for a time if they enjoy performance legitimacy, which usually results from improving economic conditions and a well-managed system of administration. But performance legitimacy is precarious. All governments make mistakes and experience crises from time to time. In those moments, opposition activists have a chance to seize the narrative and argue, “If democracy has improved the lives of people in neighboring countries, then why not here?” In a fully authoritarian regime that claims an alternate source of legitimacy, regime apologists can confront this question head-on. Supporters of the CCP in China, for instance, would respond that China is not and should not become a Western-style democracy because that system of government has its own flaws; it is incompatible with Chinese culture, history, and the current education levels of much of the population; and it would lead to social chaos and national disunity. Instead, the CCP derives its legitimacy from communist ideology, its revolutionary history, and (more recently) Chinese cultural traditions. All that said, some regimes can combine a similar emphasis on a triumphant revolutionary history with the pretense of democracy that defines CA as a regime type. This approach has been tried with some success by CA regimes in Mexico, Mozambique, Taiwan, and Zimbabwe.

While CA and other hybrid regime types have not followed a teleological path toward democracy, they have largely failed to establish a new or alternative form of stable authoritarian rule. Most of the long-lived nondemocracies still in existence today are fully authoritarian regimes in Asia and the Middle East, such as China, North Korea, Vietnam, Saudi Arabia, the United Arab Emirates, and Oman. By comparison, similarly sturdy CA regimes have been rare, and none currently standing is as old as any of the six regimes listed above. Furthermore, there are at least four plausible mechanisms linking the breakdown of CA regimes to their quasi-democratic institutions. Both the frequency of these breakdowns over the past several decades and the role played by elections in many of these cases should give pause to scholars who argue that selectively employing democratic institutions helps autocrats to stay in power. Halfway houses, it turns out, are quite shaky.

Christopher Carothers

Christopher Carothers is a Ph.D. candidate and Ashford Fellow in the Department of Government at Harvard University. His writing has appeared in the Wall Street Journal Asia, Foreign Affairs, and elsewhere.


1. See for instance Thomas Carothers, “The End of the Transition Paradigm,” Journal of Democracy 13 (January 2002): 5–21; Larry Diamond, “Elections Without Democracy: Thinking About Hybrid Regimes,” Journal of Democracy 13 (April 2002): 21–35; Steven Levitsky and Lucan A. Way, “Elections Without Democracy: The Rise of Competitive Authoritarianism,” Journal of Democracy 13 (April 2002): 51–65.

2. Steven Levitsky and Lucan A. Way, Competitive Authoritarianism: Hybrid Regimes After the Cold War (New York: Cambridge University Press, 2010), 5, 21.

3. Marina Ottaway, Democracy Challenged: The Rise of Semi-Authoritarianism (Washington, D.C.: Carnegie Endowment for International Peace, 2003), 7, 8.

4. Marc Morjé Howard and Philip G. Roessler, “Liberalizing Electoral Outcomes in Competitive Authoritarian Regimes” American Journal Of Political Science 50 (April 2006), 365.

5. See for example Barbara Geddes, “Why Parties and Elections in Authoritarian Regimes?” (unpubl. ms., 2006); Jennifer Gandhi and Adam Przeworski, “Authoritarian Institutions and the Survival of Autocrats,” Comparative Political Studies 40 (November 2007): 1279–301; Lisa Blaydes, Elections and Distributive Politics in Mubarak’s Egypt (New York: Cambridge University Press, 2011); Milan W. Svolik, The Politics of Authoritarian Rule (New York: Cambridge University Press, 2012).

6. To be clear, Levitsky and Way do not argue this.

7. Dawn Brancati, “Democratic Authoritarianism: Origins and Effects,” Annual Review of Political Science 17 (2014): 313–326.

8. “Gabon’s President Is Re-Elected, Thanks to a 99.93% Turnout in His Home Province,” Economist, 6 September 2016; “Tanzania Poll: John Magufuli of CCM Defeats Edward Lowassa,” BBC, 29 October 2015.

9. Katarina Hoije and Eric Mbog Batassi, “Bongo Bids to Entrench Power in Gabon Before Parliamentary Vote,” Bloomberg, 28 May 2018

10. “Tanzania’s Rogue President,” Economist, 15 March 2018.

11. Steven Levitsky and Lucan Way, “The Durability of Revolutionary Regimes,” Journal of Democracy 24 (July 2013): 5–17.

12. Dan Slater and Nicholas Rush Smith, “The Power of Counterrevolution: Elitist Origins of Political Order in Postcolonial Asia and Africa,” American Journal of Sociology 121 (March 2016): 1472–1516.

13. Staffan I. Lindberg, Democracy and Elections in Africa (Baltimore: Johns Hopkins University Press, 2006).

14. Some scholars place Taiwan’s democratization in the late 1980s, but others rightly note that the Kuomintang retained unfair structural advantages over the opposition in the 1990s, making the regime in this period competitive authoritarian.

15. Daniel Treisman, “Democracy by Mistake” (NBER Working Paper No. 23944, October 2017), 23.

16. See Michael McFaul, “Transitions from Postcommunism,” Journal of Democracy 16 (July 2005), 10.

17. “Cambodia: Democracy Faces Death,” Human Rights Watch, 15 November 2017,; Joshua Kurlantzick, “Cambodia Draws Closer to Outright Authoritarianism,” Council on Foreign Relations, Asia Unbound blog, 10 October 2017,

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.