Johns Hopkins University Press

Observers have expressed concerns that democratic politics is growing more personalistic. The absence of cross-national time-series data capturing personalism in democracies, however, has made it difficult to assess the validity of these observations. This study fills that gap. It covers a time-varying index of personalism in the world's democracies from 1991 to 2020. The index combines original data measuring indicators of personalism with existing data sources. We find that observers' intuitions are correct: Levels of personalism have increased in democracies in recent years. Importantly, we show that greater personalism is associated with a variety of negative outcomes, such as higher levels of populism, a higher probability of democratic erosion, and greater political polarization. In addition, we explore the potential causes of the personalist wave and find evidence that new technologies and digital tools are facilitating it. Our new index of personalism enables researchers to better explore these and other relationships. Given that a key signal of impending personalization is the leader's creation of their own political party, we encourage observers to take note of the election of such leaders as a red flag for democracy.

Personalist leaders, such as Russian president Vladimir Putin, Rwandan president Paul Kagame, and former Libyan president Muammar alQadhafi, dominate their country's political systems to such a degree that they become virtually indistinguishable from the regimes they rule. In the case of Russia, Putin has tightened his grip on power since his election in 2000 by systematically eliminating checks on his authority, including from the legislature, judiciary, regional governments, and civil society. He has also demolished any perception of a viable alternative to his leadership, even from individuals within his regime, and instead installed loyalists in key positions of power.

Personalism refers to the domination of the political realm by a single individual. The leader's personality has an outsized impact on policies and outcomes, often trumping institutions and rules. In contemporary politics, we typically associate this sort of rule with authoritarian regimes. In such a personalist autocracy, the leader governs absent the constraints of other actors: Not even the leader's political party (should it exist) or the security apparatus exert independent control. Policy choices, in turn, reflect the whims of the ruler.

Personalist rule is not a new phenomenon: For the bulk of political history, from the pharaohs of Egypt and the emperors of Rome to the czars of Russia, it was commonplace. The late twentieth century marked a departure from that norm as representative democracies emerged in [End Page 94] substantial numbers during the "third wave." As the twentieth century unfolded, not only did democracy spread but autocracies grew more diverse, and in many cases were less likely to be dominated by a single individual, as reflected in the collective-leadership structures of communist regimes in post-Mao China, Laos, and Vietnam as well as in those of the former military juntas in Argentina, Brazil, and Nigeria.

Since the end of the Cold War, however, politics has seemingly been slowly reverting to earlier times. As we have documented elsewhere, personalist dictatorships are on the rise.1 In 1988, these regimes composed 23 percent of all autocracies. By 2010, this percentage had nearly doubled to 40 percent. Although the data on authoritarian regime type end in 2010, all signs indicate that this trend has continued, with most new dictatorships exhibiting symptoms of personalism, including Turkey under Recep Tayyip Erdo¢gan and Nicaragua under Daniel Ortega.

This rapid increase is alarming: Personalist rule brings with it a host of negative outcomes compared to other types of authoritarian systems. A wide body of political-science research shows that personalist dictators pursue the most risky and belligerent foreign policies, and they are the most apt to invest in nuclear weapons and provoke interstate conflict. When these autocrats fall, their regimes are the least likely of all dictatorship types to democratize. Moreover, the departure of a personalist leader is often violent and protracted, as exemplified by the experiences of Iraq and Libya following the deaths of Saddam Hussein and Qadhafi, respectively.2

Political developments in the last decade suggest that the global trend toward personalism may no longer be confined to authoritarian systems. Across the globe, democratically elected leaders with personalist governance styles have been dominating headlines, from U.S. president Donald Trump and Brazilian president Jair Bolsonaro to Hungarian prime minister Viktor Orbán and Philippine president Rodrigo Duterte. The apparent growth in the prominence of such leaders raises the question: Is politics becoming more personalist in democracies, too? New data that capture levels of personalism in democratic states suggest that this is indeed the case. While this trend toward greater personalism could be temporary, the drivers and consequences of this dynamic indicate that there is reason for concern.

The Rise of Personalism in Democracies

All political systems feature some degree of personalism. Yet because democracies include checks and balances to varying degrees, personalism in these settings manifests itself in subtler ways than it does in authoritarian systems. For instance, a key indicator of personalism in authoritarian regimes is the extent of a leader's personal control over the security apparatus, through tactics such as the creation of private [End Page 95] paramilitary forces or appointment of friends and family members to top military posts. These types of overt indicators of personalism are unlikely in democracies, where there are typically laws in place that divide the civilian and military spheres. Ultimately, however, the underlying concept is the same: Personalist leaders hold more power relative to their political parties than nonpersonalist leaders do, such that politics reflects the leader's preferences more strongly than it does a bargaining process among multiple actors and institutions. Other scholars have conceptualized personalization similarly. In their study of Israeli politics, Gideon Rahat and Tamir Scheafer defined it as "a process in which the political weight of the individual actor in the political process increases over time, while the centrality of the political group (i.e., political party) declines."3

Personalism in authoritarian systems is associated with a wide range of negative outcomes, and our research shows that it is harmful for democratic systems, too. Greater personalism in democracies brings with it an elevated risk of political polarization, incumbent power grabs, and ultimately democratic decline and collapse. To test our ideas on the causes and consequences of personalism in democracies, we compiled new data covering 106 democracies from 1991 to 2020 to create an index of personalism.4 We collected observable and objective indicators of the relative power balance between a country's leader and his or her political party and included two indicators from the Varieties of Democracy (V-Dem) dataset: personalization of the party and government legitimation strategies. The former variable measures whether the party is primarily a vehicle for advancing the leader's political career, and the latter measures whether the leader is depicted as possessing extraordinary leadership skills. Our index does not capture leaders' personality traits or their behavior toward political actors outside the ruling party, including opposition parties, the legislature, judiciary, or media, as these are outcomes that we want to test.

Our latent index of personalism ranges from about -2 to +2 in the sample, with an average around zero. Higher values suggest greater personalism and lower values suggest less personalism. Using this index, we can compare levels of personalism across leaders and within their tenures over time. For example, our index reveals that in France personalism was low during the tenure of President François Mitterrand (-1.14 in 1992, his second year of office) and high during the tenure of President Emmanuel Macron (1.16 in 2018, his second year of office). [End Page 96]

Figure 1. P D T Note: Vertical-axis units are standard deviations of the personalism index with zero set as the average level.
Click for larger view
View full resolution
Figure 1.

Personalism in Democracies over Time

Note: Vertical-axis units are standard deviations of the personalism index with zero set as the average level.

Mitterrand governed with the backing of the established Socialist Party whereas Macron governs with the support of En Marche, a party that he created for his presidential candidacy. Comparing across countries, Macron's personalism score is also higher than that of German chancellor Angela Merkel, whose score has hovered around -0.8 throughout her time in office. She comes from a longstanding party, the Christian Democratic Union.

Figure 1 illustrates the overall trend of personalism in democracies from 1991 to 2020. The raw level of average personalism, represented by the solid line, indicates that democratic regimes overall grew less personalist until the early 2000s before returning to levels last seen in the 1990s. In regions with several third-wave democratic transitions, such as Central and Eastern Europe in the 1990s, levels of personalism were high because leaders had outsized influence over newly created political parties. In other words, where outgoing autocracies did not feature multiparty electoral competition, the parties that emerged after democratization were often new. Their leaders had never had the chance to pursue long careers working up the ranks. This had the effect of elevating the leaders' personalism scores. The dashed line in the figure adjusts the personalism index for the age of a democratic regime, allowing us to take these transitions into account. Thus a clear tendency emerges, with personalism steadily on the rise throughout the thirty-year period.

To better understand these trends, we divide our sample by region. Europe makes up the bulk (40 percent) of the observations, which include data for every year for every country in our sample, followed by Africa (17 percent), Asia (13 percent), and South America (11 percent). Other world regions make up the remaining 19 percent and are poorly [End Page 97] represented in the sample. Nearly every region in the world has seen an increase in personalism in the past decades. Even in regions, such as Africa and Asia, where personalism declined in the first decades after 1990, it has been on the rise since 2010. Though not all European countries saw personalism rise, many third-wave democracies—including the Czech Republic, Hungary, and Slovenia—did. The same is true in Asia, where personalism has increased in India, Pakistan, South Korea, and Taiwan.

Other cases showing a substantial rise in personalism include Benin in the past decade, El Salvador (with the election of Nayib Bukele), Guatemala, Mexico (where there was a substantial jump starting in 2019), Mongolia since 2006, Senegal (particularly after 2013), and Venezuela following Hugo Chávez's 1998 election. Interestingly, despite the impression given by media coverage, levels of personalism have not changed in the Philippines under Rodrigo Duterte, where they have been comparable (or lower) during his tenure to levels during the administrations of many of his predecessors, including Gloria Arroyo, Joseph Estrada, and Fidel Ramos.

Personalism is strongly correlated with populism, and the two phenomena have risen in tandem. Using our data to measure populism, defined by Cas Mudde and Cristóbal Rovira Kaltwasser as an ideology that separates society into two homogenous groups—the "pure people" and the corrupt elite—and argues that politics should express the popular will,5 we find that personalist leaders and their parties are more likely to be populist. Our measure of personalism, however, is less subjective than the concept of populism. Thus our personalism index serves as a leading indicator, or early warning, of the threats to democracy that we will discuss.

The Impact on Democracy

The rise of personalism in democratic regimes would be unremarkable if it brought few important changes. We find, however, that personalist democracies—or democracies with higher than typical personalism scores—differ in meaningful ways from their nonpersonalist counterparts, and that greater personalism is detrimental to democracy. The data show that personalism increases the chance of democratic decline by enabling the successful power grabs by incumbents—the incremental dismantling of democracy—that have become the most common way in which democracies break down. The adverse effect of personalism also holds for other forms of backsliding, whether a sharp decline or total regime collapse.

A significant, objective, and easy-to-observe predictor of personalization over time is whether a leader creates his or her own political party or rises through the ranks of an existing one.6 When we account for the fact that some leaders last longer in power than others, the data show that leaders who found their own parties are substantially more [End Page 98]

Figure 2. P C D E Note: Residualized fit from a two-way FE linear-probability model, adjusted for initial democracy and party-institutionalization levels, and democracy age; 95 percent confidence interval.
Click for larger view
View full resolution
Figure 2.

Party Creation and Democracy Erosion

Note: Residualized fit from a two-way FE linear-probability model, adjusted for initial democracy and party-institutionalization levels, and democracy age; 95 percent confidence interval.

personalist, and preside over governments that score 11 percent lower on the V-Dem polyarchy scale than those who come from within the party ranks. Similarly, personalist leaders are more than three times as likely as others to oversee a steep decline in democracy during their tenure, and the democratic regimes that they lead are nearly three times as likely to collapse (2.9 percent compared to 1 percent).7

Subsequent tests using the full personalism index, adjusting for potential confounders such as the age and quality of democracy, confirm these results. We find statistically significant effects of personalism on democratic decay, sharp democratic decline, and democratic collapse. In terms of decay, a one–standard-deviation increase in personalism is associated with, on average, a year-on-year decrease of about 1.4 percent in the level of democracy; if this average effect continued over five years, there would be an approximately 7 percent decline.

Regarding sharp declines in democratic quality (a slide of 10 percent or more from the democracy score in the year the leader entered office, which we refer to as democratic erosion) we find similarly discouraging results. The data reveal that a one–standard-deviation boost in personalism augments the risk of a steep drop-off by about 1.3 percent in any given year. Figure 2 plots this relationship. The probability of a sharp democratic decline increases from just under 1 percent at low levels of personalism to nearly 2.5 percent at higher levels. This finding remains substantive after adjusting for the potential confounding factors discussed above. There is, in sum, a meaningful and robust relationship between personalism and rapid democratic decline in the past three decades. [End Page 99]

Moreover, a one–standard-deviation increase in personalism boosts the chances of democratic collapse by roughly 1.4 percent, even after accounting for a measure of populism, which is correlated with personalism. This finding is substantively meaningful given that only 1.5 percent of all observations in the sample experienced a transition from democracy to authoritarianism.

Most democratic-collapse events since the Cold War have resulted from incremental incumbent power grabs. Some, however, have result ed from military coups, and there is anecdotal evidence to suggest that personalization also played a role in these events. In August 2020, for example, Malian president Ibrahim Boubacar Ke¦ta was ousted in a military takeover, abruptly ending democracy there. Ke¦ta had created his own party, Rally for Mali, and exhibited personalist tendencies. For example, despite having little experience in parliament, Ke¦ta's son Karim served as chair of the defense committee and Karim's father-in-law Issiaka Sidibé was the body's president. When the military intervened amid widespread protests, it linked its opposition to Ke¦ta with his propensity for personalism. Its spokesman said after the coup that "political patronage, the family management of state affairs, have ended up killing any opportunity for development in what little remains of this beautiful country."8 Observing trends in personalism, as measured by whether a leader creates his or own party, provides an early warning of risks to democracy.

How Personalism Erodes Democracy

There are a number of ways in which personalism increases a country's risk of democratic backsliding. First, personalist leaders increase polarization in the societies they govern, which existing research shows raises the risk of democratic decline. As depicted in Figure 3A below, leaders who create their own parties foster much higher levels of polarization once they are elected to power than do leaders of established parties. Moreover, leaders of established parties tend to decrease polarization after they have been in office for about four years, while long-tenured leaders who created their own party continue to increase polarization. Figure 3B shows the positive and statistically significant correlation between our personalization index and subsequent polarization after controlling for the initial degree of democracy in the year the leader assumes office as well as the democratic regime's age and initial levels of polarization and party institutionalization.

Personalism likely fuels polarization because when power is concentrated in the hands of the leader, policy choices incorporate fewer voices. Excluded, opposition groups may grow disillusioned with the political process, deepening the divide between political camps. Because our measure of personalism pre-dates a leader's assumption of power, we can ensure [End Page 100]

Figure 3A. P C S P
Click for larger view
View full resolution
Figure 3A.

Party Creation and Subsequent Polarization

that greater polarization is attributed not to his or her behavior once in office but rather to features of the prior leader-party relationship. The message is clear: When leaders come to power backed by parties that they themselves dominate, greater political polarization follows. A growing body of literature shows that polarization, in turn, is a threat to democracy. Milan Svolik finds that as societies become more polarized, voters become more willing to tolerate abuses of power and sacrifice democratic principles if doing so advances their side's interests and keeps the other side out of power.9 In politically divided societies, people develop a strong preference for their favorite candidate or party, often to the point of detesting those at the other political extreme.

Beyond polarization, several dynamics stemming from a personalist leader's relationship with his or her party also likely raise the risk of democratic decline. Typically, democracies deteriorate from within on account of leaders' actions to lessen their opponents' ability to challenge them effectively. These actions can include gerrymandering electoral districts; harassing and intimidating opponents; and monopolizing media access prior to electoral campaigns. Such power grabs are more likely to be successful when leaders govern with the support of a personalist party. First, greater personalism means that leaders have greater bargaining power vis-à-vis the rest of the party elite, making it more difficult for higher-ups to push back against the leader's efforts to consolidate control. Moreover, personalist parties tend to be weakly organized, increasing the risks for elites should they try to resist the leader's actions.

Second, a personalist leader who created his or her own party typically [End Page 101]

Figure 3B. P S P Note: Adjusted for country and year effects, democracy age, election year, and initial levels of polarization and democracy; 95 percent confidence interval.
Click for larger view
View full resolution
Figure 3B.

Personalism and Subsequent Polarization

Note: Adjusted for country and year effects, democracy age, election year, and initial levels of polarization and democracy; 95 percent confidence interval.

has substantial control over the selection of the party's candidates and key officials. Trusted family members and other loyalists are preferred to career politicians. These individuals are less likely to resist incumbent power grabs because their political fortunes are tightly tied to those of the leader. In established parties, by contrast, elites can still maintain political careers even in the leader's absence. This gives them greater incentive to act should the leader pursue a power grab.

Last, there are reasons to anticipate that the leaders and individuals who make up personalist parties are less committed to democratic institutions than those from nonpersonalist ones. In established parties, individuals rise to power by working their way up the party apparatus, in local government, or as elites in appointed positions. They have more exposure to democratic politics and how it works. They learn skills in negotiation with the opposition, policy compromise, and coalition building, which in turn mold their normative preferences for democracy. Leaders and officials within personalist parties are often newer to the game—political outsiders—and tend to lack this sort of exposure and commitment to democratic institutions. This may make them more likely to attempt a power grab in the first place.

The case of Hungary illustrates several of these dynamics. In 1995, Viktor Orbán moved the Fidesz party from the center-left to the right.10 This shift in ideology fractured the party leadership, enabling Orbán to oust higher-ups who opposed his rule. This paved the way for the rise [End Page 102] of Orbán loyalists, many of whom had not previously been part of the Hungarian political establishment and lacked substantial government experience.11 Notably, he appointed Lajos Simicska, his former high-school and university classmate, to the posts of party treasurer and head of the Hungarian tax authority from 1998 to 2002. Simicska became a central advisor to Orbán and an owner of one of the largest media conglomerates in Hungary until a falling out between the two men in 2015. After Fidesz's 2002 electoral loss, Orbán reformed the party constitution to tighten his grip on the selection process for local party bosses, parliamentary candidates, and the party's parliamentary-group leader. Installing loyalists in key positions, he squashed intraparty opposition to his rule and agenda.12

The absence of constraints on Orbán's power within Fidesz and the parliamentary supermajority that the party earned in 2010 have paved the way for Orbán to pursue unchecked power and enabled Hungary's democratic decline. He has reined in the media and promoted pro-Fidesz narratives by putting hundreds of news outlets under the supervision of a foundation monitored by his cronies.13 He has diminished judicial independence by lowering the retirement age for judges, forcing more than two hundred to resign. Fidesz has also installed political allies in the high court, election commission, state audit office, and central bank.14 Personalization and democratic backsliding form a vicious cycle as Orbán's increasing control of his party and Hungary's government magnifies his personal rule at the expense of political institutions.

Orbán's move to the political right also increased polarization within Hungary. In effect, his "New Right" created a two-bloc party system, with Fidesz and the Hungarian Democratic Forum on the center-right and the Free Democrats and the Hungarian Socialist Party converging on the center-left. The whole process deepened existing societal fissures.15 By the 2006 elections, divisions between the two parties were so sharp that demonstrations culminated in "street battles between protesters and the police."16

Personalism and Technological Change

The move toward greater personalism in democracies has been happening slowly over the last two to three decades. Researchers have pointed to impressionistic evidence of this shift, such as the substitution of leader images for party symbols during election campaigns and the media's growing propensity to mention candidates rather than the parties to which they belong.17 Even as regards parliamentary systems, scholars have offered country-specific evidence that voters are becoming more candidate-centered in their voting preferences.18 Comparable cross-national data are not available to assess the long-term trend toward personalization, but our data since 2000 indicate that it is accelerating. [End Page 103] Although personalization is likely a complex process driven by several interrelated dynamics, one factor that is contributing to the growth of personalization is leaders' use of digital tools to facilitate the consolidation of power.

Anecdotal evidence links the rise of personalization to the growth of electronic media, especially television in the 1950s and 1960s. Televised media, including the growing importance of on-air political debates during election campaigns, significantly influenced how voters viewed their leaders.19 In Russia, television played a critical role in shaping perceptions of Vladimir Putin in the 2000 presidential election—at a time when democracy was fragile and elections were more competitive than they are today. In Between Two Fires, Joshua Yaffa described how state-run media boosted Putin's image and built him up as an inevitable successor to President Boris Yeltsin.20 In the early days of Putin's tenure, state-run television facilitated his personalization of power by portraying him as a de facto commander-in-chief, overseeing military operations in Chechnya, and as uniquely qualified to lead Russia out of its troubles. Yaffa notes that during that time Putin's poll numbers rose four or five points per week. Similarly, other personalist leaders such as Hugo Chávez have used radio and television to establish direct and unmediated access to supporters, elevating the leader's personal profile above other institutions and even the ruling party.

Today, digital tools are likely amplifying these dynamics by allowing leaders to reach an even larger audience. Beyond granting the sheer power to broadcast, digital technologies—including the use of artificial intelligence (AI) to sift through vast amounts of online content or social-media bots that amplify influence campaigns and produce floods of distracting or misleading posts—create new opportunities for leaders to censor and manipulate their media environments by shutting down critical voices and more effectively "controlling the narrative." To test this proposition, we constructed a variable capturing state digital repression.21 This variable incorporates information on multiple aspects of a government's ability to monitor, censor, and shut down social media; to filter and turn off the internet; and to create social-media alternatives that are controlled by either the state or its agents.

Our results from 2001 to 2020 indicate that digital repression increases personalism, but we find no evidence of the opposite (that personalism boosts digital repression). Both digital repression and personalism increased throughout this period even after accounting for the varying capacities of states to pursue digital repression. These trends remain correlated even after adjusting for differences between countries and past observations of digital repression and capacity.

As digital repression grows, we see an increase in personalism during subsequent years. This suggests that technology lowers costs for leaders seeking to consolidate power. Greater control of the narrative may enable [End Page 104] them to lessen resistance to power grabs, mitigating the extent to which they must rely on repression to consolidate power and personalize the political system. In this way digital repression creates an environment more conducive to the personalization of power. Democracies at the low end of the digital-repression scale, such as Japan under Shinzo Abe, Senegal under Macky Sall, and Turkey under Bülent Ecevit, tended to see no increases in personalism during their leaders' tenures. Those at the high end of the scale—such as Nigeria under Muhammadu Buhari and Serbia under Aleksandar Vuèiæ—witnessed much larger increases in personalism.

While more research is needed to clarify how democratic leaders use digital tools to facilitate personalization, anecdotal evidence offers clues. In Turkey, for instance, prior to its transition to dictatorship in 2016, then–Prime Minister Erdo¢gan cracked down on social media following the Gezi Park protests of mid-2013. His heavy-handed response to a demonstration against the replacement of an Istanbul park with a shopping center sparked nationwide protests of his increasingly authoritarian rule. The government indicted 29 people for allegedly inciting violence over social media. In early 2014, the government amended Internet Law 5651 to enable telecommunications authorities to block access to specific websites that they found "discriminatory or insulting to certain members of society"—a measure that the regime would later use as a pretext to remove content from the webpages of opponents and media outlets. Additionally, the regime granted the National Intelligence Agency unfettered and indefinite access to "private data, documents, and personal information … without a court order." Google and Twitter also saw an uptick in requests from Turkish authorities to remove from their platforms content that the government considered objectionable. In the first half of 2014, "Twitter received 186 such requests, specifying 304 accounts, to remove content"—an increase from seven requests concerning thirty accounts in 2013.22 The 2016 coup attempt further accelerated Erdo¢gan's efforts to tighten control over the internet, crack down on social media, block web access, and remove online content.

In Poland, too, digital repression may be facilitating personalism, especially as the regime uses technology to amplify traditional media narratives that sow divisions in Polish society, promote the ruling Law and Justice (PiS) party, and tout cultural conservatism. In November 2015, just a month after PiS came to power, Cat@Net, a Polish troll farm backed by public subsidies from the National Disabled Rehabilitation Fund,23 began creating social-media accounts that promoted proregime narratives. PiS paired the injection of disinformation into the public space with offline efforts to muzzle media outlets critical of the regime. In January 2016, PiS passed a media law placing public broadcasting, including one of the most watched outlets, Telewizja Polska (TVP), under government control on the pretext of managing bias.24 TVP now disproportionately covers PiS officials, giving significant [End Page 105] airtime to party leader Jaros³aw Kaczyñski.25 The troll farm has amplified TVP narratives, including those supportive of Kaczyñski.26

Democratic Decline from Above

The global trend of increasing personalization, no longer confined to authoritarian regimes, is creating new risks for democracies. Based on our account, democratic backsliding today is in many ways a top-down dynamic: Politicians erode checks and balances and incrementally consolidate power, even when public support for democracy remains high. In Benin, for example, President Patrice Talon began to chip away at democratic institutions after coming to power in 2016. By the 2019 parliamentary elections, democracy was no more: New electoral rules barred all opposition candidates. This occurred despite overwhelming citizen support for democracy.27 Our findings suggest that it is an incumbent leader's relationship with the leader's own party, and not just citizens' choices, that polarizes societies and augments the risk of democratic breakdown. Should citizens seek to hold such leaders accountable, their ability to do so is often constrained due to these leaders' degradation of the integrity of the electoral process or changes to the rules of the game that make it hard to remove would-be autocrats from power.

Personalist leaders tend to dominate their parties' internal structures, weakening or eliminating intraparty resistance to the incremental incumbent power grabs that have become the most common cause of democratic decline. Our personalism index, which considers the leader's relations with other actors inside the ruling party, provides a clear, early warning of backsliding that may help democracy-support groups to prioritize their resource allocations. At the most basic level, we show that a country where the leader comes to power with the backing of a party that the leader created is at elevated risk of democratic backsliding.

Better insight into factors that can stop or slow the process of personalization is critical to ensuring that politics reflects a diversity of preferences and not just those of personalist leaders and their narrow cliques. While there are several factors behind the global growth in personalism, we offer preliminary evidence that digital technology facilitates leaders' ability to personalize politics. Repression online and in traditional media provides personalist leaders, whether in autocracies or democracies, with avenues to disseminate friendly narratives about their leadership as well as to censor or manipulate unfriendly ones. Maturing technologies will have the potential to supercharge these dynamics. New tools such as microtargeting or AI-powered algorithms can be used to reinforce support for the regime or to effectively counteract specific sources of discontent. These developments will give leaders more opportunities to amplify their own standing and disable constraints on their power.

Other explanations for democratic decline, by contrast, contend that [End Page 106] citizens' actions create conditions that leaders can exploit to consolidate control.28 If, as we suggest, the faltering of freedom around the world is flowing less from citizens' choices than from those of elite actors—namely incumbent leaders backed by personalist parties—who dismantle democracy from the top down, greater attention will need to be paid to the causes and consequences of personalism. Representation, accountability, political equality, and democracy itself will count on it.

Erica Frantz

Erica Frantz is associate professor of political science at Michigan State University.

Andrea Kendall-Taylor

Andrea Kendall-Taylor is senior fellow in and director of the Transatlantic Security Program at the Center for a New American Security and adjunct professor at Georgetown University's School of Foreign Service.

Carisa Nietsche

Carisa Nietsche is associate fellow in the Transatlantic Security Program at the Center for a New American Security.

Joseph Wright

Joseph Wright is professor of political science at Pennsylvania State University and codirector of the Global and International Studies Program.


1. See Andrea Kendall-Taylor, Erica Frantz, and Joseph Wright, "The Global Rise of Personalized Politics: It's Not Just Dictators Anymore," Washington Quarterly 40 (Winter 2017): 7–19. We use political-regime data (including information on the start and end dates of democracies and dictatorships, modes of regime transition, and authoritarian-regime types) from 1946 to 2010 from Barbara Geddes, Joseph Wright, and Erica Frantz, "Autocratic Breakdown and Regime Transitions: A New Data Set," Perspectives on Politics 12 (June 2014): 313–31. Data on democratic regimes from 2010 to 2019 come from the authors' updates.

2. See Kendall-Taylor, Frantz, and Wright, "The Global Rise of Personalized Politics," (2017) for a review of this literature.

3. Gideon Rahat and Tamir Sheafer, "The Personalization(s) of Politics: Israel, 1949–2003," Political Communication 24, no. 1 (2007): 65.

4. See Erica Frantz, Andrea Kendall-Taylor, and Joseph Wright, "Personalism in Democracies: A New Index," Luminate Working Paper, February 2021,

5. Cas Mudde and Cristóbal Rovira Kaltwasser, eds., Populism in Europe and the Americas: Threat or Corrective for Democracy? (Cambridge: Cambridge University Press, 2012).

6. We treat leaders who are not backed by an identifiable party at the time of their initial election similarly to those who create their own party because they do not face party constraints on their behavior. We do not count leaders who switch parties after they are elected or who create new parties once in office.

7. If we only look at maximum values for each leader, without accounting for the fact that some leaders survive longer in power than others, we find that 3.2 percent of those who do not create their own party erode democracy and 4 percent of them are incumbents when democracy collapses. By contrast, these figures for leaders who create their own party are both greater than 11 percent.

8. Ruth Maclean, Cheick Amadou Diourara, and Elian Peltier, "Mali Coup Leaders Pledge Democracy After Deposing President," New York Times, 19 August 2020.

9. Milan W. Svolik, "Polarization versus Democracy," Journal of Democracy 30 (July 2019): 20–32.

10. Elisabeth Bakke and Nick Sitter, "Patterns of Stability: Party Competition and Strategy in Central Europe Since 1989," Party Politics 11 (March 2005): 246.

11. Vít Hloušek, "Two Types of Presidentialization in the Party Politics of Central Eastern Europe," Italian Political Science Review 45 (November 2015): 277–99.

12. András Körösényi, Gábor Illés, and Attila Gyulai, The Orbán Regime: Plebiscitary Leader Democracy in the Making (London: Routledge, 2020).

13. Szabolcs Panyi, "A Brief History of Graft in Orbán's Hungary," Euractiv, 19 February 2020.

14. Kim Lane Scheppele, "Hungary's Constitutional Revolution," New York Times, 19 December 2011.

15. Bakke and Sitter, "Patterns of Stability," 246; András Bozóki, "Consolidation or Second Revolution? The Emergence of the New Right in Hungary," Journal of Communist Studies and Transition Politics 24 (June 2008): 191.

16. Bakke and Sitter, "Patterns of Stability," 226–227.

17. Ian McAllister, "Leaders," in Lawrence LeDuc, Richard G. Niemi, Pippa Norris, eds., Comparing Democracies: Elections and Voting in Global Perspective (Thousand Oaks, Calif.: Sage, 1996); Russell J. Dalton, Ian McAllister, and Martin P. Wattenberg, "The Consequences of Partisan Dealignment," in Russell J. Dalton and Martin P. Wattenberg, eds., Parties Without Partisans: Political Change in Advanced Industrial Democracies (Oxford: Oxford University Press, 2002).

18. Martin P. Wattenberg, The Rise of Candidate-Centered Politics (Cambridge: Harvard University Press, 1991).

19. Ian Macallister, "The Personalization of Politics," in Russel Dalton and Hans-Dieter Klingemann, eds. The Oxford Handbook of Political Behavior (Oxford: Oxford University Press, 2007).

20. Joshua Yaffa, Between Two Fires: Truth, Ambition, and Compromise in Putin's Russia (New York: Tim Duggan Books, 2020).

21. See Andrea Kendall-Taylor, Erica Frantz, and Joseph Wright, "The Digital Dictators: How Technology Strengthens Autocracy," Foreign Affairs 99 (March-April 2020).

22. "Turkey: Internet Freedom, Rights in Sharp Decline," Human Rights Watch, 2 September 2014,

23. Christian Davies, "Undercover Reporter Reveals Life in a Polish Troll Farm," Guardian, 1 November 2019.

24. Jo Harper, "Soros Boosts Anti-PiS Media Ahead of Crucial Polish Polls," Deutsche Welle, 26 (February 26, 2019),

25. Halya Coynash, "Poland's Government Uses 'Public TV' as a Propaganda Tool to Win the Elections. Will the EU React?," Kharkiv Human Rights Protection Group, 7 August 2019,

26. Nina Jankowicz, "The Growing Threat of Domestic Disinformation in Poland," Center for a New American Security, 18 August 2020,

27. Christina Scheller and Mircea Lazar, "Is Benin's Democracy Living Up to Its Reputation?" Afrobarometer Dispatch, no. 321,

28. Milan W. Svolik, "When Polarization Trumps Civic Virtue: Partisan Conflict and the Subversion of Democracy by Incumbents," Quarterly Journal of Political Science 15, no. 1 (2020): 3–31.

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.