Who Democratizes? Western-educated Leaders and Regime Transitions

Abstract

Many theories attempt to explain why some countries democratize and others do not. Existing accounts, however, focus almost exclusively on structural factors and ignore individual leaders. In this article, we argue that leaders educated at Western universities are more likely to democratize than other leaders because Western education socializes leaders to prefer democracy and creates transnational linkages that alter the strategic calculus of democratization. Utilizing an original data set on the specific colleges and universities world leaders attended, we show that Western-educated leaders significantly and substantively improve a country’s democratization prospects.

Keywords

democratization political leadership democratic institutions foreign policy international organization

Are Western-educated leaders linked to democratization? The relevance of this question is indisputable. Indeed, many policy makers assume a positive connection between Western-educated leaders and democracy. For example, US State Department initiatives such as the International Visitor Leadership Program, the Leaders for Democracy Fellowship Program, and the Fulbright Program aim to promote democracy abroad through cross-cultural academic exchange. Several elite Western universities have also launched centers and institutes—including Yale’s World Fellows Program, Harvard’s Ash Center for Democratic Governance, and Oxford’s Blavatnik School of Government—designed to groom the next generation of global democratic leaders. Moreover, it is not hard to find leaders who trumpet their Western education as a transformative political experience. Former Pakistani Prime Minister Benazir Bhutto, for instance, lauded her time at Harvard for forming “the very basis of [her] belief in democracy” (Ahmad 2012).

Despite these popular notions, no study has systematically examined whether Western-educated leaders affect democratization. Most research ignores the role of leaders almost entirely, viewing democratization solely through the lens of structural factors such as economic development, inequality, natural resources, demographics, and colonial heritage. This does not mean, however, that scholars have not speculated on the consequential role that leaders might play. Przeworski (2007), for example, suggests that a leader’s educational background could be a proxy for leadership quality, which he considers the elusive variable in research on democratization. He offers the hypothetical comparison of a developing world leader who studies at Cambridge and “absorbs the ideals of democracy” (Przeworski 2007, 161) to the stereotypical dictator. Yet he stops short of testing this proposition. In response, Stokes (2014, 42) notes that “coding the postsecondary educational careers of third world leaders sounds like a laborious task, not an impossible one.” We agree. In this article, we develop a theory linking Western-educated leaders to democratization and collect the data necessary to test it.

We argue that Western-educated leaders are more likely to democratize than other leaders for two reasons. First, Western universities socialize students to view democracy as a legitimate form of government. Both inside and outside the classroom, future leaders schooled in the West gain distinctive experiences that make them more likely to develop political preferences for democracy. Second, Western education creates transnational linkages that increase the willingness and ability of Western-educated leaders to democratize. Because they can draw on personal connections and more credibly commit to using democracy assistance for its intended purpose, Western-educated leaders are better positioned to pull international resources toward their own democratic initiatives. At the same time, their reliance on the West allows Western governments and institutions to push such leaders to democratize even when they otherwise might not.

We test our predictions with an original data set that codes the specific colleges and universities that world leaders attended for undergraduate and graduate training. Controlling for a host of factors thought to influence democracy, we find compelling support for our theory: democratization is more likely when leaders educated at Western universities are in power. This result holds under multiple model specifications, definitions of democracy, and other robustness checks. We also take several steps to minimize endogeneity concerns. Most importantly, we analyze a reduced sample of leaders who entered power via leadership transitions stemming from natural deaths in office to show that our result is not simply a function of Western-educated leaders coming into power when countries are already primed to democratize. Additionally, we account for self-selection into Western education and explore other potential confounders that might create a spurious relationship between Western-educated leaders and democratization. By addressing these issues, we mitigate the empirical challenges associated with the claim that Western-educated leaders increase a country’s democratization prospects.

Our article represents a novel contribution to both the international relations and comparative politics literatures by complementing structural accounts of democratization with a leader-centric approach. It also raises notable policy implications for the effectiveness of global democracy promotion efforts and adds a new layer to the nascent body of social science scholarship on whether and why leaders influence a broad range of political outcomes. Highlighting the importance of leaders, of course, is not intended to marginalize the salience of national-level variables in driving democracy. Rather, our point is merely that studies of democratization that ignore the role of leaders overlook an influential element in regime transitions.

The rest of this article proceeds as follows. First, we review the existing research on the causes of democratization and identify the void in the literature that our article fills. Second, we introduce our theory about the relationship between Western-educated leaders and democratization. Third, we discuss our data and research design. Fourth, we present our main empirical findings, address endogeneity concerns, and conduct a series of robustness checks. Fifth, we offer a brief case study that shows how our theory’s mechanisms unfold in practice. Last, we consider the policy implications of our results and outline an agenda for future research.

Limitations of the Existing Literature

A rich literature in political science explores the origins of democratization.¹ Almost all scholars, however, highlight factors operating at the national or international levels. Key variables in the existing literature include economic development (e.g., Przeworski et al. 2000; Boix and Stokes 2003), inequality (e.g., Acemoglu and Robinson 2006; Boix 2003), colonial legacy (e.g., Bollen and Jackman 1985), ethnic and religious demography (e.g., Rabushka and Shepsle 1972; Woodberry and Shah 2004), natural resource wealth (e.g., Ross 2012), and international context (e.g., Huntington 1991).

Although these structural characteristics are doubtlessly relevant, they miss the agency of individual leaders. This is not to say that leaders are free from domestic political constraints. On the contrary, even extremely powerful leaders rarely have the opportunity to rule unchecked. At the same time, however, all leaders exercise some degree of autonomy and often have the capacity to shape their nation’s regime type. Indeed, leaders play a key role in determining a country’s political system by dictating policies; shaping public discourse; and influencing how, when, or even if elections take place. For example, it is hard to imagine South Africa’s democratic transition without Nelson Mandela or the breakdown of the Weimar Republic without Adolf Hitler.

Historians, of course, have long subscribed to the idea that leaders can have an important impact on politics. Following Victorian-era intellectual Thomas Carlyle ([1841] 1869, 34)—who provocatively asserted that “the History of the world is but the Biography of great men”—Great Man theorists contend that only a handful of individuals have altered history. Yet such arguments are often attacked for invoking post hoc justifications for why some leaders proved extraordinarily consequential. They have also come under fire for assigning disproportionate weight to intangible leadership qualities—such as charisma and cunning—that prove difficult to measure and even harder to falsify. Thus, while historians rightly emphasize the role of leaders in politics, there is a need to study how and why they matter in a more systematic fashion.

For many decades, political scientists generally shied away from analyzing individual leaders. Ahlquist and Levi (2011, 19), for example, lament the “paucity of empirical generalizations” on the topic of leaders and leadership in political science. In recent years, however, the discipline has paid more attention to this subject.² In particular, an emerging literature examines how the characteristics of world leaders affect various outcomes including economic growth (Besley, Montalvo, and Reynal-Querol 2011), leader survival (Berlinski, Dewan, and Dowding 2012), ethnic favoritism (Kramon and Posner 2012), and foreign policy behavior (McDermott 2007; Saunders 2011; Colgan 2013; Horowitz and Stam 2014; Fuhrmann and Horowitz 2015).³

Despite these contributions, the role of education in influencing the behavior of leaders is typically omitted, with researchers more commonly focusing on military backgrounds or revolutionary experiences.⁴ This omission is surprising, given the crucial role that education plays in an individual’s life. Indeed, the idea that education is a formative experience dates back as far as Plato ([380 BC] 2012, 114), who declared in The Republic that “the direction in which education starts a man will determine his future life.” Hence, we bring a new background experience—Western education—into the literature. Our article also makes a novel contribution to the study of leaders because it focuses on a previously neglected dependent variable. Thus, we fill a gap in the literature by exploring how the characteristics of individual leaders influence democratization.

Theory

We argue that Western-educated leaders are more likely to democratize than other leaders for two reasons.⁵ First, Western education socializes leaders to embrace democratic governance. Second, Western education creates transnational linkages that alter the strategic calculus of democratization.

Socialization

A large body of research in social psychology shows that attitudes and beliefs spread through interpersonal and intergroup contact (e.g., Allport 1954; Pettigrew 1998). This idea comports with theoretical work in international relations that treats international institutions as social environments (e.g., Johnston 2001). Perhaps most relevant to our theory is research suggesting that when bureaucrats move from national politics to a regional or international institution, they often internalize the norms associated with their new environment. For example, Lewis (2005) shows how serving on the European Union’s Committee of Permanent Representatives socializes national civil servants and diplomats to accept new identities and ultimately develop new negotiating behaviors.

Higher education is an especially powerful form of interpersonal contact because it is among the most important ways in which individuals develop political beliefs and because early adulthood is arguably the most formative period of a person’s life (e.g., Caspi, Roberts, and Shiner 2005). The classes students take, the nature of the curriculum, and interactions with peers and professors all shape political values that graduates apply throughout their lives. Consequently, we conceive of institutions of higher learning as sites of socialization where future leaders adopt new views on democracy.

Our perspective aligns with other work suggesting that academic exchanges with the West might spread democratic ideas. Nye (2004), for instance, argues that states can enhance their soft power through educational exchanges that socialize others to accept the host country’s norms. Drawing on Nye’s work, Atkinson (2006, 2010) provides empirical evidence that military exchange programs with the United States are associated with improved human rights practices and democracy. Similarly, Spilimbergo (2009) finds that countries become more democratic as larger numbers of their young people study abroad in democracies. Building on this country-level research, we contend that future leaders schooled in the West should be especially likely to embrace democratic norms due to socialization both inside and outside the classroom.

Inside the classroom, Western-educated students are typically exposed to a distinctive curriculum. More than any other intellectual tradition, teachings grounded in the Western canon promote many of democracy’s principle tenets, including freedom of speech and religion, property rights, and the right to petition the government. In the words of D. Kagan (1991, section 1, p. 23), Western teachings serve as “the champion of representative democracy as the normal way for human beings to govern themselves.” By contrast, students in countries outside the West are sometimes taught a radically different perspective on democracy. Textbooks in China, for example, often denigrate democracy as a tool of foreign oppression and depict “democratization, political liberalization, a free press, and anti-government protests…[as] dangerous and destabilizing for Chinese society” (Ma 2007, 10).

Apart from the subjects studied, the very “facts” that students learn can also differ in Western and non-Western educational settings. In many states outside the West, ruling factions use education to manipulate conceptions of ethnic, religious, and political minorities. Biased historical accounts, in turn, can fuel intolerance and exacerbate in-group/out-group distinctions. For example, Nobel Peace Prize winner and current Liberian President Ellen Johnson Sirleaf (2010, 69) recalls that the even-handed history of Liberia she was taught at Harvard “was missing in the history I learned as a girl in Monrovia.” This problem is most acute where curricula riddled with flawed histories and crude stereotypes erode the inclusive culture required for democracy (e.g., Bush and Saltarelli 2000). An extreme case came in the lead up to the 1994 Rwandan genocide, where curricula designed to incite hatred of rival ethnic groups contributed to the collapse of the country’s social fabric and political system.

Outside the classroom, the activities that students pursue can further diverge in Western versus non-Western settings. For instance, student organizations dedicated to open debate on a wide range of political, economic, and social issues are a hallmark of universities in the West. These clubs bolster civic engagement and encourage individuals to become active participants in domestic and international affairs. For example, Indian Prime Minister and Cambridge University alumnus Manmohan Singh (2006) recalls that “[m]y teachers and my peers taught me to be open to argument and to be fearless and lucid in the expression of one’s opinions. In many important ways, the University of Cambridge made me.” Additionally, the structure of student groups—which are often democratically run—provides a model for future leaders to emulate when organizing their own governments. The opposite is true in many non-Western states, where authorities patrol campuses to regulate student activities and ensure that they remain under the regime’s strict purview.

Acculturation in the West also takes place off campus, as some students for the first time experience free speech, multiple religious and political views, and new forms of entertainment and media. Access to uncensored newspapers, books, websites, television, radio, and other outlets exposes students to a range of opinions on democratic politics. By contrast, experiences outside the West are often less conducive to democracy, especially in states that restrict civil liberties and enforce ideological conformity. In Russia, for instance, the Kremlin sometimes pays university students to attend pro-Putin public rallies and to harass political opponents. As a consequence of these disparate experiences both inside and outside the classroom, Western-educated leaders should be more likely to absorb democratic norms that pave the way for reform in their home countries.

Transnational Linkage

A growing literature in international relations examines the role of transnational connections in shaping various political outcomes (e.g., Keck and Sikkink 1998; Hafner-Burton, Kahler, and Montgomery 2009). Concerning the spread of democracy, scholars working in the “second image reversed” tradition (Gourevitch 1978) show that membership in international organizations helps facilitate democratic transitions (Pevehouse 2005). Most similar to our study, Levitsky and Way (2010) argue that Western linkage—defined as the overall density of ties between a country and the United States, Western Europe, and multilateral institutions—is a crucial but frequently overlooked variable in democratization.

Extending this macro-level work, we contend that Western education creates important transnational linkages because leaders schooled in the West often maintain close relations after returning to their home countries. These ties include the peers and mentors they meet at school but also extend beyond personal circles to include a broader network of like-minded individuals and groups. In addition to fostering such connections, a Western education brands developing world leaders as key points of contact for major Western institutions of power. Consequently, we claim that Western universities create transnational linkages that increase the ability and willingness of Western-educated leaders to democratize. These connections operate in two directions, both “pushing” and “pulling” leaders to democratize.

Pull factors are relevant when leaders seek assistance democratizing. Because they often maintain influential contacts in the West, Western-educated leaders are well poised to enlist the aid of foreign governments, multilateral agencies, and nongovernmental organizations in their endeavors toward democracy. Deals can be brokered through both formal and informal channels, as leaders use their connections to lobby for democracy assistance. Consider current Cote d’Ivoire President Alassane Ouattara who earned a doctorate in economics from the University of Pennsylvania and worked at the International Monetary Fund. Ouattara entered office following the rule of strongman Laurent Gbagbo and, despite his country’s post-conflict turmoil, was able to “draw on friends in high places made in western capitals during the 1980s and late 1990s” to finance a stable democracy in his homeland (Wallis 2011). By employing such contacts, Western-educated leaders can secure help—including foreign aid and diplomatic support—that accelerates democratization.

Western-educated leaders can also pull resources to their countries by more credibly committing to using democracy assistance for its intended purpose. Indeed, a common perception in Western capitals is that Western-educated leaders are not only less corrupt but also are more likely to possess the technocratic ability to implement democratic reform. One example is former Georgian President Mikheil Saakashvili, a graduate of Columbia Law School. Soon after assuming office in 2004, Saakashvili stacked his cabinet with fellow Western-educated technocrats. This dense network of ties to the West helped Saakashvili persuade Western donors that he was a trusted and capable partner. Some observers even questioned whether Saakashvili’s rapport with Washington insiders became “too personalized and too close … Saakashvili’s seductive message to Western interlocutors is that his government of thirty-something English speakers is an indispensable partner” (de Waal 2012). His strategy was undeniably effective. As one scholar concludes, “[s]eldom have so much effort and funding from the international community been directed to democracy promotion in a country that is open to democratic change” (Boonstra 2010, 1).

Push factors can generate additional incentives to democratize. Because Western-educated leaders often rely extensively on the West, they are especially susceptible to international pressure. Indeed, the status of such leaders in the West typically hinges on democratization efforts, making them hesitant not to deliver on anticipated reforms. Western states can leverage these relationships by threatening to withhold foreign aid or attaching conditions on assistance. This bargaining power is crucial because even reform-oriented leaders may be slow to democratize when doing so risks their own political survival. Consider again Georgia’s Mikheil Saakashvili. During his tenure, the World Bank (2012, 104) rated Georgia as the world’s top economic reformer and affirmed that Saakashvili’s anticorruption campaign “destroys the myth that corruption is cultural and gives hope to reformers everywhere.” Saakashvili’s commitment to democracy, however, was tested in 2007 amid growing partisan opposition at home. In response to exhortations from the White House, Saakashvili “bowed to Western pressure…by calling early elections” that ultimately confirmed his mandate (Champion and Cullison 2007, A6). This extraordinary deference only makes sense in the context of the US-Georgia relationship: the Saakashvili administration consistently ranked as a top per capita recipient of US foreign aid.⁶

Western officials and institutions can also nudge Western-educated leaders toward democracy by providing them with “soft landings” if they leave office after good faith efforts to implement democratic reforms. Lucrative jobs in multilateral bodies, as well as awards and global recognition, seem disproportionately likely for Western-educated leaders. The Mo Ibrahim Foundation’s Prize for Achievement in African Leadership, for instance, offers US$5 million and an additional US$200,000 per year for life to former African heads of state who advanced democracy and good governance.⁷ While all past African leaders are eligible, every winner since its inception—Joaquim Chissano of Mozambique, Festus Mogae of Botswana, and Pedro Pires of Cape Verde—has been Western educated. Such incentives, monetary or otherwise, can coax leaders to pursue democratic reform even when it is not in their short-term interest. Ultimately, through both push and pull effects, leaders educated in the West should be more willing and able to promote democracy than other leaders.

Given the logic of the socialization and transnational linkage mechanisms laid out earlier, the main hypothesis we test is: Western-educated leaders are more likely to pursue democratic reform than other leaders.

Data and Research Design

Sample of Cases

We rely on the Archigos data set (Goemans, Gleditsch, and Chiozza 2009) to identify the leader of each country. Because we are interested in the effect of specific leaders on democratization, the unit of analysis is an individual leader’s tenure in office, which we refer to as a “leader spell.”⁸ For example, Idi Amin ruled Uganda from 1971 to 1979, so this leader spell constitutes one observation.⁹ If a leader holds power more than once, each leader spell is coded separately. We use the following rules to determine which leader spells enter our sample of cases. First, like most studies of democratization, we focus on the post–World War II era due to data constraints on many key control variables.¹⁰ Second, we exclude leader spells of less than three years from the analysis. This minimum length of time constitutes a reasonable, if somewhat arbitrary, threshold for leaders to enact meaningful regime change.¹¹ Third, we limit our sample to leaders of non–Organization for Economic Cooperation and Development (OECD) countries because nearly all advanced industrialized nations are consolidated democracies and thus possess little variation on the dependent variable.

Dependent Variable

Political scientists offer numerous definitions of democracy. Rather than rely on a single indicator, we measure our dependent variable, democratization, in two ways. First, we use the Polity IV index (Marshall and Jaggers 2012), which rates countries from −10 (least democratic) to +10 (most democratic) to calculate the change in the level of democracy over a leader’s tenure. We determine the change in democracy associated with each leader spell by subtracting the country’s polity score in the year prior to the leader’s ascension to office from its score during the last full year the leader was in power.¹² This difference yields the total change in a country’s polity score during each leader’s tenure.

Second, to allay concerns about the Polity data,¹³ we employ a binary measure of whether a country democratizes during each leader spell. We opt for the Boix, Miller, and Rosato (2013) data set of political regimes because it offers the most comprehensive coverage of any dichotomous measure of democracy, both cross-nationally and temporally. Following Dahl’s (1971) classic categorization of democracy, this data set defines a state as democratic if it meets a minimal threshold of “contestation” (i.e., political leaders are selected through relatively free and fair elections) and “participation” (i.e., a majority of adult males has the right to vote).

Key Independent Variable

We collected an original data set on the educational experiences of world leaders by coding the specific universities, colleges, and other postsecondary educational institutions they attended.¹⁴ This allowed us to extract detailed information on the educational backgrounds of leaders that as far as we know has not been analyzed previously.¹⁵ To code the educational experiences of each leader, we consulted a variety of sources, including personal biographies and encyclopedias of world leaders (e.g., Fredriksen 2004; Lentz 1994). For each leader, we first recorded the specific institution where the leader pursued undergraduate and/or graduate studies, if any.¹⁶ We then used this information to create two Western education variables because “Western” is a contested concept. Our primary measure consists of leaders educated in the United States or the United Kingdom. The alternative definition, used as a robustness check, includes leaders educated in North America or Western Europe.¹⁷ If a leader attended university in both his or her home country and the West—for example, former Costa Rican President Oscar Arias earned an undergraduate degree from the University of Costa Rica and a doctorate in government from the University of Essex—we coded the leader as Western-educated because the socialization and transnational linkage mechanisms should apply regardless of other educational experiences. Table 1 summarizes the distribution of world leaders’ educational backgrounds.

Table 1.

The Educational Backgrounds of Leaders in Our Sample.

Category	Observations
Western education	109
Non-Western education	235
No higher education	222
Total leader spells	566

Covariates

Our statistical analysis utilizes several national-level control variables that are standard explanatory factors in studies of democratization.¹⁸ For these variables, we employ data from the year prior to the leader entering office to provide a snapshot of the conditions that each leader inherits. This technique is preferable to averaging data over each leader spell because it alleviates the concern that changes in the independent variables are endogenous to democratization (or the lack thereof).

Development

Countless studies show that high levels of income and democracy are correlated. Some scholars claim this is because economic development induces democratization (Boix and Stokes 2003), while others argue development makes established democracies more stable (Przeworski et al. 2000). As a measure of economic development, we use Gleditsch’s (2002) expanded gross domestic product per capita data.

Inequality

Two influential theories parse the connection between inequality and democratization. The first proposes that higher levels of inequality impede democratization because elites fear redistribution (Boix 2003). The second suggests that elites will be most likely to extend the franchise at middle levels of inequality (Acemoglu and Robinson 2006). We proxy inequality with capital shares data from Houle (2009).¹⁹

Oil

States that depend heavily on oil exports are often autocratic because opaque oil revenues foster corruption and undercut incentives to build democratic institutions (e.g., Ross 2012). To control for oil-dependent exporters, we use Fearon and Laitin’s (2003) data on states that received at least one-third of their export revenues from fuel products.

Ethnic fractionalization

High ethnic fractionalization may hurt democracy by encouraging politicians to engage in “ethnic outbidding” for support within their respective groups (Rabushka and Shepsle 1972). Our index of fractionalization comes from Peic and Reiter (2011).

Protestantism

Many analyses find a positive connection between large Protestant populations and democracy. This may be because Protestants are especially likely to generate institutional structures and social norms conducive to democratic politics (e.g., Woodbery and Shah 2004). Thus, we include data from Przeworski et al. (2000) on the percentage of a country’s population that is Protestant.

British colony

Research shows that British colonial heritage promotes democracy by bequeathing institutions associated with greater political freedoms, more competent political parties, and stronger civil society (Bollen and Jackman 1985). As a result, we add a dummy variable from Fearon and Laitin (2003) indicating whether each country is a former British colony.

US alliance

The structure of the international system can influence democratization by affecting the strategies great powers pursue vis-à-vis weaker states (e.g., Boix 2011). More specifically, the ideological divisions of the Cold War may have driven regime transitions throughout many parts of the world. To control for this possibility, we add a dummy variable for countries allied with the United States during the Cold War (Gibler 2009).

Aggregate student flows

Countries may be more likely to democratize as larger numbers of their young people study abroad in democracies (Spilimbergo 2009; Atkinson 2010). Consequently, we include a variable that measures the total number of students from each country studying in the United States using data from the Institute of International Education (2005).²⁰

Executive constraints

Our theory assumes that leaders have the capacity to shape their nation’s regime. The constraints a leader faces, however, are higher in some countries than in others. To control for these differences, we include Henisz’s (2002, 363) data on political constraints that estimates “the extent to which a change in the preferences of any one actor may lead to a change in government policy.”²¹

Preexisting democracy

A country’s preexisting level of democracy conditions the ability of a leader to democratize. For instance, highly autocratic countries can have a large positive change on the polity scale but not an equivalent negative change because they are already located near the lower boundary. The converse is true of highly democratic countries. Therefore, we include a variable that measures a country’s polity score in the year before a leader assumes power.²²

Missing Data

Data coverage for non-OECD countries is notoriously poor, and our analysis is not immune to this complication. We minimize listwise deletion by employing linear interpolation on control variables with missing values. While interpolation is a contentious issue, we believe this technique is appropriate because nearly all of the country-level control variables are slow changing or static. Additionally, using the leader spell as the unit of analysis raises a missing data issue. If no data exist for the country-year prior to a leader entering office—which by definition occurs for the first leader of newly formed or independent states—we use the first data point available. For example, Archigos codes Nursultan Nazarbayev as the first leader of newly born Kazakhstan starting in 1990. Because Kazakhstan did not exist in 1989, data for the country-level control variables for the Nazarbayev leader spell are from 1990.

Empirical Results

Main Findings

Using the data described above, we estimate a series of models to test our hypothesis on the relationship between Western-educated leaders and democratization. We begin by examining the change in polity scores during each leader spell, which is our dependent variable for all models in Table 2. Model 1 presents an ordinary least squares (OLS) model that includes our primary Western-educated leader variable and standard national-level controls for democratization. As predicted, the coefficient on Western-educated leader is positive and statistically significant, meaning that Western-educated leaders democratize more than other leaders. Model 2 reaches similar results when employing our alternative definition of Western education. To compare directly the democratization records of Western-educated leaders and leaders who attended universities outside the West, model 3 adds a dummy variable for leaders without university education, making leaders with non-Western education the reference category. Once again, we find a positive and statistically significant coefficient on Western-educated leader. Finally, because democracy has generally spread in “waves” in specific parts of the world and during specific time periods (e.g., Huntington 1991), model 4 reestimates model 1 with regional and decade fixed effects to ensure that unobservable heterogeneity across space or time does not skew our results. Our results remain substantively unchanged.²³

Table 2.

Western-educated Leaders and Democratization: OLS Models.

	Baseline (1)	Alternative Definition (2)	With Uneducated (3)	Fixed Effects (4)
Western-educated leader	2.384*** (0.523)	1.691*** (0.458)	1.620*** (0.529)	1.731*** (0.481)
Uneducated leader			−1.907*** (0.535)
Development	2.526 (3.546)	3.267 (3.773)	2.817 (3.349)
Oil	−0.216 (0.545)	−0.303 (0.539)	−0.172 (0.533)
Ethnic fractionalization	−0.186 (0.811)	−0.140 (0.818)	−0.270 (0.796)
Protestantism	0.022** (0.011)	0.019* (0.011)	0.020* (0.011)
British colony	−1.461*** (0.501)	−1.128** (0.494)	−1.219** (0.502)
US ally	0.192 (0.435)	0.292 (0.443)	0.303 (0.427)
US student flows	0.104** (0.047)	0.111** (0.049)	0.085* (0.048)
Executive constraints	4.278*** (1.576)	4.240*** (1.597)	4.341*** (1.557)
Preexisting democracy	−0.392*** (0.049)	−0.386*** (0.050)	−0.439*** (0.053)	−0.375*** (0.037)
Constant	−0.899 (0.559)	−1.051* (0.569)	−0.062 (0.601)	−2.160** (0.961)
n	509	509	509	533
R ²	0.201	0.191	0.228	0.272

Note: Standard errors clustered by leader within parentheses. All models use ordinary least squares (OLS) regression.

*p < .10. **p < .05. ***p < .01.

In Table 3, we explore the relationship between Western-educated leaders and democratization employing logistic regression. We use Boix, Miller, and Rosato’s dichotomous democracy variable to investigate whether a transition from autocracy to democracy occurred during each leader spell. We exclude leaders of already established democracies from these models because a leader who enters office in a democracy cannot initiate a democratic transition. As before, we estimate four different models. Model 5 compares the democratization records of Western-educated leaders to all other leaders. Model 6 uses our alternative definition of Western education. Model 7 adds a dummy variable for uneducated leaders to enable a direct comparison between Western and non-Western-educated leaders. Model 8 adds regional and decade fixed effects. In every model, the coefficient on Western-educated leader is statistically significant and positive. Therefore, Western-educated leaders are linked to democratic reform using multiple measures of democracy, verifying that our results are not limited to a particular coding scheme.

Table 3.

Western-educated Leaders and Democratization: Logit Models.

	Baseline (5)	Alternative Definition (6)	With Uneducated (7)	Fixed Effects (8)
Western-educated leader	1.572*** (0.462)	1.017*** (0.370)	1.053** (0.486)	1.498*** (0.439)
Uneducated leader			−1.343*** (0.495)
Development	5.812** (2.937)	5.968** (2.757)	5.903* (3.282)
Oil	−1.250* (0.657)	−1.281* (0.664)	−1.133* (0.657)
Ethnic fractionalization	1.047 (0.706)	1.076 (0.698)	1.074 (0.747)
Protestantism	−0.010 (0.011)	−0.013 (0.010)	−0.013 (0.011)
British colony	−0.624 (0.558)	−0.231 (0.533)	−0.533 (0.559)
US ally	1.131*** (0.410)	1.357*** (0.419)	1.269*** (0.425)
US student flows	0.036 (0.030)	0.044 (0.028)	0.028 (0.031)
Executive constraints	−1.239 (1.186)	−1.008 (1.159)	−1.442 (1.188)
Preexisting democracy	0.213*** (0.043)	0.202*** (0.044)	0.193*** (0.045)	0.150*** (0.036)
Constant	−1.967*** (0.433)	−2.230*** (0.448)	−1.546*** (0.457)	−2.129** (0.913)
n	280	280	280	292
AIC	225.523	228.812	218.527	232.437
BIC	265.505	268.795	262.144	280.235

Note: Standard errors clustered by leader within parentheses. All models use logistic regression. AIC = Akaike information criterion; BIC = Bayesian information criterion.

*p < .10. **p < .05. ***p < .01.

Substantive Effects

Before conducting further robustness checks, we assess the substantive effect of Western-educated leaders on democratization.²⁴ Using estimates from model 1 in Table 2, we first generate the predicted changes in polity scores during the tenures of Western-educated leaders relative to all other leaders (Figure 1a). Next, using estimates from model 3, we compare directly the democratization records of Western-educated leaders to leaders who attended universities outside the West and to those without higher education (Figure 1b). Both graphs indicate that the substantive effect of Western-educated leaders is considerable. In fact, the predicted disparity between Western-educated leaders and uneducated leaders is 3.6 points on the polity index. To give some meaning to this number, the change South Africa experienced during the transition from apartheid rule to the Mandela presidency was 3 points, and the difference between the polity scores of the United States and Ukraine in 2011 was 4 points.

Figure 1.

Substantive effects of leaders’ education on change in polity scores.

We also estimate the likelihood of a full transition from autocracy to democracy during the tenures of leaders with different educational backgrounds. Using estimates from model 5 in Table 3, we find that the predicted probability of a democratic transition with a Western-educated leader in power is 30 percent, while it is only 8 percent for all other leaders (Figure 2a). In other words, a democratic transition is nearly four times more likely with a Western-educated leader than with another leader who inherits the same national-level context. Similarly, using estimates from model 7, we compare the likelihood of democratization with Western-educated leaders in office relative to non-Western-educated leaders and uneducated leaders, respectively (Figure 2b). Here, we see that the predicted probability of a democratic transition is 30 percent with Western-educated leaders, whereas it is just 13 percent with non-Western-educated leaders and only 4 percent with uneducated leaders.

Figure 2.

Substantive effects of leaders’ education on probability of democratic transition.

Addressing Endogeneity Concerns

Our main results display a strong correlation between Western-educated leaders and democratization. In this section, we address several endogeneity concerns.

Endogenous leadership transitions

Arguably the most pressing endogeneity challenge is that preexisting political circumstances influence leadership transitions. Two specific issues arise here. The first involves the timing of leadership transitions. If Western-educated leaders are likely to gain office when unobservable factors (e.g., elite splits) favor democratic change, a country’s underlying conditions may be more responsible for the subsequent democratization than the leader. To address the timing of leadership transitions, we employ a technique recently developed in the economics literature that exploits plausibly exogenous leadership transitions due to leaders dying in office from natural causes (Jones and Olken 2005; Besley, Montalvo, and Reynal-Querol 2011). We execute this strategy by limiting our sample to leaders who assumed power immediately following the natural death of the previous leader.²⁵ In such cases, the timing of the transfer of power from one leader to the next is clearly unrelated to underlying political conditions, eliminating endogeneity concerns about the timing of leadership transitions.

The second issue involves leader selection. Leaders typically come to power through a competitive process that reflects the interests of powerful actors. Thus, even if the timing of the transition is random, the process for selecting the next leader is likely not. This is a particularly challenging issue, and we have no perfect solution. We minimize such concerns, however, by further subsetting the deaths-in-office sample to include only leaders who enter office through what Archigos defines as a “regular” process (e.g., a vice president replacing a president who dies of natural causes). The advantage of this approach is to exclude leaders who come into power via an “irregular” or “foreign-imposed” process, which are likely endogenous to political circumstances. Therefore, focusing only on leaders who enter office through regular means provides a reasonable, if imperfect, way to reduce concerns that leaders are being selected because of their educational background.²⁶

To implement our empirical strategy, we modified some of our prior coding rules to ensure a large enough sample for statistical analysis. First, since natural deaths in office are rare, we expanded the temporal coverage back to 1875 to take advantage of all such leadership transitions in Archigos. Data on the control variables are not available for this extended time period, but omitting them here is a worthwhile trade-off for alleviating endogeneity concerns. Second, we reduced the minimum threshold for time in office from three years to one full calendar year. This is the smallest reasonable threshold we could use, given that the democracy data sets are in a country-year format. Last, we did not restrict our sample to non-OECD world leaders because this distinction holds little meaning for much of the period under study here.

We present our results from the deaths-in-office leadership transitions sample in Table 4. Model 9 shows an OLS regression comparing the democratization records of Western-educated leaders to all other leaders.²⁷ Model 10 adds a dummy variable for uneducated leaders to enable a direct comparison between Western and non-Western-educated leaders. In models 11 and 12, we replicate these models using logistic regression with the Boix, Miller, and Rosato data. In all four models, the coefficient on Western-educated leader remains statistically significant and positive.²⁸

Table 4.

Western-educated Leaders and Democratization: Deaths-in-office Transitions.

	Baseline (9)	With Uneducated (10)	Baseline (11)	With Uneducated (12)
Western-educated leader	1.413* (0.760)	1.339* (0.737)	2.622*** (0.817)	2.664** (1.200)
Uneducated leader		−0.138 (0.577)		0.063 (1.271)
Preexisting democracy	−0.145*** (0.039)	−0.146*** (0.037)	0.018 (0.070)	0.018 (0.070)
Constant	0.222 (0.308)	0.299 (0.367)	−3.127*** (0.560)	−3.169*** (1.032)
n	127	127	89	89
R ²	.099	.100
AIC			45.798	47.796
BIC			53.264	57.750

Note: Standard errors clustered by country within parentheses. Models 9 and 10 use ordinary least squares (OLS) regression. Models 11 and 12 use logistic regression. AIC = Akaike information criterion; BIC = Bayesian information criterion.

*p < .10. **p < .05. ***p < .01.

Educational self-selection

Another endogeneity concern is that future leaders may self-select into Western education because they already harbor democratic leanings. Ideally, we would rule out this possibility quantitatively, but doing so is not an option because it would require knowing the private political preferences of all leaders before they attended university. Available evidence, however, suggests that students who study in the West—including future leaders—do so more to acquire skills and earn a prestigious degree than to seek out democratic culture. For example, in one of the most comprehensive efforts of its kind, the Institute of International Education surveyed foreign students studying at twenty-four universities in the United States to elicit their reasons for studying there. A full 83 percent of respondents said that “the reputation of academic qualifications and degrees from the US” was one of the most important factors in their decision (Institute of International Education and the Academic Cooperation Association [IIE/ACA] 2005a, 1). A similar survey of more than 20,000 foreign students attending universities in Europe found that “the most important reasons to go abroad are career-oriented” and that “[m]ost students did not seem to care about the world region where they are going to study” (IIE/ACA 2005b, 12-13).

The educational choices of foreign elites reveal a similar story. This is perhaps best evidenced by the fact that some of the world’s most notorious autocrats—such as Zimbabwe’s Robert Mugabe and Equatorial Guinea’s Teodoro Obiang Nguema Mbasogo—have sent their children and advisors to the West for training. A particularly striking example is the Chinese Communist Party, which dispatches mid- and upper-level officials to study management and public policy in the West. Even some of the most devoted Communist Party members, including five of the nine members of the Politburo’s Standing Committee, have sent their children and grandchildren to study in the United States or Western Europe. A recent inquiry into this practice concludes that earning a degree from a “brand name” university typically overrides the ideological opposition that the Chinese leadership harbors toward the West (Higgins and Fan 2012). In other words, Chinese elites select Western education despite its association with democracy rather than because of it. Hence, it seems many foreign leaders study (or send their children and advisors to study) in the West for reasons unrelated to preexisting democratic preferences.²⁹

Spurious correlation

A final concern is that the relationship between Western-educated leaders and democratization is spurious. This could occur if a confounding variable increases both the prospects for democratization and the likelihood of selecting a Western-educated leader. Here, two potential factors stand out: alliances with the West and aggregate student flows with the West. While we found only mixed evidence that these variables are linked to democratization, it is worth investigating whether they are correlated with the selection of Western-educated leaders. To check, we regress whether a country selected a Western-educated leader on whether that country was allied with the United States during the Cold War. We also regress whether a country selected a Western-educated leader on the country’s total number of students studying in the United States. In both cases, we find a statistically insignificant relationship (see Online Appendix), minimizing concerns of spurious correlation.

Qualitative Evidence

To complement our quantitative analysis, we briefly consider the life history of Lee Teng-hui, former President of Taiwan, to illustrate how the causal mechanisms in our theory unfold in practice. Highlighting Lee in particular is valuable for two main reasons. Since Lee assumed the presidency following the natural death of his predecessor, the exogenous timing of Lee’s rise to power helps us to disentangle the role he played in Taiwan’s democratization from structural factors present in the country. Furthermore, because Lee displayed antipathy for democracy before studying in the United States, we are confident he did not self-select into Western education because of a preexisting penchant for democracy.

Born in 1923, Lee earned an undergraduate diploma from National Taiwan University in 1948 and worked as an academic in Taiwan. From 1965 to 1968, Lee moved to the United States, where he pursued a doctorate in economics at Cornell University. Upon his return to Taiwan, Lee became a member of the Kuomintang party. He subsequently earned seats as an agricultural minister, mayor of Taipei, governor of Taiwan, and eventually vice president. When authoritarian leader Chiang Ching-kuo, son of longtime ruler Chiang Kai-shek, died of heart failure in 1988, Lee assumed the presidency. After entering office, Lee pushed forward one of the most sudden and dramatic democratic transitions in modern history: he allowed contested elections, curbed press censorship, and enacted new constitutional checks on the executive. Perhaps most significant was Lee’s decision to step down in 2000 in accordance with the term limits he established, representing the first time a Taiwanese leader voluntary ceded power. For his efforts, Lee is often considered “the father of Taiwan’s democracy.”

Lee’s time studying in the United States played a tremendous role in socializing the future leader to embrace democracy. But this was hardly a foregone conclusion. As a young man, Lee earned a reputation as a political radical and was a member of the Communist Party who “read everything [he] could get his hands on by Karl Marx and Friedrich Engels” (Lee 1999, 31). All of this would change at Cornell, however. Lee moved to the United States at a time of great change and social upheaval. The Civil Rights movement and anti-Vietnam War protests made American universities hotbeds of democratic activism, and Cornell was no exception. For the first time, Lee witnessed democracy up close, and it changed his political outlook forever. In Lee’s own words:

My years at Cornell from 1965 to 1968 made an indelible impression on me. It was … the time I first recognized that full democracy could engender ultimately peaceful change, and that lack of democracy must be confronted with democratic methods, and lack of freedom must be confronted by the idea of freedom before it would be possible to hasten the day of genuine democracy and freedom. I returned to my homeland determined to make my contribution toward achieving full democracy for our society (Lee 1995).

Lee’s political transformation at Cornell, however, was not the only factor that caused Taiwan to democratize. Throughout his presidency, Lee fastidiously cultivated Western connections to pull resources toward his democratic reforms. For example, when describing his most valuable lessons from studying abroad, Lee remarked, “I learned about local democracy in America, how to play golf, and how foreign aid was valuable to Taiwan” (R. C. Kagan 2007, 63). Consistent with this message, Lee rarely missed an opportunity to boost his reputation in the West, regularly welcoming foreign dignitaries, Nobel Laureates, and other celebrities to Taipei. One scholar noted Lee’s proclivity “to develop an engaging relationship with the democratic countries of … the West” and talked of his “dramatic private diplomacy” (R. C. Kagan 2007, 121). This may have been particularly important in garnering US security guarantees against China, which threatened Taiwan’s sovereignty. Lee sometimes even pushed his Western connections to the point of discomfort. In 1995, for example, Lee sparked a diplomatic firestorm between Washington and Beijing after insisting on traveling to the United States to deliver a keynote address at Cornell—an act that Chinese officials denounced given their icy relations with Taipei.

Lee’s Western connections also gave Western countries, particularly the United States, leverage when dealing with Taiwan. Indeed, Lee’s close ties to the West wedded Taiwan’s political trajectory to politics inside the Beltway. This enabled Washington officials, searching for a like-minded ally in Southeast Asia, to push for Taiwan’s democratization. One expert, for example, suggests that Taiwan’s dependence on the United States influenced “when Taiwan decided to institute democracy reform and why Taiwan’s leaders pushed it as fast as [they] did” (Copper 2005, 31). This dependence was made evident when the US explicitly threatened to withhold its sale of weapons to Taipei when Lee’s behavior did not align with American interests. In short, Lee's Western connections put Taiwan in a position where continued authoritarianism or even partial democracy were hardly viable options. By the time he left office, Lee’s claim that “[d]emocracy is thriving in my country” (Lee 1995) was undoubtedly true. As Lee’s tenure reveals, both the socialization and transnational linkage mechanisms can propel a Western-educated leader to democratize.

Conclusions

Previous studies concentrate primarily on structural variables to explain why some countries democratize and others do not. Our article, by contrast, contends that leaders also play a key role. Utilizing numerous statistical techniques, we show that Western-educated leaders are robustly linked to democratic reform. These results, however, should not be read as endorsing the view that individual agency necessarily predominates structural factors in democratization. Although leader characteristics have a substantial impact on democratization, macro-level variables—such as history, domestic politics, and international context—are also critical. The complexity inherent in the agency versus structure debate was perhaps best captured by Karl Marx ([1852] 1926, 23), who observed that “[m]en make their own history, but not just as they please.… They have to work upon circumstances as they find them.” In line with this dictum, we think that our findings should be interpreted as follows: given preexisting constraints, the educational backgrounds of leaders influence democratization. Further inquiry is needed to understand fully the interplay between leaders and structure in driving democratization.

In terms of policy implications, this article sheds light on the efficacy of global democracy promotion efforts. With debates raging under both the Bush and Obama administrations about the merits of nation building, the United States has spent billions of dollars on military interventions and other initiatives intended to export democracy abroad. The trend is not new. Woodrow Wilson once declared that the United States was entering World War I to make the world “safe for democracy,” and subsequent presidents have used military force to promote democracy in Vietnam, Cuba, Grenada, Iraq, Afghanistan, and several other countries. Given the mixed record of spreading democracy through force—as well as the considerable human and financial price of military intervention—it is worth considering other means of democracy promotion. Our evidence implies that fostering democracy through educational exchange may be an especially effective endeavor. While a long-term enterprise, there is reason to believe that the payoff is well worth the investment today.

Looking ahead, we believe this article helps open a ripe field for future research. On the topic of individual leaders and democratization, scholars could advance the literature by delving further into the mechanisms through which Western-educated leaders advance democracy and by explaining prominent outliers. In addition, there are many other background experiences and personal characteristics of leaders (e.g., age, gender, religious beliefs, and military experience) apart from education that may be relevant. When it comes to how a leader’s education shapes governance in general, democratization is but one dependent variable of interest. Indeed, the education of leaders might also influence a panoply of different outcomes, including human rights, the rule of law, economic development, foreign direct investment, and membership in international organizations. Exploring these factors will paint a more complete picture of when leaders matter and how their personal characteristics inform their approaches to governing.

Footnotes

Acknowledgments

The authors thank John Ahlquist, Melina Altamirano, Benjamin Barber, Andrew Bell, John Chin, Scott de Marchi, Christopher DeSante, Cassy Dorff, Alex Downes, Max Gallop, Karen Gift, Nathaniel Harris, Herbert Kitschelt, Laura Krcmaric, Karen Remmer, Erik Wibbels, the anonymous reviewers, and the journal editors for helpful comments.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

The supplemental materials are available at .

Notes

References

Acemoglu

Daron

Robinson

James A.

. 2006. Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press.

Ahlquist

John S.

Levi

Margaret

. 2011. “Leadership: What It Means, What It Does, and What We Want to Know About It.” Annual Review of Political Science 14 (1): 1–24.

Ahlquist

John S.

Wibbels

Erik

. 2012. “Riding the Wave: World Trade and Factor-based Models of Democratization.” American Journal of Political Science 56 (2): 447–64.

Ahmad

Manal

. 2012. “Pakistan: In the Line of Fire, Benazir Bhutto: Family Heir.” PBS. Accessed September 15, 2012. http://www.pbs.org/frontlineworld/stories/pakistan703/history/fire.html.

Allport

Gordon W.

1954. The Nature of Prejudice. Cambridge, MA: Addison-Wesley.

Ansell

Ben

Samuels

David

. 2010. “Inequality and Democratization: A Contractarian Approach.” Comparative Political Studies 43 (12): 1543–74.

Atkinson

Carol

. 2006. “Constructivist Implications of Material Power: Military Engagement and the Socialization of States, 1972–2000.” International Studies Quarterly 50 (3): 509–37.

Atkinson

Carol

. 2010. “Does Soft Power Matter? A Comparative Analysis of Student Exchange Programs 1980–2006.” Foreign Policy Analysis 6 (1): 1–22.

Beck

Neal

. 2001. “Time-series—Cross-section Data: What Have We Learned in the Past Few Years?” Annual Review of Political Science 4:271–93.

10.

Berlinski

Samuel

Dewan

Torun

Dowding

Keith

. 2012. Accounting for Ministers: Scandal and Survival in British Government 1945–2007. New York: Cambridge University Press.

11.

Besley

Timothy

Montalvo

Jose G.

Reynal-Querol

Marta

. 2011. “Do Educated Leaders Matter?” The Economic Journal 121 (554): 205–27.

12.

Besley

Timothy

Reynal-Querol

Marta

. 2011. “Do Democracies Select More Educated Leaders?” American Political Science Review 105 (3): 552–66.

13.

Boix

Carles

. 2003. Democracy and Redistribution. Cambridge, UK: Cambridge University Press.

14.

Boix

Carles

. 2011. “Democracy, Development, and the International System.” American Political Science Review 105 (4): 809–28.

15.

Boix

Carles

Miller

Michael

Rosato

Sebastian

. 2013. “A Complete Dataset of Political Regimes, 1800–2007.” Comparative Political Studies 46 (12): 1523–54.

16.

Boix

Carles

Stokes

Susan C.

. 2003. “Endogenous Democratization.” World Politics 55 (4): 517–49.

17.

Bollen

Kenneth A.

Jackman

Robert W.

. 1985. “Political Democracy and the Size Distribution of Income.” American Sociological Review 50 (4): 438–57.

18.

Boonstra

Jos

. 2010. Assessing Democracy Assistance: Georgia. Brussels, Belgium: FRIDE.

19.

Brownlee

Jason

. 2007. “Hereditary Succession in Modern Autocracies.” World Politics 59 (4): 595–628.

20.

Bush

Kenneth D.

Saltarelli

Diana

. 2000. The Two Faces of Education in Ethnic Conflict: Towards a Peacebuilding Education for Children. Florence, Italy: United Nations Children’s Fund.

21.

Byman

Daniel L.

Pollack

Kenneth M.

. 2001. “Lets Us Now Praise Great Men: Bringing the Statesman Back in.” International Security 25 (4): 107–46.

22.

Carlyle

Thomas

. (1841) 1869. Heroes and Hero-worship. London, UK: Chapman and Hall.

23.

Caspi

Avshalom

Roberts

Brent W.

Shiner

Rebecca L.

. 2005. “Personality Development: Stability and Change.” Annual Review of Psychology 56:453–84.

24.

Champion

Marc

Cullison

Alan

. 2007. “Georgian Vote Won’t End Crisis.” Wall Street Journal, November 10, 2007, A6.

25.

Chiozza

Giacomo

Goemans

H. E.

. 2011. Leaders and International Conflict. New York: Cambridge University Press.

26.

Colgan

Jeff D.

2013. “Domestic Revolutionary Leaders and International Conflict.” World Politics 65 (4): 656–90.

27.

Copper

John Franklin

. 2005. Consolidating Taiwan’s Democracy. Lanham, MD: University Press of America.

28.

Croco

Sarah E.

2011. “The Decider’s Dilemma: Leader Culpability, War Outcomes, and Domestic Punishment.” American Political Science Review 105 (3): 457–77.

29.

Dahl

Robert A.

1971. Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press.

30.

de Waal

Thomas

. 2012. “Mrs. Clinton Goes to Georgia.” The National Interest, June 4, 2012. Accessed June 3, 2012. http://nationalinterest.org/commentary/mrs-clinton-goes-georgia-7000.

31.

Downes

Alexander B.

Monten

Jonathan

. 2013. “Forced to be Free? Why Foreign-imposed Regime Change Rarely Leads to Democracy?” International Security 37 (4): 90–131.

32.

Dunning

Thad

. 2008. Crude Democracy: Natural Resource Wealth and Political Regimes. Cambridge, UK: Cambridge University Press.

33.

Fearon

James D.

Laitin

David D.

. 2003. “Ethnicity, Insurgency, and Civil War.” American Political Science Review 97 (1): 75–90.

34.

Fredriksen

John C.

2004. Biographical Dictionary of Modern World Leaders: 1900–1991. New York: Facts on File.

35.

Fuhrmann

Matthew

Horowitz

Michael C.

. 2015. “Leaders, Rebel Experience, and Nuclear Proliferation.” Journal of Politics 77 (1): 72–87.

36.

Geddes

Barbara

. 1999. “What Do We Know About Democratization after Twenty Years?” Annual Review of Political Science 2 (1): 115–44.

37.

Gerring

John

Thacker

Strom C.

. 2008. A Centripetal Theory of Democratic Governance. Cambridge, UK: Cambridge University Press.

38.

Gibler

Douglas M.

2009. International Military Alliances, 1648–2008. Washington, DC: Congressional Quarterly Press.

39.

Gleditsch

Kristian Skrede

. 2002. “Expanded Trade and GDP Data.” Journal of Conflict Resolution 46 (5): 712–24.

40.

Gleditsch

Kristian Skrede

Ward

Michael D.

. 2006. “Diffusion and the International Context of Democratization.” International Organization 60 (4): 911–33.

41.

Goemans

Henk E.

Gleditsch

Kristian Skrede

Chiozza

Giacomo

. 2009. “Introducing Archigos: A Dataset of Political Leaders.” Journal of Peace Research 46 (2): 269–83.

42.

Gourevitch

Peter

. 1978. “The Second Image Reversed: The International Sources of Domestic Politics.” International Organization 32 (4): 881–912.

43.

Hafner-Burton

Emilie M.

Kahler

Miles

Montgomery

Alexander H.

. 2009. “Network Analysis for International Relations.” International Organization 63 (3): 559–92.

44.

Henisz

Witold J.

2002. “The Institutional Environment for Infrastructure Investment.” Industrial and Corporate Change 11 (2): 355–89.

45.

Higgins

Andrew

Fan

Maureen

. 2012. “Chinese Communist Leaders Denounce U.S. Values but Send Children to U.S. Colleges.” Washington Post, May 18, 2012, http://www.washingtonpost.com/world/asia_pacific/chinese-communist-leaders-denounce-us-values-but-send-children-to-us-colleges/2012/05/18/gIQAiEidZU_story.html.

46.

Horowitz

Michael C.

Stam

Allan C.

. 2014. “How Prior Military Experience Influences the Future Militarized Behavior of Leaders.” International Organization 68 (3): 527–59.

47.

Houle

Christian

. 2009. “Inequality and Democracy: Why Inequality Harms Consolidation but Does Not Affect Democratization?” World Politics 61 (4): 589–622.

48.

Huntington

Samuel P.

1991. The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press.

49.

Institute of International Education. 2005. Open Doors: Report on International Educational Exchange, 1948-2004. New York: Institute of International Education.

50.

Institute of International Education and the Academic Cooperation Association. 2005a. Perceptions of European Higher Education in Third Countries: Country Report: USA. Washington, DC: IIE/ACA.

51.

Institute of International Education and the Academic Cooperation Association. 2005b. Perceptions of European Higher Education in Third Countries: Final Report. Washington, DC: IIE/ACA.

52.

Jervis

Robert

. 2013. “Do Leaders Matter and How Would We Know?” Security Studies 22 (2): 153–79.

53.

Johnston

Alastair Iain

. 2001. “Treating International Institutions as Social Environments.” International Studies Quarterly 45 (4): 487–15.

54.

Jones

Benjamin F.

Olken

Benjamin A.

. 2005. “Do Leaders Matter? National Leadership and Growth since World War II.” Quarterly Journal of Economics 120 (3): 835–64.

55.

Kagan

Donald

. 1991. “Western Values Are Central.” New York Times, May 4, 1991, Section 1, 23.

56.

Kagan

Richard C.

2007. Taiwan’s Statesman: Lee Teng-Hui and Democracy in Asia. Annapolis, MD: Naval Institute Press.

57.

Keck

Margaret E.

Sikkink

Kathryn

. 1998. Activists beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press.

58.

King

Gary

Tomz

Michael

Wittenberg

Jason

. 2000. “Making the Most of Statistical Analyses: Improving Interpretation and Presentation.” American Journal of Political Science 44 (2): 347–61.

59.

Kramon

Eric

Posner

Daniel N.

. 2012. “Ethnic Favoritism in Primary Education in Kenya.” Working Paper, University of California, Los Angeles.

60.

Lee

Teng-hui

. 1995. “Always in My Heart.” The Spencer T. and Ann W. Olin Lecture delivered at Cornell University Alumni Reunion. Accessed June 3, 2015. http://csis.org/files/media/csis/programs/taiwan/timeline/sums/timeline_docs/CSI_19950609.htm.

61.

Lee

Teng-hui.

1999. The Road to Democracy: Taiwan’s Pursuit of Identity. Tokyo, Japan: PHP.

62.

Lentz

Harris M.

1994. Heads of States and Governments: A Worldwide Encyclopedia of over 2,300 Leaders, 1945 through 1992. Jefferson, NC: McFarland.

63.

Lewis

Jeffrey

. 2005. “The Janus Face of Brussels: Socialization and Everyday Decision Making in the European Union.” International Organization 59 (4): 937–71.

64.

Levitsky

Steven

Way

Lucan A.

. 2010. Competitive Authoritarianism: Hybrid Regimes after the Cold War. New York: Cambridge University Press.

65.

Ying

. 2007. “China’s Stubborn Anti-democracy.” Policy Review 141:3–16.

66.

Marshall

Monty G.

Jaggers

Keith

. 2012. “Polity IV Project: Political Regime Characteristics and Transitions 1800–2010.” Accessed January 16, 2012. http://www.systemicpeace.org/polity/polity4.htm.

67.

Marx

Karl

. (1852) 1926. The Eighteenth Brumaire of Louis Bonaparte. Translated by Paul

Eden

Paul

Cedar

. New York: International.

68.

McDermott

Rose

. 2007. Presidential Leadership, Illness, and Decision Making. New York: Cambridge University Press.

69.

Nye

Joseph S.

2004. Soft Power. New York: Public Affairs.

70.

Peic

Goran

Reiter

Dan

. 2011. “Foreign-imposed Regime Change, State Power and Civil War Onset, 1920–2004.” British Journal of Political Science 41 (3): 453–75.

71.

Pettigrew

Thomas F.

1998. “Intergroup Contact Theory.” Annual Review of Psychology 49:65–85.

72.

Pevehouse

Jon C.

2005. Democracy from above: Regional Organizations and Democratization. New York: Cambridge University Press.

73.

Plato. (380 BC) 2012. Plato: The Republic and Other Works. Translated by Jowett

. New York: Knopf.

74.

Przeworski

Adam

. 2007. “Is the Science of Comparative Politics Possible?” In Oxford Handbook of Comparative Politics, edited by Boix

Carles

Stokes

Susan C.

, 147–71. Oxford, UK: Oxford University Press.

75.

Przeworski

Adam

Alvarez

Michael E.

Cheibub

Jose Antonio

Limongi

Fernando

. 2000. Democracy and Development: Political Institutions and Well-being in the World, 1950–1990. New York: Cambridge University Press.

76.

Rabushka

Alvin

Shepsle

Kenneth A.

. 1972. Politics in Plural Societies: A Theory of Democratic Instability. Columbus, OH: Merrill.

77.

Ross

Michael L.

2012. The Oil Curse: How Petroleum Wealth Shapes the Development of Nations. Princeton, NJ: Princeton University Press.

78.

Saunders

Elizabeth N.

2011. Leaders at War: How Presidents Shape Military Interventions. Ithaca, NY: Cornell University Press.

79.

Singh

Manmohan

. 2006. “Lecture Delivered by the Prime Minister to the University of Cambridge.” Accessed June 3, 2015. http://www.cam.ac.uk/news/honour-for-indian-pm.

80.

Sirleaf

Ellen Johnson

. 2010. This Child Will Be Great: Memoir of a Remarkable Life by Africa’s First Woman President. New York: HarperCollins.

81.

Spilimbergo

Antonio

. 2009. “Democracy and Foreign Education.” American Economic Review 99 (1): 528–43.

82.

Stokes

Susan C.

2014. “A Defense of Observational Research.” In Field Experiments and Their Critics, edited by Teele

Dawn Langan

, 33–57. New Haven, CT: Yale University Press.

83.

Treier

Shawn

Jackman

Simon

. 2008. “Democracy as a Latent Variable.” American Journal of Political Science 52 (1): 201–17.

84.

Wallis

William

. 2011. “Ouattara Faces Task of Uniting Ivory Coast.” Financial Times, April 12, 2011. http://www.ft.com/intl/cms/s/0/7854bc7a-652d-11e0-b150-00144feab49a.html?siteedition=intl#axzz3c1FH6qR3.

85.

Weeks

Jessica L.

2012. “Strongmen and Straw Men: Authoritarian Regimes and the Initiation of International Conflict.” American Political Science Review 106 (2): 326–47.

86.

Woodberry

Robert D.

Shah

Timothy S.

. 2004. “The Pioneering Protestants.” Journal of Democracy 15 (2): 47–61.

87.

World Bank. 2012. Fighting Corruption in Public Services: Chronicling Georgia’s Reforms. The World Bank: Washington, DC.

88.

Wright

Joseph

. 2008. “Do Authoritarian Institutions Constrain? How Legislatures Affect Economic Growth and Investment.” American Journal of Political Science 52 (2): 322–43.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.15 MB

0.00 MB